Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 6 February 2007
Published in Crop Sci 47:359-366 (2007)
© 2007 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 Similar articles in this journal
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 Google Scholar
Google Scholar
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Agricola
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Related Collections
Right arrow Soybean
Right arrow Crop Growth and Development
Right arrow Crop Physiology & Metabolism

CROP PHYSIOLOGY & METABOLISM

Accumulation of Nitrogen and Dry Matter by Soybean Seeds with Genetic Differences in Protein Concentration

D. B. Egli* and W. P. Bruening

Dep. of Plant and Soil Sciences, Univ. of Kentucky, Lexington, KY 40546-0312

* Corresponding author (degli{at}uky.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Soybean [Glycine max (L.) Merrill] yields often decline as seed protein levels increase, but the processes responsible are not clearly understood. We compared dry matter and N accumulation by individual seeds of three high protein genotypes (K 1431, KS 4402sp, NE 3396) and three commercial cultivars (Pennyrile, Flyer, Hutcheson) in the field near Lexington, KY (38° 01' N lat; 84° 35' W long) for 3 yr to determine if increasing protein concentration decreased the seed growth rate. Plants were grown in 76.2-cm rows using conventional production practices and overhead irrigation to minimize soil moisture deficits. The rate of dry matter accumulation by individual seeds (SGRDM) varied from 4.2 to 6.1 mg seed–1 d–1 and this variation was closely associated with mature seed size, but it was not affected by seed N concentration. The rate of N accumulation (SGRN, mg N seed–1 d–1) was closely associated with seed N concentration at maturity across all genotypes and years (r2 = 0.76). Thus, SGRN and SGRDM seemed to vary independently as the N concentration in mature seeds increased. The duration of seed filling (estimated by the effective filling period) was not related to mature seed N concentration. Since higher seed N concentration had no effect on the rate or duration of seed dry matter accumulation by individual seeds, the purported negative effect of seed protein levels on yield may be more closely associated with whole plant phenomena than those operating at the individual seed level.

Abbreviations: EFP, effective filling period • MG, maturity group • SGRDM, rate of dry matter accumulation per seed • SGRN, rate of N accumulation per seed


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE ABILITY of individual seeds to accumulate dry matter is a fundamental component of the yield production process in grain crops. The characteristics of seed growth are well understood (Egli, 1998) and this knowledge has contributed to a better understanding of the determination of yield. Less is known about the accumulation of N by individual seeds, in spite of the importance of N metabolism in the yield production process in soybean and other grain legumes.

Nitrogen acquisition in soybean involves both N fixation by Bradyrhizobium in the nodules and reduction of NO3– taken-up from the soil solution (Harper, 1987). Some of the N in the mature seed comes from redistribution of N from vegetative plant parts during seed filling with estimates of this contribution varying from 30 to 100% (Egli et al., 1978a; Jeppsen et al., 1978; Zeiher et al., 1982; Egli et al., 1983). Regardless of the N source, proteins in the seed are synthesized from amino acids supplied by the mother plant (Rainbird et al., 1984).

The concentration of N in mature soybean seed is determined, in part, by environmental conditions. Seed N concentration usually declines as temperature during seed filling increases (Piper and Boote, 1999; Wilson, 2004), although Wolf et al. (1982) reported increases at higher temperatures (33/28°C day/night temperatures). Seed N concentration was lower in pods near the bottom of the main stem (Collins and Cartter, 1956; Bennett et al., 2003), was influenced by N availability (Streeter, 1978; Crozat et al., 1994), and increased as source-sink ratios increased (McAlister and Krober, 1958; Openshaw et al., 1979; Miceli et al., 1995).

Seed N concentration is also under genetic control (Brim and Burton, 1979) and there is substantial variation in the germplasm collection (Wilson, 2004). Higher protein concentrations are usually associated with lower oil concentrations (Hartwig and Hinson, 1972; Brim and Burton, 1979; Helms and Orf, 1998; Wilson, 2004) and often, but not always, lower yield (Hartwig and Hinson, 1972; Brim and Burton, 1979; Helms and Orf, 1998).

Productivity and economic value of soybean are, at least partially, related to accumulation of N and protein in the seed, but a detailed evaluation of the relationship between seed protein levels and yield is limited by a lack of information at the individual seed level. Consequently, our objective was to investigate the basis for the negative relationship between soybean yield and seed protein concentration by determining the effect of seed protein levels on dry matter and N accumulation by individual seeds.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Three soybean genotypes with high seed protein concentrations [K 1431, maturity group (MG) IV; KS 4402sp, MG IV; and NE 3396, MG III] and three adapted commercial cultivars with normal protein levels [Flyer, MG IV; Pennyrile, MG IV; and Hutcheson, MG V] were grown on Spindletop Farm near Lexington KY for 3 yr. Genotypes within each protein classification were chosen to represent a range in mature seed size to facilitate the separation of possible effects of seed size and seed protein concentration on seed growth rate.

Seeds were planted on 22 May 2002 [Lanton silt loam (fine-silty, mixed, superactive, thermic Cumulic Epiaquolls)], 2 June 2003 [Donerail silt loam (fine, mixed, active, mesic Oxyaquic Argiudolls)], and 24 May 2004 [Maury silt loam (fine, mixed, semiactive, mesic Typic Paleudalfs)] at a rate of 16 to 20 seeds per m of row. Each plot consisted of five 76.2-cm rows 6 m long and there were three replications of each genotype in a randomized complete block design. Soybean had been grown recently at each location so the seeds were not inoculated with Bradyrhizobium japonicum. The plots were irrigated with an overhead sprinkler system as needed to minimize soil water deficits. Reproductive growth stages (Fehr and Caviness, 1977) were determined on two replications at weekly intervals.

Individual pods at the same stage of development (full length and containing seeds that were just starting to swell) were identified, regardless of nodal position, by marking with acrylic paint when the plants of each cultivar were approximately at growth stage R5. All pods on each genotype were marked on the same day. The plants with marked pods were in bordered rows and there were usually ≤ 3 marked pods per plant. Ten marked pods per plot were harvested at approximately weekly intervals (all marked pods on a plant were harvested on the same day) during the linear phase of seed growth (total of 4–5 samples), placed on ice and taken immediately to the laboratory where the seeds were removed from the pods, counted, dried at 60°C, weighed, and ground for analysis. A final sample (approximately 30 marked pods per plot) was taken at maturity (approximately growth stage R8). Total N was determined by a modified Kjeldahl procedure (Nelson and Sommers, 1973; Heberer et al., 1985).

Individual seed growth rate (rate of dry matter accumulation- SGRDM) and the rate of N accumulation (SGRN) were estimated for each plot by linear regression of seed dry weight (mg seed–1) or seed N content (mg N seed–1) vs. time, using only samples taken during the linear phase of seed growth. The regression analyses were significant (P ≤ 0.05) for all plots and the r2 was always > 0.90 and usually > 0.95. Seed fill duration was estimated by the effective filling period (EFP = mature seed size/seed growth rate) (Daynard et al., 1971). Genotype effects on SGRDM and SGRN were evaluated by analysis of variance (three replications) within each year. Regression analysis was used to evaluate relationships between variables. The least significant difference (LSD) was used to compare individual treatment means when the treatment effect was significant in the analysis of variance.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The genotypes in these experiments were selected to provide differences in seed protein concentration, but there was not always as much separation between the high protein and the normal genotypes at maturity as we had hoped (Table 1). There was, however, a substantial range from the high to the low genotype each year, providing the variation needed to evaluate the relationship between seed protein levels and seed growth characteristics.


View this table:
[in this window]
[in a new window]

  		Table 1. Seed growth characteristics of three high and three normal protein soybean genotypes.

 
Seed N concentration was relatively constant during the linear phase of seed growth in 2004 (Fig. 1 ) and the other 2 yr (data not shown) for three of the genotypes (NE 3396, Flyer, KS 4402sp), but not for the other three (Pennyrile, Hutcheson and K 1431). The concentration of Pennyrile and Hutcheson did not change during seed development in the other 2 yr, but the increase in K 1431 occurred in all 3 yr. The high N concentrations in the high-protein genotypes NE 3396 and K 1431 were apparent throughout seed development.


Figure 1
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 1. The effect of genotypes on soybean seed N concentration during seed development in 2004. Error bars represent ± standard error of the mean. Bars smaller than the symbols are not shown.

 
Both the seed N content (mg N seed–1) and dry matter (mg seed–1) increased at a constant rate in all 3 yr (only 2004 is shown in Fig. 2 ), consequently, linear regression analysis was used to estimate SGRDM and SGRN. There was significant genotypic variation in both traits in all 3 yr (Table 1).


Figure 2
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 2. The effect of genotypes on seed N and dry weight accumulation in 2004. Error bars represent ± standard error of the mean. Bars smaller than the symbols are not shown.

 
The SGRDM increased significantly as mature seed size increased (Fig. 3 ), as expected from the work of Egli et al. (1978b, 1981) and Guldan and Brun (1985), but, after considering these seed size effects, there was no consistent relationship between seed protein concentration and SGRDM. Seeds from high protein genotypes exhibited both higher and lower seed growth rates than expected for a specific seed size. The r2 for the relationship between the SGRN and seed size was smaller (Y = 0.043 + 0.00169X, r2 = 0.23*, n = 18, data not shown) than it was for SGRDM vs. size; there was, however, a much better relationship (significant at P = 0.05) between SGRN and mature seed N concentration (Fig. 4 ). A single regression equation described the relationship for seeds of genotypes in all 3 yr.


Figure 3
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 3. The relationship between soybean seed growth rate (SGRDM) and mature seed size. The regression equation was significant at P = 0.05.

 

Figure 4
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 4. The relationship between the rate of N accumulation by soybean seed (SGRN) and mature seed N concentration. The regression equation was significant at P = 0.001.

 
There was significant (P = 0.05) genotypic variation in the duration of seed filling (estimated by the EFP) in each of the 3 yr (Table 1), but there was no significant relationship with mature seed N concentration (Fig. 5 ).


Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 5. The relationship between the effective filling period and mature seed N concentration. The relationship was not significant (P > 0.05).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Seeds accumulate dry matter at a constant rate throughout much of their development (Fig. 2) (Egli, 1998). Nitrogen also accumulated at a constant rate in our experiments (Fig. 2) in agreement with the reports of Rubel et al. (1972) and Yazdi-Samadi et al. (1977). The seed N concentration of some of the genotypes in these experiments was relatively constant during the linear phase of seed development (Fig. 1) as reported by Yazdi-Samadi et al. (1977). Three of our genotypes, however, (Pennyrile, Hutcheson and K 1431) exhibited variation during seed development in 2004. Such variation has been noted before (Rubel et al., 1972; Miceli et al., 1995) and could be associated, in part, with environmental effects (variation in Pennyrile and Hutcheson occurred in only one of 3 yr) or it could be a characteristic of the genotype (the upward trend of K 1431 was consistent across all 3 yr). The variations in N concentration were not large enough to cause significant deviations from linearity in the accumulation of N per seed (Fig. 2).

The growth of an individual seed is dependent on raw materials, principally sucrose and a variety of amino acids, supplied by the mother plant (Egli, 1998). Storage proteins and oil are then synthesized in the seed. Penning de Vries et al. (1974) found that the energy cost of this synthesis varied among seed constituents with oil requiring the most glucose per g of product, followed by protein and then carbohydrates. Thus, the energy (reduced carbon) required to support growth of an individual seed would depend on its composition, suggesting that increasing protein concentration may decrease SGRDM.

There was no evidence in our experiments, however, that SGRDM was consistently affected by seed N concentration. The SGRDM was related to seed size, as expected (Egli et al., 1978b, 1981; Guldan and Brun, 1985), but the rates for the high protein genotypes were both higher and lower than the rate predicted by the average size effect (regression line in Fig. 3). The SGRN increased as seed N concentration increased (Fig. 4), but these higher rates apparently did nothing to compromise the ability of the seed to accumulate dry matter. The independence of SGRN and SGRDM also occurred in pea (Pisum sativum L.) when seed N concentration was manipulated by modifying the supply of N to the seed (Lhuillier-Soundele et al., 1999.)

The higher energy cost of protein synthesis is often used to explain lower levels of productivity in species with high protein seeds (Sinclair and de Wit, 1975), but apparently this relationship cannot be extended to growth of individual seeds. There are at least two possible explanations for this failure. First, some of the energy expended in protein synthesis comes from the reduction of NO3– or fixation of N2 which occurs in the mother plant, not the seed (Loomis and Conner, 1992). Second, increasing seed protein concentration often reduces oil concentration (Wilson, 2004), which, since the energy cost for oil synthesis is even greater than protein, may actually lower the glucose requirement for the synthesis of a unit weight of seed tissue, depending on the relative changes in oil and protein concentrations. If increasing seed protein concentration does not increase the amount of glucose required to support growth of an individual seed, perhaps it is not surprising that we couldn't find a negative association between SGRDW and seed N levels.

The SGRDM was not related to the mature seed N concentration, but SGRN was (Fig. 4), so seeds with higher N concentrations accumulated more N per unit dry weight accumulation than those with lower N concentrations as illustrated in Fig. 6 . This relationship seems to suggest that seeds with high N concentrations were exposed to assimilate supplies enriched in N relative to C in comparison to seeds with lower N concentrations. Culture experiments with soybean, however, (Hayati et al., 1996) and corn (Zea mays L.) (Wyss et al., 1991) demonstrated that genetic differences in seed N concentration were maintained over a range of media N levels that caused changes in seed N concentration. These findings provide strong evidence that the characteristics of the seed may be more important than the supply of N to the seed in controlling genetic variation in seed N concentration. The consistency of genetic differences across years (Table 1) also supports regulation by the seed since the genetic makeup up of the seed would not be affected by environmental conditions which could cause variation in the N supply per seed. Our experiments with individual seeds were not designed to resolve these control issues which will probably require more detailed experiments including evaluations at the whole plant level.


Figure 6
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 6. The relationship between the ratio of the rate of accumulation of N and dry matter (SGRN/SGRDM) by soybean seeds and mature seed N concentration. The regression equation was significant at P = 0.001.

 
Theoretical relationships between seed protein concentration and leaf senescence were proposed by Sinclair and de Wit (1975, 1976) and others (Frederick and Hesketh, 1994) and they argued that increasing seed N concentrations might accelerate leaf senescence and shorten the seed-filling period. We, however, found no relationship between seed N concentration and the length of the EFP (Fig. 5). Selecting for high seed protein levels increased the rate of N accumulation per seed, but it did not shorten the EFP relative to the normal cultivars. The EFPs reported here (27–38 d) are in the range of previous reports for soybean (Egli, 1981; Egli et al., 1984; Meckel et al., 1984). The failure to find an association between EFP and N concentration of individual seeds cannot be used, by itself, to discredit models linking seed N levels and senescence (Sinclair and de Wit, 1976; Frederick and Hesketh, 1994) without evaluating the effect of selection for high seed protein on the total seed N requirement (rate of N accumulation per seed x seeds m–2) and the total N available for reproductive growth (N uptake, fixation and redistribution). Evaluations at the whole plant level have, in some cases, shown that high seed N concentrations shorten seed fill duration (Salarado-Navarro et al., 1985). However, earlier work demonstrated clearly that increasing the rate of seed N accumulation per plant did not, even in the absence of nodules or N in the root zone, increase the rate of senescence or shorten the seed-filling period (Hayati et al., 1995). The seed fill duration of species with high seed protein levels is not necessarily shorter than species with low seed protein levels (Egli, 1981). Our results demonstrating that the duration of growth of individual soybean seeds was not shortened by high seed protein levels are consistent with these findings.

Increasing seed protein concentrations in soybean through breeding often results in lower yields (Hartwig and Hinson, 1972; Brim and Burton, 1979; Helms and Orf, 1998). The fundamental processes responsible for this relationship are not clear, and our results, unfortunately, do little to clarify the situation. The N requirement per unit seed growth increased (Fig. 6) with seed N concentration, but this increase had no effect on EFP, suggesting that lower yields of high protein genotypes may not be the result of shorter seed fill durations. Pod and seed number are thought to be related to the balance between assimilate supply and its use by the developing seed (Charles-Edwards et al., 1986; Egli, 1998). If assimilate use by the seed is not affected by the seed N level (SGRDM was not related to seed protein levels), seed number, the primary yield component (Egli, 1998) should not be affected by an increase in seed protein concentration (assuming that the amount of assimilate partitioned to seeds is not changed). If, however, higher seed N concentrations were associated with relatively more N in the vegetative plant, assimilate may be diverted to N acquisition and therefore would not be available to support pod and seed growth. Nitrogen acquisition would have to be active during at least the early stages of reproductive growth, a likely occurrence (Harper, 1987), for this scenario to lower yields as a result of fewer pods and seeds per unit area. Measurements of N acquisition and redistribution during reproductive growth are needed to resolve this question.

In summary, we found no relationship between seed protein levels and the rate or duration of seed dry matter accumulation in the six genotypes evaluated. There were, however, differences in the ability of the seed to accumulate N as the SGRN per unit dry weight accumulation increased significantly with seed protein concentrations. These results suggest that the negative relationship between seed protein concentrations and yield cannot be explained at the single seed level, but apparently it is a whole plant phenomenon, possibly involving reductions in the assimilate supply to the seeds as the plants accumulate more N.


   ACKNOWLEDGMENTS
 
We thank Dr. George L. Graef, University of Nebraska, and Dr. W.T. Schapaugh, Kansas State University, for supplying seed of the high protein genotypes. This project was funded, in part, by the New Crops Opportunities Center at the University of Kentucky.

Received for publication February 15, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
S. L. Naeve, T. A. O'Neill, and J. E. Miller-Garvin
Canopy Nitrogen Reserves: Impact on Soybean Yield and Seed Quality Traits in Northern Latitudes
Agron. J., May 7, 2008; 100(3): 681 - 689.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
V. O. Sadras and D. B. Egli
Seed Size Variation in Grain Crops: Allometric Relationships between Rate and Duration of Seed Growth
Crop Sci., March 19, 2008; 48(2): 408 - 416.
[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 Similar articles in this journal
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 Google Scholar
Google Scholar
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Agricola
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Related Collections
Right arrow Soybean
Right arrow Crop Growth and Development
Right arrow Crop Physiology & Metabolism

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