| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a USDA-ARS, Dale Bumpers National Rice Research Center, P.O. Box 1090, Stuttgart, AR 72160-1090
b Rice Res. & Ext. Ctr., Univ. of Arkansas, 2900 Hwy 130 E, Stuttgart, AR 72160
* Corresponding author (geizenga{at}spa.ars.usda.gov)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Abbreviations: Polymerase chain reaction (PCR) • Quantitative trait locus (QTL) • Simple sequence repeat (SSR) • Nucleotide-binding site leucine-rich repeat (NBS-LRR)
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Blast (Couch and Kohn, 2002), causes major rice yield losses worldwide with resistant cultivars and fungicides being the primary methods of disease control (Bonman, 1992; Lee, 1994). Based on the rice genome sequence, Monosi et al. (2004) determined that the rice genome carries approximately 500 nucleotide-binding site (NBS)-leucine-rich repeat (LRR) genes. Many of these NBS-LRR regions are associated with a broad spectrum of disease resistance genes (R-genes), as initially reported for the rice blast resistance genes, Pi-ta (Bryan et al., 2000) and Pi-b (Wang et al., 1999). Monosi et al. (2004) determined the distribution of the NBS-encoding genes in the ‘Nipponbare’ rice genome and identified the approximate map position of blast resistance (designated Pi-) genes, which had been previously mapped to a chromosome region based on phenotype, using various mapping populations. The U.S. rice breeding programs incorporated Pi-ta from the Vietnamese landrace Tetep into ‘Katy’ (Moldenhauer et al., 1990) and Pi-b from the Chinese cultivar Te Qing into ‘Saber’ (McClung et al., 2004). In an effort to utilize molecular markers as part of U.S. breeding programs, SSR markers associated with resistance to Pi-b, Pi-kh, Pi-ks, Pi-ta, and Pi-z were identified in the background of U.S. cultivars (Conaway-Bormans et al., 2003; Fjellstrom et al., 2004, 2006; Jia et al., 2004b).
Rice sheath blight is another major fungal disease of rice worldwide (Rush and Lee, 1992). Recently, Pinson et al. (2005) confirmed six previously identified QTLs for sheath blight resistance and identified eight new QTLs. Three of the confirmed QTLs were independent for plant height and maturity which affect sheath blight resistance. Based on identification of the aforementioned sheath blight resistance QTLs, additional studies are in progress to further map sheath blight resistance genes.
To characterize additional rice germplasm useful to U.S. rice breeders Dilday et al. (2001), Yan et al. (2002), Lee et al. (2003), and Yan et al. (2003) conducted field evaluations on approximately 1000 rice germplasm accessions recently introduced into the USA. Accessions were evaluated for desirable agronomic characteristics and for resistance to the two major diseases blast and sheath blight. Ninety-one accessions identified as blast resistant in the field tests were selected for further greenhouse tests to determine resistance to blast races commonly found in the USA (Correll et al., 2000). The objective of this research was to identify new blast resistance genes in the 91 selected resistant accessions by (i) differentiating known Pi-genes, (ii) determining relatedness of the accessions with SSR markers, and (iii) identifying associations of SSR markers with blast resistance, sheath blight resistance, and/or the morphological traits, plant height and heading date, that affect the expression of sheath blight resistance.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Genomic DNA was extracted from leaf tissue using a CTAB method (Eizenga and Phillips, 1997) or the DNeasy Plant Mini Kit (Qiagen, Valencia, CA) per the manufacturer's instructions. Three dominant markers for the resistant Pi-ta allele, one dominant marker for the susceptible pair of pi-ta dominant primers (Jia et al., 2002), and one pair of Pi-b dominant primers (Fjellstrom et al., 2004) were used to determine the presence of Pi-ta, pi-ta and Pi-b, respectively. The presence or absence of these PCR products was visualized on a 1% agarose gel.
One hundred eighty-three SSR markers were selected from the core set developed and mapped by McCouch et al. (2002). Previously, 169 of these SSR markers were used to genotype 145 U.S. rice cultivars (Lu et al., 2005) and 239 rice accessions from a diverse origin (Garris et al., 2005). Seven of the 125 SSR markers were previously shown to be associated with blast resistance genes Pi-b (RM208), Pi-kh, Pi-ks (RM144, RM224), Pi-ta (OSM89) and Pi-z (RM3431, RM5963, RM6836) in U.S. germplasm (Conaway-Bormans et al., 2003; Fjellstrom et al., 2004). Very recently, Fjellstrom et al. (2006) developed improved SSR markers for Pi-z from rice BAC AP005659. Subsequently, AP5659–1 was used to identify Pi-z in this study. The forward primers were labeled with FAM, TET, NED, or HEX fluorescent dyes and the reverse primers were unlabeled. DNA amplifications were performed using an MJ Research PTC-100 96 Plus thermal cycler. PCRs were performed in a 10 µl reaction mix containing 37.5 ng of template DNA, 1x PCR buffer, 0.025 unit of Taq DNA polymerase (Qiagen, Valencia, CA), 0.2 mmol dNTPs, and 0.8 
mol of forward and reverse primers. Information on primer sequences and PCR amplification conditions for each set of primers are available at http://www.gramene.org/ ; verified 22 May 2006. PCR products were separated by size using an ABI 3700 DNA analyzer (Applied Biosystems [ABI], Foster City, CA). SSR fragment sizing was performed with Gene Scan® software (ABI) using the Local Southern Method and default analysis settings after which alleles were identified with the Genotyper® software (ABI) and binned manually. For most markers, fractional numbers were rounded to the nearest integer and alleles differing by 1 bp were declared different. SSR data, obtained from genotyping U.S. cultivars with the same ABI 3700 DNA analyzer (Lu et al., 2005), was included for comparison.
One hundred twenty-five markers (Fig. 1 ) were used for the cluster analysis. The genetic distance matrix developed according to Nei (1972) was used to determine the clusters of genotypes applying the Unweighted Pair Group Method using Arithmetic average (UPGMA) employing NTSYSpc 2.01 (Rohlf, 1997). Calculation of the Polymorphism Information Content (PIC) for each SSR marker is described by Bolstein et al. (1980). The PIC value ascertained the relative value of each marker with respect to the number of alleles at each locus and the relative allele frequency of the alleles in the accessions included in this study.
|
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
Observing each accession's country of origin (Table 1) revealed that 35 of the 39 accessions from the Ivory Coast had Pi-b. Twenty-nine of the 35 accessions also had Pi-ta and a single accession (Tox 3749–71–1-1–3-2–2) only had Pi-ta. By contrast, Pi-ta was found in only two of the 36 accessions originating from China whereas, Pi-b was identified in ten accessions from China. Twenty-one Chinese accessions had excellent blast resistance and contained neither Pi-ta nor Pi-b.
Subsequent tests with SSR markers for Pi-z (Fjellstrom et al., 2006) and Pi-k alleles, Pi-ks and Pi-kh (Fjellstrom et al., 2004) indicated three (4642, Iac-47, Wab450–24–3-2-P18-hb) of the 27 accessions had one of these alleles (Table 1). Considering that (i) neither Pi-z nor Pi-k have the broad spectrum of resistance to U.S. blast races that is found in Pi-ta and Pi-b (Fjellstrom et al., 2004) and (ii) these three accessions were resistant to all U.S. blast races tested, there are additional Pi-genes present besides the possible Pi-z or Pi-k gene.
In addition to the leaf blast rating, the field data collected on the 91 accessions included sheath blight resistance and the agronomic characters, days to heading and plant height as shown in Table 1. Sheath blight resistance ratings ranged from 3.3 to 8.0 with 29 accessions rated between 3.3 and 5.0, indicating moderate field resistance to sheath blight. Because of the interaction between sheath blight resistance and later heading date (Pinson et al., 2005), it was determined that 20 of the 29 more resistant accessions took 95–110 d to heading as compared to less than 85 d for U.S. cultivars. Comparing sheath blight resistance ratings and plant height revealed 18 of the 29 resistant accessions ranged in height from 100–129 cm as compared to 100 cm or less for the more recent U.S. cultivars. Of the original 29 sheath blight resistant accessions, only three accessions, 4484, 46411 and Let 3137, did not have either a later heading date and/or a taller plant height.
SSR markers Identify the Diverse Background of the Resistant Accessions
To identify different novel R-genes it is essential to know how these resistant accessions are related to each other. Most likely different novel R-genes will be present in accessions that do not cluster closely together. A total of 125 SSR markers were used to genotype the 91 accessions and 11 cultivars used as controls. The distribution of the 125 SSR markers is shown in Fig. 1 along with the number of alleles and PIC value for the individual markers. The number of alleles present in these 125 markers ranged from 2 to 21 with 64 markers having eight or more alleles associated with them. The PIC values ranged from 0.208 to 0.870 with 73 markers having values greater than 0.700. These values indicate there is adequate polymorphism for the 125 markers tested.
Using Nei's genetic distance determination, the 91 accessions and 11 cultivars divided into two main clusters designated I and II (Fig. 2 ). The smaller cluster (I) divided into the U.S. cultivars (I-A) included for comparison, which divided into medium grain (Bengal, Zenith) and long grain types (all others). Clustered with the U.S. cultivars was a sub-cluster (I-B) containing a) two accessions, Wab450–1-B-P-62-hb and Wab450–24–3-2-P18-hb, derived from crosses of the Asian cultivated species, O. sativa, with the African cultivated species O. glaberrima Steud., b) three Egyptian accessions, GZ-5830–48–2-2, GZ-5578–2-1–2 and GZ-5594–23–1-2, and c) one each of accessions from Chengdu, China (02428); 460 donated by J. Tao from China; IAC47 developed in Brazil; and the single North Korean accession, Pyongyang 23. The cluster analysis indicated that these accessions were developed from a similar parentage.
|
Figure 2 also denotes (bold, italics) the aforementioned 27 accessions identified as blast resistant and having neither Pi-b nor Pi-ta. Based on the genetic distance determination, eleven of the 27 accessions were clustered closely together and were obtained from one source (J. Tao). The remaining 16 accessions were divided among five large clusters indicating these accessions had at least five different backgrounds thus, it is likely that different novel R-genes are present in these accessions.
SSR Markers Associated with Blast and Sheath Blight
The first association mapping conducted in plants associated the flowering time of 92 inbred maize lines with polymorphisms in the Dwarf8 gene using 141 SSR markers (Thornsberry et al., 2001). Following similar methods, associations were calculated for each individual SSR marker with leaf blast field ratings, the seven individual blast race ratings, sheath blight ratings, plant height and days to heading (Table 2). Forty of the 125 SSR markers had significant associations with at least one of the aforementioned traits, and one marker, RM021, had eight traits associated with it. Thirty-three of the 40 markers were associated with at least one blast trait. The range in the number of markers associated with the individual blast races was from four markers associated with IH-1 to 17 markers with IE-1K. Thirteen markers were associated with field ratings for leaf blast and 13 with sheath blight. With traits that influence sheath blight resistance, plant height, and days to heading, eight markers were associated with plant height and four with days to heading. Five markers, RM105, RM248, RM333, RM413, and RM477, were associated with both sheath blight and days to heading, and one marker, RM541, with both sheath blight and plant height. Chromosome 11 had the most, five markers associated with nine of the 11 phenotypic traits evaluated. Four of the five markers were associated with blast resistance traits.
In Fig. 1 the SSR markers that were associated with at least one of the blast traits measured are italicized, and the approximate map location of reported blast resistance (Pi-) genes, as summarized by Monosi et al. (2004), are marked. The association mapping we performed identified valid blast resistance gene locations as evidenced by the fact that of the 32 markers associated with blast traits, 20 markers were located in regions previously reported to contain one or more Pi-gene(s) and/or QTLs for blast (Ramalingam et al., 2003). Three (RM22, RM104, RM282) of these 20 markers were only located in regions of the blast QTLs. In addition, seven (RM11, RM108, RM109, RM149, RM418, RM477, RM555) of the remaining twelve markers are located in the genetic map positions of NBS-LRR identified by Monosi et al. (2004). The remaining five markers (RM7, RM228, RM333, RM334, RM478) identified by association mapping were not in regions previously identified with blast traits and/or NBS-LRR sites.
Of the seven SSR markers discovered by Fjellstrom et al. (2004) to identify blast genes Pi-b (RM208), Pi-k (RM144, RM224), Pi-ta (OSM89), and Conaway-Bormans et al. (2003) to identify Pi-z (RM3431, RM5963, RM6836) in the background of U.S. breeding lines, only one (RM3431) was associated with blast traits in this study using more diverse germplasm. RM3431, one of three markers for Pi-z, was associated with resistance to blast races IG-1 and IE-1K. Previous studies show that Pi-z confers resistance to both of these blast races (Conaway-Bormans et al., 2003). The lack of association with RM208 and OSM89, even though dominant primers for the cloned genes suggest Pi-b is present in 56 accessions and Pi-ta in 40 accessions, is most likely due to these markers being identified in southern U.S. rice breeding lines and not the diverse background of these accessions (Fjellstrom et al., 2004; Jia et al., 2004b). Lack of association with RM144 and RM224 may be due to few accessions having the Pi-kh or Pi-ks alleles and/or the influence of the diverse background. The distance from the Pi-z locus, diverse background, or an altered Pi-z locus, may explain why only RM3431 showed an association. According to Fjellstrom et al. (2006), the original Pi-z markers, RM3431, RM5963, RM6836, are not close enough to the locus and have not always worked well for U.S. breeding programs thus, they recently developed new SSR markers for Pi-z from the rice BAC AP005659.
Early studies of Pi-k indicated a large number of alleles present at this locus (Kiyosawa, 1972; McCouch et al., 1994). More recently, studies of Pi-b (Fjellstrom et al., 2004), Pi-ta (Jia et al., 2004b) and Pi-z (Hayashi et al., 2004, Fjellstrom et al., 2006) suggest multiple alleles may also be present at these loci based on gene sequence information, mapping populations and reaction to blast isolates. In addition to affecting the association identified, slightly altered alleles also may explain why the presence/absence of Pi-b as determined with RM208 and the Pi-b dominant primer results disagreed for four accessions (Table 1).
The location of SSR markers associated with sheath blight resistance, days to heading, or plant height were compared to recently summarized QTL maps for these traits (Pinson et al., 2005; Zou et al., 2000). Based on Pinson et al. (2005; S.R. Pinson, personal communication, 2005) of the 13 markers associated with sheath blight resistance (Table 2), four (RM72, RM413, RM477, RM490) have already been identified as sheath blight resistance QTLs, seven (RM341, RM251, RM435, RM541, RM105, RM33) are near sheath blight resistance QTLs, and the remaining three markers (RM248, RM21, RM206) were not close to any previously reported sheath blight resistance QTL. Of the four markers associated with plant height, RM541 and RM552 were near previously located sheath blight resistance QTLs (Pinson et al., 2005; S.R. Pinson, personal communication, 2005); RM144 was in the same region as RM206, identified in this study as associated with sheath blight; and RM154 was not related to previously reported QTLs for sheath blight resistance or plant height. Eight markers were associated with days to heading. Comparing these associations to studies by Pinson et al. (2005; personal communication, 2005), RM464 was previously associated with late heading and a sheath blight resistance QTL; RM413 and RM477 were used as markers for sheath blight resistance QTLs; and RM149, RM105, RM304 and RM333 were located near sheath blight resistance QTLs. RM248 was identified in this study as associated with sheath blight resistance, but not in Pinson et al. (2005) or Zou et al. (2000). In summary, of the 19 SSR markers associated with sheath blight resistance, heading date, or plant height, 16 were in regions previously identified in other QTL mapping studies, indicating that most of the associations identified are supported by inheritance studies of these traits.
In conclusion, from the 91 rice germplasm accessions selected from approximately 1000 accessions, 27 accessions were identified as resistant to U.S. blast races that did not have the blast resistance genes, Pi-ta or Pi-b, already incorporated into U.S. rice cultivars. Eleven of the 27 accessions were closely related to each other but the remaining 16 accessions had varying levels of genotypic diversity, including two accessions selected from crosses with O. glaberrima, one from the Philippines and one from the Ivory Coast. Thirty-two SSR markers were associated with blast resistance traits and 19 with sheath blight resistance traits. The fact that 27 of the markers associated with blast resistance traits and 17 of the markers associated with sheath blight resistance were located in regions identified in previously published inheritance studies, strengthens the validity of our association mapping results. This documented support provides encouragement that the remaining markers may prove relevant for identifying the additional resistance genes necessary for U.S. rice breeding programs.
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 31, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Plant Registrations | Soil Science Society of America Journal | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
