HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (4) Citing Articles via Google Scholar Google Scholar Articles by Cornelious, B. K. Articles by Sneller, C. H. Search for Related Content PubMed Articles by Cornelious, B. K. Articles by Sneller, C. H. Agricola Articles by Cornelious, B. K. Articles by Sneller, C. H. Related Collections Soybean Plant Genetic Resources Crop Genetics

PLANT GENETIC RESOURCES

Yield and Molecular Diversity of Soybean Lines Derived from Crosses of Northern and Southern Elite Parents

^a OARDC, Dep. of Horticulture and Crop Science, 1680 Madison Ave., Wooster, OH 44691
^b Dep. Crop, Soil, and Environmental Science, 115 Plant Science Building, University of Arkansas, Fayetteville, AR 72701

* Corresponding author (csneller{at}uark.edu)

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
Genetic diversity is low in southern United States elite soybean [Glycine max (L.) Merr.] cultivars. Multiple sources of diversity will be required to effectively diversify this gene pool. The objective in this study was to evaluate the genetic diversity and yield of familiess derived from crosses between northern elite (NE) by southern elite (SE) parents. Lines were derived from 10 crosses of NE x SE parents. Molecular markers were used to estimate genetic distance between each line and its SE parent. Yield and agronomic traits were measured in field trials from 1997 to 1999 in six of the crosses. The association of diversity with line yield, expressed relative to yield of the SE parent was determined with regression. On average, the use of NE parents reduced yield, relative to using other SE parents. Some crosses and NE parents were better than others and produced families with yield that exceeded that of their SE parent, indicating that some genes from the NE parents were superior to the genes in the SE parent. At least one line with yield either superior or similar to their SE parent was found in each cross. The finding of positive transgressive segregants in some crosses and the results of the regression analyses indicate that most of the NE parents posses some yield genes that are likely to be superior to those of the SE parents. Our approach to selecting for diversity and yield may be applicable to large introgression programs where diversity from many sources is desired.

Abbreviations: NE, northern elite • SE, southern elite • GD, genetic distance

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
SUCCESSFUL CROP IMPROVEMENT depends on genetic variability that arises from genetic diversity. The predominant use of selected elite parents in soybean cultivar breeding has lead to a narrow genetic base of commercial soybean in the United States. Recent elite public and proprietary U.S. soybean cultivars derived 90% of their parentage from only 11 ancestors (Sneller, 1994), while only 22 ancestors account for 90% of the parentage of the 258 public U.S. cultivars released from 1947 to 1988 (Gizlice et al., 1994). Gizlice et al. (1993) estimated that cultivars released in the public domain since 1983 have 50% more genes in common than public cultivars released prior to 1954.

There are strong patterns of diversity within the commercial U.S. gene pool. Elite cultivars adapted to the northern United States are quite distinct from southern cultivars (Delanney et al., 1983; Gizlice et al. 1993; Sneller, 1994). Sneller (1994) reported that the average coefficient of parentage was 0.23 among northern elite (NE) U.S. families 0.26 among southern elite (SE) U.S. cultivars, but only 0.10 between NE and SE cultivars. There appears to be less diversity among SE cultivars than among NE families. Nearly 47% of the SE parentage derives from two ancestors, CNS and S100 (Sneller, 1994; Gizlice et al., 1993).

Molecular analyses of diversity in US soybean support the conclusions of the pedigree analyses. RFLP (Keim et al., 1992; Sneller et al., 1997; Kisha et al., 1998) and SSR (Nelson et al., 1998) analyses have shown the clear separation of northern and southern cultivars and the limited diversity in the southern gene pool. In addition, RFLP analyses indicate that our current elite pool is less diverse than the ancestral pool (Kisha et al., 1998).

A lack of genetic diversity may limit breeding progress, though genetic gain in yield is still being made in U.S. soybean. Ininda et al. (1996) found the greatest mean yield and gain from selection in a population derived from many NE parents, as compared with populations comprised of 25 to 100% diverse parentage. Their results indicate that there is still sufficient diversity among a large set of NE cultivars to maintain yield progress. This may be due to the substantial diversity among the main ancestors of NE cultivars (Kisha et al., 1998). There is also evidence that a lack of diversity may limit gain from selection. Populations derived from biparental crosses of diverse northern parents were more likely to have high genetic variance for yield than were populations derived from crosses of parents that are more related (Kisha et al., 1997; Helms et al., 1997). Both studies indicate that sufficient genetic variation for yield may be obtained from some elite x elite crosses, but that limited genetic diversity between parents renders many crosses useless.

The situation in the southern United States may differ from that in the northern United States. There is less diversity among SE cultivars, the major ancestors of SE cultivars are more related to one another (Kisha et al., 1998), and southern breeders have used a more restricted set of parents in their breeding history than in the north. While gain from selection appears to be increasing in the north, it may be decreasing in the south: from 17 to 2 kg ha^-1 per year from the 1980s to the 1990s (Zhou et al., 1998). Manjarrez-Sandoval et al. (1997) predicted that there would be little or no genetic variance for yield in populations derived from southern parents whose coefficient of parentage (CP) exceeds 0.27. The average CP among maturity group IV and V elite families in 1989 to 1991 was 0.31 (Sneller, 1994).

It appears that multiple pools of diversity will need to be sampled to maximize diversity within the elite U.S. gene pool (Kisha et al., 1998; Nelson et al., 1998; Thompson et al., 1998). Pedigree and molecular analyses show that NE cultivars are genetically diverse from SE cultivars. Estimated yield potential of NE families in a southern genetic background, environment, and maturity indicate that most have yield potential that exceeds most plant introductions, and that some have yield potential approaching that of some SE cultivars (Sneller et al., 1997). The combination of diversity and yield potential should make NE cultivars a good source of useful genetic diversity for southern breeders. The objective of our study was to evaluate the genetic diversity and yield of families derived from crosses of NE families by SE soybean families.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
Crosses were made between eight SE parents and six NE parents to create ten populations (Table 1). Two of the NE parents have 50% of their parentage derived from plant introductions: HS89-3261 = LG82-8379(PI68.508 x FC04.007B) x A2943; P6906-016 = (PI80.471 x PI86.050) x (Williams 79 x A3127). Single pod descent was used to develop F₄ populations under selection for southern maturity. Populations of F_4:5 lines were grown and F_4:6 lines selected for good agronomic value, adapted maturity, and low lodging and shattering. Yield tests were conducted on 140 F_4:6 lines in 1997. From these, 43 lines were selected and tested as F_4:7 lines in 1998, and 18 of these were selected and tested as F_4:8 lines in 1999.

One hundred eighty-six dye-tagged simple sequence repeat (SSR) primer pairs from across the genome (Cregan et al., 1999) were used to assess polymorphism between the parents. Lines from different crosses were screened with different sets of primers on the basis of parental polymorphism. Polymerase chain reaction mixes contained 5 µL of 10 ng genomic DNA, 2.5 mM Mg⁺², 0.15 mM of 3' and 5' end primers that were polymorphic between both parents, 0.4mM dNTP's, 1 X polymerase chain reaction buffer containing 50 mM KCL, 10 mM Tris-HCl pH 9.0, and 0.07 units of Taq DNA polymerase in a total volume of 10 ml. Thermal cycling consisted of an initial 2 min denaturation at 94°C, 25 s denaturation at 94°C, 25 s annealing at 47°C, and 25 s elongation at 72°C for 32 cycles with a 2 min final extension at 72°C with the Hybaid OMN-E thermocycler (Hybaid Limited, Teddington, Middlesex, UK). The amplification products (10 mL/lane) were separated on a standard 6% polyacrylamide gel (19:1 crosslinking ratio) and 0.5 X TAE (Tris-acetate EDTA) running buffer, at 300 volts constant power for 100 min. The gel was stained with SYBR^© Green I nucleic acid gel stain (FMC Bio-Products, Rockland, ME) for 10 min under dark conditions. Gels were then visualized under UV light and photographed with Polaroid 667 film. For each SSR marker, a line was given a score of 0 or 1 if it was homozygous for its SE or NE parent allele, respectively. Heterogeneous lines were given a score of 0.5.

A core set of 35 randomly amplified polymorphic DNA (RAPD) primers (Thompson and Nelson, 1998) was also used to assess the diversity of lines generated from each population. Previously isolated DNA was used in a 25 µl reaction mix. Each reaction contained 5 µl of 10 ng template DNA, 2.0 mM dNTP's, 0.104 mM primer, 0.13 units of AmpliTaq, 2.0 mM MgCl₂, and 1X reaction buffer. A mineral oil overlay was used to prevent evaporation during polymerase chain reaction. Thermal cycling consisted of an initial denaturation for 2 min at 94°C, 45 s denaturation at 45°C, 45 s annealing at 38°C, 2 min elongation at 72°C, and a final extension period of 7 min at 72°C with the Hybaid OMN-E thermocycler (Hybaid Limited, Teddington, Middlesex, UK). Amplified fragments were separated on 1% agarose gels for 8 h at 80 volts with 0.5 X TBE (Tris-Borate-EDTA) buffer, stained with ethidium bromide (3, 8-diamino-5-ethyl-6-phenylphenanthridinium bromide), and visualized under ultraviolet light. Families with a marker genotype that was the same as their NE parent were scored as 1 and lines with a genotype that was the same as their SE parent were scored as 0.

The SSR and RAPD marker data were combined to calculate a genetic distance (GD) between each line and its SE parent. The GD between any line and its SE parent was calculated as

where X_iF is the score of the line for the ith marker and n is the total number of markers scored.

Each line was yield tested at two locations in 1997 and at four locations in 1998 and 1999 (Table 2). Lines were tested in six trials in 1997, two trials in 1998, and one trial in 1999 (Table 1). Not all 1997 trials were conducted at the same location (Table 2). Lines from the same cross were always tested in the same trial along with their SE parent and controls ‘Hutcheson’ and ‘Manokin’. A randomized complete block design with two replications was used in each trial. Conventional tillage, planting dates, herbicide, and irrigation treatments were used in all trials. Number of rows, row spacing, and soil type varied by location (Table 2). All plots were 6.1 m long and end trimmed to a final length of 4.9 m before harvest with a plot combine of the two or three middle rows of the four or five row plots, respectively, to obtain seed yield. Data on lodging (1 = upright, 5 = prostrate), shattering (1 = all pods intact, 5 = all pods open), stem termination, maturity (days past August 31st when 95% of pods attained their mature color), and height (cm from the ground to tip of the mainstem) were recorded.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
Eight crosses segregated for stem termination and indeterminate and determinate lines were selected from seven of these while only determinate lines were selected from the other. Only indeterminate lines were derived from the other two crosses. In 1997, 42% of the lines were determinate and 58% of the lines were indeterminate. Because of their indeterminate growth habit, many lines were taller than their SE parent. The average yield was 3265 kg ha^-1 for indeterminate and 3243 kg ha^-1 for determinate lines in 1997. The yield difference between determinate and indeterminate lines in individual crosses ranged from -252 to 264 kg ha^-1. None of these differences were significant (P > 0.05), indicating that the indeterminate growth habit of many of the lines did not bias the estimate of their yield potential in the south where most elite cultivars are determinate. Too few lines were tested per cross in 1998 or 1999 to adequately compare the yield of the indeterminate and determinate lines.

All lines in this study were selected to have a maturity between mid maturity group IV and VI controls. Average maturity of lines ranged from 24 to 43 d after 31 Aug. 1997. These averages were within the range for mid maturity group IV to maturity group VI genotypes, based on maturity of the SE parents and controls. Lodging and shattering had acceptable ratings both years.

Average yield of lines from the 10 populations ranged from 2752 to 3506 kg ha^-1 (Table 3). Yields were compared with Hutcheson to estimate the absolute yield potential of the lines. Hutcheson is a high-yielding, widely grown cultivar in the southern United States and has been a very successful parent in many yield improvement programs. On average, the SE x NE populations yielded 500 kg ha^-1 less than Hutcheson. For three crosses, the average yield of the lines was not significantly less than Hutcheson (Table 3).

For each cross, the mean yield of the lines was less than the yield of the SE parent, though the yield difference was not significant for five crosses (Table 3). Across all populations, 46% of the lines from the NE x SE crosses had yields within 90% or greater of their SE parent. The best crosses were 9641 x NK S19-90, P9501 x P6906-016, and Coker 6955 x IA2007 with over 75% of their lines yielding within 90% of their SE parent (Table 3). These data indicate that high-yielding lines can be generated at high frequencies from certain NE x SE crosses.

Assuming additive gene action, average line yield can be used to predict the yield of the NE parent, as it estimates the mid-parent value. The estimated yield of the NE parents, if they had a southern maturity and were grown in these southern environments, ranged from 1997 kg ha^-1 for A2234 to 3328 kg ha^-1 for NK S19-90 (Table 3). Even the best NE parents had estimated yields lower than the SE parents or Hutcheson.

Although SE x SE crosses yield better than the NE x SE crosses, the 1997 data showed that there is potential to produce lines with good yield from NE x SE crosses. Families from the NE parents NK S19-90 and IA2007 yielded better than lines from the NE parent A2234 as well as other NE parents.

Forty-three of 141 lines tested in 1997 were selected on the basis of their yield potential for testing in 1998 (Table 3). Six environments were used to estimate line means from the 1997–1998 analysis, thereby providing reliable estimates of yield for individual lines. Due to selection for yield in 1997, nearly 80% of the lines had yield 90% or greater than their SE parent. Almost 19% of the lines had yields that were numerically, though not significantly, greater than their SE parent. All lines yielded less than Hutcheson. As noted in the 1997 analysis, some crosses were better than others. Notably, NE parents NK S19-90 and IA2007 produced several lines with yields greater than their SE parent, which also yielded well. The NE parent P6906-16 also produced several lines with yields greater than their SE parent, though in this cross, the SE parent did not yield well (Table 3).

On the basis of relative yield, 18 lines were selected for further testing in 1999, providing 10 testing environments for these lines from 1997 to 1999. Four of the 18 lines had significantly greater yield than their SE parent (Table 4) and only one line yielded significantly less than its SE parent. Thus, true transgressive segregants were detected showing that some NE parents had yield genes that were superior to those in the SE parents to which they were crossed. Overall, only four of 141 original lines (2.8%) had a confirmed transgressive phenotype. But 12.5% of the lines from the 9641 x NK S19-90 cross had a confirmed transgressive phenotype, indicating that such lines can be readily obtained from some NE x SE crosses.

Regression analysis was conducted on 1997 yield data only because too few lines were evaluated in 1997–1998 to perform a meaningful analysis. The slope was negative in all crosses, indicating that the average effect of substituting a SE gene with a NE gene was to decrease yield. The regression was significant in five of the six crosses (Table 5). The exception was the cross of 9641 x NK S19-90 where the slope predicts that NK S19-90 would yield nearly as well as 9641 if it had a southern maturity.

In each cross, there were lines with yield that approached or exceeded the yield of the SE parent and all of these lines had large positive deviations from the regression (Table 5). The GD of the lines with the maximum positive deviation from the regression ranged from 0.41 to 0.69. The GD of these lines indicates that the high yield of these lines was not due to a fortuitous reassembly of the southern genome. Rather, the yield of these lines apparently derives from a combination of favorable alleles from the NE and SE parents. Eleven transgressive segregants were identified in 1997 and six of these had yields greater than their SE parents in the more precise 1997–1999 analysis. The transgressive lines must have positive yield genes from their NE parent and all had large positive deviations from the regression, supporting our hypothesis that lines with large positive deviations from the regression are likely to possess favorable yield genes from the diverse parent. Non-transgressive segregant lines with large positive deviations from the regression line were also noted and are also likely to possess favorable yield genes from their NE parents.

On average, the use of NE as parents reduced yield, relative to using other SE as parents. Still, every NE parent produced at least one line with yield similar to their SE parent (Table 3). The finding of positive transgressive segregants (Table 4) and the results of the regression analyses (Table 5) indicate that most of the NE parents possess some yield genes that appear superior to the yield genes of the SE parent with which they were crossed. In particular, the NE parents NK S19-90 and IA2007 appeared to possess beneficial yield genes, even when compared with high-yielding SE parents. Pedigree (Sneller, 1994) and molecular analyses (Kisha et al., 1998) show these two cultivars to be more diverse from the SE gene pool than the average NE line. Both derive considerable ancestry from PI257435, an ancestor that does not contribute to the SE pool (Sneller, 1994). The high yielding, genetically diverse lines derived from NK S19-90 and IA2007 may be quite useful in diversifying the SE gene pool. Other potentially useful lines were derived from the NE parents HS89-3261 and P6906-16 (Table 4) that derive one half their parentage from plant introductions that are novel to the SE ancestry.

Our approach to identifying useful diversity in elite x diverse crosses may have general utility to breeding. With minimal resources, we were able to identify lines from six sources that may be quite useful in increasing diversity for yield based on their good combination of yield and diversity. While from some crosses the value of the NE derived germplasm is confirmed by its trangressive nature, in others it is inferred from large positive deviations from an expected negative linear relationship of yield and GD. This approach may help to identify useful germplasm in populations that are unlikely to produce transgressive segregants either because they are too small or the yield potential of the diverse parent is very low. As the approach requires fewer lines and markers than mapping of quantitative trait loci to identify lines that are likely to posses beneficial yield genes from the diverse parents, more sources of diversity can be exploited. This is quite important for elite soybean breeding programs as molecular studies indicate that useful diversity must be acquired from many sources to create an elite gene pool that spreads over the entire genetic space, as defined by molecular markers, of Glycine max (L.) Merr. In addition, the lines selected from these investigations would make ideal parents to generate mapping populations to assign value to the diverse chromosome segments that have been introgressed.

Received for publication June 22, 2000.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 

• Cregan, P.B., T. Jarvik, A.L. Bush, R.C. Shoemaker, K.G. Lark, A.L. Kahler, N. Kaya, T.T. VanToai, D.G. Lohnes, J. Chung, and J.E. Specht. 1999. An integrated genetic linkage map of the soybean genome. Crop Sci. 39:1464–1490.[Abstract/Free Full Text]
• Delannay, X., D.M. Rodgers, and R.G. Palmer. 1983. Relative contribution among ancestral lines to North American soybean cultivars. Crop Sci. 23:944–949.[Abstract/Free Full Text]
• Gizlice, Z., T.E. Carter, Jr., and J.W. Burton. 1993. Genetic diversity in North American soybean: I. Multivariate analysis of founding stock and relation to coefficient of parentage. Crop Sci. 33:614–620.[Abstract/Free Full Text]
• Gizlice, Z., T.E. Carter, Jr., and J.W. Burton. 1994. Genetic base for North American public soybean cultivars released between 1947 and 1988. Crop Sci. 34:1143–1151.[Abstract/Free Full Text]
• Helms, T., J. Orf, G. Vallad, and P. McClean. 1997. Genetic variance, coefficient of parentage, and genetic distance of six soybean populations. Theor. Appl. Genet. 94:20–26.
• Ininda, J., W.R. Fehr, S. Cianzio, and S. Schnebly. 1996. Genetic gain in soybean populations with different percentages of plant introduction parentage. Crop Sci. 36:1470–1472.[Abstract/Free Full Text]
• Keim, P., W. Beavis, J. Schupp, and R. Freestone. 1992. Evaluation of soybean RFLP marker diversity in adapted germplasm. Theor. Appl. Genet. 85:205–212.[ISI]
• Keim, P., R.C. Shoemaker, and R.G. Palmer. 1989. Restriction fragment length polymorphism diversity in soybean. Theor. Appl. Genet. 77:786–792.
• Kisha, T., B.W. Diers, J.M. Hoyt, and C.H. Sneller. 1998. Genetic diversity among soybean plant introductions and north American germplasm. Crop Sci. 38:1669–1680.[Abstract/Free Full Text]
• Kisha, T., C.H. Sneller, and B.W. Diers. 1997. The relation of genetic distance and genetic variance in populations of soybean. Crop Sci. 37:1317–1325.[Abstract/Free Full Text]
• Manjarrez-Sandoval, P., T.E. Carter, Jr., D.M. Webb, and J.W. Burton. 1997. RFLP genetic similarity and coefficient of parentage as genetic variance predictors for soybean yield. Crop Sci. 37:698–703.[Abstract/Free Full Text]
• Nelson, R.L., P. Cregan, H.R. Boerma, T.E. Carter, C.V. Quigley, M.M. Welsh, J. Alvernaz, and R. Mian. 1998. DNA marker diversity among modern North American Chinese and Japanese soybean cultivars. p. 164. In Agronomy Abstracts, ASA, Madison, WI.
• Sneller, C.H. 1994. Pedigree analysis of elite soybean lines. Crop Sci. 34:1515–1522.[Abstract/Free Full Text]
• Sneller, C.H., J. Miles, and J.M. Hoyt. 1997. Agronomic performance of soybean plant introduction and their genetic similarity to elite lines. Crop Sci. 37:1595–1600.[Abstract/Free Full Text]
• Thompson, J.A., and R.L. Nelson. 1998. Core set of primers to evaluate genetic diversity in soybean. Crop Sci. 38:1356–1362.[Abstract/Free Full Text]
• Thompson, J.A., R.L. Nelson, and L.O. Vodkin. 1998. Identification of diverse soybean germplams using RAPD markers. Crop Sci. 38:1348–1355.[Abstract/Free Full Text]
• Zhou, X., T.E. Carter, R. Nelson, H.R. Boerma, and B.R. McCollum. 1998. Breeding progress vs. breeding effort: the road ahead in soybean variety development. p. 81. In Agronomy Abstracts, ASA, Madison, WI.

This article has been cited by other articles:

				Y.-B. Fu, G. W. Peterson, and M. J. Morrison Genetic Diversity of Canadian Soybean Cultivars and Exotic Germplasm Revealed by Simple Sequence Repeat Markers Crop Sci., September 1, 2007; 47(5): 1947 - 1954. [Abstract] [Full Text] [PDF]

				J. Yu and R. Bernardo Changes in Genetic Variance during Advanced Cycle Breeding in Maize Crop Sci., March 1, 2004; 44(2): 405 - 410. [Abstract] [Full Text] [PDF]

				C. H. Sneller Impact of Transgenic Genotypes and Subdivision on Diversity within Elite North American Soybean Germplasm Crop Sci., January 1, 2003; 43(1): 409 - 414. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (4) Citing Articles via Google Scholar Google Scholar Articles by Cornelious, B. K. Articles by Sneller, C. H. Search for Related Content PubMed Articles by Cornelious, B. K. Articles by Sneller, C. H. Agricola Articles by Cornelious, B. K. Articles by Sneller, C. H. Related Collections Soybean Plant Genetic Resources Crop Genetics