- Email: [email protected]
QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice Under Upland and Lowland Environments
Journal of Genetics and Genomics (Formerly Acta Genetica Sinica) May 2007, 34(5): 420-428
Research Article
QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice Under Upland and Lowland Environments Yongmei Guo1, 2, *, Ping Mu1, *, Jiafu Liu3, Yixuan Lu2, Zichao Li 1,① 1. Key Laboratory of Crop Genomics and Genetic Improvement of Ministry of Agriculture, Key Laboratory of Crop Heterosis and Utilization of Ministry of Education, Beijing Laboratory of Crop Genetic Improvement, China Agricultural University, Beijing 100094, China; 2. Institute of Food Crop Research, Yunnan Academy of Agricultural Sciences, Kunming 650205, China; 3. Quality and Test Center of Agricultural Administration, Yunnan Academy of Agricultural Sciences, Kunming 650201, China
Abstract: Grain cooking and nutrient qualities are the most important components of rice (Oryza sativa L.) quality. A doubled haploid (DH) population from a cross between two japonica cultivars was used to examine the phenotypic values and potential QTLs for the quality traits. The cooking and nutrient quality traits, including the amylose content (AC), the gel consistency (GC), the gelatinization temperature (GT), and the protein content (PC), in rice grown under upland and lowland environments were evaluated. Significant differences for AC, GC, GT, and PC between upland and lowland environments were detected. The phenotypic values of all four traits were higher under upland environment than lowland environment. The value of PC under upland environment was significantly higher (by 37.9%) than that under lowland environment. This suggests that upland cultivation had large effect on both cooking and nutrient qualities. A total of seven QTLs and twelve pairs of QTLs were detected to have significant additive and epistatic effects for the four traits. Significant Q × E interaction effects of two QTLs and two pairs of QTLs were also discovered. The general contribution of additive QTLs ranged from 1.91% to 19.77%. The Q × E interactions of QTLs QGt3 and QAc6 accounted for 8.99% and 47.86% of the phenotypic variation, respectively, whereas those of the 2 pairs of epistatic QTLs, QAc6-QAc11b and QAc8-QAc9, accounted for 32.54% and 11.82%, respectively. Five QTLs QGt6b, QGt8, QGt11, QGc1, and QPc2, which had relatively high general contribution and no Q × E interactions, were selected to facilitate the upland rice grain quality breeding. Keywords: upland rice; cooking quality; nutrient quality; QTL mapping; Q × E interaction effects
Water shortage has been an ever-increasing problem and the largest constraint to the development of agriculture in China. Severe water shortage has also led to a rapid decrease of irrigated rice (Oryza sativa L.) production especially in northern China. It was reported that the area of irrigated rice production has reduced by 20% (about one million hectare) in the
north, northwest, and northeast China in the past five years. Upland rice culture is a non-flooded alternative to lowland rice culture, which can save up to 50%−70% of irrigation water [1]. Owing to the large difference between upland and lowland cultural conditions, the rice quality may be affected under upland environment. Few reports on the variations of rice
Received: 2006−10−23; Accepted: 2006−11−30 This work was supported by the State Key Basic Research and Development Plan of China (973), the Hi-Tech Research and Development Program of China (863) and National Natural Science Foundation of China. * These authors contributed equally to this work. ① Corresponding author. E-mail: [email protected]; Tel&Fax: +86-10-6273 1414 www.jgenetgenomics.org
Yongmei Guo et al.: QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice under…
grain quality under upland environment are available. A study was initiated to investigate the variation and genetic patterns of rice quality traits under upland cultural conditions. Grain cooking quality including amylose content (AC), gel consistency (GC), gelatinization temperature (GT), and nutrient quality including protein content (PC) are the most important components for the grain quality in rice. The inheritance of these traits has been widely studied [2−14]. The grain cooking and nutrient qualities were both controlled by QTLs with major and minor effects. Some references only provided the presence of QTLs by comparing the different kinds of QTL mapping backgrounds[15, 16]. Thus, the exact degree of Q × E interactions can not be obtained. The objective of the present study was to analyze the effect of water stress on rice grain cooking and nutrient qualities, to locate QTLs and Q × E interactions affecting these traits under upland and lowland conditions using a doubled haploid (DH) population derived from an upland and lowland rice cross, and to provide molecular markers for grain quality breeding.
1 1. 1
Materials and Methods Materials and field conditions
A population of 116 DH lines derived from a cross between Yuefu (a Japonica lowland variety) and IRAT109 (a Japonica upland variety) was used in this study. Both lowland (flooded anaerobic soil) and upland (aerobic soil) field conditions were used in this experiment. Doubled haploid lines and the parents were planted in a randomized complete design with 2 replications under upland and lowland conditions. To protect the plant in a good growth situation, all seeds were dressed by chemical dressing, which was made of weedicide and pesticide. Seeds of each DH line were sowed directly in lowland and upland fields in early May, 2002, in Beijing. The normal fertilizer in the form of N, P2O5, and K2O in the amount of 150 kg www.jgenetgenomics.org
421
each was applied prior to plantation. Thirty plants of each DH line were grown in single rows that were spaced 30 cm apart with 5 cm space between plants. The total water input (irrigation plus rainfall) was about 1,400 mm in lowland experiment. Plants in the upland field encountered severe water stress before June (tillering stage) because of the lack of rain. Supplementary irrigation was supplied when all DH lines had a high degree of leaf rolling at noon in the field. Three times of irrigation approximately 50 mm each were provided during the growth period under upland conditions. The total water input of the upland field was about 500 mm. Three hundred kilograms urea per hectare was applied at the tiller stage to both fields. Mature seeds were collected from 5 plants in the middle row of each DH line and the parents, and then stored at room temperature for three months prior to testing. 1. 2
Trait measurement
Amylose content (AC) was measured as described by Williams et al. [17] and Juliano [18], involving a defatting step, where 95% ethanol was added to the milled rice flour prior to starch dispersion in NaOH, followed by gelatinization in a boiling water bath, and measurement of amylose based on the detection of iodine blue color at pH 4.5 to 4.8 using sodium acetate buffer. Gel consistency (GC) was measured according to the method of Cagampang et al. [19], where the flow characteristic of milled rice gel was measured in 0.2 mol/L KOH and indexed by the length (in mm) of the cold horizontal gel (using a sample of 100 g 100mesh-sieved rice flour). Alkali-spreading score (ASS), indirect measurement for GT, was measured as described by Little et al [20]. Ten milled rice grains from each parent or DH line were immersed in 1.7% potassium hydroxide solution at room temperature for 23 h. The milled rice was carefully separated using forceps, and the spreading value of rice was scored by visual assessment using the method of Jennings et al [21].
422
Journal of Genetics and Genomics
Protein content (PC) was determined by a nitrogen gas analyzer (model 528; LECO). Samples of 1 mg were placed into a quartz combustion tube in an
an average distance of 9.3 cM between adjacent markers [23]. A threshold probability of P < 0.005 was used to declare the existence of QTLs. Correlation analysis between upland and lowland environments for these traits was conducted using the SPSS 10.0 software (http://www.spss.com).
induction furnace at 900℃. Total crude protein was calculated from the nitrogen content of the processed grain where percentage nitrogen × 5.95 = percentage protein. Three replications for each sample for these traits were analyzed. 1. 3
遗传学报 Vol.34 No.5 2007
2
QTL analysis
Results
2.1
QTL analysis including additive effects, epitasis, and QTL × environment interactions effects was carried out by mixed linear model approaches using QTL Mapper version 2.0[22] and a constructed molecular linkage map. The linkage map included 94 RFLP and 71 SSR markers and covered 1,535 cM in length with
Phenotypic values and the frequency distribution for AC, GC, GT, and PC of the DH population
The four parameters of AC, GC, GT, and PC were estimated for the DH lines and the parents grown in upland and lowland environments (Table 1, Fig. 1). The analysis of variance between upland and
Table 1 Variations of cooking and nutrient quality between two parents, IRAT109 and Yufu, and among the DH population under lowland and upland environments Lowland Trait
DH population
IRAT 109
Yuefu
AC% GC/mm
5.8 90.5
GT
2.0
PC%
8.1
Upland
Mean
Range
17.4 52.0
16.21±2.05 63.67±16.19
11.6-26.4
7.0
3.73±1.96
30.0-100.0 2.0-7.0
7.96±0.97
4.9-11.2
8.1
DH population
IRAT 109
Yuefu
15.9 70.0 4.5 10.6
F value
Mean
Range
15.5 40.7
17.10±1.69 78.91±18.95
13.2-23.0
6.5 9.6
Corrlation coefficient
43.0-100.0
4.18** 6.78**
0.474** 0.253**
4.96±1.96
1.5-7.0
7.36**
0.608**
10.98±1.02
8.7-13.9
**
0.322**
27.18
**Significant at 1% level.
Fig.1
Distribution of AC, GC, GT and PC in lowland and upland environments
www.jgenetgenomics.org
Yongmei Guo et al.: QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice under…
lowland environments for the DH population showed that significant differences were detected for AC, GC, GT, and PC between upland and lowland environments. The F values were 4.18, 6.78, 7.36, and 27.18, respectively. This result indicated that AC, GC, GT, and PC were significantly influenced by field conditions. The GC values of the two parents IRAT109 and Yuefu in the upland environment were 22.65% and 21.73% lower than that in the lowland environment. In contrast, PC was 30.86% and 18.51% higher for IRAT109 and Yuefu, respectively, in the same comparison. For AC, Yuefu was 10.92% lower in upland than in lowland environment, whereas IRAT109 showed slight difference under different field conditions. The GT of the two parents responded to the field environments differently; while IRAT109 showed a large increase, Yuefu declined by 7% from lowland to upland fields. Although the differences between IRAT109 and Yuefu were relatively small in AC and PC, a large segregation of the DH population in these traits was observed. For DH population, the means for all four parameters were higher in the upland environment than in the low land environment. The value of these parameters showed an approximately normal distribution and transgressive segregation, indicating that the population was suitable for QTL mapping for these traits. The phenotypic correlations between upland and
423
lyzed, polymorphism was detected for 94 and 71, respectively. The polymorphic rate was 44.8% and 33.2%, respectively. An integrated molecular linkage map consisting of 94 RFLP markers and 71 SSR markers was constructed, which covered 1,535 cM in length with an average distance of 9.3 cM between adjacent markers. The parents of DH lines, Yuefu and IRAT109 were both Japonica variety. Yuefu was from Japan, while IRAT109 originated from Africa. Higher genetic difference was observed between these types of parents. Additive, epistatic, as well as Q × E interaction effects between upland and lowland environments associated with AC, GC, GT, and PC were detected using the phenotypic data and the constructed linkage map (Table 2, Fig. 2).
tively. This result shows that these traits were signifi-
For AC, one QTL and three pairs of QTLs were detected to have additive and epistatic effect, respectively. The QTL, QAc6, on chromosome 6 had a positive effect of 1.43 and explained 19.8% of the total variation. One pair of QTLs, QAc8 and QAc9, had a positive effect of 0.17 and explained 25.66% of the total variation. High Q × E interaction contributions of 47.86% for QAc6 and 32.54% for QAc6 and QAc11b were observed. All these results indicated that AC was affected greatly by upland environment. For GT, four QTLs and two pairs of QTLs were detected. The QTL QGt6b, had a high LOD score of 9.96 and a high general contribution of 17.24%. QGt3 had high Q × E interaction of 8.99%. One pair of QTLs, QGt1a and QGt1b,
cantly affected by the watering regime of the field,
had a negative effect of −0.76. Another pair of
but these were mainly controlled by genetic factors.
QTLs, QGt2 and QGt6a, had a positive effect of 0.56. Significant contribution to the phenotypic variation was detected in all these QTLs. One QTL and one pair of QTLs were detected for GC. The
lowland environments for these traits were determined. The correlation coefficients were 0.474, 0.253, 0.608, and 0.322 for AC, GC, GT, and PC, respec-
2. 2
Identification of QTLs for AC, GC, GT, and PC The polymorphism of markers was analyzed
QTL, QGc1, had a negative effect of −4.55 and
based on all RFLP markers and SSR markers. Among
explained 4.69% of the phenotypic variation. QGc7 and QGc12 had high epistatic effect of 6.08
the 220 RFLP markers and 216 SSR markers anawww.jgenetgenomics.org
424
Journal of Genetics and Genomics
遗传学报 Vol.34 No.5 2007
Table 2 Additive, epistatic, and QTL × environment interaction effects of QTLs associated with the cooking and nutrient quality traits in rice Trait
AC
GT
GC
PC
QTLa)
Interval
QTLa)
Interval
LOD
A/AA b)
H2A/A A%c)
AE1/ A AE1d)
AE2/ AAE2 d)
H2AE/ AAE%c)
QAc3
RM60-C814
QAc11a
RM202-RM287
4.89
0.615
2.43
−0.047
0.047
0.03
QAc6
C1004-R1962
QAc11b
G181-G320
11.46
−0.494
1.57
−0.016
0.016
32.54
QAc8
R2676-C166
QAc9
R79-R2638
4.61
0.173
25.66
−0.214
0.214
11.82
QAc6
C1004-R1962
8.61
1.429
19.77
1.572
QGt1a
C161A-RM243
QGt1b
C813-C955
QGt2
RM208-RM48
QGt6a
G2028-RM276
6.47
−0.756
5.53
−0.032
0.032
0.04
0.556
2.99
0.175
−0.175
0.31
−0.487
0.487
8.99
QGt3
RM231-RM175
6.54
0.547
5.67
RM253-RM314
9.96
0.954
17.24
QGt8
R202-R2676
4.89
−0.630
7.52
QGt11
G181-G320
3.5
0.606
6.96
QGc7
RM47-RM172
QGc1
RM259-RM84
QPc7
RM214-OSR22
QPc2
G1327-RM263
QPc12
a)
N869-R1709
RM270-C751
47.86
11.41
QGt6b
QGc12
−1.52
8.81
−6.083
8.31
4.69
−4.552
4.6
8.39
−0.638
7.05
4.28
0.320
1.91
[24] b)
QTL nomenclature followed that of McCouch et al. . A is the additive effect of QTL. The negative value indicates that the allele has a negative effect on the trait. AA is the effect of additive by additive interaction between two points. The negative value indicates that the parental two-locus genotypes have positive effect on the trait and the recombinants have negative effect. c) H2A/AA and H2AE/AAE are the contributions from the additive effect QTL, the epistatic effect QTL, the additive effect of QTL × environment interactions, and the epistatic effect of QTL × environment interactions, respectively. d) E1 and E2 represent the lowland and upland environments, respectively.
and explained 8.31% of the phenotypic variation. Also, one QTL and one pair of QTLs were detected for PC. QPc2 explained only 1.91% of the total phenotypic variation. QPc7 and QPc12 had epistatic effect of 0.64 and explained 7.05% of the phenotypic variation. 2. 3
Q × E interactions for grain cooking and nutrient quality traits
Combined analysis of multi-environment phenotypic values could directly show the degree of Q × E interactions and improve the precision of QTL detection. Using this approach, the effects and contributions of Q × E interactions for cooking and nutrient quality traits of rice under upland and lowland environments were closely examined in this study. The
degree of Q × E interactions for the grain quality QTLs detected in this study were quite different. For AC, there was a 47.86% contribution of Q × E to a QTL (QAc6) and 32.54% and 11.82% Q × E contributions to 2 QTLs, respectively (Table 2). A QTL, QGt3 showed 8.99% general contribution of Q × E interactions. Although the main effect of some of these QTLs had high general contribution, it is difficult to use these QTLs as markers in marker assisted selection in grain quality breeding because of the high degree of Q × E interactions. In contrast, QTLs such as QGt6b and QGt8 had relatively high general contributions and no significant Q × E interactions, thus becoming good candidates for molecular markers in the selection of better grain cooking and nutrient qualities.
www.jgenetgenomics.org
Yongmei Guo et al.: QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice under…
Fig. 2
Mapping locations of QTLs associated with the cooking and nutrient quality traits in rice
www.jgenetgenomics.org
425
426
3 3. 1
Journal of Genetics and Genomics
Discussion Effect on grain quality of upland environment
The area of irrigated rice was reduced in China recently because of water shortage [25,26]. Upland rice cultivation is a non-flooded alternative to lowland rice cultivation, which can save up to 50%-70% of irrigation water [1]. However, the total acreage of upland rice is still very small, mainly because of the lower yield level and the lower grain quality of rice varieties that are adapted for the upland cultivation system. Therefore, upland varieties with improved yield level and quality traits were urgently needed. Previous studies[4–8,11,14] showed that the rice grain quality traits such as AC, GC, and PC were easily influenced by different environmental factors such as field location, temperature, and solar radiation. However, there is a lack of study on the effects on the grain quality of upland environment. In the present study, the effect of upland cultivation on the grain cooking and nutrient qualities has been examined. The result indicated that both the cooking and nutrient qualities generally varied in the DH lines and their parents under upland environment. These results pointed to a great opportunity in breeding both drought resistance and high grain quality into rice varieties. 3. 2
Comparison with previous studies on QTLs for grain cooking and nutrient qualities
The inheritance of the grain cooking quality was more complicated than that of the other agronomic traits owing to the triploid endosperm, maternal, and cytoplasmic effects. Environment factors such as cultivation regimes and field conditions also have great impact on the nutrient quality of rice [27,28]. All these factors resulted in difficulties in mapping QTLs for these traits. In the present study, the use of DH population will simplify the analysis of phenotypic variation and the QTL mapping of the cooking quality traits regardless of the possible maternal and cytoplasmic effects. Moreover, QTL analysis was con-
遗传学报 Vol.34 No.5 2007
ducted using mixed liner model approaches that can improve the precision of QTL detection. As a result, more information on QTL and Q × E interactions for rice cooking quality traits was obtained in the present study. The comparison of the present study with the previous studies on QTLs for grain cooking and nutrient qualities was conducted. Previously, AC was shown to be controlled by both major and minor QTLs. Most of these studies detected a major QTL, which located near the wx gene (tightly linked with marker R1962) on chromosome 6 and had an average general contribution of 72.9% to the phenotypic variation[3, 4, 6–8, 10–14]. In this study, a QTL near R1962 was also detected as on chromosome 6 (QAc6), which accounted for 19.77% of the total variation and had high effect of Q × E interaction. Therefore, the QTL, QAc6, was probably the wx gene. The relatively lower general contribution of QAc6 (19.77%) than in the previous studies (72.9%) was mainly caused by the combined analysis between upland and lowland environments, which dissected the total contribution as the main effect contribution (19.77%) and the Q × E interaction contribution (47.86%). The result indicated that although AC was easily affected by upland and lowland environments, it was mainly controlled by a major QTL. Additionally, three pairs of epistatic QTLs were also detected in the present study, indicating the complexity of the genetics of AC. The QTL effects of GC varied in different studies. Huang et al. [6] concluded that GC was controlled by both major QTLs and minor QTLs. The major QTL was detected on chromosome 2, 6, and 7. Other studies [3, 4, 14] detected a GC QTL at the wx locus or tightly linked with wx, which had an average general contribution of 74.4% of the phenotypic variation. In contrast, Bao et al. [5], Wu et al. [11], and Li et al.[13] believed that GC was mainly controlled by two major QTLs on chromosome 2 and 7. In the present study, a QTL on chromosome 1 and two pairs of QTLs were detected. The different results of GC QTLs from these studies suggested that GC was largely affected by environwww.jgenetgenomics.org
Yongmei Guo et al.: QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice under…
ment. With an average general contribution of 78.4%, a common QTL near the gene alk on chromosome 6 for GT was detected[3, 4, 7, 12, 14]. This QTL was also detected in addition to 3 other QTLs. The number of QTLs for PC was very low and no common QTLs were found[4, 11, 12, 28, 29]. The QTL QPc2 detected in this study was not found in the previous studies. Most of the QTL studies for rice grain cooking and nutrient qualities were conducted based on the additive-dominant model and usually assumed absence of epistasis among QTLs. However, epistasis between noallelic QTLs for some complicated quantitative traits may exist. In the present study, both additive and epistatic effects of QTLs were obtained based on mixed linear model approaches. It was found that epistasis was detected in all grain cooking and nutrient quality traits (Table 2). Furthermore, epistasis was an important component for AC. This result indicated the complexity of the genetics of rice grain cooking and nutrient quality traits. References 1
Li ZC, Mu P, Li CP, Zhang HL, Li ZK, Gao YM, Wang XK．QTL mapping of root traits in a doubled haploid population from a cross between upland and lowland japonica rice in three environments. Theor Appl Genet, 2005, 110(7):
2
1244−1252. Shi CH, Zhu J, Zang RC, Chen GL. Genetic and heterosis analysis for cooking quality traits of indica rice in different
3
environments. Theor Appl Genet, 1997, 95(2): 294−300. He P, Li SG, Qian Q, Ma YQ, Li JZ, Wang WM, Chen Y, Zhu LH. Genetic analysis of rice grain quality. Theor Appl Genet,
4
5
6
1999, 98(3-4): 502−508. Tan YF, Li JX, Yu SB, Xing YZ, Xu CG, Zhong Q.The three important trait for cooking and eating quality of rice grains are controlled by a single locus in an elite rice hybrid, Shanyou 63. Theor Appl Genet, 1999, 99(3-4): 642−648. Bao JS, Zheng XW, Xia YW, He P, Shu QY, Lu X, Chen Y, Zhu LH. QTL mapping for the paste viscosity charactistics in rice (Oryza sativa L.). Theor Appl Genet, 2000, 100(2): 280−284. Huang ZL, Tan XL, Tragoonrung S, Vanavichit A. Mapping QTLs for amylase content of grain with molecular markers in rice (Oryza sativa L.). Acta Agro Sin, 2000, 26(6): 777−782 (in Chinese with an English abstract).
www.jgenetgenomics.org
7
427
Lanceras JC, Huang ZL, Naivikul O, Vanavichit A, Ruanjaichon V, Tragoonrung S. Mapping of genes for cooking and eating qualities in Thai jasmine rice (KDML105). DNA Res,
8
2000, 28(1): 93−101. Bao JS, Wu YR, Hu B. QTL for rice grain quality base on a DH population derived from parents with similar apparent
9
amylase content. Euphytica, 2002, 128(3): 317−324. Bao JS, Harold C, He P, Zhu LH. Analysis of quantitative trait loci for starch properties of rice based on an RIL popula-
tion. Acta Botanica Sinica, 2003, 45(8): 986−994. 10 Septiningsih EM, Trijatmiko KR, Moeljopawiro S, McCouch SR. Identification of quantitative trait loci for grain quality in an advanced backcross population derived from Oryza stativa variety IR64 and the wild relative O. rufipogon. Theor Appl Genet, 2003, 107(7): 1433−1441. 11 Wu CM, Sun CQ, Fu XL, Wang XK, Li ZC, Zhang Q. Study on the relationship between quality, yield characters or indica-japonica differentiation in rice. Acta Agro Sin, 2003, 29(6): 822−828 (in Chinese with an English abstract). 12 Aluko G, Martinez C, Tohme J, Castano C, Bergman C, Oard JH. QTL mapping of grain quality traits from the interspecific cross Oryza stativa× O. glaberrima. Theor Appl Genet, 2004, 109(3): 630−639. 13 Li JM, Xiao JH, Grandillo S. QTL detection for rice grain quality traits using an interspecific backcross population derived from cultivated Asian (O. sativa L.) and African (O. glaberrium S) rice. Genome, 2004, 47(4): 697−704. 14 Tian R, Jing GH, Shen LH, Wang LQ, He YQ. Mapping quantitative trait loci underlying the cooking and eating quality of rice using a DH population. Mol Breed, 2005, 15(2 ): 117−124. 15 Bao JS, He P, Li SG, Xia YW, Chen Y, Zhu LH. Comparative mapping quantitative trait loci controlling the cooking and eating quality of rice (Oryza sativa L.). Sci Agric Sin, 2000, 33(5): 8−13 (in Chinese with an English abstract). 16 Wan XY, Wan JM, Su CC, Wang CM, Shen WB, Li JM, Wang HL, Jiang L, Liu SJ, Chen LM, Yasul H, Yoshlmura A. QTL detection for eating quality of cooked rice in a population of chromosome segment substitution lines. Theor Appl Genet, 2004, 110(1): 71−79. 17 Williams VR, Wu WT, Tsai HY, Bates HG. Varietal differences in amylose content of rice starch. J Agric Food Chem, 1958, 8(1): 47−48. 18 Juliano BO. A simplified assay for milled rice amylose. Cereal Sci Today, 1971, 16: 334−336. 19 Cagampang GB, Perez CM, Juliano BO. A gel consistency test for the eating quality of rice. J Sci Food Agric, 1973, 24(3): 1589−1594. 20 Little RR, Hilder GB, Dawson EH. Differential effect of dilute alkali on 25 varieties of milled white rice. Cereal Chem,
428
Journal of Genetics and Genomics
1958, 35(1): 111−126. 21 Jennings PR, Coffman WR, Kauffman HE. Rice Improvement. International Rice Research Institute, Manila, 1979. 22 Wang DL, Zhu J, Li ZK, Paterson AH. A computer software for mapping Quantitative Trait Loci with main effects, epistatic effects and QTL × Environment interactions User manual for QTLMapper Version1.0. Texas A&M University, College Station, 1999. 23 Mu P, Li ZC, Li CP, Zhang HL, Wang XK. QTL analysis for lodging resistance in rice using a DH population under lowland and upland ecosystems. Acta Genetica Sinica, 2004, 31(7): 717−723 (in Chinese with an English abstract). 24 McCouch SR, Cho YG, Yano M, Paul E, Blinstrub M. Report on QTL nomenclature. Rice Genet Newslett, 1997, 14: 11−13. 25 Yang JC, Wang ZQ, Liu LJ, Lang YZ, Zhu QS. Growth and development characteristics and yield formation of dry-cultivated rice. Acta Agronomica Sinica, 2002, 28(1): 11−17 (in Chinese with an English abstract).
遗传学报 Vol.34 No.5 2007
26 Wang HQ, Bouman BAM, Zhao DL, Wang CG, Moya PF. Aerobic rice in northern China: Opportunities and challenges. In: Bouman BAM, Hengsdijk H, Hardy B, Bindraban, PS, Tuong TP, Ladha JK, eds. Water-wise Rice Production, International Rice Research Institute, Los Baños, (Philippines), 2002, 143−154. 27 Juliano BO, Perez CM, Kaosaad M. Grain quality characteristics of export rice in selected markets. Cereal Chem, 2002, 67(2): 192−197. 28 Li ZF, Wan JM, Xia JF, Yano M. Mapping of quantitative trait loci controlling physico-chemical properties of rice grains (Oryza sativa L.). Breed Sci, 2003, 53: 209−215. 29 Hu ZL, Li P, Zhou MQ, Zhang ZH, Wang LX, Zhu LH, Zhu
YG. Mapping of quantitative trait loci (QTLs) for rice protein and fat content using double haploid line. Euphytica, 2004, 135(1): 47−54.
水、旱条件下稻米蒸煮和营养品质性状与土壤水分环境互作分 析及其 QTL 定位 郭咏梅1,2,*, 穆 平1,*, 刘家富3, 卢义宣2, 李自超1 1. 中国农业大学农业部作物基因组学与遗传改良重点实验室，教育部作物杂种优势利用重点实验室，北京市作物遗传改良 实验室，北京 100094； 2. 云南省农业科学院粮食作物研究所，昆明 650205； 3. 云南省农业科学院农业部质量检验测试中心，昆明 650201 摘 要：以旱稻品种 IRAT109 与水稻品种越富杂交构建的 DH 群体的 116 个株系及其亲本为材料，在水、旱 2 种栽培条件 下种植，研究了稻米蒸煮和营养品质性状的变化规律，在水、旱 2 个土壤水分环境下对直链淀粉含量(AC)、胶稠度(GC)、 碱消值(GT)和蛋白质含量(PC)4 个蒸煮和营养品质性状进行 QTL 定位及 QTLs 与环境互作分析。结果表明，以上 4 个品质 性状在水、旱 2 种不同栽培条件下差异较大，说明这些性状受水分条件影响较大，旱栽条件下稻米蒸煮和营养各品质性状 均有不同程度的升高，其中蛋白质含量平均提高 37.9%。QTL 分析结果表明，4 个稻米品质性状在 2 个环境中的表现型值 都为连续分布，均存在超亲遗传类型，共检测到 7 个加性效应 QTL 与稻米蒸煮和营养品质性状 4 项指标有关，分别位于第 1、2、3、6、8、11 染色体上，单个 QTLs 对性状的贡献率在 1.91% ~ 19.77%之间。位于第 3 染色体上控制碱消值的 QGt 3， 第 6 染色体上控制直链淀粉含量的 QAc6，在 2 个不同土壤水分条件下均与环境存在显著互作，对环境互作的贡献率分别为 8.99%和 47.86%。控制直链淀粉含量的 2 对上位性 QTLs 与土壤水分环境显著互作，贡献率较大，分别为 32.54%和 11.82%。 并筛选到 5 个主效 QTL(QGt6b、QGt8、QGt11、QGc1 和 QPc2)在抗旱育种中可用于蒸煮和营养各品质性状 MAS 改良。 关键词:
旱稻；蒸煮品质；营养品质；QTL 定位；QTL与环境互作
作者简介： 郭咏梅 (1972－)，女，云南人，硕士，助理研究员。研究方向：水稻遗传育种。E-mail:[email protected]。 穆 平 (1971－)，男，山东人，博士，副教授。研究方向：水稻遗传育种。现工作单位: 莱阳农学院。 E-mail: [email protected] *表示同等贡献作者。
www.jgenetgenomics.org
Research Article
QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice Under Upland and Lowland Environments Yongmei Guo1, 2, *, Ping Mu1, *, Jiafu Liu3, Yixuan Lu2, Zichao Li 1,① 1. Key Laboratory of Crop Genomics and Genetic Improvement of Ministry of Agriculture, Key Laboratory of Crop Heterosis and Utilization of Ministry of Education, Beijing Laboratory of Crop Genetic Improvement, China Agricultural University, Beijing 100094, China; 2. Institute of Food Crop Research, Yunnan Academy of Agricultural Sciences, Kunming 650205, China; 3. Quality and Test Center of Agricultural Administration, Yunnan Academy of Agricultural Sciences, Kunming 650201, China
Abstract: Grain cooking and nutrient qualities are the most important components of rice (Oryza sativa L.) quality. A doubled haploid (DH) population from a cross between two japonica cultivars was used to examine the phenotypic values and potential QTLs for the quality traits. The cooking and nutrient quality traits, including the amylose content (AC), the gel consistency (GC), the gelatinization temperature (GT), and the protein content (PC), in rice grown under upland and lowland environments were evaluated. Significant differences for AC, GC, GT, and PC between upland and lowland environments were detected. The phenotypic values of all four traits were higher under upland environment than lowland environment. The value of PC under upland environment was significantly higher (by 37.9%) than that under lowland environment. This suggests that upland cultivation had large effect on both cooking and nutrient qualities. A total of seven QTLs and twelve pairs of QTLs were detected to have significant additive and epistatic effects for the four traits. Significant Q × E interaction effects of two QTLs and two pairs of QTLs were also discovered. The general contribution of additive QTLs ranged from 1.91% to 19.77%. The Q × E interactions of QTLs QGt3 and QAc6 accounted for 8.99% and 47.86% of the phenotypic variation, respectively, whereas those of the 2 pairs of epistatic QTLs, QAc6-QAc11b and QAc8-QAc9, accounted for 32.54% and 11.82%, respectively. Five QTLs QGt6b, QGt8, QGt11, QGc1, and QPc2, which had relatively high general contribution and no Q × E interactions, were selected to facilitate the upland rice grain quality breeding. Keywords: upland rice; cooking quality; nutrient quality; QTL mapping; Q × E interaction effects
Water shortage has been an ever-increasing problem and the largest constraint to the development of agriculture in China. Severe water shortage has also led to a rapid decrease of irrigated rice (Oryza sativa L.) production especially in northern China. It was reported that the area of irrigated rice production has reduced by 20% (about one million hectare) in the
north, northwest, and northeast China in the past five years. Upland rice culture is a non-flooded alternative to lowland rice culture, which can save up to 50%−70% of irrigation water [1]. Owing to the large difference between upland and lowland cultural conditions, the rice quality may be affected under upland environment. Few reports on the variations of rice
Received: 2006−10−23; Accepted: 2006−11−30 This work was supported by the State Key Basic Research and Development Plan of China (973), the Hi-Tech Research and Development Program of China (863) and National Natural Science Foundation of China. * These authors contributed equally to this work. ① Corresponding author. E-mail: [email protected]; Tel&Fax: +86-10-6273 1414 www.jgenetgenomics.org
Yongmei Guo et al.: QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice under…
grain quality under upland environment are available. A study was initiated to investigate the variation and genetic patterns of rice quality traits under upland cultural conditions. Grain cooking quality including amylose content (AC), gel consistency (GC), gelatinization temperature (GT), and nutrient quality including protein content (PC) are the most important components for the grain quality in rice. The inheritance of these traits has been widely studied [2−14]. The grain cooking and nutrient qualities were both controlled by QTLs with major and minor effects. Some references only provided the presence of QTLs by comparing the different kinds of QTL mapping backgrounds[15, 16]. Thus, the exact degree of Q × E interactions can not be obtained. The objective of the present study was to analyze the effect of water stress on rice grain cooking and nutrient qualities, to locate QTLs and Q × E interactions affecting these traits under upland and lowland conditions using a doubled haploid (DH) population derived from an upland and lowland rice cross, and to provide molecular markers for grain quality breeding.
1 1. 1
Materials and Methods Materials and field conditions
A population of 116 DH lines derived from a cross between Yuefu (a Japonica lowland variety) and IRAT109 (a Japonica upland variety) was used in this study. Both lowland (flooded anaerobic soil) and upland (aerobic soil) field conditions were used in this experiment. Doubled haploid lines and the parents were planted in a randomized complete design with 2 replications under upland and lowland conditions. To protect the plant in a good growth situation, all seeds were dressed by chemical dressing, which was made of weedicide and pesticide. Seeds of each DH line were sowed directly in lowland and upland fields in early May, 2002, in Beijing. The normal fertilizer in the form of N, P2O5, and K2O in the amount of 150 kg www.jgenetgenomics.org
421
each was applied prior to plantation. Thirty plants of each DH line were grown in single rows that were spaced 30 cm apart with 5 cm space between plants. The total water input (irrigation plus rainfall) was about 1,400 mm in lowland experiment. Plants in the upland field encountered severe water stress before June (tillering stage) because of the lack of rain. Supplementary irrigation was supplied when all DH lines had a high degree of leaf rolling at noon in the field. Three times of irrigation approximately 50 mm each were provided during the growth period under upland conditions. The total water input of the upland field was about 500 mm. Three hundred kilograms urea per hectare was applied at the tiller stage to both fields. Mature seeds were collected from 5 plants in the middle row of each DH line and the parents, and then stored at room temperature for three months prior to testing. 1. 2
Trait measurement
Amylose content (AC) was measured as described by Williams et al. [17] and Juliano [18], involving a defatting step, where 95% ethanol was added to the milled rice flour prior to starch dispersion in NaOH, followed by gelatinization in a boiling water bath, and measurement of amylose based on the detection of iodine blue color at pH 4.5 to 4.8 using sodium acetate buffer. Gel consistency (GC) was measured according to the method of Cagampang et al. [19], where the flow characteristic of milled rice gel was measured in 0.2 mol/L KOH and indexed by the length (in mm) of the cold horizontal gel (using a sample of 100 g 100mesh-sieved rice flour). Alkali-spreading score (ASS), indirect measurement for GT, was measured as described by Little et al [20]. Ten milled rice grains from each parent or DH line were immersed in 1.7% potassium hydroxide solution at room temperature for 23 h. The milled rice was carefully separated using forceps, and the spreading value of rice was scored by visual assessment using the method of Jennings et al [21].
422
Journal of Genetics and Genomics
Protein content (PC) was determined by a nitrogen gas analyzer (model 528; LECO). Samples of 1 mg were placed into a quartz combustion tube in an
an average distance of 9.3 cM between adjacent markers [23]. A threshold probability of P < 0.005 was used to declare the existence of QTLs. Correlation analysis between upland and lowland environments for these traits was conducted using the SPSS 10.0 software (http://www.spss.com).
induction furnace at 900℃. Total crude protein was calculated from the nitrogen content of the processed grain where percentage nitrogen × 5.95 = percentage protein. Three replications for each sample for these traits were analyzed. 1. 3
遗传学报 Vol.34 No.5 2007
2
QTL analysis
Results
2.1
QTL analysis including additive effects, epitasis, and QTL × environment interactions effects was carried out by mixed linear model approaches using QTL Mapper version 2.0[22] and a constructed molecular linkage map. The linkage map included 94 RFLP and 71 SSR markers and covered 1,535 cM in length with
Phenotypic values and the frequency distribution for AC, GC, GT, and PC of the DH population
The four parameters of AC, GC, GT, and PC were estimated for the DH lines and the parents grown in upland and lowland environments (Table 1, Fig. 1). The analysis of variance between upland and
Table 1 Variations of cooking and nutrient quality between two parents, IRAT109 and Yufu, and among the DH population under lowland and upland environments Lowland Trait
DH population
IRAT 109
Yuefu
AC% GC/mm
5.8 90.5
GT
2.0
PC%
8.1
Upland
Mean
Range
17.4 52.0
16.21±2.05 63.67±16.19
11.6-26.4
7.0
3.73±1.96
30.0-100.0 2.0-7.0
7.96±0.97
4.9-11.2
8.1
DH population
IRAT 109
Yuefu
15.9 70.0 4.5 10.6
F value
Mean
Range
15.5 40.7
17.10±1.69 78.91±18.95
13.2-23.0
6.5 9.6
Corrlation coefficient
43.0-100.0
4.18** 6.78**
0.474** 0.253**
4.96±1.96
1.5-7.0
7.36**
0.608**
10.98±1.02
8.7-13.9
**
0.322**
27.18
**Significant at 1% level.
Fig.1
Distribution of AC, GC, GT and PC in lowland and upland environments
www.jgenetgenomics.org
Yongmei Guo et al.: QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice under…
lowland environments for the DH population showed that significant differences were detected for AC, GC, GT, and PC between upland and lowland environments. The F values were 4.18, 6.78, 7.36, and 27.18, respectively. This result indicated that AC, GC, GT, and PC were significantly influenced by field conditions. The GC values of the two parents IRAT109 and Yuefu in the upland environment were 22.65% and 21.73% lower than that in the lowland environment. In contrast, PC was 30.86% and 18.51% higher for IRAT109 and Yuefu, respectively, in the same comparison. For AC, Yuefu was 10.92% lower in upland than in lowland environment, whereas IRAT109 showed slight difference under different field conditions. The GT of the two parents responded to the field environments differently; while IRAT109 showed a large increase, Yuefu declined by 7% from lowland to upland fields. Although the differences between IRAT109 and Yuefu were relatively small in AC and PC, a large segregation of the DH population in these traits was observed. For DH population, the means for all four parameters were higher in the upland environment than in the low land environment. The value of these parameters showed an approximately normal distribution and transgressive segregation, indicating that the population was suitable for QTL mapping for these traits. The phenotypic correlations between upland and
423
lyzed, polymorphism was detected for 94 and 71, respectively. The polymorphic rate was 44.8% and 33.2%, respectively. An integrated molecular linkage map consisting of 94 RFLP markers and 71 SSR markers was constructed, which covered 1,535 cM in length with an average distance of 9.3 cM between adjacent markers. The parents of DH lines, Yuefu and IRAT109 were both Japonica variety. Yuefu was from Japan, while IRAT109 originated from Africa. Higher genetic difference was observed between these types of parents. Additive, epistatic, as well as Q × E interaction effects between upland and lowland environments associated with AC, GC, GT, and PC were detected using the phenotypic data and the constructed linkage map (Table 2, Fig. 2).
tively. This result shows that these traits were signifi-
For AC, one QTL and three pairs of QTLs were detected to have additive and epistatic effect, respectively. The QTL, QAc6, on chromosome 6 had a positive effect of 1.43 and explained 19.8% of the total variation. One pair of QTLs, QAc8 and QAc9, had a positive effect of 0.17 and explained 25.66% of the total variation. High Q × E interaction contributions of 47.86% for QAc6 and 32.54% for QAc6 and QAc11b were observed. All these results indicated that AC was affected greatly by upland environment. For GT, four QTLs and two pairs of QTLs were detected. The QTL QGt6b, had a high LOD score of 9.96 and a high general contribution of 17.24%. QGt3 had high Q × E interaction of 8.99%. One pair of QTLs, QGt1a and QGt1b,
cantly affected by the watering regime of the field,
had a negative effect of −0.76. Another pair of
but these were mainly controlled by genetic factors.
QTLs, QGt2 and QGt6a, had a positive effect of 0.56. Significant contribution to the phenotypic variation was detected in all these QTLs. One QTL and one pair of QTLs were detected for GC. The
lowland environments for these traits were determined. The correlation coefficients were 0.474, 0.253, 0.608, and 0.322 for AC, GC, GT, and PC, respec-
2. 2
Identification of QTLs for AC, GC, GT, and PC The polymorphism of markers was analyzed
QTL, QGc1, had a negative effect of −4.55 and
based on all RFLP markers and SSR markers. Among
explained 4.69% of the phenotypic variation. QGc7 and QGc12 had high epistatic effect of 6.08
the 220 RFLP markers and 216 SSR markers anawww.jgenetgenomics.org
424
Journal of Genetics and Genomics
遗传学报 Vol.34 No.5 2007
Table 2 Additive, epistatic, and QTL × environment interaction effects of QTLs associated with the cooking and nutrient quality traits in rice Trait
AC
GT
GC
PC
QTLa)
Interval
QTLa)
Interval
LOD
A/AA b)
H2A/A A%c)
AE1/ A AE1d)
AE2/ AAE2 d)
H2AE/ AAE%c)
QAc3
RM60-C814
QAc11a
RM202-RM287
4.89
0.615
2.43
−0.047
0.047
0.03
QAc6
C1004-R1962
QAc11b
G181-G320
11.46
−0.494
1.57
−0.016
0.016
32.54
QAc8
R2676-C166
QAc9
R79-R2638
4.61
0.173
25.66
−0.214
0.214
11.82
QAc6
C1004-R1962
8.61
1.429
19.77
1.572
QGt1a
C161A-RM243
QGt1b
C813-C955
QGt2
RM208-RM48
QGt6a
G2028-RM276
6.47
−0.756
5.53
−0.032
0.032
0.04
0.556
2.99
0.175
−0.175
0.31
−0.487
0.487
8.99
QGt3
RM231-RM175
6.54
0.547
5.67
RM253-RM314
9.96
0.954
17.24
QGt8
R202-R2676
4.89
−0.630
7.52
QGt11
G181-G320
3.5
0.606
6.96
QGc7
RM47-RM172
QGc1
RM259-RM84
QPc7
RM214-OSR22
QPc2
G1327-RM263
QPc12
a)
N869-R1709
RM270-C751
47.86
11.41
QGt6b
QGc12
−1.52
8.81
−6.083
8.31
4.69
−4.552
4.6
8.39
−0.638
7.05
4.28
0.320
1.91
[24] b)
QTL nomenclature followed that of McCouch et al. . A is the additive effect of QTL. The negative value indicates that the allele has a negative effect on the trait. AA is the effect of additive by additive interaction between two points. The negative value indicates that the parental two-locus genotypes have positive effect on the trait and the recombinants have negative effect. c) H2A/AA and H2AE/AAE are the contributions from the additive effect QTL, the epistatic effect QTL, the additive effect of QTL × environment interactions, and the epistatic effect of QTL × environment interactions, respectively. d) E1 and E2 represent the lowland and upland environments, respectively.
and explained 8.31% of the phenotypic variation. Also, one QTL and one pair of QTLs were detected for PC. QPc2 explained only 1.91% of the total phenotypic variation. QPc7 and QPc12 had epistatic effect of 0.64 and explained 7.05% of the phenotypic variation. 2. 3
Q × E interactions for grain cooking and nutrient quality traits
Combined analysis of multi-environment phenotypic values could directly show the degree of Q × E interactions and improve the precision of QTL detection. Using this approach, the effects and contributions of Q × E interactions for cooking and nutrient quality traits of rice under upland and lowland environments were closely examined in this study. The
degree of Q × E interactions for the grain quality QTLs detected in this study were quite different. For AC, there was a 47.86% contribution of Q × E to a QTL (QAc6) and 32.54% and 11.82% Q × E contributions to 2 QTLs, respectively (Table 2). A QTL, QGt3 showed 8.99% general contribution of Q × E interactions. Although the main effect of some of these QTLs had high general contribution, it is difficult to use these QTLs as markers in marker assisted selection in grain quality breeding because of the high degree of Q × E interactions. In contrast, QTLs such as QGt6b and QGt8 had relatively high general contributions and no significant Q × E interactions, thus becoming good candidates for molecular markers in the selection of better grain cooking and nutrient qualities.
www.jgenetgenomics.org
Yongmei Guo et al.: QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice under…
Fig. 2
Mapping locations of QTLs associated with the cooking and nutrient quality traits in rice
www.jgenetgenomics.org
425
426
3 3. 1
Journal of Genetics and Genomics
Discussion Effect on grain quality of upland environment
The area of irrigated rice was reduced in China recently because of water shortage [25,26]. Upland rice cultivation is a non-flooded alternative to lowland rice cultivation, which can save up to 50%-70% of irrigation water [1]. However, the total acreage of upland rice is still very small, mainly because of the lower yield level and the lower grain quality of rice varieties that are adapted for the upland cultivation system. Therefore, upland varieties with improved yield level and quality traits were urgently needed. Previous studies[4–8,11,14] showed that the rice grain quality traits such as AC, GC, and PC were easily influenced by different environmental factors such as field location, temperature, and solar radiation. However, there is a lack of study on the effects on the grain quality of upland environment. In the present study, the effect of upland cultivation on the grain cooking and nutrient qualities has been examined. The result indicated that both the cooking and nutrient qualities generally varied in the DH lines and their parents under upland environment. These results pointed to a great opportunity in breeding both drought resistance and high grain quality into rice varieties. 3. 2
Comparison with previous studies on QTLs for grain cooking and nutrient qualities
The inheritance of the grain cooking quality was more complicated than that of the other agronomic traits owing to the triploid endosperm, maternal, and cytoplasmic effects. Environment factors such as cultivation regimes and field conditions also have great impact on the nutrient quality of rice [27,28]. All these factors resulted in difficulties in mapping QTLs for these traits. In the present study, the use of DH population will simplify the analysis of phenotypic variation and the QTL mapping of the cooking quality traits regardless of the possible maternal and cytoplasmic effects. Moreover, QTL analysis was con-
遗传学报 Vol.34 No.5 2007
ducted using mixed liner model approaches that can improve the precision of QTL detection. As a result, more information on QTL and Q × E interactions for rice cooking quality traits was obtained in the present study. The comparison of the present study with the previous studies on QTLs for grain cooking and nutrient qualities was conducted. Previously, AC was shown to be controlled by both major and minor QTLs. Most of these studies detected a major QTL, which located near the wx gene (tightly linked with marker R1962) on chromosome 6 and had an average general contribution of 72.9% to the phenotypic variation[3, 4, 6–8, 10–14]. In this study, a QTL near R1962 was also detected as on chromosome 6 (QAc6), which accounted for 19.77% of the total variation and had high effect of Q × E interaction. Therefore, the QTL, QAc6, was probably the wx gene. The relatively lower general contribution of QAc6 (19.77%) than in the previous studies (72.9%) was mainly caused by the combined analysis between upland and lowland environments, which dissected the total contribution as the main effect contribution (19.77%) and the Q × E interaction contribution (47.86%). The result indicated that although AC was easily affected by upland and lowland environments, it was mainly controlled by a major QTL. Additionally, three pairs of epistatic QTLs were also detected in the present study, indicating the complexity of the genetics of AC. The QTL effects of GC varied in different studies. Huang et al. [6] concluded that GC was controlled by both major QTLs and minor QTLs. The major QTL was detected on chromosome 2, 6, and 7. Other studies [3, 4, 14] detected a GC QTL at the wx locus or tightly linked with wx, which had an average general contribution of 74.4% of the phenotypic variation. In contrast, Bao et al. [5], Wu et al. [11], and Li et al.[13] believed that GC was mainly controlled by two major QTLs on chromosome 2 and 7. In the present study, a QTL on chromosome 1 and two pairs of QTLs were detected. The different results of GC QTLs from these studies suggested that GC was largely affected by environwww.jgenetgenomics.org
Yongmei Guo et al.: QTL Mapping and Q×E Interactions of Grain Cooking and Nutrient Qualities in Rice under…
ment. With an average general contribution of 78.4%, a common QTL near the gene alk on chromosome 6 for GT was detected[3, 4, 7, 12, 14]. This QTL was also detected in addition to 3 other QTLs. The number of QTLs for PC was very low and no common QTLs were found[4, 11, 12, 28, 29]. The QTL QPc2 detected in this study was not found in the previous studies. Most of the QTL studies for rice grain cooking and nutrient qualities were conducted based on the additive-dominant model and usually assumed absence of epistasis among QTLs. However, epistasis between noallelic QTLs for some complicated quantitative traits may exist. In the present study, both additive and epistatic effects of QTLs were obtained based on mixed linear model approaches. It was found that epistasis was detected in all grain cooking and nutrient quality traits (Table 2). Furthermore, epistasis was an important component for AC. This result indicated the complexity of the genetics of rice grain cooking and nutrient quality traits. References 1
Li ZC, Mu P, Li CP, Zhang HL, Li ZK, Gao YM, Wang XK．QTL mapping of root traits in a doubled haploid population from a cross between upland and lowland japonica rice in three environments. Theor Appl Genet, 2005, 110(7):
2
1244−1252. Shi CH, Zhu J, Zang RC, Chen GL. Genetic and heterosis analysis for cooking quality traits of indica rice in different
3
environments. Theor Appl Genet, 1997, 95(2): 294−300. He P, Li SG, Qian Q, Ma YQ, Li JZ, Wang WM, Chen Y, Zhu LH. Genetic analysis of rice grain quality. Theor Appl Genet,
4
5
6
1999, 98(3-4): 502−508. Tan YF, Li JX, Yu SB, Xing YZ, Xu CG, Zhong Q.The three important trait for cooking and eating quality of rice grains are controlled by a single locus in an elite rice hybrid, Shanyou 63. Theor Appl Genet, 1999, 99(3-4): 642−648. Bao JS, Zheng XW, Xia YW, He P, Shu QY, Lu X, Chen Y, Zhu LH. QTL mapping for the paste viscosity charactistics in rice (Oryza sativa L.). Theor Appl Genet, 2000, 100(2): 280−284. Huang ZL, Tan XL, Tragoonrung S, Vanavichit A. Mapping QTLs for amylase content of grain with molecular markers in rice (Oryza sativa L.). Acta Agro Sin, 2000, 26(6): 777−782 (in Chinese with an English abstract).
www.jgenetgenomics.org
7
427
Lanceras JC, Huang ZL, Naivikul O, Vanavichit A, Ruanjaichon V, Tragoonrung S. Mapping of genes for cooking and eating qualities in Thai jasmine rice (KDML105). DNA Res,
8
2000, 28(1): 93−101. Bao JS, Wu YR, Hu B. QTL for rice grain quality base on a DH population derived from parents with similar apparent
9
amylase content. Euphytica, 2002, 128(3): 317−324. Bao JS, Harold C, He P, Zhu LH. Analysis of quantitative trait loci for starch properties of rice based on an RIL popula-
tion. Acta Botanica Sinica, 2003, 45(8): 986−994. 10 Septiningsih EM, Trijatmiko KR, Moeljopawiro S, McCouch SR. Identification of quantitative trait loci for grain quality in an advanced backcross population derived from Oryza stativa variety IR64 and the wild relative O. rufipogon. Theor Appl Genet, 2003, 107(7): 1433−1441. 11 Wu CM, Sun CQ, Fu XL, Wang XK, Li ZC, Zhang Q. Study on the relationship between quality, yield characters or indica-japonica differentiation in rice. Acta Agro Sin, 2003, 29(6): 822−828 (in Chinese with an English abstract). 12 Aluko G, Martinez C, Tohme J, Castano C, Bergman C, Oard JH. QTL mapping of grain quality traits from the interspecific cross Oryza stativa× O. glaberrima. Theor Appl Genet, 2004, 109(3): 630−639. 13 Li JM, Xiao JH, Grandillo S. QTL detection for rice grain quality traits using an interspecific backcross population derived from cultivated Asian (O. sativa L.) and African (O. glaberrium S) rice. Genome, 2004, 47(4): 697−704. 14 Tian R, Jing GH, Shen LH, Wang LQ, He YQ. Mapping quantitative trait loci underlying the cooking and eating quality of rice using a DH population. Mol Breed, 2005, 15(2 ): 117−124. 15 Bao JS, He P, Li SG, Xia YW, Chen Y, Zhu LH. Comparative mapping quantitative trait loci controlling the cooking and eating quality of rice (Oryza sativa L.). Sci Agric Sin, 2000, 33(5): 8−13 (in Chinese with an English abstract). 16 Wan XY, Wan JM, Su CC, Wang CM, Shen WB, Li JM, Wang HL, Jiang L, Liu SJ, Chen LM, Yasul H, Yoshlmura A. QTL detection for eating quality of cooked rice in a population of chromosome segment substitution lines. Theor Appl Genet, 2004, 110(1): 71−79. 17 Williams VR, Wu WT, Tsai HY, Bates HG. Varietal differences in amylose content of rice starch. J Agric Food Chem, 1958, 8(1): 47−48. 18 Juliano BO. A simplified assay for milled rice amylose. Cereal Sci Today, 1971, 16: 334−336. 19 Cagampang GB, Perez CM, Juliano BO. A gel consistency test for the eating quality of rice. J Sci Food Agric, 1973, 24(3): 1589−1594. 20 Little RR, Hilder GB, Dawson EH. Differential effect of dilute alkali on 25 varieties of milled white rice. Cereal Chem,
428
Journal of Genetics and Genomics
1958, 35(1): 111−126. 21 Jennings PR, Coffman WR, Kauffman HE. Rice Improvement. International Rice Research Institute, Manila, 1979. 22 Wang DL, Zhu J, Li ZK, Paterson AH. A computer software for mapping Quantitative Trait Loci with main effects, epistatic effects and QTL × Environment interactions User manual for QTLMapper Version1.0. Texas A&M University, College Station, 1999. 23 Mu P, Li ZC, Li CP, Zhang HL, Wang XK. QTL analysis for lodging resistance in rice using a DH population under lowland and upland ecosystems. Acta Genetica Sinica, 2004, 31(7): 717−723 (in Chinese with an English abstract). 24 McCouch SR, Cho YG, Yano M, Paul E, Blinstrub M. Report on QTL nomenclature. Rice Genet Newslett, 1997, 14: 11−13. 25 Yang JC, Wang ZQ, Liu LJ, Lang YZ, Zhu QS. Growth and development characteristics and yield formation of dry-cultivated rice. Acta Agronomica Sinica, 2002, 28(1): 11−17 (in Chinese with an English abstract).
遗传学报 Vol.34 No.5 2007
26 Wang HQ, Bouman BAM, Zhao DL, Wang CG, Moya PF. Aerobic rice in northern China: Opportunities and challenges. In: Bouman BAM, Hengsdijk H, Hardy B, Bindraban, PS, Tuong TP, Ladha JK, eds. Water-wise Rice Production, International Rice Research Institute, Los Baños, (Philippines), 2002, 143−154. 27 Juliano BO, Perez CM, Kaosaad M. Grain quality characteristics of export rice in selected markets. Cereal Chem, 2002, 67(2): 192−197. 28 Li ZF, Wan JM, Xia JF, Yano M. Mapping of quantitative trait loci controlling physico-chemical properties of rice grains (Oryza sativa L.). Breed Sci, 2003, 53: 209−215. 29 Hu ZL, Li P, Zhou MQ, Zhang ZH, Wang LX, Zhu LH, Zhu
YG. Mapping of quantitative trait loci (QTLs) for rice protein and fat content using double haploid line. Euphytica, 2004, 135(1): 47−54.
水、旱条件下稻米蒸煮和营养品质性状与土壤水分环境互作分 析及其 QTL 定位 郭咏梅1,2,*, 穆 平1,*, 刘家富3, 卢义宣2, 李自超1 1. 中国农业大学农业部作物基因组学与遗传改良重点实验室，教育部作物杂种优势利用重点实验室，北京市作物遗传改良 实验室，北京 100094； 2. 云南省农业科学院粮食作物研究所，昆明 650205； 3. 云南省农业科学院农业部质量检验测试中心，昆明 650201 摘 要：以旱稻品种 IRAT109 与水稻品种越富杂交构建的 DH 群体的 116 个株系及其亲本为材料，在水、旱 2 种栽培条件 下种植，研究了稻米蒸煮和营养品质性状的变化规律，在水、旱 2 个土壤水分环境下对直链淀粉含量(AC)、胶稠度(GC)、 碱消值(GT)和蛋白质含量(PC)4 个蒸煮和营养品质性状进行 QTL 定位及 QTLs 与环境互作分析。结果表明，以上 4 个品质 性状在水、旱 2 种不同栽培条件下差异较大，说明这些性状受水分条件影响较大，旱栽条件下稻米蒸煮和营养各品质性状 均有不同程度的升高，其中蛋白质含量平均提高 37.9%。QTL 分析结果表明，4 个稻米品质性状在 2 个环境中的表现型值 都为连续分布，均存在超亲遗传类型，共检测到 7 个加性效应 QTL 与稻米蒸煮和营养品质性状 4 项指标有关，分别位于第 1、2、3、6、8、11 染色体上，单个 QTLs 对性状的贡献率在 1.91% ~ 19.77%之间。位于第 3 染色体上控制碱消值的 QGt 3， 第 6 染色体上控制直链淀粉含量的 QAc6，在 2 个不同土壤水分条件下均与环境存在显著互作，对环境互作的贡献率分别为 8.99%和 47.86%。控制直链淀粉含量的 2 对上位性 QTLs 与土壤水分环境显著互作，贡献率较大，分别为 32.54%和 11.82%。 并筛选到 5 个主效 QTL(QGt6b、QGt8、QGt11、QGc1 和 QPc2)在抗旱育种中可用于蒸煮和营养各品质性状 MAS 改良。 关键词:
旱稻；蒸煮品质；营养品质；QTL 定位；QTL与环境互作
作者简介： 郭咏梅 (1972－)，女，云南人，硕士，助理研究员。研究方向：水稻遗传育种。E-mail:[email protected]。 穆 平 (1971－)，男，山东人，博士，副教授。研究方向：水稻遗传育种。现工作单位: 莱阳农学院。 E-mail: [email protected] *表示同等贡献作者。
www.jgenetgenomics.org