| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Biochemistry, 109 Life Sci. Ctr., Univ. of Missouri, Columbia, MO 65211-7700
b Dep. of Statistics, Iowa State Univ., Ames, IA 50011
c Agronomy Dep., Iowa State Univ., Ames, IA 50011
d USDA-ARS, Corn Insect and Crop Genetics Research Unit, Iowa State Univ., Ames, IA 50011
* Corresponding author (charlsond{at}missouri.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: Hb2, Broad-sense heritability • IDC, iron-deficiency chlorosis • MAS, marker-assisted selection • MLG, molecular linkage group • QTL, quantitative trait loci • RFLP, restriction fragment length polymorphism • SSR, simple sequence repeat
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The best method of preventing IDC is to plant IDC-resistant cultivars (Fehr, 1982). Although resistance to IDC in soybean has been improved through conventional breeding approaches (Cianzio, 1991; Cianzio and Voss, 1994; Hintz et al., 1987), IDC-resistant lines may have lower yield potential than IDC-susceptible cultivars (Fehr, 1982, 1983).
Selection for IDC resistance has been based on foliar-chlorosis symptoms observed in plantings on calcareous soils (Cianzio et al., 1979). Complex polygenic inheritance of IDC resistance and genotype x environment interactions may result in inaccurate assessment of resistance, which reduces the value of phenotypic rating and decreases efficiency of breeding to improve the trait (Lin et al., 2000b). Therefore, if an environment-independent approach could be devised, there might be the potential to improve breeding efficiency for IDC.
Quantitative trait loci for IDC resistance have been identified using restriction fragment length polymorphism (RFLP) markers in two soybean populations in field and hydroponic evaluations (Lin et al., 1997, 2000a). Lack of common RFLP markers in the two populations prompted Lin et al. (2000b) to conclude that the examined molecular markers would be inefficient in MAS. Instead, they proposed use of SSR markers due to high occurrence of polymorphisms and the availability of large number of SSR markers in the public domain.
The mapping populations used by Lin et al. (1997)(2000a) were developed by using highly IDC-susceptible genotypes crossed to highly IDC-resistant genotypes. However, there is no information about the associations of QTLs and IDC resistance in actual breeding populations, namely when the parents used in the cross possess a number of desirable agronomic traits along with moderate IDC resistance. To simulate a practical breeding program, a population was developed as described above, using parents with intermediate IDC scores and a number of desirable agronomic traits, to examine the association between QTLs and IDC using SSR markers. Preliminary results from this population, using 1-yr data, indicated a possible association between Satt481 and IDC resistance (Charlson et al., 2003). On this basis, our objective was to continue examining whether SSR markers located in previously reported IDC QTLs would be associated with IDC resistance in the breeding population.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The cross, population development, and seed increases were conducted at Isabela, PR, on noncalcareous soils. Several F1 seeds were harvested during winter 1997, which were planted during the 1998–1999 season in Puerto Rico. Seven F1 plants were grown, and the F2 seed of each F1 plant was harvested individually. While maintaining the identity of the F2 lines, 145 F2 plants resulting from the seven F1 plants were self-fertilized for two successive generations to produce 145 F2:4 lines for 2000 and 2001 field evaluations. The F2:4 lines were advanced to the F2:5 generation for 2002 field evaluations.
Phenotypic Evaluation of IDC Resistance
The parents, F2–derived lines (F2:4 and F2:5 lines), and five IDC check cultivars (BSR101, Century 84, Corsoy 79, Kenwood 94, and Williams 82) were evaluated for IDC resistance on calcareous soils at two Iowa locations, Ames and Humboldt, during the summers of 2000, 2001, and 2002. The soil at each location is described as Harps loam (fine-loamy, mixed, superactive, mesic Typic Calciaquolls) and soil pH averages 8.0. In 2000, plots were planted at Ames on 22 May and Humboldt on 23 May. In 2001, plots were planted at Ames 8 June and Humboldt on 18 June. In 2002, plots were planted at Ames 29 May and Humboldt 30 May. Each genotype was randomly assigned to a plot. Plots were 75-cm-long, single rows planted with 25 seeds at a rate of one seed per 3 cm. At each location, plots were organized in a randomized complete block design with three replications.
Evaluation of IDC resistance was conducted at the V4 stage of growth (Fehr and Caviness, 1981) by visually rating foliar chlorosis using a scale of 1.0 to 5.0 with 0.5 increments, where 1.0 represented no yellowing of the leaves and 5.0 indicated severe yellowing with necrosis and plant death (Cianzio et al., 1979). Each plot was given a chlorosis score on the basis of the average amount of chlorosis observed for all plants in the plot.
Analyses of variance were conducted with all main and interaction effects considered random. If an ANOVA indicated that significant (P 
0.05) differences existed among F2–derived lines, transgressive segregation was tested with Student's t test (
= 0.05) by comparing means of the most- and least-resistant F2–derived lines to the means of each parent. Differences between parental means also were tested with the Student's t test ( = 0.05). Broad-sense heritabilities 
were estimated from expected mean squares (Fehr, 1987).
To identify lines with superior IDC resistance, a selection-threshold value was determined by first ranking the 145 F2–derived lines on the basis of mean chlorosis scores at each location and combined over locations. The mean chlorosis score of the tenth-most resistant F2–derived line was used as a selection-threshold value. Subsequently, F2–derived lines with mean chlorosis scores equal to or less than the selection-threshold value were considered to possess superior IDC resistance.
Molecular Marker Evaluation
For each parent and F2 line, leaf samples from 10 to 15 plants were harvested at the V4 stage of growth (Fehr and Caviness, 1981). Using a method adapted from Saghai-Maroof et al. (1984), DNA was extracted from macerated, freeze-dried leaf tissue with extraction buffer [1% (w/v) CTAB (cetyltrimethylammonium bromide), 50 mM Tris (2,3-Dibromo-1-propanol phosphate), 50 mM EDTA (ethylenediamine tetra-acetic acid), 0.7 NaCl, and 1 mM 1,10 phenanthroline]. The DNA was isolated from the extraction buffer with 24 parts chloroform and 1 part isoamyl alcohol solution and precipitated with 2/3 volume of isoproponal. The precipitated DNA was rinsed with 80% (v/v) ethanol and 15 mM ammonium acetate solution. The DNA was air-dried before resuspension in TE buffer (0.01 mM Tris, 0.01 mM EDTA) with 0.003% of RNase A.
One hundred and eight SSR markers (Table 1) genetically linked to eight previously identified QTLs (MLGs A1, B1, B2, G, H, I, L, and N) for IDC (Cregan et al., 1999; Lin et al., 1997, 2000a) were tested on P9254 and A97–770012 using procedures adapted from Cregan and Quigley (1997). For each marker, polymerase chain reactions (PCRs) were conducted in 20 µL-volumes of sterile, double-distilled water containing the following reagents: 0.2 mM of dATP (5'-adenylate triphosphate), dCTP (2'-deoxycytidine-5'-triphosphate), dGTP (2'-deoxyguanosine 5'-triphosphate), and dTTP (5'-thymidylate triphosphate); 1.5 mM MgCl2; 20 mM Tris-HCl, 50 mM KCl; 0.5 U TAQ Polymerase; 45 ng genomic DNA; and 1.5 pmol of each SSR primer (forward and reverse). Forward and reverse primer sequences were obtained from SoyBase (http://www.soybase.org, verified 14 July 2005) for each SSR marker. Each reaction was subjected to touch-down PCR (Don et al., 1991) (PTC-200 DNA Engine, MJ Research, Inc.; Walthan, MA) using thermal-cycle conditions consisting of 94°C for 120 s followed by 12 cycles of 94°C for 60 s (denaturation), 52°C with a 1°C decrease for each successive cycle (range of 47 to 52°C) (annealing), and 68°C for 60 s (extension). Touch-down cycles were immediately followed by 35 cycles of 94°C for 30 s (denaturation), 47°C for 30 s (annealing), and 68°C for 30 s (extension), and followed with 68°C for 300 s. Amplification products were separated by electrophoresis on 2.5% (w/v) agarose gels incubated in buffer [40 mM Tris and 1 mM EDTA dissolved in 0.1% (v/v) glacial acetic acid] for 1.5 h at 180 V. Gels then were incubated in 0.0001% (w/v) ethidium bromide for 20 min before visualization of amplification products with ultraviolet light.
|
0.01) were eliminated from the analysis.
Since all polymorphic markers identified in our study were associated with previously reported IDC QTLs, data from these markers were combined for analysis. Significant (P 0.05) associations of genotypic classes with mean chlorosis score was determined with Type III (Partial) sums of squares in multiple-locus ANOVAs to identify SSR markers genetically linked with chlorosis expression. This approach gives the unique contribution for each marker after the correlated effects of the other markers are removed and this statistical approach reduces the rate of false positives relative to single-locus ANOVA (Snedecor and Cochran, 1989; Steel et al., 1997). Coefficients of determination (R2) were reported for each marker significantly (P 0.05) associated with chlorosis scores. For mean analysis, mean chlorosis score of selected F2–derived lines was compared with the overall mean of the 145 F2–derived lines using Dunnett's test ( = 0.05).
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
0.05) differences among genotypes, and a significant interaction (P 0.05) of location x year at individual locations and in the combined data (analysis not shown). At both locations, ranking of the five check cultivars for IDC resistance was similar to the previously established ranking (data not shown). This result indicated that foliar chlorosis expression resulted from iron deficiency.
In the overall analysis of chlorosis scores of the F2–derived lines, there were no significant differences between locations and among years (Table 2). However, the location x year interaction was significant (P 0.01), suggesting that environment is important in chlorosis expression. Within locations, years were significantly (P 0.05) different, however, the absence of a line x year interaction suggested relative differences in chlorosis scores among F2–derived lines were similar among years.
|
0.01) differences among F2–derived lines within locations and in the overall analysis (Table 2). Transgressive segregation was also observed in the population (Table 2 and Fig. 1)
. Chlorosis score means of the F2–derived lines ranged from 1.28 to 2.61 at Ames, 1.67 to 3.44 at Humboldt, and 1.64 to 2.89 for the overall data. At each location and across locations, mean chlorosis scores of the each parent were significantly ( = 0.05) different from the most-IDC susceptible line in the population at Ames, Humboldt, and overall (analysis not shown). In contrast, the parents were not different from the most-IDC resistant line at Humboldt or overall, except at Ames, where both parents differed ( = 0.05) in mean chlorosis score with the most IDC-resistant line (analysis not shown).
|
|
Selection-threshold values were 1.50 at Ames, 1.89 at Humboldt, and 1.72 for overall data (Table 3). Thirteen lines at Ames, 14 at Humboldt, and 11 lines overall expressed chlorosis scores equal to or less than the corresponding selection-thresholds. There were three lines (060, 095, and 130) identified as possessing superior IDC resistance at both locations and overall (Table 3). Only five lines at Ames and three at Humboldt were selected as having superior IDC resistance using the overall mean at each location.
Molecular Marker Evaluation
Of the 108 SSR markers genetically linked to previously identified QTLs for IDC resistance (Lin et al., 1997, 2000a), 24 markers (22.2%) were polymorphic between P9254 and A97–770012. All markers except one (Satt288) followed Mendelian segregation. Multiple-locus ANOVA indicated significant (P 0.05) F values for Ames (1.48), Humboldt (1.59), and overall data (1.78). Of the 23 markers, only three were significantly (P 0.05) associated with chlorosis scores using Ames, Humboldt, or overall data. The significant markers were Satt211 (MLG A1), Satt481 (MLG L), and Sat_104 (MLG I) (Table 4). Significant (P 0.05) R2 values for the three markers ranged from 3.9 to 11.5%, with the greatest R2 value for Satt481.
|
0.05) different only for Humboldt, where lines homozygous for P9254 allele (AA) showed the greatest IDC resistance (Table 4). For Satt481, the greatest IDC resistance was observed in lines homozygous for the A97–770012 allele (BB) at both locations and with the overall data. For Sat_104, the heterozygous (AB) genotype demonstrated the greatest IDC resistance in the overall mean. However, lines homozygous for the A97–770012 allele (BB) and heterozygous (AB) as Sat_104 demonstrated the same chlorosis scores at Humboldt. In a practical breeding situation, only the homozygotic state of Sat_104 would be useful in a self-pollinated and highly, genetically homozygous crop such as soybean.
The mean chlorosis scores of lines selected with individual markers were compared with the mean of the 145 F2–derived lines within location and overall data (Table 5). Although selection of lines homozygous for the resistant allele at Satt211 and Sat_104 had no effect on population means, F2–derived lines chosen with Satt481 alone demonstrated a significantly lower ( = 0.05) mean score of 0.11 to 0.14 units relative to the mean of the entire population of F2–derived lines.
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our study reflects the complex genetics of IDC resistance. Two genetic mechanisms for IDC resistance have been previously identified in soybean: a major gene with modifier genes and classical polygenic inheritance (Lin et al., 1997). In our population, the normal distribution in chlorosis scores observed in the F2–derived lines suggested polygenic inheritance of IDC resistance, which may explain why major QTLs for resistance were not identified in the study.
Another likely explanation for identifying only one useful marker is that the desirable resistance alleles, especially genes contributing major effects to phenotypic expression, could have become fixed in germplasm as breeders develop cultivars. Considering the parents used in our study represented elite lines with comparable IDC resistance, the possibility exists that the parents already shared favorable alleles for the major genes. Fixation of major genes may be supported by the lack of markers associated with resistance in MLG N, where a major gene for IDC resistance had been previously found, accounting for up to 80% of the total variation (Lin et al., 1997, 2000a). In our population, this major QTL was not detected, suggesting that we may be searching for modifier genes with minor effects on chlorosis expression, which would be difficult to detect. The low R2–values observed for the three significant markers suggest that these markers mapped modifier genes with minor effects on resistance.
Although previous studies have demonstrated significant importance of the environment on genotype selection (Lin et al., 1997, 2000a), this was not statistically detected in our study. Still, environmental effects were evident in our population. Of the eleven-most resistant lines selected on the basis of chlorosis score means, only three lines were selected at both locations and with the overall mean. These observations suggest the role played by environment on chlorosis expression, which would also explain the intermediate heritability values for IDC observed in our study.
Despite the fact that Satt481 made a small contribution to the overall phenotypic variation in chlorosis score means, it is interesting to note that 73% of the top eleven-most resistant lines were homozygous resistant for Satt481. This allele was contributed by the A97–770012 parent, which had been developed specifically for IDC resistance. Lastly, although the selected lines were homozygous resistant at Satt481, a normal distribution in chlorosis score means was still observed for the lines, suggesting that several unidentified genes may be involved in IDC in the population.
Usefulness of Satt481 in a practical breeding situation will also be dependent on if Satt481 could be used in other breeding populations, and if additional markers could be identified. Work is underway to identify specific genes for IDC resistance using microarray experiments (O‘Rourke and Shoemaker, 2005, personal communication). This genetic information could lead to gene-specific markers that may improve the effectiveness of MAS to increase IDC resistance in soybean.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Received for publication August 26, 2004.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Ishii and K. Yonezawa Optimization of the Marker-Based Procedures for Pyramiding Genes from Multiple Donor Lines: II. Strategies for Selecting the Objective Homozygous Plant Crop Sci., September 1, 2007; 47(5): 1878 - 1886. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
