Seed banks and molecular maps: unlocking genetic potential from the wild

Nearly a century has been spent collecting and preserving genetic diversity in plants. Germplasm banks-living seed collections that serve as repositories of genetic variation-have been established as a source of genes for improving agricultural crops. Genetic linkage maps have made it possible to study the chromosomal locations of genes for improving yield and other complex traits important to agriculture. The tools of genome research may finally unleash the genetic potential of our wild and cultivated germplasm resources for the benefit of society.

Computational and experimental analysis of microsatellites in rice (Oryza sativa L.): frequency, length variation, transposon associations, and genetic marker potential

A total of 57.8 Mb of publicly available rice (Oryza sativa L.) DNA sequence was searched to determine the frequency and distribution of different simple sequence repeats (SSRs) in the genome. SSR loci were categorized into two groups based on the length of the repeat motif. Class I, or hypervariable markers, consisted of SSRs > or =20 bp, and Class II, or potentially variable markers, consisted of SSRs > or =12 bp <20 bp. The occurrence of Class I SSRs in end-sequences of EcoRI- and HindIII-digested BAC clones was one SSR per 40 Kb, whereas in continuous genomic sequence (represented by 27 fully sequenced BAC and PAC clones), the frequency was one SSR every 16 kb. Class II SSRs were estimated to occur every 3.7 kb in BAC ends and every 1.9 kb in fully sequenced BAC and PAC clones. GC-rich trinucleotide repeats (TNRs) were most abundant in protein-coding portions of ESTs and in fully sequenced BACs and PACs, whereas AT-rich TNRs showed no such preference, and di- and tetranucleotide repeats were most frequently found in noncoding, intergenic regions of the rice genome. Microsatellites with poly(AT)n repeats represented the most abundant and polymorphic class of SSRs but were frequently associated with the Micropon family of miniature inverted-repeat transposable elements (MITEs) and were difficult to amplify. A set of 200 Class I SSR markers was developed and integrated into the existing microsatellite map of rice, providing immediate links between the genetic, physical, and sequence-based maps. This contribution brings the number of microsatellite markers that have been rigorously evaluated for amplification, map position, and allelic diversity in Oryza spp. to a total of 500.

Molecular mapping of rice chromosomes

We report the construction of an RFLP genetic map of rice (Oryza sativa) chromosomes. The map is comprised of 135 loci corresponding to clones selected from a PstI genomic library. This molecular map covers 1,389 cM of the rice genome and exceeds the current classical maps by more than 20%. The map was generated from F2 segregation data (50 individuals) from a cross between an indica and javanica rice cultivar. Primary trisomics were used to assign linkage groups to each of the 12 rice chromosomes. Seventy-eight percent of the clones assayed revealed RFLPs between the two parental cultivars, indicating that rice contains a significant amount of RFLP variation. Strong correlations between size of hybridizing restriction fragments and level of polymorphism indicate that a significant proportion of the RFLPs in rice are generated by insertions/delections. This conclusion is supported by the occurrence of null alleles for some clones (presumably created by insertion or deletion events). One clone, RG229, hybridized to sequences in both the indica and javanica genomes, which have apparently transposed since the divergence of the two cultivars from their last common ancestor, providing evidence for sequence movement in rice. As a by product of this mapping project, we have discovered that rice DNA is less C-methylated than tomato or maize DNA. Our results also suggest the notion that a large fraction of the rice genome (approximately 50%) is single copy.

Mapping quantitative trait loci for yield, yield components and morphological traits in an advanced backcross population between Oryza rufipogon and the Oryza sativa cultivar Jefferson

An advanced backcross population between an accession of Oryza rufipogon (IRGC 105491) and the U.S. cultivar Jefferson (Oryza sativa ssp. japonica) was developed to identify quantitative trait loci (QTLs) for yield, yield components and morphological traits. The genetic linkage map generated for this population consisted of 153 SSR and RFLP markers with an average interval size of 10.3 cM. Thirteen traits were examined, nine of which were measured in multiple environments. Seventy-six QTLs above an experiment-wise significance threshold of P<0.01 (corresponding to an interval mapping LOD>3.6 or a composite interval mapping LOD>3.9) were identified. For the traits measured in multiple environments, 47% of the QTLs were detected in at least two environments. The O. rufipogon allele was favorable for 53% of the yield and yield component QTLs, including loci for yield, grains per panicle, panicle length, and grain weight. Morphological traits related to the domestication process and/or weedy characteristics, including plant height, shattering, tiller type and awns, were found clustered on chromosomes 1 and 4. Comparisons to previous studies involving wild x cultivated crosses revealed O. rufipogon alleles with stable effects in multiple genetic backgrounds and environments, several of which have not been detected in studies between Oryza sativa cultivars, indicating potentially novel alleles from O. rufipogon. Some O. rufipogon-derived QTLs, however, were in similar regions as previously reported QTLs from Oryza sativa cultivars, providing evidence for conservation of these QTLs across the Oryza genus. In addition, several QTLs for grain weight, plant height, and flowering time were localized to putative homeologous regions in maize where QTLs for these traits have been previously reported, supporting the hypothesis of functional conservation of QTLs across the grasses.

Identification of trait-improving quantitative trait loci alleles from a wild rice relative, Oryza rufipogon

Wild species are valued as a unique source of genetic variation, but they have rarely been used for the genetic improvement of quantitative traits. To identify trait-improving quantitative trait loci (QTL) alleles from exotic species, an accession of Oryza rufipogon, a relative of cultivated rice, was chosen on the basis of a genetic diversity study. An interspecific BC2 testcross population (V20A/O. rufipogon//V20B///V20B////Ce64) consisting of 300 families was evaluated for 12 agronomically important quantitative traits. The O. rufipogon accession was phenotypically inferior for all 12 traits. However, transgressive segregants that outperformed the original elite hybrid variety, V20A/Ce64, were observed for all traits examined. A set of 122 RFLP and microsatellite markers was used to identify QTL. A total of 68 significant QTL were identified, and of these, 35 (51%) had beneficial alleles derived from the phenotypically inferior O. rufipogon parent. Nineteen (54%) of these beneficial QTL alleles were free of deleterious effects on other characters. O. rufipogon alleles at two QTL on chromosomes 1 and 2 were associated with an 18 and 17% increase in grain yield per plant, respectively, without delaying maturity or increasing plant height. This discovery suggests that the innovative use of molecular maps and markers can alter the way geneticists utilize wild and exotic germplasm.

RFLP mapping of genes conferring complete and partial resistance to blast in a durably resistant rice cultivar

Moroberekan, a japonica rice cultivar with durable resistance to blast disease in Asia, was crossed to the highly susceptible indica cultivar, CO39, and 281 F7 recombinant inbred (RI) lines were produced by single seed descent. The population was evaluated for blast resistance in the greenhouse and the field, and was analyzed with 127 restriction fragment length polymorphism (RFLP) markers. Two dominant loci associated with qualitative resistance to five isolates of the fungus were tentatively named Pi-5(t) and Pi-7(t). They were mapped on chromosomes 4 and 11, respectively. To identify quantitative trait loci (QTLs) affecting partial resistance, RI lines were inoculated with isolate PO6-6 of Pyricularia oryzae in polycyclic tests. Ten chromosomal segments were found to be associated with effects on lesion number (P < 0.0001 and LOD > 6.0). Three of the markers associated with QTLs for partial resistance had been reported to be linked to complete blast resistance in previous studies. QTLs identified in greenhouse tests were good predictors of blast resistance at two field sites. This study illustrates the usefulness of RI lines for mapping a complex trait such as blast resistance and suggests that durable resistance in the traditional variety, Moroberekan, involves a complex of genes associated with both partial and complete resistance.

Molecular-genetic maps for group 1 chromosomes of Triticeae species and their relation to chromosomes in rice and oat

Group 1 chromosomes of the Triticeae tribe have been studied extensively because many important genes have been assigned to them. In this paper, chromosome 1 linkage maps of Triticum aestivum, T. tauschii, and T. monococcum are compared with existing barley and rye maps to develop a consensus map for Triticeae species and thus facilitate the mapping of agronomic genes in this tribe. The consensus map that was developed consists of 14 agronomically important genes, 17 DNA markers that were derived from known-function clones, and 76 DNA markers derived from anonymous clones. There are 12 inconsistencies in the order of markers among seven wheat, four barley, and two rye maps. A comparison of the Triticeae group 1 chromosome consensus map with linkage maps of homoeologous chromosomes in rice indicates that the linkage maps for the long arm and the proximal portion of the short arm of group 1 chromosomes are conserved among these species. Similarly, gene order is conserved between Triticeae chromosome 1 and its homoeologous chromosome in oat. The location of the centromere in rice and oat chromosomes is estimated from its position in homoeologous group 1 chromosomes of Triticeae.

Genetic and physical analysis of the rice bacterial blight disease resistance locus, Xa21

Nearly isogenic lines (NILs) of rice (Oryza sativa) differing at a locus conferring resistance to the pathogen Xanthomonas oryzae pv. oryzae were surveyed with 123 DNA markers and 985 random primers using restriction fragment length plymorphism (RFLP) and random amplified polymorphic DNA (RAPD) analysis. One chromosome 11 marker (RG103) detected polymorphism between the NILs that cosegregated with Xa21. All other chromosome 11 DNA markers tested were monomorphic between the NILs, localizing the Xa21 introgressed region to an 8.3 cM interval on chromosome 11. Furthermore, we identified two polymerase chain reaction (PCR) products (RAPD2148 and RAPD818) that detected polymorphisms between the NILs. Genomic sequences hybridizing with RAPD818, RAPD248 and RG103 were duplicated specifically in the Xa21 NIL. All three markers cosegregated with the resistance locus, Xa21, in a F2 population of 386 progeny. Based on the frequency with which we recovered polymorphic Xa21-linked markers, we estimated the physical size of the introgressed region to be approximately 800 kb. This estimation was supported by physical mapping (using pulsed field gel electrophoresis) of the sequences hybridizing with the three Xa21-linked DNA markers. The results showed that the three Xa21-linked markers are physically close to each other, with one copy of the RAPD818 sequences located within 60 kb of RAPD248 and the other copy within 270 kb of RG103. None of the enzymes tested generated a DNA fragment that hybridized with all three of the markers indicating that the introgressed region containing the resistance locus Xa21 is probably larger than 270 kb.

Chromosomal regions associated with segregation distortion of molecular markers in F2, backcross, doubled haploid, and recombinant inbred populations in rice (Oryza sativa L.)

Chromosomal regions associated with marker segregation distortion in rice were compared based on six molecular linkage maps. Mapping populations were derived from one interspecific backcross and five inter-subspecific (indica/japonica) crosses, including two F2 populations, two doubled haploid (DH) populations, and one recombinant inbred (RI) population. Mapping data for each population consisted of 129-629 markers. Segregation distortion was determined based on chi-square analysis (P < 0.01) and was observed at 6.8-31.8%, of the mapped marker loci. Marker loci associated with skewed allele frequencies were distributed on all 12 chromosomes. Distortion in eight chromosomal regions bracketed previously identified gametophyte (ga) or sterility genes (S). Distortion in three other chromosomal regions was found only in DH populations, where japonica alleles were over-represented, suggesting that loci in these regions may be associated with preferential regeneration of japonica genotypes during anther culture. Three additional clusters of skewed markers were observed in more than one population in regions where no gametophytic or sterility loci have previously been reported. A total of 17 segregation distortion loci may be postulated based on this study and their locations in the rice genome were estimated.

Genome-wide prediction models that incorporate de novo GWAS are a powerful new tool for tropical rice improvement

To address the multiple challenges to food security posed by global climate change, population growth and rising incomes, plant breeders are developing new crop varieties that can enhance both agricultural productivity and environmental sustainability. Current breeding practices, however, are unable to keep pace with demand. Genomic selection (GS) is a new technique that helps accelerate the rate of genetic gain in breeding by using whole-genome data to predict the breeding value of offspring. Here, we describe a new GS model that combines RR-BLUP with markers fit as fixed effects selected from the results of a genome-wide-association study (GWAS) on the RR-BLUP training data. We term this model GS + de novo GWAS. In a breeding population of tropical rice, GS + de novo GWAS outperformed six other models for a variety of traits and in multiple environments. On the basis of these results, we propose an extended, two-part breeding design that can be used to efficiently integrate novel variation into elite breeding populations, thus expanding genetic diversity and enhancing the potential for sustainable productivity gains.

Comparative mapping in grasses. Wheat relationships

Conventionally, the genetics of species of the family Gramineae have been studied separately. Comparative mapping using DNA markers offers a method of combining the research efforts in each species. In this study, we developed consensus maps for members of the Triticeae tribe (Triticum aestivum, T. tauschii, and Hordeum spp.) and compared them to rice, maize and oat. The aneuploid stocks available in wheat are invaluable for comparative mapping because almost every DNA fragment can be allocated to a chromosome arm, thus preventing erroneous conclusions about probes that could not be mapped due to a lack of polymorphism between mapping parents. The orders of the markers detected by probes mapped in rice, maize and oat were conserved for 93, 92 and 94% of the length of Triticeae consensus maps, respectively. The chromosome segments duplicated within the maize genome by ancient polyploidization events were identified by homoeology of segments from two maize chromosomes to regions of one Triticeae chromosome. Homoeologous segments conserved across Triticeae species, rice, maize, and oat can be identified for each Triticeae chromosome. Putative orthologous loci for several simply inherited and quantitatively inherited traits in Gramineae species were identified.

Inferences on the genome structure of progenitor maize through comparative analysis of rice, maize and the domesticated panicoids

Corn and rice genetic linkage map alignments were extended and refined by the addition of 262 new, reciprocally mapped maize cDNA loci. Twenty chromosomal rearrangements were identified in maize relative to rice and these included telomeric fusions between rice linkage groups, nested insertion of rice linkage groups, intrachromosomal inversions, and a nonreciprocal translocation. Maize genome evolution was inferred relative to other species within the Panicoideae and a progenitor maize genome with eight linkage groups was proposed. Conservation of composite linkage groups indicates that the tetrasomic state arose during maize evolution either from duplication of one progenitor corn genome (autoploidy) or from a cross between species that shared the composite linkages observed in modern maize (alloploidy). New evidence of a quadruplicated homeologous segment on maize chromosomes 2 and 10, and 3 and 4, corresponded to the internally duplicated region on rice chromosomes 11 and 12 and suggested that this duplication in the rice genome predated the divergence of the Panicoideae and Oryzoideae subfamilies. Charting of the macroevolutionary steps leading to the modern maize genome clarifies the interpretation of intercladal comparative maps and facilitates alignments and genomic cross-referencing of genes and phenotypes among grass family members.

Identification of quantitative trait loci for yield and yield components in an advanced backcross population derived from the Oryza sativa variety IR64 and the wild relative O. rufipogon

A BC(2)F(2) population developed from an interspecific cross between Oryza sativa (cv IR64) and O. rufipogon (IRGC 105491) was used in an advanced backcross QTL analysis to identify and introduce agronomically useful genes from this wild relative into the cultivated gene pool. The objectives of this study were: (1) to identify putative yield and yield component QTLs that can be useful to improve the elite cultivar IR64; (2) to compare the QTLs within this study with previously reported QTLs in rice as the basis for identifying QTLs that are stable across different environments and genetic backgrounds; and (3) to compare the identified QTLs with previously reported QTLs from maize to examine the degree of QTL conservation across the grass family. Two hundred eighty-five families were evaluated in two field environments in Indonesia, with two replications each, for 12 agronomic traits. A total of 165 markers consisting of 131 SSRs and 34 RFLPs were used to construct the genetic linkage map. By employing interval mapping and composite interval mapping, 42 QTLs were identified. Despite its inferior performance, 33% of the QTL alleles originating from O. rufipogon had a beneficial effect for yield and yield components in the IR64 background. Twenty-two QTLs (53.4%) were located in similar regions as previously reported rice QTLs, suggesting the existence of stable QTLs across genetic backgrounds and environments. Twenty QTLs (47.6%) were exclusively detected in this study, uncovering potentially novel alleles from the wild, some of which might improve the performance of the tropical indica variety IR64. Additionally, several QTLs for plant height, grain weight, and flowering time detected in this study corresponded to homeologous regions in maize containing previously detected maize QTLs for these traits.

Microsatellite marker development, mapping and applications in rice genetics and breeding

Microsatellites are simple, tandemly repeated di- to tetra-nucleotide sequence motifs flanked by unique sequences. They are valuable as genetic markers because they are co-dominant, detect high levels of allelic diversity, and are easily and economically assayed by the polymerase chain reaction (PCR). Results from screening a rice genomic library suggest that there are an estimated 5700-10,000 microsatellites in rice, with the relative frequency of different repeats decreasing with increasing size of the motif. A map consisting of 120 microsatellite markers demonstrates that they are well distributed throughout the 12 chromosomes of rice. Five multiple copy primer sequences have been identified that could be mapped to independent chromosomal locations. The current level of genome coverage provided by these simple sequence length polymorphisms (SSLPs) in rice is sufficient to be useful for genotype identification, gene and quantitative trail locus (QTL) analysis, screening of large insert libraries, and marker-assisted selection in breeding. Studies of allelic diversity have documented up to 25 alleles at a single locus in cultivated rice germplasm and provide evidence that amplification in wild relatives of Oryza sativa is generally reliable. The availability of increasing numbers of mapped SSLP markers can be expected to complement existing RFLP and AFLP maps, increasing the power and resolution of genome analysis in rice.

Development of microsatellite markers and characterization of simple sequence length polymorphism (SSLP) in rice (Oryza sativa L.)

Microsatellite markers containing simple sequence repeats (SSR) are a valuable tool for genetic analysis. Our objective is to augment the existing RFLP map of rice with simple sequence length polymorphisms (SSLP). In this study, we describe 20 new microsatellite markers that have been assigned to positions along the rice chromosomes, characterized for their allelic diversity in cultivated and wild rice, and tested for amplification in distantly related species. Our results indicate that the genomic distribution of microsatellites in rice appears to be random, with no obvious bias for, or clustering in particular regions, that mapping results are identical in intersubspecific and interspecific populations, and that amplification in wild relatives of Oryza sativa is reliable in species most closely related to cultivated rice but becomes less successful as the genetic distance increases. Sequence analysis of SSLP alleles in three related indica varieties demonstrated the clustering of complex arrays of SSR motifs in a single 300-bp region with independent variation in each. Two microsatellite markers amplified multiple loci that were mapped onto independent rice chromosomes, suggesting the presence of duplicated regions within the rice genome. The availability of increasing numbers of mapped SSLP markers can be expected to increase the power and resolution of genome analysis in rice.

Comparative mapping in grasses. Oat relationships

The development of RFLP linkage maps in hexaploid and diploid oat allows us to study genetic relationships of these species at the DNA level. In this report, we present the extension of a previously developed diploid oat map (Avena atlantica x A. hirtula) and its molecular-genetic relationships with wheat, rice and maize. Examination of 92-99% of the length of the oat genome map with probes common to Triticeae species, rice or maize showed that 84, 79 and 71%, respectively, was conserved between these species and oat. Generally, the orders of loci among chromosomes homoelogous to oat chromosomes A and D were the most conserved and those of chromosomes homoeologous to oat chromosome G were the least conserved. Conservation was observed for blocks ranging from whole chromosomes 101 cM long to small segments 2.5 cM long containing two loci. Comparison of the homoeologous segments of Triticeae, rice and maize relative to oat indicated that certain regions have been maintained in all four species. The relative positions of major genes governing traits such as seed storage proteins and resistance to leaf rusts have been conserved between cultivated oat and Triticeae species. Also, the locations of three vernalization/or photoperiod response genes identified in hexaploid oat correspond to the locations of similar genes in homoeologous chromosomes of wheat, rice or maize. The locations of the centromeres for six of the seven oat chromosomes were estimated based on the homoeologous segments between oat and Triticeae chromosomes.

Sequence divergence of rice microsatellites in Oryza and other plant species

To determine the basis of genetic variation at microsatellite loci, eleven primer pairs, developed to amplify microsatellite markers in rice, were evaluated for their ability to amplify a PCR product and for both electromorphic and sequence-based polymorphism of the resulting products in 12 plant samples, including representatives from six different species within the genus Oryza and one genotype each from Zea (maize), Triticum (wheat) and Arabidopsis. PCR amplification was reliable in the four O. sativa samples as well as in the closely related Oryza relatives with AA genomes, while only 73% (8/11) of primers amplified in the BB/CC and CC genomes of Oryza, and 27% (3/11) amplified in the other genera. Three out of seven DNA fragments that were amplified from all genera were determined to be orthologous to their rice counterparts. A total of 115 amplicons were detected using polyacrylamide gel electrophoresis and these clustered into 74 distinct electromorphs. Sequencing of 108 amplicons revealed size homoplasy, exposing 13 new sequence-based variants. Allelic diversity within a species was predominantly due to changes in the number of repeats in the microsatellite region, but the frequency of insertions/deletions (indels) and base substitutions increased as the genetic distance between samples increased. This study suggests that electromorph size polymorphism is an adequate measure of genetic difference in studies involving closely-related individuals, but that when phylogenetic or evolutionary inferences are being made over longer time scales, evaluation of SSR variation at the sequence level is essential.

Fine mapping of a grain weight quantitative trait locus on rice chromosome 8 using near-isogenic lines derived from a cross between Oryza sativa and Oryza rufipogon

A quantitative trait locus (QTL) for grain weight (GW) was detected near SSR marker RM210 on chromosome 8 in backcross populations derived from a cross between the Korean japonica cultivar Hwaseongbyeo and Oryza rufipogon (IRGC 105491). The O. rufipogon allele increased GW in the Hwaseongbyeo background despite the fact that O. rufipogon was the small-seeded parent. Using sister BC(3)F(3) near-isogenic lines (NILs), gw8.1 was validated and mapped to a 6.1 cM region in the interval between RM42 and RM210 (P < or = 0.0001). Substitution mapping with eight BC(3)F(4) sub-NILs further narrowed the interval containing gw8.1 to about 306.4 kb between markers RM23201.CNR151 and RM30000.CNR99. A yield trial using homozygous BC(3)F(4) sister sub-NILs and the Hwaseongbyeo recurrent parent indicated that the NIL carrying an O. rufipogon chromosome segment across the entire gw8.1 target region out-yielded its sister NIL (containing Hwaseongbyeo chromosome in the RM42-RM210 interval) by 9% (P=0.029). The higher-yielding NIL produced 19.3% more grain than the Hwaseongbyeo recurrent parent (P=0.018). Analysis of a BC(3)F(4) NIL indicated that the variation for GW is associated with variation in grain shape, specifically grain length. The locus, gw8.1 is of particular interest because of its independence from undesirable height and grain quality traits. SSR markers tightly linked to the GW QTL will facilitate cloning of the gene underlying this QTL as well as marker-assisted selection for variation in GW in an applied breeding program.

Cloning and mapping of variety-specific rice genomic DNA sequences: amplified fragment length polymorphisms (AFLP) from silver-stained polyacrylamide gels

An efficient technique for cloning DNA from silver-stained denaturing polyacrylamide gels was developed to allow the isolation of specific bands obtained from selective restriction fragment amplification (SRFA). This method proved as reliable as cloning radioactively labelled SRFA bands from the same gels. Rice DNA was used as a template, both with and without [32P]dCTP, using the same PCR profiles. Amplified products were separated using denaturing polyacryamide gel electrophoresis and visualized either by silver staining of gels or by autoradiography of 32P-labelled products. We cloned specific polymorphic SRFA bands directly from the denaturing polyacrylamide gels with one round of PCR amplification and confirmed that the sequences of the bands from silver-stained gels were identical to the corresponding 32P-labelled bands. The bands that were chosen represented amplified fragment length polymorphisms (AFLPs) between japonica and indica rice varieties. We studied the ability of two cloned AFLP bands to serve as heritable genetic markers by mapping them as RFLPs in an interspecific rice population and found that they represented single-copy DNA at unique loci in the rice genome.

Locating genes associated with root morphology and drought avoidance in rice via linkage to molecular markers

This research was undertaken to identify and map quantitative trait loci (QTLs) associated with five parameters of rice root morphology and to determine if these QTLs are located in the same chromosomal regions as QTLs associated with drought avoidance/tolerance. Root thickness, root:shoot ratio, root dry weight per tiller, deep root dry weight per tiller, and maximum root length were measured in three replicated experiments (runs) of 203 recombinant inbred lines grown in a greenhouse. The lines were from a cross between indica cultivar Co39 andjaponica cultivar Moroberekan. The 203 RI lines were also grown in three replicated field experiments where they were drought-stressed at the seedling, early vegetative, and late-vegetative growth stage and assigned a visual rating based on leaf rolling as to their degree of drought avoidance/tolerance. The QTL analysis of greenhouse and field data was done using single-marker analysis (ANOVA) and interval analysis (Mapmaker QTL). Most QTLs that were identified were associated with root thickness, root/shoot ratio, and root dry weight per tiller, and only a few with deep root weight. None were reliably associated with maximum root depth due to genotype-by-experiment interaction. Root thickness and root dry weight per tiller were the characters found to be the least influenced by environmental differences between greenhouse runs. Correlations of root parameters measured in greenhouse experiments with field drought avoidance/tolerance were significant but not highly predictive. Twelve of the fourteen chromosomal regions containing putative QTLs associated with field drought avoidance/tolerance also contained QTLs associated with root morphology. Thus, selecting for Moroberekan alleles at marker loci associated with the putative root QTLs identified in this study may be an effective strategy for altering the root phenotype of rice towards that commonly associated with drought-resistant cultivars.

QTL mapping in rice

In the past 10 years, interest in applying the tools of molecular genetics to the problem of increasing world rice production has resulted in the generation of two highly saturated, molecular linkage maps of rice, and the localization of numerous genes and quantitative trait loci (QTLs). Primary studies have identified QTLs associated with disease resistance, abiotic stress tolerance and yield potential of rice in a range of ecosystems. The ability to identify, manipulate and potentially clone individual genes involved in quantitatively inherited characters, combined with the demonstrated conservation of numerous linkage blocks among members of the grass family, emphasizes the contribution of map-based genetic analyses both to applied and to basic crop research.

Nuclear- and chloroplast-microsatellite variation in A-genome species of rice

Simple sequence length polymorphism analysis was carried out to reveal microsatellite variation and to clarify the phylogenetic relationships among A-genome species of rice. Total DNA from 29 cultivars (23 Oryza sativa and 6 O. glaberrima) and 30 accessions of wild A-genome species (12 O. rufipogon, 5 O. glumaepatula, 2 O. longistaminata, 6 O. meridionalis, and 5 O. barthii) was used as a template for PCR to detect 24 nuclear and 10 chloroplast microsatellite loci. Microsatellite allelic diversity was examined based on amplified banding patterns. Microsatellites amplified clearly in all 59 accessions, with an average of 18.4 alleles per locus. The polymorphism information content (PIC) value ranged from 0.85 to 0.94, with an average of 0.89. At the species level, high average PIC values were observed in O. sativa (0.79) and O. rufipogon (0.80). For chloroplast microsatellites, the average number of alleles per locus and the average PIC value were 2.9 and 0.38, respectively. While the magnitude of diversity was much greater for nuclear microsatellites than for chloroplast microsatellites, they showed parallel patterns of differentiation for each taxonomic group. Using the ratio of common alleles (estimated as size of amplified fragments) as a similarity index, the average percentages of common microsatellite alleles were calculated between taxa. For both nuclear and chloroplast microsatellites, O. sativa showed the highest similarity values to O. rufipogon, and O. glaberrima was most similar to O. barthii. These data support previous evidence that these cultivars originated from the corresponding wild ancestral species.Key words: simple sequence length polymorphism, SSLP, microsatellite marker, rice, Oryza sativa, allelic diversity, phylogenetics.

Frequency of microsatellite sequences in rice (Oryza sativa L.)

This study was undertaken to estimate the relative frequencies of 13 microsatellite motifs in the rice genome as a basis for efficient development of a microsatellite map. Two dinucleotide, seven trinucleotide, and four tetranucleotide repeat motifs were end labelled and used as hybridization probes to screen genomic and cDNA libraries of rice, cv. IR36. Optimal washing temperatures for identification of clones containing specific microsatellite motifs were estimated based on washing temperatures near Td (dissociation temperature; Td = Tm - 7.6 degrees C). Sequencing of 20 putatively positive clones corresponding to each of 4 microsatellite motifs suggested that while Td provides a useful predictor of washing stringency for most of the repeats studied, those with a very high GC or AT content were most prone to error. The results from screening the rice genomic library suggest that there are an estimated 1360 poly(GA)n and 1230 poly(GT)n microsatellites in the rice genome, and that the relative frequency of different repeats decreased with increasing size of the motif. The most frequently observed microsatellites in the cDNA library were the same as for genomic library, but no poly(CGG)n, poly(ATC)n, or tetranucleotide motifs were observed among cDNA in this study.

Transcriptional repression by Oshox1, a novel homeodomain leucine zipper protein from rice

This paper describes the characterization of Oshox1, a cDNA clone from rice encoding a member of the homeodomain-leucine zipper (HD-Zip) class of putative transcription factors. Oshox1 maps to chromosome 10 and belongs to a family of related rice genes. Two-hybrid assays showed that Oshox1 protein can homodimerize, but can also form heterodimers with an Arabidopsis HD-Zip protein. This suggests that protein-protein interactions may also occur between different HD-Zip proteins in rice, which would provide enormous versatility for generating specific gene-control mechanisms. Oshox1 mRNA could be detected in various rice tissues at different developmental stages, with highest levels in embryos, shoots of seedlings, and leaves of mature plants. Transgenic expression of Oshox1 in Arabidopsis retarded growth and affected leaf size and shape, indicative of a role as developmental regulator. In vitro and in vivo DNA-binding studies revealed that Oshox1 interacts with the pseudopalindromic sequence CAAT(C/G)ATTG, confirming that the protein represents a transcription factor. Oshox1 was found to repress reporter gene activity in rice suspension cells, most likely by a mechanism of active transcriptional repression. Repression was strictly dependent on the presence of upstream Oshox1 binding sites in the reporter gene constructs and a function of the N-terminal region of Oshox1, preceding the homeodomain.

Leveraging natural diversity: back through the bottleneck

Plant breeders have long recognized the existence of useful genetic variation in the wild ancestors of our domesticated crop species. In cultivated rice (Oryza sativa), crosses between high-yielding elite cultivars and low-yielding wild accessions often give rise to superior offspring, with wild alleles conferring increased performance in the context of the elite cultivar genetic background. Because the breeding value of wild germplasm cannot be determined by examining the performance of wild accessions, a phylogenetic approach is recommended to determine which interspecific combinations are most likely to be useful in a breeding program. As we deepen our understanding of how genetic diversity is partitioned within and between cultivated and wild gene pools of Oryza, breeders will have increased power to make predictions about the most efficient strategies for utilizing wild germplasm for rice improvement.