Genetic Identification of Quantitative Trait Loci for Contents of Mineral Nutrients in Rice Grain

Transcription

1 Journal of Integrative Plant Biology 2008 Genetic Identification of Quantitative Trait Loci for Contents of Mineral Nutrients in Rice Grain Ana Luisa Garcia-Oliveira, Lubin Tan, Yongcai Fu and Chuanqing Sun (Department of Plant Genetics and Breeding and State Key Laboratory of Plant Physiology and Biochemistry, China Agricultural University Beijing , China; National Centre for Evaluation of Agricultural Wild Plant (Rice Beijing , China); and Beijing Key Laboratory of Crop Genetic Improvement and Genome of Ministry of Agriculture, Beijing , China) Abstract In present study, Fe, Zn, Mn, Cu, Ca, Mg, P and K contents of 85 introgression lines (ILs) derived from a cross between an elite indica cultivar Teqing and the wild rice (Oryza rufipogon) were measured by inductively coupled argon plasma (ICAP) spectrometry. Substantial variation was observed for all traits and most of the mineral elements were significantly positive correlated or independent except for Fe with Cu. A total of 31 putative quantitative trait loci (QTLs) were detected for these eight mineral elements by single point analysis. Wild rice (O. rufipogon) contributed favorable alleles for most of the QTLs (26 QTLs), and chromosomes 1, 9 and 12 exhibited 14 QTLs (45%) for these traits. One major effect of QTL for zinc content accounted for the largest proportion of phenotypic variation (11% 19%) was detected near the simple sequence repeats marker RM152 on chromosome 8. The co-locations of QTLs for some mineral elements observed in this mapping population suggested the relationship was at a molecular level among these traits and could be helpful for simultaneous improvement of these traits in rice grain by marker assisted selection. Key words: introgression lines; mineral elements; Oryza sativa; Oryza rufipogon; quantitative trait loci. Garcia-Oliveira AL, Tan L, Fu Y, Sun C (2008). Genetic identification of quantitative trait loci for contents of mineral nutrients in rice grain. J. Integr. Plant Biol. doi: /j x Available online at Humans require at least 49 nutrients for their normal growth and development, and the demand for most nutrients are supplied by cereals, particularly rice due to its staple role (Welch and Graham 2004). Among these nutrients, mineral elements play numerous beneficial roles due to their direct or indirect effect in both plant and human metabolism and the deficiencies or insufficient intakes of these nutrients leads to several dysfunctions and diseases in humans. Studies have indicated widespread occurrence of deficiencies for mineral elements such as anemia for iron and osteoporoses for calcium in most developing countries as well as developed countries Received 20 Mar Accepted 15 May 2008 Supported by the Project of Conservation and Utilization of Agricultural Wild Plants of the Ministry of Agriculture of China, the National High-Tech Research and Development ( 863 ) Program of China (2006AA100101), and the National Natural Science Foundation of China ( ). Author for correspondence. Tel (Fax): ; <suncq@cau.edu.cn>. C 2008 Institute of Botany, the Chinese Academy of Sciences doi: /j x (Welch and Graham 1999). The numbers indicate that around two billion people suffer from iron deficiency, while prevalence of zinc deficiency is much harder to quantify due to the lack of a reliable and easy clinical assay (FAO 2004). In addition, other mineral deficiencies such as calcium are also associated with malnutrition and have reached worrying levels with data suggesting that roughly three million people over the age of 50 years suffer from osteoporosis (van Staa et al. 2001). Recent epidemiological studies found that whole-grain intake (such as brown rice), is linked to disease prevention against cancer, cardiovascular disease, diabetes and obesity (Slavin 2003). In the past, much emphasis was placed on the enhancement of yield to increase the availability of food for resource-poor peoples. In the recent past, due to more awareness regarding the importance of mineral elements in human diets, breeders started to pay more attention to the improvement of nutrient qualities of major food grain crops especially mineral elements (Zhang et al. 2004). Many researchers have already studied genetic variation for mineral elements in cereal grains such as rice (Gregorio et al. 2000; Zhang et al. 2004), wheat (Cakmak et al. 2000; Ortiz-Monasterio and Graham 2000; Balint et al. 2001) and maize (Arnold and Bauman 1976; Arnold et al. 1977; Banziger and Long 2000) and reported the narrow genetic base

2 2 Journal of Integrative Plant Biology 2008 for mineral elements especially for iron, zinc, manganese and copper in cultivated rice (Banziger and Long 2000; Beebe et al. 2000; Chavez et al. 2000; Gregorio et al. 2000; Ortiz-Monasterio and Graham 2000; Stangoulis et al. 2007). However, at the molecular level (quantitative trait loci [QTL] analysis) information on cereals is very limited. Recently, Stangoulis et al. (2007) identified some QTLs under controlled environment for Fe, Zn, Mn and P in rice grain using a doubled-haploid population. Several researchers have successfully exploited the wild related species to incorporate the variability in cultivated rice for the improvement of qualitative and quantitative traits of agronomic importance (Brar and Khush 1997). Thus, to overcome the bottleneck for improvement of rice grain quality, use of wild relatives such as Oryza rufipogon can be candidate sources for broadening the genetic base of rice cultivars. In the present study, we mapped QTLs responsible for the accumulation of micro (iron, zinc, manganese and copper) and macro (calcium, magnesium, potassium and phosphorus) mineral elements in rice grain using introgression lines (ILs) under field conditions. Results Trait segregation and field performances The overall mean and range of each trait measured in grains of the 85 ILs grown during both 2005 and 2006 are presented in Table 1. ANOVA showed the significant genotype by environment interactions for these mineral elements. A wide range of variation was observed for levels of all of the mineral elements studied in this set of population (Table 1). Among micro-elements, Zn and Mn were observed in highest quantities with a combined mean value of 27.1 and 16.6 μg/g, respectively, whereas Fe and Cu were found in the lowest quantities with a mean performance of 9.6 and 5.3 μg/g, respectively. Fe and Mn showed high performance during 2005, whereas the reverse trend was observed in the case of Zn and Cu (Table 1). Among macroelements, P and K was observed as predominant elements with mean values of 3010 μg/g and 2297 μg/g, respectively, whereas Mg and Ca was found in the lowest quantities with mean values of 1132 and μg/g, respectively. Among macro-elements, the contribution of P and K was observed highest in 2006, whereas Mg and Ca showed opposite trends. Skewness and Kurtosis were measured to describe the nature of distribution (Table 1). All eight traits had platykurtic distribution (kurtosis value < 3). Heritability and correlation analysis A wide range of heritability in a broad-sense was observed for these elements (Table 2). Heritability estimates for Fe, Zn, Mn, Cu, Ca, Mg, P and K were observed 72.8%, 40.6%, 55.9%, 85.6%, 70.6%, 54.2%, 46.4% and 19.1%, respectively. Microelements exhibited medium to high levels of heritability, whereas very low to high levels of heritability were observed for macroelements. Micro elements exhibited a weak correlation with each other. Only during 2005 and 2006 did Fe content show very weak correlation from negative Cu to positive Zn and Mn, respectively, whereas Zn showed a fragile correlation in a positive direction with Cu and Mn in 2005 and 2006, respectively. In contrast, a highly positive correlation was observed among macro-elements except Ca with K. However, no negative correlation was observed between micro and macro elements (Table 2). QTL detection for mineral elements A total of 31 putative QTLs associated with these mineral elements were detected on all chromosomes except chromosome 7. Out of 31 putative QTLs, only 17 QTLs were observed during both years (Table 3). Chromosomes 1 and 9 had the Table 1. Mean (standard deviation), range, skewness and kurtosis for nutritional traits in 85 rice introgression lines and recipient parent grown during summer 2005 and 2006 Trait Teqing Introgression lines (μg/g DW) Pooled Pooled Range Skewness Kurtosis Micro- Fe ± ± ± element Zn ± ± ± Mn ± ± ± Cu ± ± ± Macro- Ca ± ± ± element Mg ± ± ± P ± ± ± K ± ± ± Ca, calcium; Cu, copper; DW, dry weight; Fe, iron; K, potassium;.mg, magnesium; Mn, manganese; P, phosphorus; Zn, zinc.

3 QTLs for Mineral Nutrients in Rice 3 Table 2. Pearson correlation coefficients and heritability in broad-sense (h 2 B%) among mineral elements measured in 85 introgression lines (ILs) derived from Teqing Oryza rufipogon during 2005 (above diagonal) and 2006 (below diagonal) Traits Micro-element Macro-element h 2 B(%) Fe Zn Mn Cu Ca Mg P K Micro-element Fe Zn Mn Cu Macro-element Ca Mg P K Ca, calcium; Cu, copper; Fe, iron; K, potassium; Mg, magnesium; Mn, manganese; P, phosphorus; Zn, zinc. highest number of QTLs (five QTLs each) for these mineral elements, whereas the lowest number of QTLs was observed on chromosomes 6 and 11 (one QTL each) (Figure 1). At least one QTL and as many as seven QTLs were detected for different traits. The position of QTLs for micro and macro elements along with linked marker, degree and direction of additive effect, probability level and phenotypic variance explained by each QTL detected during 2005 and 2006 were given in Table 3. Micro-elements A total of 10 QTLs were detected for micro elements covering all of the chromosomes except chromosomes 4, 7 and 11. Out of 10 QTLs, six were observed during both years. Iron For iron content, one QTL located on chromosome 2 accounted for 5% and 7% phenotypic variation in 2005 and 2006, respectively. During 2006, a minor QTL was also detected near marker RM296 on chromosome 9. The favorable allele for iron content at chromosome 2 was contributed by O. rufipogon, whereas recipient parent Teqing was accommodated at chromosome 9. Zinc Three QTLs for Zn content were identified on chromosomes 5, 8 and 12. The QTL near marker RM152 on chromosome 8 accounted for the largest proportion of phenotypic variation (11 19%) for Zn content over both years, whereas the QTL that was located on chromosome 12 accounted for 9% phenotypic variation and was detected only during The O. rufipogon alleles enhanced the Zn content at these loci with a range of μg/g. On the other hand a minor QTL for Zn content was contributed by Teqing at chromosome 5 with an additive effect of μg/g. Manganese Four QTLs explaining 5 11% phenotypic variation for Mn content were found on chromosomes 1, 2, 3, and 10. Two of them, located on chromosomes 3 and 10 near markers RM5488 and RM590, respectively, were found in both years with opposing allelic effects. In contrast, QTLs located on chromosomes 1 and 2 were found only during a single year and accounted for 11% and 5% phenotypic variation for Mn content, respectively. The O. rufipogon-derived alleles enhanced Mn content at these loci. Copper Only one QTL accounting for 6% phenotypic variation for Cu content was identified during both years near marker RM204 on chromosome 6 and the favorable allele at this locus was accommodated by O. rufipogon. Macro-elements A total of 21 putative QTLs associated with macro elements were identified on all chromosomes except chromosomes 2, 6 and 7. It is noteworthy that more than 50% of QTLs (11 QTLs) were located only on chromosomes 1, 9 and 12, and the O. rufipogon-derived alleles enhanced macro elements at these loci in the Teqing background. Calcium Seven QTLs were observed for Ca content and O. rufipogon contributed the Ca-enhancing alleles at these QTLs except QTL qca11-1 located near the marker RM202 on chromosome 11. Ca QTLs located on chromosomes 1, 4, 5, 9, 10, 11 and 12 and explained a range of 5 14% total phenotypic variation. Five QTLs, qca1-1, qca4-1, qca5-1, qca9-1 and qca12-1 were detected in both years. QTL, qca1-1, located near marker RM6480 on chromosome 1 exhibited the largest proportion of phenotypic variation (9 14%) for Ca content. Magnesium Five QTLs were detected one each of chromosomes 1, 3, 5, 9 and 12 explaining phenotypic variation from 5 16% for Mg content. Three of these QTLs were detected in a single year only (qmg1-1 and qmg12-1 in 2005, and qmg9-1 in 2006) and the Mg content increasing alleles came from O. rufipogon at these loci. Only two QTLs that were located near markers RM5488 and RM598 exhibited stable performances across the years with opposing allelic effects of Teqing and O. rufipogon, respectively. Phosphorus Five QTLs enhancing P content were found on chromosomes 1, 3, 8, 9 and 12. Only two of them explained >10% phenotypic variation for P in both years, located near markers RM212 and RM201 on chromosomes 1 and 9, respectively. However, three QTLs accounting for 8 9% phenotypic

4 4 Journal of Integrative Plant Biology 2008 Table 3. Quantitative trait loci (QTLs) for micro- and macro-elements detected by single point analysis Traits Locus Marker Chromosome P PV (%) Add a P PV (%) Add a Micro-element Fe qfe2-1 RM qfe9-1 RM Zn qzn5-1 RM qzn8-1 RM qzn12-1 RM Mn qmn1-1 RM qmn2-1 RM qmn3-1 RM qmn10-1 RM Cu qcu6-1 RM Macro-element Ca qca1-1 RM qca4-1 RM qca5-1 RM qca9-1 RM qca10-1 RM qca11-1 RM qca12-1 RM Mg qmg1-1 RM qmg3-1 RM qmg5-1 RM qmg9-1 RM qmg12-1 RM P qp1-1 RM qp3-1 RM qp8-1 RM qp9-1 RM qp12-1 RM K qk1-1 RM qk4-1 RM qk8-1 RM qk9-1 RM a Positive and negative signs of the estimates indicate that Oryza rufipogon and Teqing contributed towards higher value alleles for the corresponding traits, respectively. Ca, calcium; Cu, copper; Fe, iron; K, potassium;.mg, magnesium; Mn, manganese; P, phosphorus; Zn, zinc. variation for P content were also observed during 2005, but these QTLs were not detected in All of the favorable alleles controlling levels of P were derived from the donor parent (O. rufipogon). Potassium Four QTLs one each on chromosomes 1, 4, 8 and 9 were associated with quantitative variation for K content in the present study. Two QTLs (qk4-1 and qk9-1) of them located near markers RM1113 and RM3787 on chromosome 4 and 9, respectively were found in both years, whereas QTLs (qk1-1 and qk8-1) detected near markers RM5 and RM3572 on chromosomes 1 and 8 were observed only during 2006 and 2005, respectively. These QTLs explained a range of 4 14% of total phenotypic variation for levels of K and the favorable alleles at these loci were contributed by O. rufipogon. Discussion The presence of a mineral deficiency in humans is manifested by a wide spectrum of complex systems resulting in less

5 QTLs for Mineral Nutrients in Rice 5 Figure 1. Chromosomal locations of putative quantitative trait loci (QTLs) for contents of mineral elements detected in introgressed lines (ILs) population derived from a cross between indica cultivar Teqing and the wild rice Oryza rufipogon. Note: Symbols indicate the positions of putative QTL for corresponding minerals. attention on systematic breeding efforts for mineral elements in major food crops. Several new functions of mineral elements identified over the past few decades created a new interest in biofortification of major food crops with enhanced levels of mineral elements. Several reports indicated the narrow genetic variability for mineral elements in cultivated rice, whereas a higher level of mineral elements was observed in wild rice O. rufipogon (Cheng et al. 2005). Therefore, the exploration of a new source of genetic variability for the improvement of rice grain quality to the extent of its wild relatives, common wild rice (O. rufipogon) was highly relevant in the present context. Trait variation and interrelationships The mean performances of ILs population were observed to be significantly superior over elite cultivar Teqing for most of the traits except for K content, which had higher levels in Teqing. Thus, the results obtained in the present investigation clearly exhibit the favorable effect of introgressed genetic variability and suggested that inter-specific crosses, especially O. rufipogon had enormous genetic potential for improving the nutritional values of rice cultivars. The information regarding a heritable portion of a trait in any breeding program is highly relevant. Although, heritability estimates are limited to experimental material and setup, and may differ widely in the same crop and same trait (Hill et al. 1998; Holland et al. 2002). Heritability estimates observed for these nutritional traits are in good agreement with results from previous studies in rice (Zhang et al. 2004) and maize (Arnold and Bauman 1976; Arnold et al. 1977). In addition to ANOVA, the lower level of heritability estimates also indicated the strong environmental effect on these traits. In order to develop a breeding selection scheme for simultaneous improvement of multiple traits, the results obtained in the present investigation from interrelationship studies among these nutritionally important mineral elements were encouraging. Most of the traits studied in this experiment were significantly positively correlated or independent except for Fe with Cu (Table 2). This may point to common molecular mechanisms controlling

6 6 Journal of Integrative Plant Biology 2008 the uptake and metabolism of these minerals in grains (Vreugdenhil et al. 2004). QTLs for mineral elements from wild rice Advances in molecular marker technology have resulted in numerous QTL analysis studies for simple agronomic traits in rice, which has led to the identification of harboring genes under these loci. Yet, mineral accumulation in seeds continues to be a perplexing phenomenon that seems to be controlled by poly-genes (Grusak and DellaPenna 1999). However, despite substantial genetic variability for these mineral elements in germplasm of major food crops, the information at the molecular level is limited and to date, only a few QTL analysis studies have been reported in Arabidopsis thaliana (Bentsink et al. 2003; Vreugdenhil et al. 2004, 2005), beans (Guzman-Maldonado et al. 2003), and more recently in rice (Stangoulis et al. 2007). To resolve the complexity of quantitative traits, introgression lines have key advantages in reducing the polygenic traits by dissecting them into a set of monogenic loci (Peleman and Van-der-Voort 2003). Presently, introgression lines become a useful experimental material for large-scale identification, finemapping, cloning and molecular characterization of QTLs in rice (Li et al. 2005; Tian et al. 2006). Furthermore, use of wild relatives of cultivated plant species such as wild rice (O. rufipogon) not only provides unique genetic resources to study the genetics and molecular biology, but also may help to identify the rare alleles for these mineral elements. A notable aspect of this study was that out of 31 putative QTLs for mineral elements, 26 QTLs (83.9%) had trait-improving alleles derived from the wild rice (O. rufipogon) and only three chromosomes (chromosomes 1, 9 and 12) contributed 45% of chromosomal regions (14 QTLs) for these traits. The co-locations of QTL for some mineral elements observed in this mapping population also indicated the relationship at the molecular level among these traits. Vreugdenhil et al. (2004) also reported co-location of QTLs for levels of mineral elements in Arabidopsis thaliana. Although, if the QTL clusters identified in this study are further confirmed, particularly located on chromosomes 9 and 12, they could be used in an efficient markerassisted selection program to facilitate increasing levels of mineral elements in rice grain. The co-locations of QTLs suggest the importance of these regions for mineral accumulation in rice grain and may be due to physiological coupling of the accumulation of certain minerals or tight linkage of different genes. Quantitative trait loci detected for Zn content in this study on chromosome 12 near the simple sequence repeats (SSR) marker RM235 was also previously reported in this region (Stangoulis et al. 2007). However, this QTL accounted for 9% phenotypic variation for levels of Zn and was only detected during One of the reasons for the differences in power of QTL detection in both studies may be due to the strong environmental effect on these traits. Stangoulis et al. (2007) conducted the experiment in controlled conditions (glass house), whereas we grew the experimental material in open field conditions. However, both studies suggested the stability of this QTL across environments. Due to the wide prevalence of anemia, the biofortification of rice grain with enhanced amounts of Fe content is becoming a hot issue in rice growing as well as consuming countries. In the present study we detected a QTL for Fe content near the SSR marker RM6641 on chromosome 2. Furthermore, Stangoulis et al. (2007) also reported a major QTL explaining 16.5% of phenotypic variation for Fe content on chromosome 2 using a double haploid population derived from a cross between an elite indica variety IR64 and an upland japonica variety Azucena. The favorable allele at this QTL was contributed by Azucena, whereas in the present investigation the Fe enhancing allele came from wild rice (O. rufipogon). The results obtained by Stangoulis et al. (2007) and in the present study indicated that indica subspecies may have inferior alleles for Fe content at this region on chromosome 2. Gregorio et al. (2000) also previously reported three loci explaining 19 30% variation for Fe content on chromosomes 7, 8, and 9 in rice. In the present study we did not detect any QTL for Fe content on chromosomes 7 and 8, but a minor QTL was observed on chromosome 9, also indicating the consensus region for Fe content on chromosome 9 in rice. The number of QTL detected in each study depends on the genetic diversity among parents, population size and the number of markers tested (Brondani et al. 2002). It is noteworthy that QTLs detected in the present investigation for K on chromosomes 1 and 4 (Wu et al. 1998), for P on chromosomes 1 and 12 (Ni et al. 1998; Wissuwa et al. 1998; Ming et al. 2001; Wissuwa and Ae 2001a, 2001b) and for Mn on chromosome 10 (Wang et al. 2002) were also previously reported in different parts (roots and shoots) of rice plants. Thus the association of common chromosomal regions indicates the same physiological process in sink to source. However, little information is available regarding the candidate genes that play a critical role for the transport of these metals. Because, most functional studies for these candidate genes have been carried out in yeast, but their actual functions in plants are still in question, especially from source to sink (Narayanan et al. 2007). Clemens (2001) suggested that maintaining cation homeostasis in plants requires a coordinated network of several genes associated with metal uptake, transportation, trafficking and sequestration mechanisms. In contrast to the sequence homology, further fine mapping and analysis of near isogenic lines (NILs) could provide a more precise role of co-localization of QTL for these traits. Thus, it is difficult to conclude from these suggestive co-localizations of QTL for mineral elements and it indicates coincidental or may be results of closely linked genes. Finally the information reported from the present investigation may help to solve the genetic complexity of mineral accumulations in rice grain.

7 QTLs for Mineral Nutrients in Rice 7 Materials and Methods Plant materials The introgressed population used in the present investigation was derived from three consecutive backcrosses with a recipient parent Teqing (Oryza sativa ssp. indica), an elite indica cultivar, whereas an accession of common wild rice (O. rufipogon Griff.) introduced from Yunnan, China was used as donor parent. The development of ILs population and genomic contribution of O. rufipogon in the total genome of each line were explained by Tan et al. (2007). A total of 85 ILs with normal maturity and fertility, and having sufficient amounts of seeds during both years were included in this study. Field evaluation The 85 ILs, along with the recurrent parent Teqing, were evaluated in randomized complete block design at the Experiment Station of China Agricultural University, Beijing (E116 28, N39 54 ) during summer 2005 and Each genotype was planted in five-row plots containing 11 plants per row, placed 15 cm apart. The field management conditions were similar to those under normal rice production. At maturity, ten plants per plot were randomly selected from the central three rows and harvested separately. To obtain a representative sample for trait measurements, seeds from individual plants within the plot were mixed. Eight mineral elements; namely, iron (Fe), zinc (Zn), Manganese (Mn), calcium (Ca), copper (Cu), magnesium (Mg), potassium (K) and phosphorus (P) were measured. Sample preparation and extraction of mineral elements from rice grains A representative sample from each line was manually dehulled in a mini rice huller (Kett TR-100, Tokyo, Japan) and washed with deionized double-distilled water to remove small husks and other possible contaminations. The dehulled rice grain samples were covered in individual paper bags and kept in an oven at 75 C overnight. About 15 g of air-dried sample from each genotype was ground in a small rice miller and stored in plastic bags. Mineral elements were extracted from a 0.5-g representative ground sample using a standard wet-ashing digestion method (AOAC 1995): 5 ml nitric acid (HNO 3 ) of analytical grade was added into each microwave vessel containing rice samples and the samples were kept overnight at room temperature. After overnight digestion, 1 ml of hydrogen peroxide (H 2 O 2 )was added to each sample. For final digestion, the temperature was ramped up to 120 C in first 5 min, raised again to 150 C in next 3 min, finally reaching 185 C in 4 min. Each digested sample was cooled down for 20 min in a microwave accelerate reaction system (model MARS) (CEM Corporation, North Carolina, USA). Each sample was transferred into a 25-mL volumetric flask and final volume was made up to 25 ml with MilliQ water. To avoid possible contamination, each sample was stored in polypropylene flasks. Micro- and macro-nutrients measurement Inductively coupled argon plasma (ICAP) spectrometer (model Iris Intrepid II XSP, Thermo Electron Corporation, California, USA) was carried out to measure the content of mineral elements in rice grain samples. To control the variation within samples, each sample was measured thrice and the mean of triplicate observation was used to represent the content of mineral elements in each sample. For error control among samples, all samples were measured in a set of 20 samples containing two blanks and two standard reference samples with known concentrations of mineral elements. Standard reference samples of rice grains were used to compensate for possible matrix interference variations during microwave digestion, whereas external standards were used to construct the standard curve for calibration of ICP. The content of each sample was converted into the amount of mineral elements (μg/g) in each sample using Microsoft Excel. Genotypic data and genetic linkage map Genomic DNA extraction and SSR analysis were reported earlier by Tan et al Initially, 120 polymorphic SSR markers were used to construct the genetic linkage map with MAP MANAGER QTX ( in BC 3 F 1 mapping population. In subsequent generations, an additional 59 polymorphic SSR markers were included. Finally, 179 SSR markers were used to generate the molecular linkage map (Tan et al. 2007). Trait analysis Individual (2005 and 2006) as well as pooled ANOVA and Pearson correlation coefficients were carried out for all studied traits; namely, micro and macro-nutrients content (μg/g dry seed) using the PROC GLM procedure of SAS (SAS Institute, version 8.0, 1999). Significance of correlation coefficients (r) at P = 0.05 or 0.01 is indicated by or, respectively. Heritability in a broad-sense was estimated as explained by Singh and Chaudhary (1985). QTL analysis Quantitative trait loci mapping was carried out by the single point analysis method using MAP MANAGER QTX software. Model QTXb17 was used to detect association between phenotype and the corresponding marker genotype (Manly et al. 2001).

8 8 Journal of Integrative Plant Biology 2008 The statistical threshold for single-point analysis was P < For the main QTL effects, positive and negative signs of the estimates indicate that O. rufipogon and Teqing contributed toward higher value alleles for the corresponding traits, respectively. QTLs were named according to McCouch et al. (1997), in which a two- or three-letter abbreviation is followed by the number of the chromosome where the QTL is found and a terminal suffix, separated by a period, providing a unique identifier to distinguish multiple QTL on a single chromosome. References AOAC (1995). Official Methods of Analysis of AOAC. 16th edn. AOAC International, Arlington. Arnold JM, Bauman LF (1976). Inheritance of and interrelationship among maize kernel traits and elemental contents. Crop Sci. 16, Arnold JM, Bauman LF, Aycock HS (1977). Interrelations among protein, lysine, oil, certain mineral element concentrations and physical kernel characteristics in tow maize populations. Crop Sci. 17, Balint AF, Kovacs G, Erdei L, Sutka J (2001). Comparison of the Cu, Zn, Fe, Ca and Mg contents of the grains of wild, ancient and cultivated wheat species. Cereal Res. Commu. 29, Banziger M, Long M (2000). The potential for increasing iron and zinc density of maize through plant-breeding. Food Nutr. Bull. 21, Beebe S, Gonzalez A, Rengifo J (2000). Research on trace minerals in the common bean. Food Nutr. Bull. 21, Bentsink L, Yuan K, Koornneef M, Vreugdenhil D (2003). The genetics of phytate and phosphate accumulation in seeds and leaves of Arabidopsis thaliana, using natural variation. Theor. Appl. Genet. 106, Brar DS, Khush GS (1997). Alien introgression in rice. Plant Mol. Biol. 35, Brondani C, Rangel PHN, Brondani RPV, Ferreira ME (2002). QTL mapping and introgression of yield-related traits from Oryza glumaepatula to cultivated rice (Oryza sativa) using microsatellite markers. Theor. Appl. Genet. 104, Cakmak I, Ozkan H, Braun HJ, Welch RM, Romheld V (2000). Zinc and iron concentrations in seeds of wild, primitive, and modern wheat. Food Nutr. Bull. 21, Chavez AL, Bedoya JM, Iglesias C, Ceballos H, Roca W (2000). Iron, carotene, and ascorbic acid in cassava roots and leaves. Food Nutr. Bull. 21, Cheng ZQ, Huang XQ, Zhang YZ, Qian J, Yang MZ, Wu CJ et al. (2005). Diversity in the content of some nutritional components in husked seeds of three wild rice species and rice varieties in Yunnan Province of China. J. Integr. Plant Biol. 47, Clemens S (2001). Molecular mechanisms of plant metal tolerance and homeostasis. Planta 212, Food and Agriculture Organization of the United Nations (2004). Undernourishment around the World. The state of food insecurity in the world website. Available on-line: 007/y5650e/y5650e00.htm. Gregorio GB, Senadhira D, Htut T, Graham RD (2000). Breeding for trace mineral density in rice. Food Nutr. Bull. 21, Grusak MA, DellaPenna D (1999). Improving the nutrient composition of plants to enhance human nutrition and health. Ann. Rev. Plant Physiol. Plant Mol. Biol. 50, Guzman-Maldonado S, Martinez O, Acosta-Gallegos JA, Guevara- Lara F, Paredes-Lopez O (2003). Putative quantitative trait loci for physical and chemical components of common bean. Crop Sci. 43, Hill J, Becker HC, Tigerstedt PMA (1998). Quantitative and Ecological Aspects of Plant Breeding. Chapman and Hall, London. Holland JB, Nyquist WE, Cervantes-Martinez CT (2002). Estimating and interpreting heritability for plant breeding: an update. Plant Breed. Rev. 22, Li ZK, Fu BY, Gao YM, Xu JL, Ali J, Lafitte HR et al. (2005). Genomewide introgression lines and their use in genetic and molecular dissection of complex phenotypes in rice (Oryza sativa L.). Plant Mol. Biol. 59, Manly KF, Cudmor Jr RH, Meer JM (2001). Map Manager QTX, crossplatform software for genetic mapping. Mamm. Genome 12, McCouch SR, Chen X, Panaud O, Temnykh S, Xu YB, Cho YG et al. (1997). Microsattelite marker development, mapping and applications in rice genetics and breeding. Plant Mol. Biol. 35, Ming F, Zheng X, Mi G, Zhu L, Zhang F (2001). Detection and verification of quantitative trait loci affecting tolerance to low phosphorus in rice. J. Plant Nutr. 24, Narayanan NN, Vasconcelos MW, Grusak MA (2007). Expression profiling of Oryza sativa metal homeostasis genes in different rice cultivars using a cdna macroarray. Plant Physiol. Biochem. 45, Ni JJ, Wu P, Senadhira D, Huang N (1998). Mapping QTLs for phosphorus deficiency tolerance in rice (Oryza sativa L.). Theor. Appl. Genet. 97, Ortiz-Monasterio JL, Graham RD (2000). Breeding for trace minerals in wheat. Food Nutr. Bull. 21, Peleman JD, Van-der-Voort JR (2003). Breeding by design. Trends Plant Sci. 8, Slavin J (2003). Why whole grains are protective: biological mechanisms. Proc. Nutr. Soc. 62, Stangoulis JCR, Huynh BL, Welch RM, Choi EY, Graham RD (2007). Quantitative trait loci for phytate in rice grain and their relationship with grain micronutrient content. Euphytica 154, Singh RK, Chaudhary BD (1985). Biometrical Methods in Quantitative Genetic Analysis. Kalyani Publishers, New Delhi. Tan LB, Liu FX, Xue W, Wang GJ, Ye S, Zhu ZF et al. (2007). Development of Oryza rufipogon and O. sativa introgressed lines and assessment for yield-related quantitative trait loci. J. Integr. Plant Biol. 49, Tan LB, Zhang PJ, Fu YC, Liu FX, Wang XK, Sun CQ et al. (2004). Identification of quantitative trait loci controlling plant height and days

9 QTLs for Mineral Nutrients in Rice 9 to heading from Yuanjiang common wild rice (Oryza rufipogon Griff.) Acta Genet. Sin. 31, Tian F, Li DJ, Fu Q, Zhu ZF, Fu YC, Wang XK et al. (2006). Construction of introgression lines carrying wild rice (Oryza rufipogon Griff.) segments in cultivated rice (O. sativa L.) background and characterization of introgressed segments associated with yieldrelated traits. Theor. Appl. Genet. 112, van Staa TP, Dennison EM, Leufkens HG, Cooper C (2001). Epidemiology of fractures in England and Wales. Bone 29, Vreugdenhil D, Aarts MGM, Koormneef M, Nelissen H, Ernst WHO (2004). Natural variation and QTL analysis for cationic mineral content in seeds of Arabidopsis thaliana. Plant Cell Environ. 27, Vreugdenhil D, Aarts MGM, Koormneef M (2005). Exploring natural genetic variation to improve plant nutrient content. In: Broadley MR, White J, eds. Plant Nutritional Genomics. Blackwel Publishing, Oxford. pp Wang YX, Wu P, Wu YR, Yan XL (2002). Molecular marker analysis of manganese toxicity tolerance in rice under greenhouse conditions. Plant Soil 238, Welch R, Graham RD (1999). A new paradigm for world agriculture: meeting human needs. Productive, sustainable, nutritious. Field Crops Res. 60, Welch RM, Graham RD (2004). Breeding for micronutrients in staple food crops from a human nutrition perspective. J. Exp. Bot. 55, Wissuwa M, Ae N (2001a). Genotypic variation for tolerance to phosphorus deficiency in rice and the potential for its exploitation in rice improvement. Plant Breed. 120, Wissuwa M, Ae N (2001b). Further characterization of two QTLs that increase phosphorus uptake of rice (Oryza sativa L.) under phosphorus deficiency. Plant Soil 237, Wissuwa M, Yano M, Ae N (1998). Mapping of QTLs for phosphorusdeficiency tolerance in rice (Oryza sativa L.). Theor. Appl. Genet. 97, Wu P, Ni JJ, Luo AC (1998). QTLs underlying rice tolerance to lowpotassium stress in rice seedlings. Crop Sci. 38, Zhang MW, Guo BJ, Peng ZM (2004). Genetic effects on Fe, Zn, Mn and P content in indica black pericarp rice and their genetic correlations with grain characteristics. Euphytica 135, (Handling editor: Ai-Min Zhang)