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


     

Published online 1 July 2008
Published in Crop Sci 48:1341-1349 (2008)
© 2008 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, B.
Right arrow Articles by Ishibashi, T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Zhang, B.
Right arrow Articles by Ishibashi, T.
Agricola
Right arrow Articles by Zhang, B.
Right arrow Articles by Ishibashi, T.
Related Collections
Right arrow Soybean
Right arrow Cell Biology & Molecular Genetics
Right arrow Crop Genetics

Quantitative Trait Loci Mapping of Seed Hardness in Soybean

Bo Zhanga, Pengyin Chena,*, Charles Y. Chenb, Dechun Wangc, Ainong Shia, Anfu Houa and Tetsuaki Ishibashia

a Dep. of Crop, Soil, and Environmental Sciences, 115 Plant Science Building, Univ. of Arkansas, Fayetteville, AR 72701
b USDA-ARS, National Peanut Research Lab., Dawson, GA 39842
c Dep. of Crop and Soil Sciences, A384-E Plant and Soil Science Building, Michigan State Univ., East Lansing, MI 48824

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


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Soybean [Glycine max (L.) Merr.] seeds with undesirable texture cause processing complications in soyfood production. Seed hardness is an important quality attribute for food-grade soybeans. The objective of this study was to identify quantitative trait loci (QTLs) associated with seed hardness in soybean. Three generations of F2–derived lines (159 F2:3, F2:4, and F2:5 lines) from a soft (‘SS-516’) x hard (‘Camp’) soybean cross were grown in a replicated test in Fayetteville, AR, in 2003, 2004, and 2005. Pressure-cooked samples from each line were tested for seed hardness using a texture analyzer. A total of 874 simple sequence repeat (SSR) markers were used to screen the parents; 177 out of 236 polymorphic markers between the parents showed polymorphism in the F2:3 lines. A linkage map for seed hardness was established using 148 SSR markers, 15 of which were new and added to the current public soybean genetic linkage map. All identified markers were placed on 19 linkage groups (LGs) and covered 1363.7 cM of the soybean genome with an average distance of 9.6 cM between markers. Broad-sense heritability was estimated to be 0.56 for seed hardness. Two stable QTLs across environments (Ha1 and Ha2, p < 0.00001) were identified near Satt229 on LG L and Satt531 on LG D1a, respectively, for the average seed hardness over 3 yr. Ha1 had a logarithmic odds score of 6.17 with R2 = 12.7%; Ha2 had a score of 5.08 with R2 = 36.1%. A dominance-by-dominance interaction was detected between Ha1 and Ha2, explaining 7.9% of the phenotypic variance. Research is under way to confirm the identified QTLs for soybean seed hardness in multiple populations with different genetic backgrounds.

Abbreviations: CIM, composite-interval mapping • MAS, marker-assisted selection • MIM, multiple-interval mapping • QTL, quantitative trait locus • RFLP, restriction fragment length polymorphism • SMA, single-marker analysis • SSR, simple sequence repeat


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
SOYBEAN [Glycine max (L.) Merr.] not only is an important protein source for livestock and poultry but is also used for edible vegetable and industrial oil in the United States. (Rao et al., 2002). The demand for various soyfoods has been increasing due to their nutriceutical and phamaceutical implications such as reducing blood serum cholesterol and the risk of cardiovascular diseases (Physicians Laboratory, 2006). Soyfood has been consumed in Asia for over 1000 yr and is becoming increasingly popular around the globe. Therefore, breeding specialty soybeans for the soyfood market may be pivotal to the global soybean industry.

Food-grade soybeans are grouped into two major classes on the basis of soyfood manufacture. One is for whole-soy foods including natto, vegetable soybeans, and soy nuts; the other is for cracked-soy foods such as tofu, miso (fermented soup-base paste), soymilk, and soy sauce. Natto is a typical and popular Japanese fermented soyfood made of small seeds with soft and sticky texture, and it has higher isoflavone (genistein and genistin) content than soymilk or tofu (Fukutake et al., 1996). Vegetable soybean, also called edamame in Japan and mao dou in China, is a green large-seeded soybean with a sweet, nutty flavor, and usually harvested at the R6 stage (Fehr et al., 1971). The important quality properties of vegetable soybean are seed color, texture, and size (Mbuvi and Litchfield, 1995). Soy nuts are roasted soybean seeds requiring soft texture after roasting and a low percentage of antinutritional factors such as trypsin inhibitors (Chauhan et al., 2003). Apparently, seed texture is an important factor in sensory evaluation of seed characteristics for whole-soy foods consumption. Seed texture is also important for cracked-soy foods because the texture of raw seeds and soyfood products will have an impact on the manufacturing process and consumer acceptance (Mullin and Xu, 2001).

Seed hardness is an important quality attribute for soyfoods because it affects water absorption, seed coat permeability, cookability, and overall texture. For example, seed hardness was found to be significantly correlated to steamed seed hardness in a natto quality-trait study (r = 0.75; Chandler et al., 2000). However, testing for seed hardness is costly and time-consuming and often not practical for breeders' selection at the early stages of a breeding program. The suitability of potential cultivars or seed product for the natto market is usually determined by professional testers based on sensory evaluations. At least 0.91 kg of soybean seeds is required for each sensory test in which 10 well-trained researchers taste natto products in a 4-wk period with four replications of each product (Wei and Chang, 2004). In addition, there is no standard method in testing soybean seed hardness using scientific instruments. A puncture test is one of the simplest and commonly used instrumental measurements for food texture (Bourne, 2002), but it can give rise to large experimental errors if only a single sample is tested. The shear test also has limited utility without an accepted standard testing procedure. A compression test is rarely used because gases are easily trapped in most products (Wheeler et al., 1994).

Several hardness testing methods were compared using a 16-blade shear cell, a single blade, a 2-mm probe, a 75-mm cylinder, and pea rigs (16 2-mm probes) with different amounts of seed (Zhang et al., 2008). All these methods require a texture analyzer and are not feasible for breeders to use in selection because measuring seed hardness is difficult and time-consuming.

Although seed hardness is one of the major traits evaluated for food-grade soybeans, little information is available on genetic analysis of seed hardness. Keim et al. (1990) identified five restriction fragment length polymorphism (RFLP) markers associated with seed germination affected by hard seeds in a population derived from a cross between G. max and G. soja. These markers accounted for a total of 71% of variation in hard seededness, which affects seed germination. Keim et al. (1990) also stated that three interactions between five QTLs represented 38% of the total hard seededness variation.

In soybean, DNA markers have been widely and frequently used in identification of QTLs for major seed quality traits such as protein, oil, and seed size. However, research on soybean seed hardness is lacking. Information on genetic control of seed hardness will help understand this important seed quality trait and accelerate the process of breeding specialty soybeans for the soyfood market. The objective of this study was to identify QTLs associated with seed hardness, which would improve breeding efficiency with marker-assisted selection (MAS).


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plant Material
A total of 159 lines representing three generations of F2–derived lines (F2:3, F2:4, F2:5) were developed from a cross of ‘SS-516’ x ‘Camp’. Both parents were small-seeded cultivars specifically developed for the natto market. Camp, released by Virginia Polytechnic Institute and State University, is hard, with a hardness of 580 to 640 N per 30 g cooked seeds (maximum force in newtons to compress 30 g cooked seeds). SS-516, released by the Southern States seed company, is soft, with a hardness of 340 to 400 N per 30 g cooked seeds. SS-516 has white flower and broad leaf, whereas Camp has purple flower and narrow leaf. In crossing, a single plant of SS-516 was used as female to mate with a single plant of the pollen parent Camp in a greenhouse in spring of 2002. Four F1 plants were grown in the field and identified as true hybrids in summer of 2002. The four hybrid plants were harvested in bulk to form the F2 population. The F2 plants were then grown in the field and harvested individually to derive the F2:3 lines in a winter nursery in Costa Rica. Subsequently, each line was maintained as a separate entry using bulked seed for phenotyping at the F4 and F5 generations.

All the field experiments were conducted at the University of Arkansas Agricultural Experiment Station, Fayetteville, AR. In summer 2003, 2004, and 2005, F2:3, F2:4, F2:5 lines were planted in the field with a randomized complete block design with two replications. Each line was grown in a single-row plot 1.5 m in length, with a 0.95-m row spacing. Both Camp and SS-516 were included as checks in the test. A bulk leaf sample consisting of a young leaf per plant from each F2:3 line was collected at the V5 growth stage (Fehr et al., 1971) and stored at –80°C for DNA extraction. Plants in each plot were harvested in bulk, and a 50-g sample was used for hardness analysis.

Seed Hardness Testing
Fifty grams of unbroken seeds with uniform seed size from each F2–derived line was weighed and soaked in heat-resistant plastic boxes containing 250 mL water at ambient temperature for 16 h. Seeds were recovered from the soaking water with a sieve and blot-dried on paper towels. Stone seeds, which did not absorb water during soaking, were picked from the soaked seeds. Soaked-seed samples were pressure-cooked at 121.1°C and 1.2 kg cm–2 for 20 min. Hardness tests of 30 g cooked seeds from each entry were conducted in two replications using a TMS Texture System (TMS-2000, Food Technology Corp., Sterling, VA) equipped with a 16-blade shear cell. The maximum force to compress cooked seeds in newtons was determined as seed hardness (Song et al., 2003).

Molecular Marker Screening
Total genomic DNA was isolated from ground frozen leaf tissue using the CTAB (hexadecyltrimethyl ammonium bromide) method (Kisha et al., 1997). Polymerase chain reactions (PCRs) were performed as described by Cornelious et al. (2005). The amplified product was loaded in a 6% (w/v) nondenaturing polyacrylamide gel and stained with ethidium bromide (Wang et al., 2003). Two parents were screened with 874 simple sequence repeat (SSR) markers randomly distributed on 20 linkage groups (LGs). Two hundred and thirty-six polymorphic markers between parents were used to genotype the F2:3 lines.

Data Analysis
Phenotypic data on seed hardness was analyzed using JMP 5.0 software (SAS Institute, 2002) to remove outliers (studentized residual > 3 or < –3). Shapiro-Wilks' (W) test was used to test the mapping population for normal distribution. ANOVA was used to assess the difference of hardness among lines over 3 yr (2003 to 2005). Heritability of seed hardness was estimated using H2 = {sigma}2g/[{sigma}2g + ({sigma}2ge/e) + ({sigma}2/re)] (Nyquist, 1991), where H2 is heritability, {sigma}2g is genotypic variance, {sigma}2ge/e is genotype x environment interaction variance, {sigma}2 is error variance, r is number of replications, and e is number of environments, which is the number of years in this study.

The SSR marker segregation data of F2:3 lines were used to construct a genetic map by using JoinMap 3.0 (Van Ooijen and Voorrips, 2001) with the Kosambi function (Kosambi, 1944). Markers that exhibited unusual segregation ratios in the mapping population were excluded from the linkage analysis. Linkage parameter was set as recombination fractions of 0.40, likelihood (logarithmic odds [LOD]) score of 3.0, and the error detection ratio of 5%. QTL Cartographer (Basten et al., 1999) with single-marker analysis (SMA), composite-interval mapping (CIM), and multiple-interval mapping (MIM) analysis were used to identify QTLs associated with seed hardness by comparing genotyping data of F2:3 lines and phenotyping data of three sets of F2–derived lines (F2:3, F2:4, F2:5). In SMA, p < 0.001 was used as the threshold for significant markers. In the CIM analysis, the empirical significance threshold was determined by 1000-time permutation with a walk speed of 1 cM and significance level of 0.05 (Churchill and Doerge, 1994). Multiple-interval mapping analysis was used to estimate the optimum positions and effects of QTLs as well as QTL epitasis interactions. The MIM model c(n) = ln(n) was selected with a walk speed of 1 cM. Mapchart (Voorrips, 2002) was conducted to create the LOD plots according to the data from QTL Cartographer.


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Seed Hardness
The seed hardness of 159 F2–derived lines in each year from 2003 to 2005 and 3-yr average showed continuous variation and normal distribution, suggesting that seed hardness is a quantitative trait controlled by multiple genes (Fig. 1 ). ANOVA results showed that the seed hardness between two of three years and the interaction of genotype and year were significantly different (data is not shown). Hardness data in each year and combined in 3 yr were analyzed. The seed hardness of F2:3 lines ranged from 246.5 to 744.5 N with a mean of 438.8 N, the F2:4 lines ranged from 393.5 to 972 N with a mean of 677.7 N, and the F2:5 lines ranged from 297.5 to 843 N with a mean of 544.7 N. The 3-yr average of the F2–derived lines ranged from 368.5 to 799.5 N with a mean of 544.7 N. The broad-sense heritability (H2) estimate using variance components for the seed hardness in 3-yr average data of the mapping population was 0.56.


Figure 1
View larger version (21K):
[in this window]
[in a new window]

  		Figure 1. Distribution of seed hardness of F2:3 lines in 2003, F2:4 lines in 2004, and F2:5 in 2005, and average seed hardness over 3 yr.

 
Linkage Map Construction
Of the 874 SSR markers screened, 236 markers showed polymorphism between the parents, Camp and SS-516. One hundred and seventy-seven polymorphic markers (75%) that were easy to score and had good amplification segregated in the mapping population, of which 148 markers were integrated into all but LG D1b of the 20 LGs with a total genomic coverage of 1363.7 cM (Table 1 ). The grouping and position of most markers were consistent with the public soybean linkage map (Song et al., 2004). Each LG contained an average of eight markers with an average interval of 9.6 cM. Linkage groups G, L, and M had the best coverage with more markers for their relatively long genetic distance. There were 18 markers on LG G, for 101.5-cM genomic coverage, and LGs L and M included 16 and 15 markers for 80.4 and 114-cM genomic coverage, respectively. Fifteen new markers that are currently not on the published public consensus map were mapped on the linkage map of this study (Table 1). Specifically, five markers (Satt704, Satt098, Satt401, Sat_210, and Satt616) were added on LG G, four (Sct_029, Satt176, Satt040, and Satt093) on LG F, two (Satt714 and Satt265) on LG L, and one on each of LGs B1 (Sat_257), C1 (Sat_327), H (Satt246), and N (Satt084) (Table 1).


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

  		Table 1. Soybean genetic map of seed hardness including 148 simple sequence repeat (SSR) markers with estimated position on respective linkage groups (LG).

 
Quantitative Trait Loci Analysis
Quantitative trait loci for seed hardness were analyzed separately in each year and jointly across 3 yr. In the SMA, 34 markers in 2003, 5 markers in 2004, and 68 markers in 2005 were significantly associated with seed hardness (p < 0.001). Twenty-nine markers were shared between 2003 and 2005, two markers were shared between 2003 and 2004, and three markers were shared between 2004 and 2005 (Table 1). Sat_110 and Satt531 on LG D1a were shared among 2003, 2004, and 2005 (Table 2 ).


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

  		Table 2. Shared single sequence repeat markers significantly associated with seed hardness in F2–derived lines (F2:3, F2:4, and F2:5) from soybean cross SS-516 x Camp from 2003 to 2005.

 
In the CIM and MIM analysis, empirical significance threshold was computed as LOD score of 5.26 after 1000 permutations in the mapping population in 2003. However, none of markers that were significantly associated with seed hardness in 2003 had an LOD score as high as or higher than 5.26. Satt229 (72.3 cM) on LG L had the highest LOD score of 4.47 among all the markers, and it was significantly associated with seed hardness in 2003 (p < 0.001) in SMA (Fig. 2a ). In 2004, the LOD score of 3.73 was used as the empirical significance threshold. A putative QTL (LOD = 4.5) was located at 120.4 cM on LG D1a, which is close to Satt531 at 112.4 cM (Fig. 2b). This QTL accounted for 24.4% of the seed-hardness variation. In 2005, the same two QTLs were identified to be significantly associated with seed hardness in the mapping population (Fig. 2c and d). One QTL resides at 72.3 cM on LG L where Satt229 is exactly located (p < 0.00001 in SMA) in our genetic linkage map (Fig. 2c). The other QTL was located at 110.3 cM on LG D1a between Satt532 (95.3 cM) and Satt531 (112.4 cM; p < 0.000001 in SMA) (Fig. 2d). Thus, Satt531 is likely the marker tightly linked to this QTL on LG D1a. These two QTLs accounted for 9.4 and 8.2% of phenotypic variation, respectively, which is lower than the variation explained by Satt229 (10.7%) in 2003 and by Satt531 (24.4%) in 2004. The slight discrepancy in QTL effect is probably because these two QTLs exhibited epistatic interaction on seed hardness in this population, and this interaction occurred under specific environments.


Figure 2
View larger version (28K):
[in this window]
[in a new window]

  		Figure 2. Logarithmic odds score plots for seed-hardness QTLs on linkage groups (LGs) L and D1a. Markers with *, **, ***, and **** are significantly associated with seed hardness in single-marker analysis at p = 0.01, 0.001, 0.0001, and 0.00001 levels, respectively. Linkage group L (a) shows a QTL for seed hardness in 2003; LG D1a (b) shows a QTL for seed hardness in 2004; LG L (c) and LG D1a (d) show QTLs for seed hardness in 2005.

 
In summary, Satt229 was marginally associated with the QTLs for seeds hardness on LG L, with a slightly lower LOD score than the threshold in 2003. Satt229 was not significantly associated with seed-hardness variation in 2004 (LOD = 1.37) but was tightly linked to the seed hardness QTLs on LG L in the mapping population in 2005. Satt531 was significantly associated with seed-hardness variation but had a low LOD score (3.38) in 2003. However, Satt532 was tightly linked to the QTLs on LG D1a in 2004 and 2005.

Two QTLs were identified near Satt229 on LG L and Satt531 on LG D1a, respectively, for the average seed-hardness over 3 yr (Fig. 3a and b ). The first QTL, designated Ha1, was located at 72.3 cM on LG L and 0.0 cM from Satt229. Ha1 had an LOD score of 6.17, which was higher than the empirical significance threshold of 4.13 with R2 = 12.7%. The other QTL, named Ha2, was located 136.4 cM on LG D1a, 24 cM from Satt531 and 18.5 cM from Satt184. Ha2 had an LOD score of 5.08 with R2 = 36.1%.


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		Figure 3. Logarithmic odds score plots for seed-hardness QTLs on linkage groups L (Ha1) and D1a (Ha2), using average seed hardness data over 3 yr. Markers with *, **, ***, and **** are significantly associated with seed hardness in single-marker analysis at the 0.01, 0.001, 0.0001, and 0.00001 levels, respectively.

 
MIM models further confirmed the two QTLs for seed hardness, defined their position on the linkage map, and revealed their epistatic interaction effect on phenotypic variation. Ha1 was still mapped at 72.3 cM on LG L, while Ha2 was relocated at 128.4 cM on LG D1a, 16 cM away from Satt531. A dominance-by-dominance interaction was detected between Ha1 and Ha2, and it explained 7.9% of the phenotypic variance for seed hardness.


   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Soyfood manufacturers prefer soft small-seeded soybeans for natto production. Breeding populations such as the mapping population used in this study will enable breeders to select superior natto lines with desired seed texture. In this study, there were significant differences in seed hardness among lines in the population in 3 yr, with significant interaction of year by genotype. The mean seed hardness in 2004 was higher than that in both 2003 and 2005, perhaps because there was less rainfall in August and September of 2004 (2.97 cm) at the R3–R7 stages of soybean development as compared with 2003 (16.81 cm) and 2005 (13.31 cm). Lack of rainfall was most likely the reason for harder seeds in 2004 than 2003 and 2005. The broad-sense heritability of seed hardness (0.56) on entry-mean basis was moderately high, similar to the heritability of other seed quality traits such as aspartic acid (0.57, Panthee et al., 2006) and oil (Brummer et al., 1997), and lower than that of protein (Brummer et al., 1997). Selection for proper seed texture can be achieved through genetic manipulation and breeding selection, although environmental conditions also play an important role in seed hardness. Since there is very limited information on genetic analysis of seed hardness in soybean, results from this study provide a fundamental basis for future research on genetic improvement of seed hardness of soybean.

Of the 874 SSR markers tested, 235 (26.9%) were polymorphic between the two parents, and 177 (20.2%) markers showed polymorphism in the F2:3 population. This polymorphic ratio was in agreement with 16 to 30% polymorphism reported for other soybean populations (Arahana et al., 2001). In this study, the genetic linkage map covered 1363.7 cM of the soybean genome, which was equivalent to 54.0% genetic distance of the public linkage map (Song et al., 2004). Furthermore, 15 newly identified markers in the present study provided reference of markers added to the public linkage map, which improved the likelihood of identifying QTLs associated with seed hardness and would facilitate QTL discovery for other traits in the future. In the consensus map, Satt184 (17.5 cM) and Satt531 (40.9 cM) were 23.4 cM apart on LG D1a. Sat_353 (36.2 cM) was the only SSR marker located between Satt531 and Satt184 (Song et al., 2004). However, Sat_353 did not show polymorphism among lines in the mapping population of this study. Additional markers in the region between Satt531 and Satt184 will be needed to reduce the interval of the Ha2 region in the future.

Useful QTLs in MAS should be stable across environments. Evaluating a QTL across several years can be used to assess stability (Brummer et al., 1997). The two QTLs identified in this study were detected using seed hardness data from 3 yr. Ha2 was more stable than Ha1 because Ha2 was detected in both 2004 and 2005, while Ha1 was detected in 2005 and was only a candidate QTL in 2003. However, both QTLs were detected if pooled data were used over the 3 yr. In addition, one of the seed coat hardness QTLs in Soybase, Hrd Sd 1–3, is in close proximity to Ha1. Therefore, both QTLs should be useful, once confirmed, in a marker-assisted breeding program. Similarly, QTLs for protein and oil were detected in five populations in 3 yr and in three populations in 2 yr (Brummer et al., 1997). Breeders may combine the two QTLs in MAS in early generations to improve the selection efficiency and accelerate breeding process for seed hardness under different environments.

Quantitative trait locus regions usually contain loci for more than one trait (Wang et al., 2004). Satt229, linked to Ha1 on LG L, also flanked a QTL for oil content (Hyten et al., 2004). The other flanking marker of this oil QTL was Satt373, which is 8.1 cM from Satt229 in our linkage map. Several seed-size QTLs were detected upstream of Satt229 (93.9 cM) in the consensus map (Hoeck et al., 2003). Satt006 (92.0 cM) and Sat_099 (78.2 cM) on LG L explained 32.5% and 33.7% of seed size variation (Hoeck et al., 2003). Quantitative trait locus oil-5 was located between Satt184 and Satt179 on LG D1a (Hyten et al., 2004). Satt531 (40.9 cM) was in the middle of Satt184 (17.5 cM) and Satt179 (56.2 cM) on LG D1a in the consensus map. Thus, Ha2, tightly linked to Satt531 in our study, was located within this region. Panthee et al. (2006) also identified a seed-size QTL near Satt184 with R2 of 11.3, and a lysine QTL near Satt184 with R2 of 13.1. These regions of LGs L and D1a, with multiple QTL clusters for seed-quality traits, may be attributed to genetic linkage and/or pleiotropy. Nevertheless, it is possible that some of the molecular markers identified in these regions can be used in MAS for multiple traits.

A dominance by dominance epistatic interaction between Ha1 and Ha2 existed in controlling seed hardness and explained 7.9% of the phenotypic variation, which is useful in breeding for proper seed hardness. Both QTLs may be needed in a correct allelic form to give rise to maximum genetic effect on seed hardness in a given line. Similarly, three interactions between five QTLs were identified to represent 38% of the total hard seededness variation in a previous study (Keim et al., 1990). The gene controlling hard seededness in soybean at the i locus (between pT-153a and pA-111 markers) was thought to epistatically manipulate other genes (Bernard and Weiss, 1973). Thus, the interaction between pA-111 and the other three markers confirmed the regulatory function of the i locus. Continued research efforts are needed to validate the two QTLs for seed hardness indentified in this study so that they can be effectively used in MAS for seed hardness of specialty soybeans for natto production.


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

Received for publication October 2, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 J HeredHome page
B. Zhang, P. Chen, A. Shi, A. Hou, T. Ishibashi, and D. Wang
Putative Quantitative Trait Loci Associated with Calcium Content in Soybean Seed
J. Hered., March 1, 2009; 100(2): 263 - 269.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, B.
Right arrow Articles by Ishibashi, T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Zhang, B.
Right arrow Articles by Ishibashi, T.
Agricola
Right arrow Articles by Zhang, B.
Right arrow Articles by Ishibashi, T.
Related Collections
Right arrow Soybean
Right arrow Cell Biology & Molecular Genetics
Right arrow Crop Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome