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


     

Published online 27 May 2005
Published in Crop Sci 45:1345-1352 (2005)
© 2005 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 Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
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 Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dai, W.-M.
Right arrow Articles by Cai, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Dai, W.-M.
Right arrow Articles by Cai, R.
Agricola
Right arrow Articles by Dai, W.-M.
Right arrow Articles by Cai, R.
Related Collections
Right arrow Rice
Right arrow Nutrients
Right arrow Crop Genetics

GENOMICS, MOLECULAR GENETICS & BIOTECHNOLOGY

Genetic Dissection of Silicon Content in Different Organs of Rice

Wei-Min Daia,b, Ke-Qin Zhangb, Bin-Wu Duanb, Kang-Le Zhengb, Jie-Yun Zhuangb and Run Caia,*

a Plant Science Dep., Shanghai Jiaotong Univ., Shanghai 201101, China
b Rice Product Quality Inspection and Supervision Center, China National Rice Research Inst., Hangzhou 310006, China

* Corresponding author (cairun{at}sjtu.edu.cn)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Quantitative trait loci (QTLs) analysis for Si content in rice (Oryza sativa L.) was performed to study the inheritance of Si content in rice. A recombinant inbred line (RIL) population of 244 lines derived from the indica rice cross ‘Zhenshan 97B’ (ZS97B)/‘Milyang 46’ (MY46) were grown in 2003 with two replications. Silicon content in the hull (HUS), flag leaf (FLS), and stem (STS) were measured. On the basis of a linkage map consisting of 207 DNA marker loci, a total of 10 QTLs showing significant additive effects and 14 digenic interactions having significant additive-by-additive (AA) epistatic effects were detected. The number of QTLs and AA interactions for individual traits were four and four for HUS, four and four for FLS, and two and six for STS, respectively. General contributions to the phenotype variance due to additive effects and AA effects were 29.3 and 18.6% for HUS, 14.8 and 13.6% for FLS, and 8.6 and 28.6% for STS, respectively. This indicates that gene actions at both the one- and two-locus levels play an important role to genetically control Si content in rice. In addition, it was shown that the detection of QTLs at the one-locus level, as well as magnitude and direction of the additive effect, might be influenced greatly by digenic interactions involving loci linked to the given QTL. By comparing these results with other reports, genetic association between QTLs for Si content and QTLs for yield component traits and lodging-tolerance related traits were evident.

Abbreviations: AA, additive-by-additive • FLS, mg Si per 100 mg dry weight in the flag leaf • HUS, mg Si per 100 mg dry weight in the hull • MY46, ‘Milyang 46’ • QTL, quantitative trait locus • RIL, recombinant inbred line • STS, mg Si per 100 mg dry weight in the stem • ZS97B, ‘Zhenshan 97B’


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
SILICON IS THE SECOND most abundant constituent in soil. It constitutes 28% of the total soil weight in soil, which is only lower than oxygen that is 47% (Datnoff and Snyder, 2001). Great variation is evident among different plant species rooting in soil, and rice is known to be the most effective Si-accumulating species (Idris et al., 1975; Lee et al., 1987; Balasta and Perez, 1989; Takahashi et al., 1990; Takahashi, 1995). While rice has a general high ability to absorb and accumulate Si, Si deposition in rice varies greatly among different organs of the plant (Alina, 1984). Silicon accumulation into various plant organs varies among rice genotypes (Majumder et al., 1985; Winslow, 1992; Winslow et al., 1997). Silicon content in different organs of a rice plant generally ranged from high to low, in descending rank in the hull, leaf, leaf sheath, culm, and root (Zhu, 1985).

Numerous experiments have shown that Si deposition in the plant tissues could improve yield, biotic stress, and abiotic stress of rice plants (Okuda and Takahashi, 1964; Snyder et al., 1986; Ma and Takahashi, 1990; Epstein, 1994, 1999; Savant et al., 1997). As in China, where half of the paddy fields are of Si deficient, Si deficiency in paddy fields has become a worldwide constraint for sustainable rice production (Ma, 1990, 1994, 1997). However, widespread application of Si fertilizer is hindered by the high cost of the material and labor (Alvarez and Datnoff, 2001). It is conceivable that development of rice varieties with high Si deposition would be a better strategy than amending the soil with Si (Savant et al., 1997; Deren, 2001).

Diallel analysis showed that genotype differences on Si content in rice were controlled by polygenes (Majumder et al., 1985). While Si content in a given rice genotype might change with variation of Si fertilization and/or Si in soil, the relative ranking of Si content among different rice genotypes remained fairly stable across different environments and locations (Deren, 2001; Deren et al., 1992). Dissection of QTLs is a powerful tool to study the inheritance of complex traits (Xu, 2002), yet the genes controlling Si content in rice have not been mapped. Using an indica rice RIL population to determine QTLs conditioning Si content in rice, the objectives of this study are: (i) to provide evidence for multigenic control of Si content in rice; (ii) to discover QTLs with major effects on Si content in rice; and (iii) to discover QTLs for Si content in developing near isogenic lines (NIL) that can be used for fine mapping.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Rice Material
The rice population used was ZS97B/MY46 RIL population, in which ZS97B and MY46 are maintainer and restorer lines of the commercial three-line rice hybrid Shanyou 10, respectively. In 2003, 244 F8 lines and the parental lines were transplanted in the paddy field of China National Rice Research Institute, Hangzhou, China, in two replications with randomized complete block design. Twelve plants per replication were planted, and main stems of the middle five plants of each replication were harvested at maturity. The soil is acid (pH of 5.5 in 1:1 H2O; upper 15 cm). Plant–available Si in soil is 9.13 mg/100 g measured with acetate buffer pH 4.0 (Snyder, 2001).

Measurement of Silicon Content
Each main stem was separated into panicle, flag leaf, and stem (culm and leaf sheath). They were dried in an oven at 70°C for 7 d. Hull was collected after husking dry seeds. Hull, flag leaf, and stem of each line in each replication were mixed and ground, respectively, followed by drying at 60°C for 48 h. They were then weighted, digested with NaOH, and dissolved in ddH2O. Silicon concentration was determined by the colorimetric molybdenum blue method (Elliott, 1991; Snyder, 2001). Silicon content in rice was measured as the amount of Si in milligrams per 100 mg dry weight in the hull (HUS), flag leaf (FLS), and stem (STS). Mean values across two replications were taken for data analysis.

Data Analysis
The RIL population was applied in our previous QTL mapping study for yield traits in rice (Zhuang et al., 2002). Survey of parental polymorphism was conducted using more RFLP and SSLP markers. Polymorphic markers were employed to assay the RILs. Linkage analysis was performed with MAPMAKER/EXP 3.0 (Lander et al., 1987; Lincoln et al., 1992). An updated linkage map consisting of 207 DNA marker loci was constructed and applied for the present QTL mapping study.

QTLMAPPER 1.60 of the mixed linear model (Wang et al., 1999, 2003; http://statgen.ncsu.edu/zhu/index.html; verified 7 Feb. 2005) was used to determine QTLs conditioning the three traits analyzed. In addition to having a power for the detection of main-effect QTLs similar to interval mapping (Zhuang et al., 2000), this method has advantages of the control of background genetic variation and simultaneous detection of main-effect QTLs and digenic interactions (Cao et al., 2001b; Liao et al., 2001; Zhang et al., 2001). In this study, important markers and marker pairs were selected using stepwise regression analysis (P < 0.001). Background genetic variation due to main and epistatic effects of important markers was controlled. The threshold of LOD > 3.0 was chosen for claiming a putative QTL. The significance of additive effect and AA epistatic effect was further tested by running the submenu Bayesian test (P < 0.001). With the submenu contributions, output from the Bayesian test was applied to determine the partial determination coefficients (R2) of each QTL and each epistasis, as well as the total phenotypic variance explained by all the QTLs and epistasis detected for the given trait, respectively. The QTLs showing significant additive effects were designated as proposed by McCouch et al. (1997). When two or more QTLs were detected on the same chromosome, QTL numbering was suffixed following the chromosome number from top to bottom of the chromosome.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Trait Performance
Silicon content of ZS97B and MY46 measured in mg/100 mg dry wt. were 13.50 and 15.28 in the hull, 7.67 and 7.60 in the flag leaf, and 5.68 and 7.79 in the stem, respectively. Normal distribution was observed for each of the three traits analyzed (Fig. 1), indicating that the data were feasible for QTL analysis.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 1. Frequency distribution of the three traits analyzed in ZS97B/MY46 recombinant inbred line population.

 
Significant positive correlation was observed between each pair of the three traits. Moderate correlation was observed between FLS and STS (r = 0.49, P < 0.01), while low correlation was shown between FLS and HUS (r = 0.20, P < 0.01), and between HUS and STS (r = 0.13, P < 0.05). This implied that genetic control of Si content in ZS97B/MY46 RIL population might change greatly from organ to organ, although a proportion of genes conditioning Si content in rice may have similar effects on different organs.

QTLs Showing Significant Additive Effects for Silicon Content
For the three traits analyzed in ZS97B/MY46 RIL population, a total of 10 QTLs showing significant additive effects were detected on chromosomes 1, 5, 6, 11, and 12 (Table 1, Fig. 2). For ease of description, QTLs showing significant additive effects were hereafter referred to as QTLs.


View this table:
[in this window]
[in a new window]
  		Table 1. Quantitative trait loci (QTLs) exhibiting significant additive effects for Si contents in ZS97B/MY46 recombinant inbred line population.

 


View larger version (41K):
[in this window]
[in a new window]
  		Fig. 2. Most likely positions of quantitative trait loci (QTL) conditioning Si contents in rice detected in Zhenshan 97B/Milyang 46 recombinant inbred line population. The blank bars indicate chromosomes, and the solid portions indicate approximate positions of the centromeres according to Temnykh et al. (2000). For each chromosome, loci showing significant additive effects and additive-by-additive epistatic effects are shown on the left-hand and right-hand sides, respectively. Epistatic loci involved in a same interaction are indicated by the same number. FLS, silicon in the flag leaf; HUS, silicon in the hull; STS, silicon in the stem.

 
Four QTLs were detected for HUS and explained 29.3% of the phenotypic variance. The QTL qHUS-6 on the short arm of chromosome 6 explained 17.4% of the phenotypic variance, which was much larger than the proportion of 1.6 to 7.7% explained by other QTLs. The allele for increasing Si content in the hull was from the maternal parent ZS97B at qHUS-1-1 and qHUS-11, and from the paternal parent MY46 at qHUS-6 and qHUS-1-2.

Four putative QTLs were detected for FLS and explained 14.8% of the phenotype variance. These QTLs were distributed on chromosomes 1, 5, 11, and 12, and the contribution of a single QTL to the phenotype variance ranged as 1.9 to 6.4%. The allele for increasing Si content in the flag leaf was from the paternal parent at qFLS-1, and from the maternal parent at the other three QTLs.

Two putative QTLs for STS were detected and explained 8.6% of the phenotype variance. The QTL qSTS-1 contributed 6.3% to the phenotype variance, and its allele for increasing Si content in the stem was from the paternal parent. The QTL qSTS-5 had an opposite direction of additive effect and contributed 2.4% to the phenotype variance.

By comparing QTLs detected for different traits, it was found that three of the four QTLs for HUS shared similar genomic regions with QTLs for either FLS or STS. In intervals RM246 to RG101 and RZ536 to RM217 on the long arm of chromosomes 1 and 11, respectively, QTLs for HUS and FLS were detected and each pair of QTLs had the same direction of additive effects. In interval RM151 to RG532 on the short arm of chromosome 1, QTLs for HUS and STS were detected, but they had opposite directions of additive effects. No common genomic regions were observed between QTLs for FLS and STS.

Significant Additive-by-Additive Interactions for Silicon Content
A total of 14 digenic interactions having significant AA effects were detected for the three traits analyzed in ZS97B/MY46 RIL population, including four for HUS, four for FLS, and six for STS (Table 2). Epistatic loci involved in the interactions were distributed on all the rice chromosomes except chromosome 10 (Fig. 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Significant additive-by-additive (AA) interactions detected for Si contents in ZS97B/MY46 recombinant inbred line population.

 
The general contributions of the digenic interactions to the phenotype variance were 18.6, 13.6, and 28.6% for HUS, FLS, and STS, respectively. For each of the three traits, one digenic interaction acted for increasing the value of the recombinant type and the others were in favor of the parental type.

On comparison of genomic locations between QTLs and epistatic loci detected for the same trait, no intervals containing both types of loci were detected. Nevertheless, a number of epistatic loci were detected in genomic regions adjacent to intervals harboring QTLs for the same trait. Genotypes of markers most closely linked to the peak LOD location in these regions were used to analyze the interference of digenic interaction on QTL effects.

For HUS, a locus linked to qHUS-6 interacted with a locus in RG667 to RM201 on chromosome 9. It was shown that the additive effect measured at RM111 for the epistatic locus on chromosome 6 (Fig. 3A) and at RZ516 for qHUS-6 (Fig. 3B) was larger in the presence of maternal than paternal genotype at RM201. Similarly, the QTL located on the long arm of chromosome 5 had a larger additive effect in the presence of maternal than paternal genotype at RM294B on chromosome 1, no matter the effect was measured at B10A for the epistatic locus (Fig. 3C) or at RM164 for qSTS-5 (Fig. 3D).



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 3. Average phenotype values in groups classified based on the genotype combinations of DNA markers linked to pairwise epistatic genes. 1 = maternal homozygote, 2 = paternal homozygote. RZ516 in Panel B, RM164 in Panel D, and RG101 in Panels H, I, and J represent qHUS-6, qSTS-5, and qFLS-1, respectively.

 
Three epistatic loci, each involved in an interaction for FLS, were detected in locations linked to qFLS-1. One locus was located in interval RZ19 to RG229 next to the interval harboring qFLS-1. It interacted with a locus in interval RM263 to RM6 on chromosome 2, resulting in magnitude change of the additive effects with same direction at RG229 (Fig. 3E) and at qFLS-1 (Fig. 3H). Additive effects at the other two epistatic loci represented by RZ154 (Fig. 3F) and RZ538 (Fig. 3G) changed their directions with the genotype alternation of their counterparts, but the interactions had little effect on qFLS-1 (Fig. 3I, 3J). This is understandable since RZ154 and RZ538 were located at either side of qFLS-1, they interacted with loci in linkage, and the two interactions had opposite directions.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
With the advent of DNA marker technology, genetic dissection of QTLs underlining complex traits have been extensively performed in rice (Xu, 2002) and other crop species. A few studies have worked on the genetic basis of genotype differences on the content of major elements such as N, P, and K (Loudet et al., 2003; Cao et al., 2001a; Lin et al., 2004), but little work related to nonessential elements was reported. Although Si has not been classified as an essential element for higher plants, it is evident that Si plays an important role in facilitating plant growth and development (Savant et al., 1997; Ma and Takahashi, 2002). Moreover, Si is the only element that does not cause serious injury when accumulated in excess amounts in plants (Ma et al., 2001).

In the present study, a RIL population derived from the indica cross ZS97B/MY46 was applied for mapping QTLs conditioning Si content in the hull, flag leaf, and stem in rice. A total of 10 QTLs having significant additive effects and 14 digenic interactions having significant AA effects were detected. This first study of QTL mapping for Si content in rice provides evidence in support of the conclusion drawn by classical studies that genotype differences on Si content in rice were controlled by polygenes (Majumder et al., 1985).

The importance of epistasis as the genetic basis of complex trait has been suggested by classical quantitative genetic studies and supported by recent QTL mapping studies (Eshed and Zamir, 1996; Li et al., 1997; Li, 1998; Cao et al., 2001b; Zhang et al., 2001; Liao et al., 2001). In the ZS97/MY46 RIL population, the general contribution to the phenotype variance of Si content in rice due to AA effects was comparable with the contribution due to additive effects, although the relative contribution of the two types of effects varied among traits. As compared with the general contributions due to additive effects, the contributions due to AA effects were lower for HUS, similar for FLS, and higher for STS. In addition, it was shown that the additive effect of a QTL might be influenced by digenic interactions involving loci linked to the given QTL.

Among the three traits analyzed in this study, HUS and FLS had more in common in terms of the genomic location and effect direction of QTLs having significant additive effects. The QTLs qHUS-1-2 and qFLS-1 were detected in the same interval, so were qHUS-11 and qFLS-11. The QTLs in each interval had the same direction of additive effects for the two traits. On the other hand, QTLs for FLS and STS shared no common intervals, and qHUS-1-1 and qSTS-1 detected in the same interval had opposite direction of additive effects. These results were in contrast to the result from simple correlation analysis in which correlation between FLS and STS was much higher than the other two pairs. This might result from the interference of digenic interaction. For example, the additive effects at RM294B for STS (Fig. 3C, 3D) and at RZ154 for FLS (Fig. 3F) were masked by digenic interactions. When the background loci were from the same parent, they would have same directions of additive effects. Since simple correlation analysis was based on trait performance of each line which had a unique genotype combination, it would be less affected by the masking effects.

Another important characteristic of quantitative traits is the influence of environmental effects and genotype x environment (GE) interactions on their phenotypic performance. It was shown that Si content in rice of a given genotype changed with different native Si in soil, and the content increased in response to Si fertilization (Winslow, 1992; Deren, 2001). Nevertheless, it was observed in these studies that relative Si content of different rice genotypes were rather consistent under different Si fertilization and across different locations. This indicated that the difference on Si content among rice genotypes are largely ascribed to the genetic effect of QTLs, rather than to environmental effects and GE effects. Our present study is aiming to have a preliminary insight into the locations and effects of QTLs conditioning Si content in rice. The NILs that can be used for fine mapping were developed based on the QTL mapping results obtained in this study.

It was shown that the increase of Si content in the hull was related to the increase of 1000-grain weight and spikelet fertility in rice (Savant et al., 1997; Datnoff and Snyder, 2001). The RIL population used in this study was applied in a QTL mapping studies for grain yield and its components traits (Zhuang et al., 2002). The QTLs qTGWT-1-1 and qTGWT-6 for 1000-grain weight were detected in locations similar to qHUS-1-1 and qHUS-6, respectively. The alleles from the paternal parent MY46 acted for decreasing and increasing trait value at qTGWT-1-1 and qTGWT-6, respectively, which was in accordance with qHUS-1-1 and qHUS-6.

An important beneficial effect of Si in rice is alleviating biotic and abiotic stresses (Epstein, 1994, 1999; Suzuki, 1997; Datnoff et al., 1991). As compared with the work reported by Kashiwagi and Ishimaru (2004), seven of the 10 QTLs showing significant additive effects for Si content in ZS97B/MY46 RIL population were located in genomic regions where QTLs for lodging resistance and related traits were detected in their study. The QTLs qHUS-1-1, qSTS-1, and qFLS-12 had locations similar to QTLs conditioning stem diameter, qHUS-6, qHUS-11, and qFLS-11 to QTLs for pushing resistance, and qFLS-5 to QTLs for pushing resistance and lodging resistance by typhoon. In 2003, the RIL population of ZS97B/MY46 was also tested for traits related to lodging tolerance. The length and thickness of the first and second internodes were measured, respectively, and QTLs for each of the four traits were detected in a location similar to qSTS-1. The alleles in favor of lodging tolerance were from the paternal allele (Zhang et al., 2005), which was in the same direction to additive effect at qSTS-5. Further studies are needed to clarify the role of such genetic associations for the beneficial effects of Si on lodging tolerance in rice.


   ACKNOWLEDGMENTS
 
We grateful acknowledge the Cornell Group in the USA and Japanese Genome Research Program for providing the DNA probes. This work was supported by the Chinese 863 Program and the Rockefeller Foundation International Rice Biotechnology Program.

Received for publication August 25, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 


Related articles in Crop Science:

THIS ISSUE IN CROP SCIENCE

Crop Science 2005 45: vii. [Full Text]  



This article has been cited by other articles:


Home page
 Plant Physiol.Home page
J. F. Ma, N. Yamaji, K. Tamai, and N. Mitani
Genotypic Difference in Silicon Uptake and Expression of Silicon Transporter Genes in Rice
Plant Physiology, November 1, 2007; 145(3): 919 - 924.
[Abstract] [Full Text] [PDF]

Home page
 ANN BOT (LOND)Home page
S.-H. Cheng, J.-Y. Zhuang, Y.-Y. Fan, J.-H. Du, and L.-Y. Cao
Progress in Research and Development on Hybrid Rice: A Super-domesticate in China
Ann. Bot., October 1, 2007; 100(5): 959 - 966.
[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 Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
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 Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dai, W.-M.
Right arrow Articles by Cai, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Dai, W.-M.
Right arrow Articles by Cai, R.
Agricola
Right arrow Articles by Dai, W.-M.
Right arrow Articles by Cai, R.
Related Collections
Right arrow Rice
Right arrow Nutrients
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