Crop Science 43:141-151 (2003)
© 2003 Crop Science Society of America
CROP PHYSIOLOGY & METABOLISM
Grain Position Affects Grain Macronutrient and Micronutrient Concentrations in Wheat
Daniel F. Calderini*,a and
Ivan Ortiz-Monasteriob
a Dep. Producción Vegetal, Facultad de Agronomía, Univ. de Buenos Aires, Argentina
b CIMMYT (International Maize and Wheat Improvement Center), Apdo. Postal 6-641, 06600 Mexico, D.F., Mexico
* Corresponding author (danielcalderini{at}uach.cl)
 |
ABSTRACT
|
|---|
Little is known about nutrient distribution within the spike of bread wheat (Triticum aestivum L.). This knowledge is important for determining breeding strategies aimed at increasing grain yield without affecting nutritional quality. The objective of this study was to gain a better understanding of how grain position affects nutrient concentration, dry matter distribution, and water dynamics of grains. An experiment using two sowing dates was performed under field conditions. Dry weight and concentrations of macronutrients (Ca, Mg, K, P, and S) and micronutrients (Cu, Fe, Mn, and Zn) in grains from the basal (BS), central (CS), and apical spikelets (AS) of two cultivars and one synthetic hexaploid line were determined. Grain water dynamics and nutrient and dry matter concentrations were also measured throughout the grain-filling period for the second sowing date. Genotypes showed different distributions of dry matter in different grain positions. Grain macronutrient and micronutrient concentrations in all genotypes decreased at grain positions more distal from the rachis. This reduction was as great as 30% (Ca) but varied by nutrient (e.g., Zn = -18%; S = -10%; K = +1%). Grain water content did not differ between grains. The observed differences in grain weight and nutrient concentration between grain positions could have important implications for wheat breeding. They suggest that it might be more effective to select for higher grain yield by increasing individual grain weight rather than grain number, a strategy that, in addition, would be less likely to affect the balance of nutrient concentrations within the spike.
Abbreviations: AS, apical spikelets BS, basal spikelets CS, central spikelets G14, grain positions 1 to 4 S12, sowing dates 1 and 2
 |
INTRODUCTION
|
|---|
DURING THE TWENTIETH CENTURY, plant breeding and improved crop management techniques successfully increased grain yield of bread wheat (Slafer et al., 1994; Calderini and Slafer, 1998; Calderini and Slafer, 1999) and, in so doing, effectively reduced food shortages and malnutrition. The future poses a similar challenge: by 2020, the demand for wheat is expected to be 40% greater than its current level of 552 million tons per year (Rosegrant et al., 1997). Now, however, a unique opportunity exists for agriculture to invest in developing more nutrient-dense staple food crops (Underwood, 2000) that could help reduce not only energy but also nutrient malnutrition. Welch and Graham (1999) proposed a new balanced nutrition paradigm for crop production. They point out that although the world's food supply in recent years has been sufficient, it does not promote an adequate nutritional balance (Anonymous, 1992; World Bank, 1994).
Breeding staple crops with high nutrient concentration in the grain has been proposed as a low cost, sustainable strategy for reducing mineral deficiencies in humans (more than 2 billion people suffer from deficiencies worldwide) and crops (Welch and Graham, 2000). To explore this possibility, the International Maize and Wheat Improvement Center (CIMMYT) and other international centers and research institutions have been participating in the Consultative Group on International Agricultural Research (CGIAR) micronutrient project since 1995. The project is studying the feasibility of developing micronutrient-dense cultivars of wheat, rice (Oryza sativa L.), maize (Zea mays L.), phaseolus beans (Phaseolus vulgaris L.), and cassava (Manihot esculenta Crantz). Its objectives are to determine: (i) available genetic variability that could be exploited in future breeding programs; (ii) the genetics and physiology of the selected traits; and (iii) the bioavailability of Fe and Zn in the best wheat selections (Bouis et al., 2000). This study, part of the broader CGIAR project, focused on the physiology of nutrient concentration in wheat grain. A better understanding of nutrient physiology could help design wheat breeding strategies aimed at increasing grain yield without reducing grain nutrient content. Nutrient-rich grain will, in turn, ensure a balanced nutrition in consumers, especially those who rely on wheat for a major portion of their calorie intake.
Wheat breeding has produced significant reductions in N (Austin et al., 1980; Slafer et al., 1990; Calderini et al., 1995; Ortiz-Monasterio et al., 1997) and P (Calderini et al., 1995) concentrations in the grain due to biomass dilution. Reductions in grain N concentrations as a result of plant breeding have also been reported for maize (Duvick, 1996) and sunflower (Helianthus annuus L.) (López Pereira et al., 2000). Therefore, it could be expected that nutrient concentration in individual grains has been affected similarly by dilution, that is, the heavier the grain the greater the dilution effect. Individual grains of modern wheat cultivars, however, have shown significant variation in N concentration depending on their position within the spike. Simmons and Moss (1978) and Herzog and Stamp (1983) found that lighter grains (mainly those in distal positions on the spikelet) had lower N concentrations than heavier grains (generally in proximal positions on basal-central spikelets), which suggests that grain position can have considerable impact on nutrient concentration. Therefore, future breeding strategies aimed at increasing grain yield while maintaining or increasing grain nutrient concentrations should take into account both the biomass dilution effect and nutrient distribution within the spike.
In the past, wheat yield gains have been associated with increases in the number of grain rather than in grain size (Loss and Siddique, 1994; Slafer et al., 1994; Calderini et al., 1999). Increases in grain number have been due mainly to improved grain set in more distal positions of the spikelet (Slafer et al., 1994). Miralles and Slafer (1995) showed there was a greater proportion of grains in distal positions in semidwarf wheat than in standard-height isogenic lines. The weight of these distal grains was clearly lower than that of grains from florets more proximal to the rachis. This trend suggests that continued selection for additional grains at more distal positions on the spikelets without parallel selection for higher nutrient concentration will result in continued reductions in nutritional quality and a greater imbalance in nutrient concentrations among grains. The effect of grain position on N concentration has been evaluated, but the manner and extent to which the position within the spikelet affects grain concentration of other macronutrients have not been reported.
The finding that grain N concentration is variable within the spike (Simmons and Moss, 1978; Herzog and Stamp, 1983) supports the hypothesis that grain concentration of other macronutrients (e.g., P) is also affected by grain position. This hypothesis, however, might not hold for micronutrients, which are less demanded by growing grains (Kochian, 1991). Ortiz-Monasterio and Graham (2000) found a very weak association (r2 < 0.33) between Fe and Zn grain concentrations and year of release of CIMMYT cultivars, and a strong (r2 = 0.72) linear relationship between grain yield and year of release in the same set of cultivars. To the best of our knowledge, there are no reports on how grain distribution within the wheat spike affects grain micronutrient concentrations.
It has been proposed that dry matter distribution influences N partitioning within the spike (Herzog and Stamp, 1983), since distal grains achieve less weight than those located closer to the rachis (Bremner and Rawson, 1978; Slafer and Savin, 1994; Miralles and Slafer, 1995; Kruk et al., 1997). At the same time, the dry matter and water dynamics of growing grains are closely related in wheat cultivars (e.g., Sofield et al., 1977; Schnyder and Baum, 1992; Calderini et al., 2000). Little is known, however, about the relationship between dry matter distribution and macro- and micronutrient concentration in grains in different positions and how they accumulate during grain filling. Synthetic hexaploid wheats (T. durum x Aegilops tauschii) can be used to study these relationships, since they have shown differences in dry matter distribution compared with modern cultivars (Calderini and Reynolds, 2000). Synthetic wheats may also be a source of variability for grain nutrient concentration, given that A. tauschii, a parent of the synthetics, has shown higher grain concentrations of micronutrients than T. aestivum (Ortiz-Monasterio and Graham, 2000).
The objectives of this work were (i) to study the concentration of macronutrients (Ca, Mg, K, P, and S) and micronutrients (Cu, Fe, Mn, and Zn) in grains set at different positions within the spike in two currently available wheat cultivars and a synthetic hexaploid line (T. durum x A. tauschii), and (ii) to evaluate macro- and micronutrient accumulation in grains together with dry matter distribution and water dynamics within the spike during grain filling.
 |
MATERIALS AND METHODS
|
|---|
This study was performed under field conditions at CIMMYT, El Batán, Mexico (19°31' N, 98°50' W; 2249 m above sea level) in 1998. The treatment design was a factorial combination of two sowing dates, 22 May (S1) and 17 June (S2), and three wheat genotypes, two high yielding CIMMYT cultivars (Bacanora T88 and Rayón F89) and one hexaploid synthetic line [68.111/RGB-4//WARD/3/FGO/4/RABI/5/Ae. sq. (878)], also developed at CIMMYT. The synthetic was chosen because its phenology and 1000-grain weight are similar to those of Bacanora T88 and Rayón F89.
Experimental plots consisted of two 5-m long raised beds spaced 80-cm apart, with two rows 20-cm apart on each bed. Seeding rate was one seed per 3 cm. Plots were arranged in a split-plot design with three replications. Main plots were assigned to sowing dates and subplots to genotypes. Mineral N was applied at 150 kg N ha-1 as urea and incorporated during land preparation (see results of presowing soil chemical analysis in Table 1). The plots were surface irrigated at sowing, and irrigation was continued as required until physiological maturity. Weeds were removed periodically by hand, while diseases (powdery mildew) and insects (aphids) were controlled by applying Tebuconazole {
-[2-(4 chlorophenyl) ethyl-
-(1-1-dimethylethyl)-1H-1,2,4 triazol-1-ethanol} and Triadimephon [1(4 chloro phenoxyl-3,3 dimethyl 1-(1H-1,2,4 thriazol-1-il)-2-butomene), respectively.
The dates when the crop reached booting, heading, anthesis, and physiological maturity were recorded using the scale proposed by Zadoks et al. (1974). At maturity, plants within a 2-m row (ca. 65 plants) in each plot were harvested. A subsample of 10 main-shoot spikes with similar spikelet number and size was collected to record grain weight and determine grain macronutrient (Ca, Mg, K, P, and S) and micronutrient (Cu, Fe, Mn, and Zn) concentrations. In addition, 75 similar spikes were tagged at heading (Zadoks 55) in the S2 treatment to study fresh and dry weight, and grain macronutrient and micronutrient concentrations during grain filling. For this purpose, 10 to 17 spikes, depending on grain size, were harvested weekly. The number of spikes harvested was adjusted to meet requirements for determining grain concentration of each nutrient.
Spikes were divided into three fractions containing five fertile spikelets each to study the effect of grain position upon grain weight and macronutrient and micronutrient concentrations. Two spikelets were excluded (the lowest and uppermost) because they generally carry only one grain instead of two or more as the other spikelets. Spikelets were pooled for the BS, CS, and AS fractions, and grains were sub-divided into first (G1) and second (G2) floret positions for all spikelets. In CS, grains of the third (G3) and fourth (G4) florets were also evaluated. Macronutrient and micronutrient concentrations were quantified by digesting samples with nitric/perchloric acid and analyzing them using ARL inductively coupled plasma spectrometry (Inductively Coupled Plasma Optical Emission Spectrometer ARL 3580 B, Switzerland) (Zarcinas et al., 1987).
For the purpose of comparing grain weight and grain nutrient concentration between grain positions, relative grain weight and relative grain nutrient concentration (percentage change from CS-G1) were calculated as follows:
 | [1] |
where RGW and RNC are relative grain weight and relative grain nutrient concentration, respectively; Gi is grain weight or grain nutrient concentration at a particular grain position; and CS-G1 is grain weight or grain nutrient concentration in G1 of CS.
Data were subjected to an analysis of variance utilizing a split-split-plot design, where sowing date was the main plot, genotype was the subplot, and grain position was the sub-subplot. Bacanora T88 and Rayón F89 set four grains in CS, while the synthetic line set just three. Therefore, two analyses of variance were performed. ANOVA 1 evaluated all grain positions shared by the three genotypes (i.e., all grain positions except G4 in CS). ANOVA 2 evaluated all grain positions including G4 for those genotypes that set it (i.e., all grain positions in Bacanora T88 and Rayón F89).
 |
RESULTS
|
|---|
General
Sowing date modified date of anthesis; S2 delayed anthesis by about one month with respect to S1 (Table 2). Within each sowing date, however, genotypes reached anthesis on similar dates. Bacanora T88, Rayón F89, and the synthetic line had comparable grain-filling periods and reached physiological maturity on similar dates. The only exception was Rayón F89 in S1, which showed slightly longer grain filling than did Bacanora T88 and the synthetic. Average temperature during grain filling did not differ between the two sowing dates.
View this table:
[in this window]
[in a new window]
|
Table 2. Date of anthesis, physiological maturity, average temperature during the period between anthesis and physiological maturity (A-PM), and the number of fertile spikelets per spike (Spt Sp-1) of Bacanora T88, Rayón F89, and the synthetic line at the first (S1) and the second (S2) sowing dates (S).
|
|
The number of spikelets on the main spike varied both with sowing date (P < 0.01) and genotype (P < 0.001). More spikelets per spike were produced in S2 (Table 2). Bacanora T88 showed the greatest number of spikelets per spike, while the synthetic line produced the fewest at both S1 and S2. Bacanora T88 developed almost two more spikelets per spike than the synthetic line, while Rayón F89 was intermediate for this trait.
Grain Weight
ANOVA 1 revealed that grain weight was affected by sowing date and grain position (Table 3). This analysis included all grain positions except G4, which was not present in the synthetic line, for the three genotypes.
View this table:
[in this window]
[in a new window]
|
Table 3. Mean squares for grain weight (GW) and grain concentration of macronutrients (Ca, Mg, K, P, and S) and micronutrients (Cu, Fe, Mn, and Zn) for comparisons between grain positions shared by Bacanora T88, Rayón F89, and the synthetic line (ANOVA 1) and for comparisons between grain position 4 (ANOVA 2) and the other grain positions in Bacanora T88 and Rayón F89 (the genotypes which carried G4 in central spikelets).
|
|
There was no genotype effect on grain weight (Table 3, ANOVA 1 and 2). The effect of sowing date on grain weight was due to an increase in grain weight of the synthetic line at S2 (7% on average for all grain positions), since there was no sowing date effect on Bacanora T88 and Rayon F89 (Tables 3 and 4). When only Bacanora T88 and Rayón F89 were analyzed, including G4 (Table 3, ANOVA 2), grain position affected grain weight, but no significant effect of sowing date or genotype was found. Both analyses (Table 3, ANOVA 1 and 2) showed interactions between genotype and grain position. Interaction between sowing date and grain position was also demonstrated in the ANOVA 2 analysis.
View this table:
[in this window]
[in a new window]
|
Table 4. Grain weight at grain positions 1 (G1), 2 (G2), 3 (G3), and 4 (G4) from the rachis in basal (BS), central (CS), and apical (AS) spikelets (Spt) for Sowing Date 1 (S1) and 2 (S2) for Bacanora T88, Rayón F89, and the synthetic line.
|
|
For all three genotypes, G2 in CS produced the heaviest grains, while more distal spikelet positions had the lightest grains, especially G4 in Bacanora T88 and Rayón F89 (Table 4). The weight of G4 grains in Rayon F89 S2 was as much as half the weight of G2 grains (Table 4). In general, grain weight was greater in CS than in BS or AS (Table 4). Although these trends were similar in all genotypes, the synthetic line showed a more symmetrical distribution of grain weight among grain positions, since grains in AS had lower weight than those in BS of Bacanora T88 and Rayón F89 (Table 5). Grain weight of G2 relative to G1 in the synthetic line was double that of G2/G1 in Bacanora T88 and Rayón F89.
View this table:
[in this window]
[in a new window]
|
Table 5. Relative grain weight (percentage change from CS-G1) at grain positions 1 (G1), 2 (G2), 3 (G3), and 4 (G4) from the rachis in basal (BS), central (CS), and apical (AS) spikelets (Spt) of Bacanora T88, Rayón F89, and the synthetic line. Values are the means of both sowing dates.
|
|
Grain Macronutrient Concentration
Grain position was the most important source of variation for grain concentration of Ca, Mg, K, P, and S (Table 3, ANOVA 1 and 2). However, genotype also had a significant effect on Ca, Mg, and P concentrations when common grain positions in Bacanora T88, Rayón F89, and the synthetic line were analyzed together. Calcium was the only macronutrient that varied between Bacanora T88 and Rayón F89. Sowing date only affected K concentration (Table 3, ANOVA 1). In the case of Ca and K, there was significant genotype x grain position interaction in both ANOVA 1 and 2.
Results showed that P and K were the macronutrients with the highest grain concentration in all three genotypes; S and Mg reached intermediate levels, and Ca showed the lowest concentration (Table 6). Grain concentration of Mg and P was
20% higher in the synthetic line, and there were no differences between Bacanora T88 and Rayón F89 for these macronutrients. The synthetic line showed the lowest concentration of Ca, while Bacanora T88 showed the highest (
16 and 24% higher than Rayón F89 and the synthetic line, respectively). There was little variation among genotypes for grain K and S concentrations.
View this table:
[in this window]
[in a new window]
|
Table 6. Grain macronutrient (M) concentrations in grain positions 1 (G1), 2 (G2), 3 (G3), and 4 (G4) from the basal (BS), central (CS), and apical (AS) spikelets (Spt) for sowing Date 1 (S1) and 2 (S2) for Bacanora T88, Rayón F89, and the synthetic line.
|
|
Comparison among spikelet groups revealed that AS generally had lower macronutrient concentration than did CS and BS. One exception was Ca, which was lowest in BS. Generally, grain Ca concentration was similar in AS and CS (Table 6). Grain macronutrient concentration between spikelets showed average differences of 4% (S) to 12% (Ca) across genotypes and sowing dates. Differences among genotypes were also observed. Bacanora T88 had the largest difference in grain macronutrient concentration (average of all nutrients) between spikelets (10%), while the synthetic line had the smallest difference (5%) (Table 6). Rayón was intermediate at 8%.
The clearest effect of grain position on macronutrient concentration was found between grains in CS (Table 6). In Fig. 1, the values were expressed relative to G1 (data shown are averages of S1 and S2) to facilitate viewing the differences in macronutrient concentration among grain positions. Except for K, grain macronutrient concentration decreased as distance from the rachis increased. The magnitude of the reduction, however, varied by nutrient; for example, Ca, Mg, P and S decreased by 30, 25, 17, and 10% in G4 of Bacanora T88 and Rayón F89, while K showed similar or higher concentrations in distal positions of CS (Fig. 1).

View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1. The concentration of macronutrients (Ca, Mg, K, P and S) in wheat grain at grain positions G2, G3 and G4 of central spikelets. Data are presented relative to G1 concentrations in Bacanora T88, Rayón F89, and the synthetic line. Closed symbols indicate significant (P < 0.05) differences from G1.
|
|
Grain Micronutrient Concentration
Grain concentrations of Cu, Fe, Mn, and Zn were affected by genotype and grain position (Table 3, ANOVA 1 and 2). Genotype x grain position interaction was significant only for Mn. Sowing date only affected Zn concentration, and Mn was the only micronutrient that showed sowing date x grain position interaction.
For all genotypes, Fe and Mn achieved the highest concentrations; Zn was intermediate, and Cu was the lowest. The most consistent effect of genotype on grain micronutrient concentration was evident in the difference between the synthetic line and cultivars Bacanora T88 and Rayón F89. The synthetic line had higher concentrations of Cu, Fe, Mn, and Zn than the other genotypes in almost all grain positions, while differences in concentration between Bacanora T88 and Rayón F89 were rare (Table 7). The synthetic line accumulated around 15, 20, 30, and 20% more Cu, Fe, Mn, and Zn, respectively, in the grain than the two cultivars (averages for sowing date and grain position, excluding G4). Differences between Bacanora T88 and Rayón F89 were always <5% regardless of whether G4 was included in the analysis. Sowing date had a smaller effect on grain micronutrient concentrations. The difference between sowings was <6% even for Cu (3.8%) and Zn (5.3%), the only micronutrients affected by sowing date (Tables 3 and 7).
View this table:
[in this window]
[in a new window]
|
Table 7. Grain micronutrient (m: Cu, Fe, Mn, and Zn) concentrations at grain positions 1 (G1), 2 (G2), 3 (G3), and 4 (G4) from the rachis in basal (BS), central (CS), and apical (AS) spikelets (Spt) for both sowing dates (S), sowing date 1 (S1), and 2 (S2) of Bacanora T88, Rayón, and the synthetic line.
|
|
Grain position affected grain micronutrient concentration (Table 7). The lowest concentration was generally found in the most distal grains of CS (G3 or G4, depending on the genotype). There was no grain position effect, however, on the highest grain micronutrient concentrations. Generally, proximal grains of BS, CS, and AS had similar concentrations, and these positions showed higher values than distal grains of CS.
The values of all grain positions were expressed relative to G1 CS (averages of S1 and S2, shown in Table 7) in Table 8 to facilitate comparing genotypes and grain positions for grain micronutrient concentration. The effect of spikelet position was most evident in Bacanora T88, in which grain concentrations of Cu, Fe, and Zn were lower in AS than in CS. Almost no effect of spikelet position was detected in Rayón and the synthetic line, yet micronutrient concentration was greatly reduced in distal grains of CS. This effect of position was similar among genotypes despite differences in grain set at G3 and G4, and dry matter distribution between grains (Tables 4 and 5). Although there were no clear differences among micronutrients trends, the effect of grain position was greater for Mn and Zn (where G4 decreased
20% relative to G1 in Bacanora T88 and Rayón F89) than for Fe and Cu concentrations (where the decrease in Bacanora T88 and Rayón F89 was
15%).
View this table:
[in this window]
[in a new window]
|
Table 8. Grain micronutrient concentrations (m) at grain positions 1 (G1), 2 (G2), 3 (G3), and 4 (G4) from the basal (BS), central (CS), and apical (AS) spikelets (Spt) relative to grain position G1 (central spikelet) within each genotype. Values are the mean of both sowing dates.
|
|
Water Dynamics and Macro- and Micronutrient Concentrations in Grains
There was a continuous reduction in grain water concentration in all genotypes during grain filling (Fig. 2). Grain position did not affect water concentration, as most of the sampling dates showed similar values for G1, G2, G3, and G4. Similarly, macronutrient concentration generally declined throughout grain filling. The decline was greater for Ca than for P and S. These differences were related to the kinetics of nutrient deposition in the grains. For example, Ca, P, and S concentrations (averaged across genotypes) on the first sampling date were 70, 30, and 28% of their values at physiological maturity, compared with 25% for dry matter.

View larger version (33K):
[in this window]
[in a new window]
|
Fig. 2. Water (W), Ca, Mg, K, P, and S concentrations in developing grains of Bacanora T88, Rayón F89, and the Synthetic. Samples were from the second sowing date. G1, G2, G3, G4 indicate grain positions within the CS, G1 is the grain position closest to the rachis. Values in parentheses indicate relative differences (%) between grain positions G3 or G4 (depending on genotype) and G1 averaged across all sampling dates.
|
|
Figure 2 shows that variation among genotypes for macronutrient concentration at harvest was evident in grain filling. Similarly, the differences between G1 and G3 (or G4, depending on the genotype) observed at maturity were also apparent during grain filling. Thus, the effect of grain position on Ca, Mg, P, and S concentration was established early in grain filling, as were the consistent values of K concentration between grain positions.
In agreement with data obtained for macronutrients, the effect of grain position on micronutrient concentration in grains at maturity was also found during grain filling (Fig. 3). In addition, micronutrient concentration in the synthetic line tended to be higher than in the two cultivars during development and at maturity (Fig. 3). Interestingly, Fe showed a different dynamic compared with other micronutrients. Concentrations of Cu, Mn, and Zn decreased (by an average of 30, 38, and 38%, respectively), while Fe concentration was nearly constant during grain filling (Fig. 3). This result implies that Fe unloading in grains had a similar relative rate as dry matter, while Cu, Mn, and Zn were unloaded at lower relative rates.

View larger version (35K):
[in this window]
[in a new window]
|
Fig. 3. Copper, Fe, Mn, and Zn concentrations in developing grains of Bacanora T88, Rayón F89, and the synthetic line. Samples were from the second sowing date. G1, G2, G3, G4 indicate grain positions within the CS, G1 is the grain position closest to the rachis. Values in parentheses indicate relative differences (%) between grain positions G3 or G4 (depending on genotype) and G1 averaged across all sampling dates.
|
|
 |
DISCUSSION
|
|---|
To cope with the objectives of the present study, genotypes were chosen on the basis of similar crop cycle and grain weight to reduce variation associated with grain development. The similitude showed by the genotypes for these traits was confirmed in the experiment (Tables 2 and 3). In addition, other criteria for choosing the genotypes were that they had different patterns of dry matter distribution within the spike and nutrient concentration in grains. These differences were also confirmed during the experiment as genotype x grain position interaction demonstrated that genotypes differed in dry matter distribution within the spike and in element concentration in grains (Table 3).
Genetic Variability
As expected, genotype had a significant effect on grain macronutrient (Ca, Mg, and P) and micronutrient (Cu, Fe, Mn, and Zn) concentrations. This effect was mainly the result of the higher concentrations reached in the synthetic line. Similar results have been reported for micronutrients evaluated in a wide range of synthetic lines (Ortiz-Monasterio and Graham, 2000). The present study confirms that there is variability for nutrient concentration in wheat grains, and that synthetic hexaploids may be a useful source for increasing nutrient concentrations. Nonetheless, exceptions to this general pattern were found for some macronutrients (Ca, K, and S).
Vertical vs. Horizontal Axis Spike Variability in Grain Nutrient Concentration
Grain position significantly affected grain weight in both the vertical and horizontal axes of the spike. Grains located in AS and the G3 and G4 positions of CS were particularly low in weight. Clear differences in grain nutrient concentrations both between and within spikelets were observed. The present study showed that the greatest differences in grain nutrient concentrations were found within spikelets. Thus, grains along the rachis were less affected in their nutrient concentration than those along the rachilla (Tables 6 and 8). Similar results have been found for grain N concentration between wheat spikelets (Simmons and Moss, 1978; Herzog and Stamp, 1983). Differences in grain N concentration between spikelets were also found in two-rowed barleys, where each spikelet carries only one grain (Ellis and Marshall, 1998). Although significant, these differences were quite small. In wheat, inter-spikelet differences in N concentration were smaller than those between proximal and distal grains within a spikelet (Simmons and Moss, 1978; Herzog and Stamp, 1983).
Breeding progress had different effect on macronutrient (N and P) and micronutrient (Fe and Zn) concentration in grains as N and P concentrations of high yielding wheat cultivars were reduced (Austin et al., 1980; Slafer et al., 1990; Calderini et al., 1995; Ortiz-Monasterio et al., 1997) while Fe and Zn were not significantly affected by the dilution effect (Ortiz-Monasterio and Graham, 2000). However, modern cultivars evaluated in the present study showed similar patterns of distribution within the spike for most of macronutrients and micronutrients; that is, distal positions within spikelets showed lower nutrient concentration at maturity and during all grain filling period. Moreover, water dynamics during this period showed that grain water concentration did not vary with grain position (Fig. 2), in agreement with previous reports (Calderini et al., 2000). In addition, the synthetic line showed similar nutrient and water patterns to cultivars. These results suggest that the lower nutrient concentrations found in distal grains could be related with the transport of nutrients from the rachis to distal positions. The ratio of phloem vs. xylem concentration of nutrients and the mobility of the different nutrients within the phloem (see Kochian, 1991) may prompt studies aimed at understanding the causes of the observed differences in nutrient concentration in individual grains.
Breeding Implications
The differences in grain weight and nutrient concentration between grain positions within the spike observed in this study could have important implications for future wheat breeding strategies. For example, if plant breeders continue to strive for increased wheat yields by selecting for grain set in distal positions of the spike, the inherent lesser grain weight potential at these distal positions will likely limit advances in grain yield. Moreover, the added grains will have progressively lower mineral concentrations compared with proximal grains of CS. This conclusion is supported by the positive associations (P < 0.05) between nutrient concentration and grain weight found in Bacanora T88 and Rayón F89 for most macronutrients: Ca (r = 0.41), Mg (r = 0.68), P (r = 0.82), and S (r = 0.77); and micronutrients: Cu (r = 0.56), Fe (r = 0.71), Mn (r = 0.65), and Zn (r = 0.59). The magnitude of this effect, however, would likely be different for each nutrient. For example, a greater imbalance between grain positions would be expected for Ca, Mg, and Mn than for S and Cu, considering the differences in nutrient concentrations among grain positions. On the other hand, no effect is expected for K.
From a practical perspective, a greater number of smaller grains would have negative consequences for test weight and milling efficiency. In addition, small seeds (Richards and Lukacs, 2002) and low seed Zn (Rengel and Graham, 1995) and Mn (Longnecker et al., 1991) contents negatively affect seedling vigor and resistance to infection during the seedling stage (Genc et al., 2000; McCay Buis et al., 1995).
A promising breeding strategy would be to select for grain yield by increasing average grain weight. This would take advantage of the fact that yield of modern wheat cultivars is sinkrather than sourcelimited during grain filling (Slafer and Savin, 1994; Richards, 1996; Kruk et al., 1997). The distribution of grain nutrient concentration within the spike was similar among the genotypes examined in this study (Fig. 1 and Table 8). But only a few significant (P < 0.05) associations were found between nutrient concentrations and grain weight in the synthetic line (Cu: r = 0.39; S: r = 0.43), which implies that any imbalance in nutrient concentration between grain positions is probably more closely related to spike architecture than to dry matter distribution among grains. If so, the alternative strategy of increasing individual grain weight rather than grain number would be less likely to impact the balance of grain nutrient concentrations within the spike. To this end, the use of synthetic hexaploids in breeding programs could offer a potent source for increasing grain weight without sacrificing grain nutritional quality (Villareal et al., 1994; Mujeeb-Kazi and Hettel, 1995; Calderini and Reynolds, 2000).
 |
ACKNOWLEDGMENTS
|
|---|
We especially thank Dr. A. Mujeeb-Kazi (CIMMYT) for providing the synthetic hexaploid line used in this study, as well as Dr. A. Barneix (University of Buenos Aires), Dr. R. Graham (University of Adelaide), and Dr. R. Welch (Cornell University) for critically reviewing the manuscript. We also thank Alma McNab (CIMMYT) for revising the English usage, and Magda Lobnik, Tirso Rojo, José Luís Miranda, and CIMMYT's Wheat Physiology Lab for their important technical assistance. This work was supported by an overseas scholarship of the University of Buenos Aires, Argentina (Programa Thalmann).
Received for publication July 20, 2001.
 |
REFERENCES
|
|---|
- Anonymous. 1992. World declaration and plan of action for nutrition: Int. Conf. on Nutrition, Rome. 511 Dec. 1992. Food and Agricultural Organization of the United Nations, Rome, and World Health Organization, Geneva.
- Austin, R.B., J. Bingham, R.D. Blackwell, L.T. Evans, M.A. Ford, C.L. Morgan, and M. Taylor. 1980. Genetic improvement in winter wheat yield since 1900 and associated physiological changes. J. Agric. Sci. (Cambridge) 94:675689.
- Bouis, H.E., R.D. Graham, and R.M. Welch. 2000. The Consultative Group on International Agricultural Research (CGIAR) micronutrient project: Justification and objectives. UNU Food Nutr. Bull. 21:374381.
- Bremner, P.M., and H.M. Rawson. 1978. The weights of individual grains of the wheat ear in relation to their growth potential, the supply of assimilate and interaction between grains. Aust. J. Plant Physiol. 5:6172.
- Calderini, D.F., L.G. Abeledo, and G.A. Slafer. 2000. Physiological maturity in wheat based on kernel water and dry matter. Agron. J. 92:895901.[Abstract/Free Full Text]
- Calderini, D.F., and M.P. Reynolds. 2000. Changes in grain weight as a consequence of de-graining treatments at pre- and postanthesis in synthetic hexaploid lines of wheat (Triticum durum x T. tauschii). Aust. J. Plant Physiol. 27:183191.
- Calderini, D.F., M.P. Reynolds, and G.A. Slafer. 1999. Genetic gains in wheat yield and associated physiological changes during the twentieth century. p. 351377. In E.H. Satorre and G.A. Slafer (ed.) Wheat: Ecology and physiology of yield determination. Haworth Press, New York.
- Calderini, D.F., and G.A. Slafer. 1998. Changes in yield and yield stability in wheat during the 20th Century. Field Crops Res. 57:335347.
- Calderini, D.F., and G.A. Slafer. 1999. Has yield stability changed with genetic improvement of wheat yield? Euphytica 107:5159.[ISI]
- Calderini, D.F., S. Torres León, and G.A. Slafer. 1995. Consequences of wheat breeding on nitrogen and phosphorus yield, grain nitrogen and phosphorus concentration and associated traits. Ann. Bot. (London) 76:315322.[Abstract/Free Full Text]
- Duvick, D.N. 1996. What is yield? p. 332335. In G.O. Edmeades et al. (ed.) Developing drought- and low N-tolerant maize. CIMMYT, Mexico D.F., Mexico.
- Ellis, R.P., and B. Marshall. 1998. Growth, yield and grain quality of barley (Hordeum vulgare L.) in response to nitrogen uptake. Plant development and rate of germination. J. Exp. Bot. 49:10211029.[Abstract/Free Full Text]
- Genc, Y., G.K. McDonald, and R.D. Graham. 2000. Effect of seed zinc content on early growth of barley (Hordeum vulgare L.) under low adequate soil zinc supply. Aust. J. Agric. Sci. 51:3746.
- Herzog, H., and P. Stamp. 1983. Dry matter and nitrogen accumulation in grains at different ear positions in gigas semidwarf and normal spring wheats. Euphytica 32:511520.
- Kochian, L.V. 1991. Mechanisms of micronutrients uptake and translocation in plants. p. 229296. In J.J. Mortvedt et al. (ed.) Micronutrients in agriculture. 2nd ed. SSSA Book Ser. 4. SSSA, Madison, WI.
- Kruk, B.C., D.F. Calderini, and G.A. Slafer. 1997. Grain weight in wheat cultivars released from 1920 to 1990 as affected by postanthesis defoliation. J. Agric. Sci. 128:273281.
- Longnecker, N.E., N.E. Marcar, and R.D. Graham. 1991. Increased manganese content of barley seeds can increase grain yield in manganese-deficient conditions. Aust. J. Agric. Res. 42:10651074.
- López Pereira, M., N. Trápani, and V.O. Sadras. 2000. Genetic improvement of sunflower in Argentina between 1930 and 1995. Part III. Dry matter partitioning and grain composition. Field Crops Res. 67:215221.
- Loss, S.P., and K.H.M. Siddique. 1994. Morphological and physiological traits associated with wheat yield increases in Mediterranean environments. Adv. Agron. 52:229276.
- McCay Buis, T.S., D.M. Huber, R.D. Graham, J.D. Phillips, and K.E. Miskin. 1995. Manganese seed content and take-all of cereals. J. Plant Nutr. 18:17111721.
- Miralles, D.J., and G.A. Slafer. 1995. Individual grain weight responses to genetic reduction in culm length in wheat as affected by sourcesink manipulations. Field Crops Res. 43:5566.
- Mujeeb-Kazi, A., and G.P. Hettel. 1995. Utilizing wild grass biodiversity in wheat improvement. 15 years of wild cross research at CIMMYT. CIMMYT Res. Rep. No. 2. CIMMYT, Mexico D.F., Mexico.
- Ortiz-Monasterio, J.I., and R.D. Graham. 2000. Breeding for trace minerals in wheat. UNU Food Nutr. Bull. 21(4):392396.
- Ortiz-Monasterio, J.I., K.D. Sayre, S. Rajaram, and M. McMahon. 1997. Genetic progress in wheat yield and nitrogen use efficiency under four nitrogen rates. Crop Sci. 37:898904.[Abstract/Free Full Text]
- Rengel, Z., and R.D. Graham. 1995. Wheat genotypes differ in Zn efficiency when grown in chelate-buffered nutrient solution. I. Growth. Plant Soil 176:307316.
- Richards, R.A. 1996. Increasing the yield potential in wheat: Manipulating sources and sinks. p. 134149. In M.P. Reynolds et al. (ed.) Increasing yield potential in wheat: Breaking the barriers. CIMMYT, Mexico D.F., Mexico.
- Richards, R.A., and Z. Lukacs. 2002. Seedling vigour in wheat-sources of variation for genetic and agronomic improvement. Aust. J. Agric. Res. 53:4150.
- Rosegrant, M.W., M.A. Agcaolili-Sombilla, R.V. Gerpacio, and C. Ringler. 1997. Global food markets and US exports in the twenty-first century: Illinois World Food and Sustainable Agriculture Program Conf., Meeting the Demand for Food in the 21st Century: Challenges and Opportunities. 28 May 1997. Univ. of Illinois, Urbana-Champaign, IL.
- Schnyder, H., and U. Baum. 1992. Growth of the grain of wheat (Triticum aestivum L.). The relationship between water content and dry matter accumulation. Eur. J. Agron. 1:5157.
- Simmons, R., and D.N. Moss. 1978. Nitrogen and dry matter accumulation by kernels formed at specific florets in spikelets of spring wheat. Crop Sci. 18:139143.[Abstract/Free Full Text]
- Slafer, G.A., F.H. Andrade, and F.E. Feingold. 1990. Genetic improvement of bread wheat (Triticum aestivum L.) in Argentina: Relationships between nitrogen and dry matter. Euphytica 50:6371.
- Slafer, G.A., E.H. Satorre, and F.H. Andrade. 1994. Increases in grain yield in bread wheat from breeding and associated physiological changes. p. 168. In G.A. Slafer (ed.) Genetic Improvement of Field Crops. Marcel Dekker, New York.
- Slafer, G.A., and R. Savin. 1994. Grain mass change in a semi-dwarf and a standard-height wheat cultivar under different sink-source relationships. Field Crops Res. 37:3949.
- Sofield, I., I.F. Wardlaw, L.T. Evans, and S.Y. Zee. 1977. Nitrogen, phosphorus and water contents during grain development and maturation in wheat. Aust. J. Plant Physiol. 4:799810.
- Underwood, B.A. 2000. Overcoming micronutrient deficiencies in developing countries: Is there a role for agriculture? UNU Food Nutr. Bull. 21:356360.
- Villareal, R.L., A. Mujeeb-Kazi, E. Del Toro, J. Crossa, and S. Rajaram. 1994. Agronomic variability in selected Triticum turgidum x T. tauschii synthetic hexaploid wheats. J. Agron. Crop Sci. 173:307317.
- Welch, R.M., and R.D. Graham. 1999. A new paradigm for world agriculture: Meeting human needs Productive, sustainable, nutrious. Field Crops Res. 60:110.
- Welch, R.M., and R.D. Graham. 2000. A new paradigm for world agriculture: Productive, sustainable, nutritious, healthful food systems. UNU Food Nutr. Bull. 21:361366.
- World Bank. 1994. The challenge of dietary deficiencies of vitamins and minerals. p. 613. In Anonymous enriching lives: Over-comming vitamin and mineral malnutrition in developing countries. World Bank, Washington, DC.
- Zadoks, J.C., T.T. Chang, and C.F. Konzak. 1974. A decimal code for the growth stages of cereals. Weed Res. 14:415421.
- Zarcinas, B.A., B. Cartwright, and L.R. Spouncer. 1987. Nitric acid digestion and multielement analysis of plant material by inductively coupled plasma spectrometry. Commun. Soil. Sci. Plant Anal. 18:131146.
This article has been cited by other articles:

|
 |

|
 |
 
B. L. Duggan and D. B. Fowler
Yield Structure and Kernel Potential of Winter Wheat on the Canadian Prairies
Crop Sci.,
May 18, 2006;
46(4):
1479 - 1487.
[Abstract]
[Full Text]
[PDF]
|
 |
|