Genetic Variability for Phytic Acid Phosphorus and ... - PubAg - USDA

Published online December 2, 2005

Genetic Variability for Phytic Acid Phosphorus and Inorganic Phosphorus in Seeds of Soybeans in Maturity Groups V, VI, and VII D. W. Israel,* P. Kwanyuen, and J. W. Burton When beginning a breeding program for low PA, it will be useful to know the concentrations of PA-P in cultivars currently in production and available natural variation for this trait. While such information is available for soybeans produced in the Corn Belt (Raboy et al., 1984), concentrations of PA are not published for soybeans that are produced in southeastern or midsouthern regions of the USA. Likewise, concentrations of Pi in those soybeans are also not published. Because Pi is much higher in seeds of the low-PA mutant, a qualitative colorimetric assay for Pi is being used in breeding programs as a rapid screen for the mutant genotype. As with PA, it is useful to know the natural variation for Pi in breeding lines. The extent to which environment and genotype may interact to affect seed Pi concentration is also not known. The objectives of the research reported here were (i) to determine the variation in PA-P and Pi concentrations found in soybean breeding lines and cultivars typical of southern soybean production, (ii) to assess the extent of genotype by environment interaction for both Pi and PA-P concentrations in soybean seeds, and (iii) to examine relations among seed concentrations of PA-P, Pi, and protein.

Reproduced from Crop Science. Published by Crop Science Society of America. All copyrights reserved.

ABSTRACT Phytic acid (PA; myo-inositol 1,2,3,4,5,6 hexakisphosphate) in soybean [Glycine max (L.) merr.] meal is a major source of P in animal excreta, a serious environmental pollutant. Genetic mutants in which seed PA is reduced by 60% have been developed. The objectives were to assess (i) natural variation in seed PA-P and inorganic phosphorus (Pi) concentrations in soybean breeding lines and cultivars of Maturity Groups (MGs) V, VI, and VII; (ii) genotype 3 environment (G 3 E) interactions for Pi and PA-P, and (iii) relations among PA-P, Pi, and seed protein concentrations. Three sets of cultivars and breeding lines were tested separately in two or three environments. Variation among lines was highly significant, ranging from 3.77 to 5.07 g kg 1 PA-P and from 0.19 to 0.37 g kg 1 Pi. The G 3 E interactions were highly significant for Pi concentration, but significant variation for PA-P concentration was observed only among cultivars, not across environments nor as G 3 E interactions. Rank correlation coefficients for Pi concentrations between environments were large (0.65–0.88), suggesting that the G 3 E interactions were due to differences in average Pi concentration in various environments. Variation in seed protein was highly significant in all three sets, but protein was not correlated with PA-P and was correlated with Pi (r 5 0.56) only in the MG V breeding lines test. Therefore, genetic relationships between protein and either PA-P or Pi could not be established. Significant natural genetic variation indicates that PA level of potential adapted parents may be useful in breeding low-PA soybeans.

M

in soybean seeds, between 68 and 78%, is stored as PA (Raboy et al., 1984). Phytic acid P is not bioavailable to either livestock or human consumers of soybeans. Thus, it is a major source of P in animal excreta and an environmental pollutant. It is a particular problem in areas where there is concentrated swine and poultry production, as both consume large quantities of protein meal. Phytic acid also binds to nutritionally important minerals, particularly Zn, making them unavailable to humans and nonruminant livestock (Thompson, 1989; Raboy et al., 1984). A low-PA soybean has been developed by mutagenesis which has a 60% reduction in PA-P and an increase in Pi compared with ‘Athow’, (Wilcox et al., 2000). This mutant is being used by soybean breeders to develop low-PA cultivars. Low-PA maize (Zea mays L.) has been shown to provide more bioavailable P for swine than regular maize (Veum et al., 2001), and it is expected that low-PA soybean will have a similar effect when fed in swine and poultry rations. UCH OF THE PHOSPHORUS

MATERIALS AND METHODS Materials used in this study were soybean cultivars and breeding lines which are a sample of the germplasm adapted to the midsouthern and southeastern states (Table 1). All were in MGs V, VI, or VII. While these materials are not exhaustive of the gene pool being used by soybean breeders in southeastern states, they do represent a genetically diverse sample of materials found in public soybean breeding programs. The MG V cultivars were ‘Hutcheson’ (Buss et al., 1988), ‘P9594’, and ‘Holladay’ (Burton et al., 1996). P9594 was developed by Pioneer HiBred International, Inc. Maturity Group V breeding lines were developed by the USDA-ARS at Raleigh, NC, and included N00–111, N00–116, N00–19, N00–322, N00– 39, N00–506, and N99–522. Lines developed elsewhere were MD97–6065, selected by William Kenworthy, University of Maryland; S99–2176 was developed by Grover Shannon, University of Missouri; and TN97–271 developed by Fred Allen and Vince Pantalone at the University of Tennessee. The MG VI cultivars tested included ‘Balbuena’, ‘Brim’ (Burton et al., 1994), ‘Satelite’, and ‘Young’ (Burton et al., 1987). Balbuena was developed in Mexico. Satelite was developed by USDA-ARS in North Carolina and publicly released in 2001. The MG VI breeding lines developed by USDAARS at Raleigh, NC, included N00–377, N00–483, N00–560, N00–634, N00–672, N99–32, and N99–813. Lines developed elsewhere included R97–1053, developed by Clay Sneller and

D.W. Israel, USDA-ARS, 3131 Williams Hall, Raleigh, NC 27695; P. Kwanyuen and J.W. Burton, USDA-ARS, 3127 Ligon Street, Raleigh, NC 27607. Received 25 Jan. 2005. *Corresponding author (dan_israel@ ncsu.edu). Published in Crop Sci. 46:67–71 (2006). Crop Breeding, Genetics & Cytology doi:10.2135/cropsci2005.0076 ª Crop Science Society of America 677 S. Segoe Rd., Madison, WI 53711 USA

Abbreviations: G 3 E, genotype 3 environment; HPLC, high performance liquid chromatography; MG, Maturity Group; PA, phytic acid; PA-P, phytic acid phosphorus; Pi, inorganic phosphorus.

67

68

CROP SCIENCE, VOL. 46, JANUARY–FEBRUARY 2006

Table 1. Breeding line and cultivar pedigrees.


Pedigrees Maturity Group V lines and cultivars MD97-6065 N00-111 N00-116 N99-19 N00-39 N00-322 N00-506 N00-522 S99-2176 TN97-271 ‘Holladay’ ‘Hutcheson’ ‘P9594’ Maturity Group VI lines and cultivars N00-377 N00-483 N00-560 N00-634 N00-672 N99-32 N99-813 R97-1053 Sc96-1624 N93-1264 ‘Balbuena’ ‘Brim’ ‘Roy’ ‘Satelite’ ‘Young’ Maturity Group VII line and cultivar N90-845 ‘Vernal’

Manokin/Holladay Dillon/N93-66 Clifford/N94-696 N90-516/N92-598 Clifford/N91-386 N91-78/N93-54 N90-516/N92-598 N92-32/K1309 S93-1591/S94-1808 N86-7687/Hutcheson N77-179/Johnston V68-1034/Essex Pioneer 9592/Asgrow 6297 Au92-916/N90-845 Clifford/N94-696 Dillon/N93-66 Centennial/N93-1170 Clifford/N91-386 N92-598/Haskell N93-1264/Holladay Pioneer 9592/Northrup King S59-60 Sc89-181/Northrup King S75-55 Brim/PI 416937 Cajeme/Tamazula S-80 Young/N82-1198 Holladay/Brim N90-2013/C1726/2/2*N88-431/3/2* Brim/4/Soyola Essex/Davis Brim/N80-777 D77-12244/Bedford

Pengyin Chen, University of Arkansas; and SC96–1624, developed by Emerson Shipe at Clemson University. Two MG VII entries were included in the tests. One was a cultivar, ‘Vernal’ (Hartwig, 1993). The other was a breeding line N90–845 (Table 1).

Experimental Field Design The MGV and MGVI breeding line tests were grown in a randomized complete block design with two replications (complete blocks) at three North Carolina locations (Clayton, Plymouth, and Clinton) in 2002. Entries were planted in threerow plots. Rows were 0.9 m wide and 6 m long. At maturity, 4.9 m was harvested from the center row. Seed yield (kg ha 1), and concentration of protein, PA, and Pi were determined for each plot. Total seed (PA-P 1 Pi) in kg ha 1 was calculated as yield 3 (PA-P 1 Pi concentration)/1000. Entries in the MGV test included the 10 breeding lines plus two cultivars, Hutcheson and Pioneer 9594. Entries in the MGVI test included the nine breeding lines. Nine cultivars plus N90–845 were also grown in a randomized complete block design with four replications at Clayton, NC, in 2002 and 2003, using plot size identical to that previously described. The tests at Clayton were grown on Plinthic Paleudult soils with P indices of 114 in 2002 and 81 in 2003. Tests at Plymouth were grown on Typic Umbraquilt soils with a P index of 94. Application of phosphatic fertilizer is not recommended when the P index is greater than 50. The soil type at the Clinton test site is not known. Here, the P index was 30 for one test and 22 for the other. Soil amendments were applied according to soil test recommendations.

Unless otherwise stated, all experiments and procedures were performed at room temperature. Soybean seed samples were ground in a centrifugal grinding mill equipped with a 24-tooth rotor and a 1.0-mm stainless steel ring sieve with the motor speed set at 15c000 rpm. This setting produced ground samples with a uniform particle size , 0.5 mm. For convenience, extraction was done in a 20-mL vial with 0.5 M HCl in a ratio of 1:20 (w/v) for 1 h while stirring. In our sample preparation, 0.5 g of sample and 10 mL of 0.5 M HCl were used throughout. Approximately 2 mL of crude extract from each sample were centrifuged at 18c000 3 g for 10 min in a microcentrifuge. An aliquot of 1-mL supernatant containing PA was then filtered with a 1-mL tuberculin syringe and a 13-mm/0.45-Am syringe filter. Filtered samples can be stored at 4jC for several days before high performance liquid chromatography (HPLC) analysis. Chromatography was performed on the binary HPLC system with a 50- by 4.6-mm PL-SAX strong anion exchange column (Polymer Laboratories, Amherst, MA) equipped with a 7.5- by 4.6-mm guard column. Elution of PA was achieved with a 30-min linear gradient of 0.01 M 1-methylpiperazine, pH 4.0 to 0.5 M NaNO3 in 0.01 M 1-methylpiperazine, pH 4.0, at a flow rate of 1 mL min 1, as previously described by Rounds and Nielsen (1993) with modifications. Wade’s color reagent (Wade and Morgan, 1955), consisting of 0.015% (w/v) FeCl3 and 0.15% (w/v) 5-sulfosalicylic acid (also at flow rate of 1 mL min 1) and PA eluted from the column, were mixed in a mixing tee with inline check valves for both eluants installed before the mixing tee to prevent backflow. The postcolumn reaction was allowed to take place in a 0.05- by 210-cm PEEK tubing at the combined flow rate of 2 mL min 1. The absorbance was monitored at 500 nm while the detector signals and/or PA peaks were processed and integrated by the chromatographic data acquisition system.

Inorganic Phosphorus Analysis Inorganic P in seed was determined by a modification of the microtitre plate assay method described by Larson et al. (2000). This involves extraction of 100 mg of dry seed ground to pass a 1-mm screen with 3.0 mL of 12.5% w/v trichloroacetic acid containing 25 mM MgCl2, reaction of extracts with Chen’s reagent (Chen et al., 1956) (1 vol. 0.02 M ammonium-molybdate, 1 vol. 10% w/v ascorbic acid, 1 vol. 3.0 M sulfuric acid, and 2 vol. distilled water) in wells of microtitre plates and determination of absorbance at 882 nm in each well using a plate reader. Total protein was measured by the Dumas reductive combustion method coupled with thermal conductivity detection.

Statistical Analysis The SAS GLM procedure (SAS Institute, 1999) was used for the statistical analysis. Line and cultivar entries (genotype) in the tests were considered fixed. Locations and years were considered to be random environments. To test for genotype effects, the Type III mean square of genotype 3 year or genotype 3 location was used as the error term. Residual mean square was used for testing the effect genotype 3 locations or genotype 3 year interaction. The SAS Corr procedure with line and cultivar means was used to provide simple correlation coefficients.

Phytic Acid Analysis

RESULTS AND DISCUSSION

Determination of PA-P content in soybean lines was performed as previously described (Kwanyuen and Burton, 2005).

There was significant variation among the MG V and VI lines in concentrations of both PA-P and Pi (Table 2).

69

ISRAEL ET AL.: PHYTIC ACID P AND INORGANIC P IN SOYBEAN SEED

Table 2. Mean squares from a multilocation (or year) ANOVA of seed inorganic phosphorus (Pi) and phytic acid phosphorus (PA-P) concentrations in two sets of breeding lines and nine cultivars.. Maturity Group V lines


Environments (E) Reps/E Genotypes (G) G3E Error

Maturity Group VI lines

Cultivars

df

Pi

PA-P

Protein

df

Pi

PA-P

Protein

df

Pi

PA-P

Protein

2 6 11 22 63

0.0380* 0.0054 0.0285** 0.0055** 0.0023

1.2695 0.6833 0.5989** 0.1055 0.0816

8764** 48 1042** 197** 58

2 6 8 15 40

0.1287** 0.0033 0.0090** 0.0018** 0.008

0.5453 1.8109 0.4525** 0.0582 0.0981

433 163 698** 86 56

1 6 8 8 48

0.0020 0.0006 0.0029 0.0014** 0.0004

0.4398 0.4238 1.2477** 0.2176* 0.0860

1165* 266 3899** 267** 52

* Significant at the 0.05 probability level. ** Significant at the 0.01 probability level. . Maturity Group V and VI tests were conducted in three locations in 1 yr, and a cultivar test was conducted in one location for 2 yr.

Table 3. Average inorganic phosphorus (Pi), phytic acid phosphorus (PA-P), and protein concentrations (g kg 1 dry weight) in seeds of two sets of soybean cultivars and breeding lines averaged across three replications at each of three locations in North Carolina. Pi

PA-P g kg

Maturity Group V lines and cultivars ‘Hutcheson’ MD97-6065 N00-111 N00-116 N00-19 N00-39 N00-322 N00-506 N99-522 S99-2176 TN97-271 ‘P9594’ LSD (0.05) Maturity Group VI lines N00-377 N00-483 N00-560 N000-634 N00-672 N99-813 N99-32 R97-1053 Sc96-1624 LSD (0.05)

Protein 1

0.37 0.29 0.23 0.36 0.26 0.31 0.26 0.21 0.19 0.35 0.26 0.26 0.073

4.87 4.54 4.30 4.80 4.11 4.99 4.57 4.47 4.36 4.71 4.80 4.47 0.32

385 369 396 393 368 385 383 375 359 392 380 383 14

0.24 0.30 0.26 0.29 0.29 0.23 0.26 0.31 0.24 0.070

4.09 4.97 5.05 4.66 4.61 4.60 4.70 5.06 4.51 0.24

384 376 390 387 389 374 365 397 386 12

soy71’ had PA-P concentrations of 4.26, 4.26, 4.51, and 4.57 g kg 1, respectively. Average concentrations of Pi were also significantly different among breeding lines (Table 2). Cultivar differences were significant each year but not combined across years. The highest concentration in breeding line tests was 0.37 g kg 1 and the lowest concentration was 0.19 g kg 1. Among the cultivars, Hutcheson had the highest concentration of Pi (0.30 g kg 1) and Brim had the lowest (0.24 g kg 1) (Table 4). Differences in total P (PA-P 1 Pi) were significant among the entries of the two line tests, but not the cultivar test (Tables 5, 6). This suggests that there are real genotypic differences in ability to either take up P from the soil, or to translocate P to the developing seed. Of the two seed P forms, only Pi concentration was significantly affected by environment (Table 2). Among MG V lines, standard deviations of means ranged from 0.04 to 0.08 g kg 1 and among MG VI lines, standard deviations of means ranged from 0.06 to 0.10 g kg 1. For cultivars, Pi concentration was not affected by environment (Table 2), and standard deviations of means ranged from 0.02 to 0.04 g kg 1. Genotype 3 environment interactions in the two groups of breeding lines were significant for Pi but not

Table 4. Average inorganic phosphorus (Pi), phytic acid phosphorus (PA-P), and protein concentration and yield and total (PA-P 1 Pi) for soybean cultivars grown at Clayton, NC, in 2002 and 2003. Cultivar

Maturity Group

Pi

PA-P

Protein

Balbeuna Brim Holladay Hutcheson Satelite Vernal Roy Young N98-845 LSD (0.05)

VI VI V V VI VII VI VI VII

0.26 0.24 0.24 0.30 0.25 0.25 0.24 0.28 0.27 ns.

g kg dry wt. 5.07 4.39 3.77 4.51 4.10 4.35 3.80 4.42 4.19 0.54

PA-P 1 Pi

Yield

1

kg ha 420 441 386 409 453 390 429 429 413 18

2199 2753 2738 2560 2319 2471 2914 2515 3403 ns

1

11.7 12.7 11.0 12.3 10.1 11.3 11.8 11.8 15.2 ns

- ns 5 not significant.

The average concentrations of PA-P for the lines were between 4.09 and 5.06 g kg 1 dry seed weight (Table 3). The cultivar with the lowest PA-P was Holladay at 3.77 g kg 1 dry weight, (Table 4), and the cultivar with the highest concentration was Balbuena at 5.07 g kg 1. Raboy et al. (1984) found similar concentrations between 3.9 and 6.45 g kg 1 among 38 Corn Belt cultivars. In that experiment, the standard cultivars at the time of publication, ‘Williams 79’, ‘Wells II’, ‘Della’, and ‘Am-

PA-P (Table 2). Rank correlation coefficients between locations for Pi were positive and large (0.65 to 0.88) for both tests. Thus, the significant genotype 3 environment interaction was mostly due to differences in magnitude of seed Pi concentrations in the various environments. Since the three test sites either had soil P levels in the excessive range or received fertilizer P when the soil P index was below 50, differences in soil P availability would not seem to account for these differences.

70

CROP SCIENCE, VOL. 46, JANUARY–FEBRUARY 2006

Table 5. Mean squares from a multilocation (or year) ANOVA of seed yield and total [phytic acid phosphorus (PA-P) 1 inorganic phosphorus (Pi)] in two sets of breeding lines. Maturity Group V.


Environment (E) Rep/E Lines Lines 3 E Error

Maturity Group VI.

Cultivars

df

Yield

PA-P 1 Pi

df

Yield

PA-P 1 Pi

df

Yield

PA-P 1 Pi

2 6 11 22 63

10 644 487** 179 124 842 699* 283 929** 121 975

308.91** 19.91 16.59* 6.24 3.86

2 6 8 15 40

5 151 994** 382 901 168 452 128 261 197 412

93.12 29.66 6.71* 2.52 5.03

1 6 8 8 48

416 753 191 091 1 026 230 389 974* 186 606

166.91* 7.02 12.37 4.70 3.39

. Maturity Group V and VI tests were conducted in three locations in 1 yr and cultivar test was conducted in one location for 2 yr. * Significant at the 0.05 probability level. ** Significant at the 0.01 probability level.

Table 6. Average seed yield and total [phytic acid phosphorus (PA-P) 1 inorganic phosphorus (Pi)] in seeds of two sets of soybean cultivars and breeding lines averaged over two replications at each of three locations in North Carolina. PA-P 1 Pi

Yield kg ha Maturity Group V cultivars and lines. ‘Hutcheson’ MD97-6065 N00-111 N00-116 N00-19 N00-39 N00-322 N00-506 N99-522 S99-2176 TN97-271 ‘P9594’ LSD (0.05) Maturity Group VI linesN00-377 N00-483 N00-560 N00-634 N00-672 N99-32 N99-813 R97-1053 Sc96-162 LSD (0.05)

1

2492 2387 2852 2321 2469 2298 2789 3049 3179 2640 2968 2965 502

13.0 11.5 13.0 11.9 10.7 12.1 13.5 14.5 14.9 13.1 15.1 14.1 2.4

2772 2604 2754 2498 2693 2888 2911 2771 2587 NS

12.8 13.9 14.3 12.6 13.2 14.2 13.7 14.8 11.5 1.6

Table 7. Simple correlation coefficients among means for seed concentrations of phytic acid phosphorus (PA-P), inorganic phosphorus (Pi), and protein. Maturity Group V Maturity Group VI Cultivars

Pi PA-P Pi PA-P Pi PA-P

PA-P

Protein

0.72**

0.56* 0.49 0.37 0.21 0.07 0.10

0.63 0.40

* Significant at the 0.05 probability level. ** Significant at the 0.01 probability level.

Raboy and Dickinson (1993) found no G 3 E interaction, but found significant environmental effects for PA-P, which they concluded was primarily due to soil P availability. In our studies, environmental effects were significant for only Pi and not PA-P (Table 2). It seems then that PA-P is very stable over environments for a given genotype, suggesting that it is under tight genetic control, and that P taken up by the plant is preferentially used for PA synthesis in soybean seeds. Simple correlation coefficients between Pi and PA-P were 0.72 and 0.63 in the MGV and MGVI tests, respec-

tively (Table 7). This contrasts with studies of the lowPA mutant, in which strong reciprocal relationships between Pi and PA-P concentrations in seeds have been observed (Larson et al., 2000; Wilcox et al., 2000). In this situation, mutations in the PA biosynthetic pathway cause large decreases in PA-P and large increases in Pi concentrations in seeds. Raboy et al. (1984) also found significant positive correlations (r 5 0.74 and 0.50) between PA-P and protein concentrations. Variation in seed protein concentration was highly significant in the three sets of materials tested in our experiments (Table 2). The correlation coefficients between PA-P and protein were positive but small, between 0.49 and 0.10 (Table 7). Correlation coefficients between protein and Pi were 0.56, 0.37, and 0.07. On the basis of these results, it is impossible to say whether any genetic relationship exists between protein and either PA-P or Pi in seeds. But given the highly significant variation among genotypes for both protein and PA-P, such a relationship seems unlikely. Consideration of the PA-P concentration of adapted parents may be useful in designing a breeding program for low-PA soybeans using a low-PA mutant. It is not known what the variation in PA signifies from a plant physiological perspective. But it is quite likely that a cultivar like Holladay or ‘Roy’, with a naturally low level of PA-P, may be the best parental choice for pedigree or backcross breeding. The physiological consequences to the seed and the plant of reducing PA-P by 60% are currently under intense investigation. In these materials, the radical shift in P storage in seeds from organic to inorganic forms may be detrimental. Some very low-PA soybean lines have reduced seedling emergence in field plantings (Hulke et al., 2004) . But it is not known at this time whether or not it is the low PA per se causing poor emergence. Given the benefits of this trait for improved soybean meal digestibility, efforts to overcome reduced emergence through breeding will continue. ACKNOWLEDGMENTS We thank the United Soybean Board for financial support of this project. We thank Drs. Pengyin Chen, William Kenworthy, Grover Shannon, Emerson Shipe, Vince Pantalone, and Pioneer HiBred Seed company for providing materials included in this research. We thank Peggy Longmire, Martin Friedrichs, and Takia Harris for their assistance in performing chemical analyses. We thank Connie Bryant for her assistance in preparation of this manuscript. We thank Earl Huie, Bobby McMillen, and Fred Farmer for their work in managing the field experiments.

ISRAEL ET AL.: PHYTIC ACID P AND INORGANIC P IN SOYBEAN SEED


REFERENCES Burton, J.W., C.A. Brim, and M.F. Young. 1987. Registration of ‘Young’ soybean. Crop Sci. 27:1093. Burton, J.W., T.E. Carter, Jr., and E.B. Huie. 1994. Registration of ‘Brim’ soybean. Crop Sci. 34:301. Burton, J.W., T.E. Carter, Jr., and E.B. Huie. 1996. Registration of ‘Holladay’ soybean. Crop Sci. 36:467. Buss, G.R., H.M. Camper, and C.W. Roane. 1988. Registration of ‘Hutcheson’ soybean. Crop Sci. 28:1024. Chen, P.S., T.Y. Toribara, and H. Warner. 1956. Microdeterminations of phosphorus. Anal. Chem. 28:1756–1758. Hartwig, E.E. 1993. Registration of ‘Vernal’ soybean. Crop Sci. 33:1101. Hulke, B.S., W.R. Fehr, and G.A. Welke. 2004. Agronomic and seed characteristics of soybean with reduced phytate and palmitate. Crop Sci. 44:2027–2031. Kwanyuen, P., and J.W. Burton. 2005. A simple and rapid procedure for PA determination in soybean and soy products. J. Am. Oil Chem. Soc. 82:81–85. Larson, S.R., J.N. Rutger, K.A. Young, and V. Raboy. 2000. Isolation and genetic mapping of a non-lethal rice (Oryza sativa L.) low phytic acid I mutation. Crop Sci. 40:1397–1405.

71

Raboy, V., and D.B. Dickinson. 1993. Phytic acid levels in seeds of Glycine max and G. soja as influenced by phosphorus status. Crop Sci. 33:1300–1305. Raboy, V., D.B. Dickinson, and F.E. Below. 1984. Variation in seed total phosphorus, phytic acid, zinc and calcium, magnesium, and protein among lines of Glycine max and G. soja. Crop Sci. 24:431–434. Rounds, M.A., and S.S. Nielsen. 1993. Anion-exchange high-performance liquid chromatography with post-column detection for the analysis of phytic acid and other inositol phosphates. J. Chromatogr. 653:148–152. SAS Institute. 1999. SAS/STAT user’s guide. v. 8. SAS Inst., Cary, NC. Thompson, L.U. 1989. Nutritional and physiological effects of phytic acid. p. 410–431. In J.E. Kinsells and W.G. Soucie (ed.) Food proteins. Am. Oil Chemist Soc., Champaign, IL. Veum, T.L., D.R. Ledoux, V. Raboy, and D.S. Ertl. 2001. Low-phytic acid corn improves nutrient utilization for growing pigs. J. Anim. Sci. 79:2873–2880. Wade, H.E., and D.M. Morgan. 1955. Fractionation of phosphates by paper ionophoresis and chromatography. Biochem. J. 60:264–270. Wilcox, J.R., G.S. Premashandra, K.A. Young, and V. Raboy. 2000. Isolation of high seed inorganic P, low-phytate soybean mutants. Crop Sci. 40:1601–1605.