Crop Science 41:63-67 (2001)
© 2001 Crop Science Society of America
CROP BREEDING, GENETICS & CYTOLOGY
Family and Line Selection for Reduced Palmitate, Saturates, and Linolenate of Soybean
Leon G. Streita,
Walter R. Fehrb,
Grace A. Welkeb,
Earl G. Hammondb and
Silvia R. Cianziob
a Dep. of Research and Product Development, Pioneer Hi-Bred International, Inc., Johnston, IA 50131
b Dep. of Food Science and Human Nutrition, Iowa State Univ., Ames, IA 50011
Corresponding author (wfehr{at}iastate.edu)
 |
ABSTRACT
|
|---|
Development of soybean [Glycine max (L.) Merr.] cultivars with reduced palmitate and stearate will lower the total saturated fatty ester content of the seed oil, and reduction of linolenate will improve its oxidative stability. The objective of this study was to compare the family and line methods of selection for reduced palmitate, palmitate + stearate (saturates), and linolenate in four populations segregating for the major alleles fap1 and fap3 for reduced palmitate or the fan1(A5) and fan2 for reduced linolenate. Four random F3-derived lines from 21 F2 families from each population were evaluated in a plant-row-yield test in 1995 and replicated trials at four locations in 1996. For the family method, the mean palmitate, saturates, and linolenate of the four F3-derived lines of each F2 family was used to identify families from which to select individual lines. For the line method, lines were selected without regard to the family structure. The fatty ester contents of the selected and unselected lines based on data from one environment were compared with their mean fatty ester contents in the other environments. The number of lines selected for each of the traits by the family method was less than for the line method in all populations. There was a greater percentage of lines incorrectly rejected by the family method than by the line method. For development of cultivars with reduced palmitate, saturates, and linolenate, breeding methods that rely on family performance would not be more effective or efficient than methods that ignore family structure.
|
INTRODUCTION
|
|---|
COMBINING OIL QUALITY TRAITS in soybean can improve the nutritional and functional quality of the oil. A reduction in palmitate and stearate would enhance the nutritional quality by lowering total saturated fatty esters, and reduction of linolenate would improve the oxidative stability of the oil (Dutton et al., 1951; Smouse, 1979; Mounts et al., 1988; White and Miller, 1988). Soybean genotypes with 
40 g kg-1 palmitate have been developed by combining the fap1 and fap3 alleles (Erickson et al., 1988; Fehr et al., 1991; Schnebly et al., 1994). Genotypes with 25 g kg-1 linolenate were obtained by combining the fan1(A5) and fan2 alleles for reduced linolenate (Hammond and Fehr, 1983; Fehr and Hammond, 1996). Although there are major genes for reduced palmitate and linolenate, the traits can be considered quantitatively inherited due to environmental effects and the influence of modifying genes (Graef et al., 1988; Fehr et al., 1992; Horejsi et al., 1994; Ndzana et al., 1994; Schnebly et al., 1994; Rebetzke et al., 1998). Stearate content also is considered a quantitative trait because no major genes have been reported for reduction of the fatty ester.
Falconer (1960) suggested that family selection is favored when the heritability of a trait is low and the number of families is large. For any quantitatively inherited trait, selection can be conducted among and within families during inbreeding by the pedigree or early generation testing methods (Fehr, 1987). Alternatively, one or more seeds from selected plants can be bulked and selection practiced among lines without regard to family structure, as would be done with the single-seed-descent and bulk methods. Breeding methods that involve family selection require more record keeping, labor, and land than methods based solely on line selection (Fehr, 1987).
The only comparison of family and line selection for altered fatty ester content in soybean was made by Bravo et al. (1999). They concluded that breeding methods that rely on family performance would not be more efficient or effective than methods that ignore family structure for the development of cultivars with elevated palmitate. The objective of this study was to compare the family and line methods of selection for reduced palmitate, saturates, and linolenate in soybean.
|
MATERIALS AND METHODS
|
|---|
Four populations were developed for this study. The parent lines YA7343Z006 and AX8154A370 had the fap1 fap1 fap3 fap3 genotype for reduced palmitate, and the parent cultivars 9282 and 9322 had the fan1(A5) fan1(A5) fan2 fan2 genotype for reduced linolenate. Each of the parents was crossed to F2:3 lines selected from three populations that had palmitate of 
71 g kg-1 and linolenate of 30 g kg-1. The F2:3 lines were assumed to have the genotype fap1 fap1 fap3 fap3 fan1(A5) fan1(A5) fan2 fan2 based on their contents of palmitate and linolenate.
The crosses to form the four populations were made in March 1994 at the Iowa State University–University of Puerto Rico soybean breeding nursery at Isabela, PR. The soil type was Coto clay (very-fine, koalinitic, isohyperthermic, Typic Haplorthox). The crosses of F2:3 lines with 9282 were collectively designated AX11056, with 9322 were AX11063, with YA7343Z006 were AX11080, and with AX8154A370 were AX11104. The F1 seeds were planted in Puerto Rico in May 1994, and each F1 plant was harvested and threshed individually. Five F2 seeds from each F1 plant were analyzed by gas chromatography, as described by Hammond (1991), to verify from the segregation for fatty ester content that each plant was a hybrid. Within each cross, F2 seeds from confirmed hybrid F1 plants were bulked.
A total of 700 F2 seeds from each population were cut into two parts with a razor blade, and the portion lacking the embryonic axis was analyzed for fatty ester content during August 1994. In October 1994, 264 F2 seeds for AX11056, 186 F2 seeds for AX11063, 160 F2 seeds for AX11080, and 97 F2 seeds for AX11104 were planted in Puerto Rico in 102-cm-wide rows at 20 seeds m-1 of row. All seeds planted had <80 g kg-1 saturates (palmitate + stearate) and <30 g kg-1 linolenate. The F2 plants were harvested individually, and a bulk of five F3 seeds from each F2 plant was analyzed for fatty ester content. The 50 F2 plants with the least saturates and linolenate contents from each population were selected. There were 12 seeds of each F2:3 line planted in Puerto Rico in January 1995 in 61-cm-long rows spaced 102 cm apart. Five random F3 plants from each line were harvested individually.
The five F3:4 lines of each F2 family were grown in a plant-row-yield test at Johnston, IA, in 1995. The soil type is a Waukegan loam (fine-loamy, over sandy or sandy skeletal, mixed, superactive, mesic Typic Hapludoll). A plot was a single row 108 cm long, with a 77-cm spacing between rows and a 92-cm alley between the end of plots. The seeding rate was 33 seeds m-1 of row. Each F3:4 line was harvested individually in bulk, and a sample of seven random seeds was analyzed by gas chromatography to determine fatty ester composition.
For the replicated tests in 1996, four F3:5 lines from 21 of the 50 F2 families were randomly chosen from each population. The 84 lines of each population were grown as a separate experiment. The 84 lines of each experiment were subdivided into 21 blocks, with each block consisting of the four F3-derived lines from an F2 family. Blocks and entries within blocks were randomized in each replication. The four experiments were planted as a randomized complete-block design with two replications at Ames, Atlantic, and Washington, IA, and Bethany, MO, in 1996. The soil types are a Nicollet loam (fine-loamy, mixed, mesic Aquic Hapludoll) at Ames, a Marshall silt loam (fine-silty, mixed, superactive, mesic Typic Hapludoll) at Atlantic, a Mahaska silty clay loam (fine, smetitic, mesic Aquertic Argiudoll) at Washington, and a Haig silt loam (fine, smetitic, mesic Vertic Argiaquoll) at Bethany. A plot consisted of paired rows 3.7 m long, with a 77-cm row spacing and a 92-cm alley between the end of plots. The seeding rate was 31 seeds per m-1 of row. After harvest, the fatty ester content of each plot was determined by gas chromatography on a seven-seed bulk sample.
For comparison of the family and line methods, selection was practiced independently for 38 g kg-1 palmitate, 70 g kg-1 saturates, and 35 g kg-1 linolenate, and for the combination of 70 g kg-1 saturates and 35 g kg-1 linolenate. The criteria for palmitate and linolenate were chosen so that 50% of the lines would be selected when averaged across populations and selection environments. The use of 70 g kg-1 saturates was based on the approximate content of palmitate and stearate that could be in an oil that would meet the standard established by the Food and Drug Administration for an oil that could be labeled as low in saturated fat (U.S. Food and Drug Administration, 1994).
For the family method, the mean content of the four F3-derived lines within a F2 family was determined. Within families that met the selection criterion, lines that met the selection criterion were selected. For the line method, individual lines that met the criterion were selected without regard to the family performance.
Selection for each of the four traits was conducted independently in each of the four environments based on the two individual replications and the mean of the two replications. The performance of selected and unselected lines in one environment was compared with their appropriate mean fatty ester concentration in the other three environments. Acceptance and rejection errors were calculated for both methods of selection. Acceptance error occurred when lines were chosen on the basis of one or two replications of testing in one environment, but the lines did not meet the selection criterion based on the mean of the other three environments. Rejection error occurred when lines were not chosen because they did not meet the selection criterion based on one or two replications of testing in the selection environment, but the lines met the criterion based on the mean of the other three environments.
The data from each experiment were analyzed as a randomized complete-block design for individual environments and across environments. All variables in the analysis of variance were considered random effects. The analyses of variance were performed by the general linear models procedure of the SAS software package (release 6.12) (SAS Institute, 1992). Variance components, heritability estimates, and their standard errors for each population were calculated from the combined analyses of variance across environments (Hallauer and Miranda, 1995).
|
RESULTS AND DISCUSSION
|
|---|
There were significant (P < 0.05) differences among lines for palmitate, saturate, and linolenate in the four populations (Tables 1 and 2). The differences among families were significant (P < 0.05) in the four populations for linolenate and in all of the populations, except AX11104, for palmitate and saturates. The variation among lines with the fap1 fap1 fap3 fap3 genotype for palmitate and the fan1(A5) fan1(A5) fan2 fan2 genotype for linolenate was attributed to the action of modifying genes (Graef et al., 1988; Fehr et al., 1992; Horejsi et al., 1994; Schnebly et al., 1994; Walker et al., 1998). The variation in stearate was attributed to multiple genes with small effects.
View this table:
[in this window]
[in a new window]
|
Table 1. Means and ranges of F2 families and F3-derived lines for palmitate, saturates, and linolenate contents for four soybean populations averaged across four environments in 1996
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Variance component and heritability estimates and their standard errors for palmitate, saturates, and linolenate contents of 84 F3-derived soybean lines from four populations evaluated in four environments
|
|
The heritability estimates on a plot basis averaged across populations were 0.44 for palmitate, 0.43 for saturates, and 0.48 for linolenate (Table 2). The lower heritability estimates for AX11056 and AX11080 were associated with less variability among genotypes for the three traits than in the other two populations.
The number of lines selected for each of the traits by the family method was less than for the line method in all populations because a line could not be chosen if it was in a family that did not meet the selection criterion (Tables 3 to 6). The average percentage of lines selected for palmitate was 62% by the family and 70% by the line method, for saturates was 45% by the family and 56% by the line method, for linolenate was 51% by the family and 60% by the line method, and for both saturates and linolenate was 21% by the family and 30% by the line method. The lower percentage of lines selected for both saturates and linolenate compared with single trait selection indicated that combining the two traits would reduce the frequency of acceptable lines in a segregating population.
View this table:
[in this window]
[in a new window]
|
Table 3. Errors associated with selection by the family and line methods for 38 g kg-1 palmitate content among four F3-derived soybean lines from each of 21 families in four populations averaged across four selection environments
|
|
View this table:
[in this window]
[in a new window]
|
Table 6. Errors associated with selection by the family and line methods for 70 g kg-1 saturates content among 35 g kg-1 linolenate content among four F3-derived soybean lines from each of 21 families in four populations averaged across four selection environments
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Errors associated with selection by the family and line methods for 70 g kg-1 saturates content among four F3-derived soybean lines from each of 21 families in four populations averaged across four selection environments
|
|
View this table:
[in this window]
[in a new window]
|
Table 5. Errors associated with selection by the family and line methods for 35 g kg-1 linolenate content among four F3-derived soybean lines from each of 21 families in four populations averaged across four selection environments
|
|
Even though fewer lines were chosen by the family method, the acceptance error was similar for the two selection methods (Tables 3 to 6). The average percentage of lines selected without error for palmitate was 80% by the family and 79% by the line method, for saturates was 66% by the family and 63% by the line method, for linolenate was 70% by the family and 72% by the line method, and for selection of both saturates and linolenate was 42% by the family and 40% by the line method. The greater acceptance error associated with selection for both saturates and linolenate indicated that selection for the two traits would be less reliable than for the individual traits.
The frequency of rejection error was higher for the family than for the line method (Tables 3 to 6). The average percentage of lines incorrectly rejected for palmitate was 33% by the family and 24% by the line method, for saturates was 45% by the family and 35% by the line method, for linolenate was 44% by the family and 33% by the line method, and for selection of both saturates and linolenate was 70% by the family and 58% by the line method. The greater rejection error for the family method reflected the limitation that a line could not be selected if its family exceeded the selection criterion. The greater rejection error for selection of both saturates and linolenate further reflected the lower reliability of selection than for the individual traits.
The reliability of selection based on one replication was similar to that of selection based on the mean of two replications (Tables 3– 6). Averaged across selection methods, populations, traits, and selection environments, the use of one replication resulted in 50% of the lines being selected, 36% acceptance error, and 42% rejection error. The use of two-replication means resulted in 49% of the lines being selected, 36% acceptance error, and 43% rejection error. Use of a single replication at a location should suffice for the evaluation of palmitate, saturates, and linolenate.
The study indicated no advantage for maintaining family structure in the selection of lines with reduced palmitate, saturates, linolenate, or both saturates and linolenate. Use of F2 family means lowered the number of lines selected, did not reduce acceptance error, and increased the rejection error. Our results were similar to those reported by Bravo et al. (1999) who indicated that the single-seed-descent or bulk methods would be more efficient than the early generation testing or pedigree methods for selection of lines with elevated palmitate. The single-seed-descent and bulk methods are less expensive to conduct, require less land and labor, and require less record keeping than the early generation testing or pedigree methods (Fehr, 1987).
Although the development of cultivars with both reduced saturates and linolenate would be possible, combining the two traits would lower the effectiveness of selection compared with that for one of the traits. An additional complication in selecting for the two traits is the development of populations with an acceptable number of segregates. To recover an adequate number of lines with reduced saturates and linolenate for this study, it was necessary to cross lines with the fap1 fap1 fap3 fap3 fan1(A5) fan1(A5) fan2 fan2 genotype for reduced saturates and linolenate to a parent with the Fap1 Fap1 Fap3 Fap3 fan1(A5) fan1(A5) fan2 fan2 genotype for normal saturates and reduced linolenate or a parent with the fap1 fap1 fap3 fap3 Fan1(A5) Fan1(A5) Fan2 Fan2 genotype for reduced saturates and normal linolenate. In the breeding program at Iowa State University, crosses of lines with both reduced saturates and linolenate to parents with normal saturates and linolenate produced <1% of acceptable lines for the two traits because of the number of major and minor genes that were segregating for the two traits. When high-yielding conventional cultivars are not used to form single-cross populations because of the expense associated with screening large numbers of segregates, it will be difficult to develop cultivars with reduced saturates and linolenate that have comparable yield to cultivars with conventional fatty ester concentrations.
|
NOTES
|
|---|
Journal Paper No. J-18789 of the Iowa Agric. and Home Econ. Exp. Stn., Ames; Projects No. 2799 and 3107 and supported by the Hatch Act, State of Iowa, and Pioneer Hi-Bred International, Inc.
Received for publication March 6, 2000.
|
REFERENCES
|
|---|
- Bravo, J.J., W.R. Fehr, G.A. Welke, E.G. Hammond, and S.R. Cianzio. 1999. Family and line selection for elevated palmitate in soybean. Crop Sci. 39:679–682.[Abstract/Free Full Text]
- Dutton, H.J., C.R. Lancaster, C.D. Evans, and I.C. Cowan. 1951. The flavor problem of soybean oil: VIII. Linolenic acid. J. Am. Oil Chem. Soc. 28:115–118.[ISI]
- Erickson, E.A., J.R. Wilcox, and J.F. Cavins. 1988. Inheritance of altered palmitic acid in two soybean mutants. J. Hered. 79:465–468.[Abstract/Free Full Text]
- Falconer, D.S. 1960. Introduction to quantitative genetics. Ronald Press Co., New York.
- Fehr, W.R. 1987. Principles of cultivar development: Theory and technique. Macmillian Publ., New York.
- Fehr, W.R., and E.G. Hammond. 1996. Soybean having low linolenic acid content and method of production. U.S. Patent Number 5534425. Date issued: 9 July.
- Fehr, W.R., G.A. Welke, E.G. Hammond, D.N. Duvick, and S.R. Cianzio. 1991. Inheritance of reduced palmitic acid content in seed oil of soybean. Crop Sci. 31:88–89.
- Fehr, W.R., G.A. Welke, E.G. Hammond, D.N. Duvick, and S.R. Cianzio. 1992. Inheritance of reduced linolenic acid content in soybean genotypes A16 and A17. Crop Sci. 32:903–906.[Abstract/Free Full Text]
- Graef, G.L., W.R. Fehr, L.A. Miller, E.G. Hammond, and S.R. Cianzio. 1988. Inheritance of fatty acid composition in a soybean mutant with low linolenic acid. Crop Sci. 28:55–58.[Abstract/Free Full Text]
- Hallauer, A.R., and J.B. Miranda. 1995. Quantitative genetics in maize breeding. Iowa State Univ. Press, Ames, IA.
- Hammond, E.G. 1991. Organization of rapid analysis of lipids in many individual plants. p. 321–330. In H.F. Linskens and J.F. Jackson (ed.) Modern methods of plant analysis. Vol. 12. Springer-Verlag, Berlin.
- Hammond, E.G., and W.R. Fehr. 1983. Registration of A5 germplasm line of soybean. Crop Sci. 23:192.[Free Full Text]
- Horejsi, T.F., W.R Fehr, G.A. Welke, D.N. Duvick, E.G. Hammond, and S.R. Cianzio. 1994. Genetic control of reduced palmitate content in soybean. Crop Sci. 34:331–334.[Abstract/Free Full Text]
- Mounts, T.L., K. Warner, G.R. List, R. Kleiman, W.R. Fehr, E.G. Hammond, and J.R. Wilcox. 1988. Effect of altered fatty acid composition on soybean oil stability. J. Am. Oil Chem. Soc. 65:624–628.[ISI]
- Ndzana, X., W.R. Fehr, G.A. Welke, E.G. Hammond, D.N. Duvick, and S.R. Cianzio. 1994. Influence of reduced palmitate content on agronomic and seed traits of soybean. Crop Sci. 34:646–649.[Abstract/Free Full Text]
- Rebetzke, G.J., J.W. Burton, T.E. Carter, Jr., and R.F. Wilson. 1998. Genetic variation for modifiers controlling reduced saturated fatty acid content in soybean. Crop Sci. 38:303–308.[Abstract/Free Full Text]
- SAS Institute. 1992. SAS guide for personal computers. 6th ed. SAS Inst., Cary, NC.
- Schnebly, S.R., W.R. Fehr, G.A. Welke, E.G. Hammond, and D.N. Duvick. 1994. Inheritance of reduced and elevated palmitate in mutant lines of soybean. Crop Sci. 34:829–833.[Abstract/Free Full Text]
- Smouse, T.H. 1979. A review of soybean oil reversion flavor. J. Am. Oil Chem. Soc. 56:747A–751A.
- U.S. Food and Drug Administration. 1994. Code of federal regulations. Title 21, Volume 2, Parts 100–169. 21CFR101.75, p. 127–129. U.S. Gov. Print Office, Washington, DC.
- Walker, J.B., W.R. Fehr, G.A. Welke, E.G. Hammond, D.N. Duvick, and S.R. Cianzio. 1998. Reduced-linolenate content associations with agronomic and seed traits of soybean. Crop Sci. 38:352–355.[Abstract/Free Full Text]
- White, P.J., and L.A. Miller. 1988. Oxidative stabilities of low-linolenate, high-stearate, and common soybean oils. J. Am. Oil Chem. Soc. 65:1334–1338.
This article has been cited by other articles:
|
|
|
|
 
W. R. Fehr
Breeding for Modified Fatty Acid Composition in Soybean
Crop Sci.,
December 18, 2007;
47(Supplement_3):
S-72 - S-87.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. G. Streit, W. R. Fehr, and G. A. Welke
Family and Line Selection for Seed Yield of Soybean
Crop Sci.,
March 1, 2001;
41(2):
358 - 362.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
