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


     

Published online 1 July 2008
Published in Crop Sci 48:1355-1364 (2008)
© 2008 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 Gao, Y.
Right arrow Articles by Buss, G.R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gao, Y.
Right arrow Articles by Buss, G.R.
Agricola
Right arrow Articles by Gao, Y.
Right arrow Articles by Buss, G.R.
Related Collections
Right arrow Soybean
Right arrow Phosphorus
Right arrow Crop Genetics

Validation of Low-Phytate QTLs and Evaluation of Seedling Emergence of Low-Phytate Soybeans

Y. Gao, R.M. Biyashev, M.A. Saghai Maroof*, N.M. Glover, D.M. Tucker and G.R. Buss

Dep. of Crop and Soil Environmental Sciences, Virginia Tech, Blacksburg, VA 24061

* Corresponding author (smaroof{at}vt.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
M153-1-4 is a low-phytate (LP) soybean [Glycine max (L.) Merr.] mutant derived from chemical mutagenesis, possibly from a single gene mutation. However, the inheritance of LP in CX1834, as descended from M153-1-4, has been shown to involve two independent loci which were subsequently mapped to linkage groups (LGs) L and N. The objectives of this study were to (i) validate the two previously identified LP QTLs in CX1834 using 208 F2:5 lines of CX1834-1-6 x V99-3337, (ii) characterize associations of LP QTLs with phytate content in a set of F2:5 and backcross (BC) lines as derived from phenotypic or genotypic selections, and (iii) evaluate the impact of LP on seedling emergence. Overall, the previously reported LP QTLs were validated in this study. However, observations of natural genetic variation and transgressive segregants suggest that additional LP QTLs could exist, potentially contributed by V99-3337, the other parent of the validation population. A perfect correlation was found between phenotypic (phytate-based) and genotypic (marker-based) selections for LP lines in this study. However, findings of LP marker genotypes in the high phytate (HP) phenotypic selections suggest a potential linkage between HP and LP genes on LG L in CX1834. LP lines did not have uniformly low emergence in this study. Under some environments, HP and LP lines had little difference in seedling emergence in the field.

Abbreviations: ANOVA, Analysis of variance • BC, backcross • cM, centiMorgan • dd, double-distilled • HP, high or highest phytate • IM, interval mapping • LG, linkage group • LOD, logarithm of the odds • LP, low or lowest phytate • MAS, marker-assisted selection • MGs, maturity groups • Pi, inorganic phosphorus • QTL, quantitative trait loci • SSR, simple sequence repeat


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
ABOUT 75% OF THE TOTAL phosphorus (P) in conventional soybean [Glycine max (L.) Merr.] seed exists as phytic acid or, in its complex anion form, phytate salt. It has been estimated that about 50% of the total elemental P applied every year in world agriculture ends up in the phytate molecule (Dilger and Adeola, 2006; Karr-Lilienthal et al., 2005; Lott et al., 2000). However, phytate P in animal feed cannot be readily digested by non-ruminant animals such as swine and poultry because they do not produce sufficient phytase to breakdown phytate. Therefore, phytate P in animal feed is largely excreted in manure, causing phosphorus pollution in the environment (Raboy, 2001). To meet nutritional requirements for optimal productivity, soy feed must be supplemented with a form of available P or must be treated with the enzyme phytase, either of which leads to extra production cost in poultry and swine industries. Genetic reduction of phytate content in soybean lines would result in better digestibility by non-ruminant animals, improved feed efficiency, and decreased P excretion (Dilger and Adeola, 2006; Powers et al., 2006).

Induced by chemical mutagenesis, two sources of low phytate soybean mutants, LR33 (Sebastian et al., 2000) and M153-1-4 (Wilcox et al., 2000), have been used to develop LP soybean lines (Hitz et al., 2002; Oltmans et al., 2004; Walker et al., 2006). A 50% reduction in seed phytate level was found in mutant LR33 (Hitz et al., 2002) and 80% reduction was observed in the mutant line M153-1-4 (Wilcox et al., 2000). The LP mutations in LR33 as well as in maize, barley, and rice had each been previously mapped to a single locus (Hitz et al., 2002; Larson et al., 2000; Larson et al., 1998; Raboy et al., 2000). The low-phytate phenotype of M153-1-4 is also considered to be a single gene mutation on the basis of data from Wilcox et al. (2000) (Oltmans et al., 2004; Walker et al., 2006). However, the genetic inheritance of low phytate in CX1834, an F3:5-derived line from ‘Athow’ x M153-1-4, has been characterized as being controlled by two independent recessive genes, pha1pha1 and pha2pha2 (Oltmans et al., 2004). Utilizing an indirect inorganic phosphorus (Pi) assay, Walker et al. (2006) subsequently identified two QTLs contributing to the low-phytate phenotype in CX1834-1-2 (a sister line of CX1834-1-6) on LG N and L near Satt237 and Satt527, respectively. Therefore, origins of these two low phytate genes, pha1pha1 and pha2pha2, must be addressed (Oltmans et al., 2004; Walker et al., 2006) and genetic inheritance of low phytate in both M153-1-4 and CX1834 needs validation. As QTL mapping results may vary across studies because of the use of different genotypes, environments, and sampling variation (Cabezas et al., 2006; Ha et al., 2007), confirmation and validation of seed phytate composition QTLs is required.

Despite desirable characteristics of LP soybean lines, some inherent defects, including reduced seedling emergence, have been observed to be associated with the LP trait in soybean seeds (Raboy, 2007). However, emergence tests with soybeans derived from different sources of LP mutants have shown varying degrees of reduction in seedling emergence (Hulke et al., 2004; Meis et al., 2003; Oltmans et al., 2005). Some LP lines with improved seedling emergence have been reported (Spear and Fehr, 2007). However, questions concerning impacts of LP on seedling emergence still remain, particularly when sources of LP are different.

The study reported here was undertaken with the following three objectives, (i) validate two previously identified LP QTLs (Walker et al., 2006) in CX1834, using 208 advanced F2:5 lines of CX1834-1-6 x V99-3337 and a direct phytate assay protocol (Gao et al., 2007), (ii) verify associations of the two LP QTLs in CX1834 with phytate content in backcross progenies as identified by markers closely linked to the LP QTLs and in inbred lines phenotypically selected for LP and HP contents, and (iii) evaluate the impact of LP on seedling emergence with different sources of low-phytate lines.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Plant Material
The LP soybean line, CX1834-1-6 (Oltmans et al., 2004; Walker et al., 2006; Wilcox et al., 2000), obtained from J.R. Wilcox, USDA-ARS and Purdue University, was crossed with V99-3337, a Virginia Tech experimental line derived from ‘Hutcheson’ (Buss et al., 1988). F1 seeds were planted in 2002 at Kentland Farm (Blacksburg, VA) and were confirmed to be true hybrids by DNA markers. One F1 plant designated as 02-512-1 was allowed to self and also was backcrossed to V99-3337 as the female parent. The derived F2 and BC1F1 seeds were harvested in the fall and planted in the field in 2003. A total of 208 F2 individuals were planted, from which 208 F2:5 families were developed from 2003 to 2005 and used as the validation mapping population in this study for LP QTL. From these 208 F2:5 families, the five highest phytate (HP) lines (designated as H-1, to -5) and the five lowest phytate (LP) lines (designated as L-1 to -5) were selected for characterizing associations of phytate content with LP QTLs and seedling emergence. In 2003, a total of 69 BC1F2 plants were genotyped at the Satt237, Satt339, Satt527, and Satt561 marker loci, which were indentified to be closely linked to LP content in CX1834-1-2 through inorganic phosphorus assays (Walker et al., 2006). One of the 69 BC1F2 plants was selected for containing homozygous CX1834-1-6 alleles at all of the four marker loci and designated as 03-516-4. In 2004, six BC1F4 lines (04-1272, 04-897-1 to -5) were developed from BC1F2 03-516-4. In 2005, 21 BC1F5 lines were developed from 04-1272 and 04-897-1 to -5 by single plant selections.

Controls included in this study were two LP mutants CX1834-1-6 and M766 (Wilcox et al., 2000), five commercial cultivars, Essex, Hutcheson, Lee 68, Mn1401, and Williams, five Virginia Tech experimental lines, V71-370, V97-7158, V99-3337, V99-5089, and V99-8060, three plant introductions, PI 87013, PI 96983, and PI 200508, and one Arkansas experimental line, R95-1705. Among the five Virginia Tech experimental lines, V99-5089 is a LP line with phytate levels in between CX1834 and M766 (Gao et al., 2007). The maturity groups (MGs) of each of these controls are summarized in Table 1 .


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

  		Table 1. Phytate and total P levels and seedling emergence of selected soybean lines used in the study.

 
Molecular Marker Assay and Data Analysis
A total of 30 simple sequence repeat (SSR) markers from LGs L and N were screened for polymorphism between the parents CX1834-1-6 and V99-3337. These markers were chosen based on estimated locations on the integrated soybean linkage maps of Cregan et al. (1999) and Song et al. (2004) to span the high inorganic phosphorus QTL detected by Walker et al. (2006). Among 30 markers, 17 (57%) were polymorphic between the two parents and assayed on the entire population, with eight marker loci on LG L and nine on LG N. Phytate content was assayed for each of the 208 F2:5 families. Leaf tissue samples from individual F2 plants were collected for DNA extractions. All genotyping procedures including DNA isolation, polymerase chain reactions, polyacrylamide gel-electrophoresis and SSR assays were performed according to Yu et al. (1994) and Saghai Maroof et al. (1994). Minitab 14 (2004) was used for analysis of variance (ANOVA), significance tests, and graphical analysis. Mapmaker 3.0b (Lander et al., 1987) was used for linkage group construction at a LOD 3.0 using the "group" command with a maximum Haldance distance of 50 centiMorgans (cM). Genetic linkage maps were imported into QTL Cartographer 2.5 (Wang et al., 2007) and were used for locating putative QTL by interval mapping. An empirical LOD threshold for interval mapping (IM) was determined to be 4.8 by l000 permutation tests (Churchill and Doerge, 1994). Estimates of LOD peaks and variance explained were obtained from the Zmapqtl program run in model 3 of QTL cartographer 2.5.

Phytic Acid and Total Phosphorus Assay
Total P assays were conducted according to Gao et al. (2007). The colorimetric (Wade reagent) method (Gao et al., 2007) was used for all phytate assays with minor modifications for higher throughput. Briefly, a sample of 50 to 75 random seeds from each soybean line were ground with a Cyclone Sample Mill with a 0.5 mm mesh screen (UDY Corporation, Fort Collins, CO). Samples of 0.50 g of fine ground powder were thoroughly mixed with 10 mL of 2.4% HCl and centrifuged at 1000 g (Sorvall RT6000B, Du Pont, Newtown, CT) at 10°C for 10 min. Then, 0.5 mL of the crude extracts was transferred to 2 mL micro-centrifuge tubes and 0.5 mL 20% NaCl was added on top. The NaCl-treated crude extracts were precipitated at 4°C for 60 min and then were centrifuged at 13200 rpm on an Eppendorf microcentrifuge for 15 min (Eppendorf 5415D centrifuge, Hamburg, Germany). Subsequently, 120 µL of clear supernatants were added to 14-mL Falcon tubes containing 2.88 mL double-distilled (dd) H2O. For color development, 1 mL Wade reagent (0.03% FeCl3·6H2O + 0.3% sulfosalicylic acid) was added to the 3 mL diluted clear supernatant. Absorbance of color reaction products was read at 500 nm on a DU 640 spectrophotometer (Beckman Coulter, Fullerton, CA) with dd H2O as blanks. Calibration standards were prepared using phytic acid sodium salt from corn (Sigma, St. Louis, MO) containing 0, 1.12, 2.24, 3.36, 5.6, 7.84, and 11.2 mg L–1 phytate-P and the same concentration of NaCl and HCl as samples. The final phytate content of each sample (mg phytate g–1 Dwt.) = 3.5484 x 2 x 10–3 x Y (mg L–1) x 25 x 10 (ml) g–1 Dwt., where Y was the sample concentration calculated from the regression between the decrease of absorbance at 500 nm and the concentrations in the calibration standards; 3.5484 was the conversion factor between phytate P (186 g per mole of phytate) and phytate (MW 660); 2, 25, and 10 were dilution factors. CX1834-1-6 was included as the reference sample for every 36 samples and all samples from the same population were assayed in the same run.

Evaluation of Seedling Emergence of LP Lines
Lines varying in their phytate levels were included in seedling emergence tests in the field, including 16 control genotypes with natural genetic variation for phytate content ranging from 8- 18 mg g–1, the five LP (ranging from 7–8 mg g–1) and five HP (ranging from 19–21 mg g–1) extremes phenotypically selected from the F2:5 QTL validation population, and the 21 LP BC lines (designated as S-1 to S-13 and S-15 to S-22) as identified by markers linked to LP QTLs. Emergence was evaluated at Blacksburg, VA in 2006 and at both Blacksburg and Warsaw, VA in 2007. All emergence tests were conducted in a randomized complete block (RCB) design and planted in a density of 75 seeds per 3.05-m length row with 0.76 m between rows. CX1834-1-6, M766, Essex, Hutcheson, Lee 68, V71-370, V99-3337, V99-5089, PI 87013, PI 96983, and PI 200508, L-1 to 5 and H-1 to 5, S-1 to S-13, and S-15 to S-22 were included in the 2006 trails. All seeds used in the 2006 tests were from 2005 planting at Blacksburg, VA except for that V99-3337 and Lee 68 were from 2005 planting at Warsaw, VA. All of the above mentioned lines were evaluated again in 2007, using bulk seeds harvested from the 2006 evaluation plots. Five additional genotypes, Williams, Mn1401, V97-7158, V99-8060, and R95-1705, were also included for emergence in 2007. Seeds used for planting these five genotypes in 2007 also came from 2006 planting at Blacksburg, VA. For both years, the control genotypes, LP and HP lines were planted as one row per plot and a total of three plots/replications per line at each location. In 2006, each of the 21 LP BC lines was planted as four adjacent rows per plot and one plot/replication per line as a separate experiment in Blacksburg, VA. In 2007, seeds harvested from these 21 LP BC lines (S-1 to S-13 and S-15 to S-22) in 2006 were planted as one row per plot and a total of six plots/replications for each line at both Blacksburg and Warsaw, VA. The 2006 plots at Blacksburg and 2007 plots at Warsaw were planted by hand and the 2007 plots at Blacksburg were planted by machine. The soil at Blacksburg, VA is a Hayter loam (fine-loamy, mixed, active, mesic Ultic Hapludalf); at Warsaw, VA, the soil is a Kempsville Loam (fine-loamy, siliceous, thermic Typic Hapludult). Seedling emergence data were collected at the V2–V4 stage as defined by Fehr and Caviness (1977). Emergence percentage was calculated by counting the number of plants in a plot, then dividing it by the number of seeds planted and multiplying by 100. Minitab 14 (2004) was used for significance tests and graphical analysis.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Validation of LP QTL
CX1834-1-6 (8.6 ± 0.4 mg g–1) was found in the low phytate range while V99-3337 (13.3 ± 0.8 mg g–1) was below the average of normal- to high-phytate genotypes (Table 1) of all soybean lines (6.2 to 17.9 mg g–1) assayed for phytate content in this study. About 50% of total P in CX1834-1-6 was in the form of phytate P. In contrast, 60% of total P in V99-3337 was phytate P and the majority of control soybean genotypes had 70 to 80% of total P as phytate P (Table 1) agreeing with Raboy et al. (1984). Other LP lines found in this study include V99-5089 (9.9 ± 0.5 mg g–1) (Table 1) and a set of BC lines derived from CX1834 x V99-3337 by marker-based selection (5.9 to 8.8 mg g–1) (Table 2 ). Seed phytate content for inbred lines and homozygous genotypes was stable over different years (data not shown) also agreeing with previous reports (Israel et al., 2006).


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

  		Table 2. Phytate content and seedling emergence of (CX1834 x V99-3337) x V99-3337 backcross progenies derived from a BC1F2 plant that was identified to contain homozygous CX1834 alleles at Satt237, Satt339, Satt527, and Satt561 marker loci.

 
Analysis of variance (ANOVA) using data from the validation population detected significant (P < 0.001) association of LG N (Satt387, Satt236, Satt521, Satt339, Satt237, Sat091, and Satt257) and L markers (Satt156, Satt448, Satt527, Satt561, Satt166, and Sat099) with variation for phytate content. According to single-factor ANOVA, individuals with a homozygous CX1834 (13.6 to 14.8 mg g–1) allele at any one of these loci had significantly (P < 0.01) lower mean phytate than those with homozygous V99-3337 (16.1 to 17.2 mg g–1) or heterozygous (15.1 to 15.6 mg g–1) alleles at the same loci. Two LGs were constructed with Mapmaker 3.0b (Fig. 1 ) based on the above SSR marker data. Putative QTL locations were identified by interval mapping using QTL Cartographer 2.5 (Wang et al., 2007). One LP QTL was detected on LG L near Satt527/Satt561, contributing 12% of the total phytate variation (Fig. 1) in the population. Another LP QTL was detected on LG N near Satt237/Satt339, accounting for 28% of the total variation in phytate content (Fig. 1). Map positions and order of the eight LG L markers were in agreement with Song et al. (2004). Satt339 and Satt237 were reversed from the order reported by Song et al. (2004) on LG N in the CX1834-1-6 x V99-3337 population, but the order agreed with maps from Cregan et al. (1999) and Walker et al. (2006). Overall, the LP QTLs detected through direct mapping of the phytate phenotype in this study were located in the same map intervals as the two high-inorganic phosphorus QTLs detected by Walker et al. (2006). Interactions between LGs L and N were detected by ANOVA in this QTL validation population, accounting for about 12.9% of total phytate variation, further agreeing with Walker et al. (2006). However, the phenotypic variation explained by the two QTLs detected in this study were 28% by Satt237/Satt339 and 12% by Satt527/Satt561, compared to the respective 41 and 11% as reported by the previous study. Differences in genetic backgrounds, types of mapping populations, and phytate assay procedures employed may have contributed to the differences in the QTL effects between the two studies for the LG N QTL. The V99-3337 parent had a medium-low level of phytate (13.3 ± 0.8 mg/g) (Table 1) suggesting that minor LP QTLs could also exist in V99-3337 at unknown loci inherited from Hutcheson (13.4 ± 0.2 mg/g). Possibly these different LP QTLs from both CX1834-1-6 and V99-3337 parents may have been combined in our validation population. Additional contribution from V99-3337 LP loci to the total phytate variation in this validation population may explain the different QTL effect from Walker et al. (2006). In their mapping population, CX1834 is reported as the sole LP source contributing to phytate variation (Walker et al., 2006). In addition, phytate levels were directly assayed in the current study while measurements of Pi were used to estimate the association of phytate phenotype and LP QTLs by Walker et al. (2006). Although an inverse relationship between Pi and phytate contents exists (Raboy et al., 2001; Wilcox et al., 2000), the correlation is not always high, but depends rather on total P content differences among samples (Gao et al., 2007; Kwanyuen and Burton, 2005). Use of Pi content as an indicator of phytate content also was found to be valid only at certain Pi levels (Gao et al., 2007). Therefore, it is expected that the direct and indirect phytate measurements would produce some differences in the phytate-marker association analysis. Regardless of the differences detected with LG N QTL effect, the two LP QTLs identified by Walker et al. (2006) near Satt527/Satt561 and Satt237/Satt339 on LGs L and N, respectively, were validated in this study. However, phytate variations in this validation population cannot be solely explained by the two LP QTLs in CX1834. Existence of additional LP QTLs other than the two in CX1834 was suggested by strong transgressive segregation for both HP and LP extremes in the validation population (Fig. 2 ), which did not appear in the mapping population of Walker et al. (2006). Approximately 5% of the validation population had phytate content less than 8 mg g–1, which was significantly lower than that of CX1834-1-6 (8.58 ± 0.43 mg g–1), and above 70% had phytate content greater than that of V99-3337 (13.27 ± 0.77 mg g–1) (Fig. 2). Similar results were also found with different combinations of marker genotypes at the two LP QTLs (Fig. 3 ). Some individuals with homozygous CX1834 alleles (cc) at both LG N (Satt 339) and LG L (Satt527) had significantly lower phytate content than CX1834. Likewise, some individuals with homozygous V99-3337 allele (vv) at both of the two LP QTLs corresponded to significantly (P = 0.01) higher mean phytate levels than that of V99-3337. The unexpected high number of HP transgressive segregants in the validation population could have been contributed by the recombination of CX1834 and V99-3337 HP alleles at multiple loci in some progenies, the genetic effects of which could have been masked by LP gene(s) in their original parental backgrounds. Likewise, additional minor LP genes in V99-3337 may have combined with the CX1834 LP genes leading to additional reduction of phytate content and hence LP transgressive segregants. Possibly, additional LP QTLs other than the known QTLs in CX1834 (Walker et al., 2006) and LR33 (Hitz et al., 2002) exist in soybean as natural genetic variation for phytate (Gao et al., 2007; Israel et al., 2006; Raboy et al., 1984) and other genomic regions governing soybean phytate biosynthesis (Yuan et al., 2007) have been reported. To verify the hypothesis of the existence of additional LP QTLs, we are currently mapping additional markers for the entire genome in populations derived from two sources of LP lines, V99-5089 and CX1834. In addition, results observed in this study suggest that the intermediate phytate line, V99-3337, as inherited from Hutcheson could also be utilized to develop LP soybean lines.


Figure 1
View larger version (14K):
[in this window]
[in a new window]

  		Figure 1. Phytate content QTL position and logarithm of the odds (LOD) score plots from interval mapping (IM) of linkage groups L and N in the QTL validation population CX1834-1-6 x V99-3337. The dashed line parallel to the vertical axis indicates the LOD threshold of 4.8. Values along linkage maps, shown in centiMorgans, are map distances between adjacent markers (not to scale).

 

Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		Figure 2. Frequency distribution of seed phytate content in 208 F2:5 families of CX1834-1-6 x V99-3337.

 

Figure 3
View larger version (22K):
[in this window]
[in a new window]

  		Figure 3. Phytate levels for various marker genotypic classes based on Satt527 (LG L) and Satt339 (N) markers in the CX1834-1-6 x V99-3337 population (cc = homozygous alleles from CX1834-1-6; vv = homozygous alleles from V99-3337; cv = heterozygous alleles from both CX1834-1-6 and V99-3337. Bars represent two standard errors. Letters A, B...D on top of bars indicate significant phytate content differences at a 0.01 level of significance between the designated classes.

 
Identification of Low-Phytate Lines with LP QTLs
Double homozygous CX1834 alleles at the two LP QTLs (based on markers Satt 237, Satt339, Satt527, and Satt561) were present in one BC1F2 plant, 516-4, out of a total of 69. Several BC1F4 lines derived from this plant and their BC1F5 progenies (S-1–S-22) had phytate content (6–9 mg g–1) similar to or significantly lower than that of CX1834 (8–9 mg g–1) (Table 2), which agree with phytate levels previously detected in similar LP progenies derived from CX1834 (Oltmans et al., 2005). Therefore, in this study, marker-based selection at the early generation (BC1F2) worked perfectly in identifying LP lines with phytate levels equivalent to or lower than that of CX1834. Scaboo et al. (2005) used Satt237 and Satt561 markers to select for LP soybean lines among 187 F5:7 recombinant inbred lines derived from CX1834. They analyzed the association of the two marker genotypes with phytate phenotypes as measured by an inorganic P visual assay for each corresponding line. However, according to Scaboo et al. (2005) only 23 and 21% of their selections were true high-inorganic phosphorus or LP types when MAS was based on homozygous CX1834 alleles at only one of the Satt237 (LG N) and Satt561 (LG L) marker loci, respectively. Their MAS efficiency increased to 50% when both loci were simultaneously used for selection. Scaboo et al. (2005) subsequently ascribed their results to potential existence of additional LP QTLs on genomic regions other than LGs L and N. Our findings discussed above agree with Scaboo et al. (2005) that additional minor LP QTLs probably exist. However, markers closely linked to LP QTLs on LGs L and N are still considered to contribute to the major genetic effect of LP lines derived from CX1834. To ensure successful MAS, Satt237, Satt339, Satt527, and Satt561 in combination may be required as recombination between Satt339/Satt237 and between Satt527/Satt561 can occur but a double recombination would be extremely low. Given the relatively low efficiency of using four markers to select for a single trait, markers more closely linked to LP genes are needed, and should be easier to develop as the draft soybean genome sequence is currently available.

Association between Phytate Levels and LP QTLs
To investigate associations of phytate phenotypes with the two LP QTLs, the five lowest phytate (L-1 to L-5) and five highest phytate lines (H-1 to H-5) were selected from the F2:5 families of CX1834 x V99-3337 and subsequently genotyped at the two LP QTLs for Satt237, Satt339, Satt527, and Satt561 marker loci using bulked DNA samples from 20 random plants of each line (Table 3 ). Mean phytate content of L-1 to -5 (7.6 ± 0.4) was significantly lower than CX1834 (8.6 ± 0.4) and the mean of H-1 to -5 (19.53 ± 0.59) was significantly higher than V99-3337 (13.3 ± 0.8). All LP selections, L-1 to L-5, were found to have a 100% correspondence with the homozygous CX1834 marker allele (cc) at both LP QTLs (Table 3). The five HP selections, H-1 to H-5, however, were found to carry homozygous V99-3337 allele (vv) only at Satt237 and Satt339 on LG N. On LG L, three of the five HP selections (H-1, -2, and -4) corresponded to the CX1834 genotype and two (H-3 and-5) contained heterozygous marker alleles. As all 10 LP and HP selections were based on phytate phenotypic assays, selections of HP transgressive extremes have possibly resulted in a selection of genes for HP that are located at completely different loci from the two known LP QTLs. These new HP loci could be located on LG L linked to CX1834 allele at Satt527/Satt561 loci. The potential linkage found between HP and LP genes on LG L in CX1834 in this study will not impede the effectiveness of using markers closely linked to both LP QTLs to select for LP soybeans derived from CX1834, as demonstrated in this study. However, this finding would explain why homozygous CX1834 alleles at both pha1pha1 and pha2pha2 are needed for an LP phenotype (Oltmans et al., 2004). It would also provide evidence for the assumption of Walker et al. (2006) that an LP QTL on LG L has existed in ancestors of M153-1-4, but reduced phytate phenotype in M153-1-4 did not appear until a single gene mutation occurred at the QTL on LG N. Further genetic characterization of these HP and LP selections along with their respective parents and ancestors will address questions of the origins of the two LP QTLs in CX1834, the number of gene mutations involved in M153-1-4, and the utility of natural genetic variation for phytate on LG L.


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

  		Table 3. Phytate content, marker genotypes, and seedling emergence of 10 F2:5 lines selected as five lowest- (L-1 to -5) and five highest-phytate (H-1 to -5) extremes of the CX1834-1-6 x V99-3337 QTL validation population.

 
Seedling Emergence of LP Lines
Variations for seedling emergence was found to exist among normal phytate lines, and varied from 43 ± 8% in PI 87013 to 87 ± 4% in Essex in 2006 at Blacksburg and from 21 ± 7% in Mn1401 to 66 ± 11% in Williams in 2007 at Warsaw (Table 1 and Fig. 4A ). LP lines did not have uniformly low emergence, and varied from 14 ± 12% in L-5 to 81 ± 17% in L-1 (Table 3 and Fig. 4B) and from 12 to 87% in the 21 LP BC lines (Table 2 and Fig. 4C) in 2006. Similar ranges of emergence variations were also reported by Hulke et al. (2004) among similar LP BC lines. Under one environment (in 2007 at Blacksburg), no significant difference in emergence was observed between HP and LP lines, with an average of 86% (Fig. 4). While another environment (in 2007 at Warsaw), HP selections as a whole performed better than LP selections (Fig. 4B) agreeing with Hulke et al. (2004) and Oltmans et al. (2005). However, similar or even greater decreases of emergence were observed in some normal phytate genotypes (Mn1401, Hutcheson, PI 87013, PI 96983, and V71-370) compared to the LP lines (CX1834, V99-5089, M766, L-1 to -5 and some BC lines). Therefore, overall our data suggests that LP genotypes may not be the only causal factor for reduced seedling emergence. Instead, reduced emergence of some LP lines could be associated with genotype by environment interactions. Some LP F2:5 and BC selections used in this study such as L-1 and S-10 and S-15 had significantly lower phytate content than that of the LP parent, CX1834, (Tables 2 and 3) but had similar emergence to that of V99-3337 and the average of all HP genotypes in Table 1 in 2006 at Blacksburg. Therefore, agreeing with Spear and Fehr (2007), inbreeding or backcrossing to the HP parent could improve seedling emergence of LP lines derived from CX1834. However, improved emergence of these LP breeding lines did not appear to be stably inherited in this study. To further advance understanding of impacts of LP on emergence and dissect environment and genotype interaction on emergence, LP lines with normal emergence in some environments in this study need to be evaluated in a broader range of environments in the future, particularly under stress conditions.


Figure 4
View larger version (39K):
[in this window]
[in a new window]

  		Figure 4. Seedling emergence variation for different soybean genotypes at two Virginia locations (Blacksburg, and Warsaw) in 2006 and 2007. (A) Seedling emergence of control genotypes with phytate content varying from 8 to 18 mg g–1. (B) Seedling emergence (%) of five low- (L-1 to L-5, phytate content ranging from 7–8 mg g–1) and five high-phytate (H-1 to H-5, phytate content ranging from 19–21 mg g–1) F2:5 lines from CX1834-1-6 x V99-3337 population. Data presented are means ± SE of three replications. (C) Seedling emergence (%) of 21 (S-1 to S-13 and S-15 to S-22) backcross lines, with phytate levels ≤ CX1834 and containing homozygous alleles at the Satt237/Satt339 and Satt527/Satt561 marker loci at the BC1F2 generation of V99-3337 x (CX1834-1-6 x V99-3337). Emergence data presented for 2006 are means of four adjacent rows as one replication and those for 2007 are means of six replications.

 
In summary, we have validated the two low-phytate QTLs responsible for the low-phytate phenotype in CX1834 as previously identified by Walker et al. (2006). However, transgressive segregation for both HP and LP extremes occurred in the current validation population, presumably because of the contribution of additional minor LP and HP QTLs by V99-3337 and CX1834, respectively. Genome-wide marker analysis is needed to map these additional phytate QTLs. Existing variation for LP was observed in commonly used soybean germplasm in this study and should be further explored to potentially augment the currently available LP mutants. Progenies derived from marker-based selection for CX1834 genotypes at the two LP QTLs at the BC1F2 generation all contained phytate levels lower than or similar to CX1834 in the current study. All individuals phenotypically selected as LP at the F2:5 generation corresponded to homozygous CX1834 alleles at both of the LP QTLs. In this study, a potential linkage between some unknown HP genes with the LP QTL on LG L in CX1834 was observed, which may provide an explanation for why homozygous alleles at both pha1pha1 and pha2pha2 are needed for LP phenotype. Two years of seedling emergence tests demonstrated that regardless of the sources of LP, either natural genetic variation or mutation, LP lines do not have uniformly low emergence, and under some environments, no difference is observed between HP and LP lines. Inheritance of improved emergence in LP lines needs to be evaluated over a broad range of environments.


   ACKNOWLEDGMENTS
 
The study reported here was supported by the United Soybean Board.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication November 19, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 The Plant GenomeHome page
J. D. Gillman, V. R. Pantalone, and K. Bilyeu
The Low Phytic Acid Phenotype in Soybean Line CX1834 Is Due to Mutations in Two Homologs of the Maize Low Phytic Acid Gene
The Plant Genome, July 1, 2009; 2(2): 179 - 190.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
M.A. S. Maroof, N. M. Glover, R. M. Biyashev, G. R. Buss, and E. A. Grabau
Genetic Basis of the Low-Phytate Trait in the Soybean Line CX1834
Crop Sci., January 28, 2009; 49(1): 69 - 76.
[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 Gao, Y.
Right arrow Articles by Buss, G.R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gao, Y.
Right arrow Articles by Buss, G.R.
Agricola
Right arrow Articles by Gao, Y.
Right arrow Articles by Buss, G.R.
Related Collections
Right arrow Soybean
Right arrow Phosphorus
Right arrow Crop Genetics

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