본 논문은 국제벼연구소(International Rice Research Institute: IRRI)의 분자육종체계의 현황을 살펴보고, 국제적으로 시장중 심의 육종체계로 변화함에 따라, 새로운 분자마커 이용 육종 법이 갖추어야 할 조건에 대해 논하였다. 벼에 있어서 신품종 개량을 위한 분자생물학의 도움을 받는 분자육종법은 재해 및 병충해 저항성과 수량 및 품질 제고에 전반적으로 활용되고 있으며, 관련 계통 및 품종이 속속 개발 되고 있다. 국제벼연구소는 기후변화 및 자원부족에 대응하면 서도 수량과 품질의 저하가 없는 쌀 품종을 개발하기 위해, 분자육종법을 적극 활용하고 있다. 지역적으로 시장구조별로 특화시킨 품종개발프로필(VDP)을 개발하고, 그에 맞는 형질을 발굴하여, 빠르게 우수형질에 도 입하기 위해서 분자표지를 이용한 육종이 활용될 수 있다. 각 형질은 유전적으로 우수하고 분명한 효과를 보이는 것으 로서, 분자마커는 해당 형질에 강하게 연관되어야 유용하다. 성공적인 분자표지육종법을 위해서, 고효율 유전분석법, 생장 촉진 및 세대촉진법, 연관마커 신속 설계 및 분석, 중간계통의 효율적인 표현형 검정법 개발이 필수적이다. 최근에는 차세대유전체분석법 (NGS)의 발달로 저렴하게 계 통의 유전체 분석이 용이해졌고, 이를 위해서 생물정보학과 유 전체학의 도움을 받는 유전체선발법이 개발되었다. 향후 성공적인 대량 신속 계통육성을 위한 분자육종법의 도 입을 위해 필요한 기술과 연구 제목을 제안하면, (1) 경제적 가치평가 기술과 합리적 효과 판정의 적용 (2) 체계적인 기본 식물 관리, (3) 온실 및 농장 전산화, (4) 계통육성시스템과 유전분석시스템의 일체화, (5) 유전분석 비용의 경제화, (6) 대 량 형질 및 유전자 추출 연구, (7) 다형질간 상호작용 연구, (8) 세대촉진법, (9) 육종가를 위한 결정지원소프트웨어 개발, (10) 집단개량 및 집단선발법의 체계화, 그리고 (11) 인위적 변 이 기술과의 접목 등이다. 분자육종법을 이용한 품종개량의 개발이익은 경제사회적으 로 그 효과가 적절히 평가될 때만이 궁극적으로 소비자와 농 민에게 이익을 제공하는 의미를 갖게 될 것이다.

New QTLs were identified for high grain yield with long panicle in rice. A total of 137 F15 recombinant inbred lines (RILs) derived from a cross between Dasanbyeo (Tongil) and TR22183 (japonica), together with the parents were evaluated for 16 agronomic traits at IRRI in dry and wet seasons under different phosphorus (P) and irrigation conditions. A linkage map was constructed using 236 polymorphic markers in 384-plex Bead Xpress indica-japonica single nucleotide polymorphism (SNP) platform. P and water effects were significant in both wet and dry seasons. Both parents and RILs showed varying degree of sensitivities to scarcities in water and phosphorus in terms of panicle length. Collocating with 20 yield-related QTLs, the panicle QTLs on chromosomes 1 (pl01) ,2 (pl02), 9 (pl09), and 11 (pl11) under low P and rainfed conditions were identified. RILs with TR22183 allele at pl11 showed longer panicle length under low P input rainfed condition in dry and wet seasons. The whole-genome sequences of the two varieties are being compared to design the molecular markers for fine-mapping and candidate gene identification. Based on Nipponbare MSU 7.0 annotation, a total of 1464 genes with predicted function were identified within the four QTL regions. Candidate genes identified in other studies for QTLs under low P and water conditions, such as calmodulin and dehydrin genes, were targeted for designing molecular markers for fine-mapping and expression analysis. Pyramiding the panicle length QTLs correlating with yield QTLs will provide an opportunity of improving yield traits.

Tongil (IR667-98-1-2) rice, developed in 1972, is a high-yielding rice variety derived from a three-way cross between indica and japonica. Tongil contributed to staple food self-sufficiency of Korea, an achievement that was termed the ‘Korean Green Revolution’. In this study, we analyzed the nucleotide-level genome structure of Tongil rice and compared it to those of the parental varieties. A total of 17.3 billion Illumina Hiseq reads, 47× genome coverage, were generated from Tongil rice. Three parental accessions, two indica and one japonica types, of Tongil rice were also sequenced for approximately 30x genome coverage. A total of 2,149,991 SNPs were detected between Tongil and Nipponbare; the average SNP frequency of Tongil was 5.77 per kb. Genome composition based on the SNP data by comparing with the three parental genome sequences on sliding window of Nipponbare genome sequence revealed that 91.8% of the Tongil genome originated from the indica parents and 7.9% from the japonica parent, different from the theoretical expectation in a three-way cross, i.e., 75% indica and 25% japonica parental origins on average. Copy number of SSR motifs, ORF gene distribution throughout the whole genome, gene ontology (GO) annotation, yield-related QTLs or gene locations, and polymorphic transposon insertions were also comparatively analyzed between Tongil and parents using sequence-based tools. The results indicated that each genetic factor was transferred from parents into Tongil in proportion to the whole-genome composition. The Tongil rice is the first successful superior cultivar derived from indica × japonica hybridization in Korea. Defining of genome structure demonstrates that the Tongil genome is composed mostly of the indica genome with a small proportion of japonica genome introgression. This work was supported by a grant from the Next-Generation BioGreen 21 Program (Plant Molecular Breeding Center No. PJ008125), Rural Development Administration, Republic of Korea.

We are currently developing a high-throughput single nucleotide polymorphism (SNP) genotyping service at IRRI to accelerate progress in rice breeding by providing rapid and cost-effective marker services. SNP marker development and validation is being performed based on cloned genes and QTLs, GWAS hits, and whole genome sequence data to identify predictive SNP markers at important genes for key traits for the breeding programs. Trait-based and targeted SNPs are being deployed in sets of 24 and 96 SNPs on a Fluidigm EP1 system. At the same time, 384 SNP sets and a 6K SNP chip developed by Susan McCouch at Cornell University are being used for higher density genome scans on an Illumina system. Genotyping by sequencing (GBS) approaches with 96 and 384 barcoded samples per sequence lane are also being evaluated in comparison to SNP array technology based on the number of loci, call rates, turnaround times, and cost per sample. An efficient sample processing workflow with an integrated LIMS is also being optimized to enable high throughput genotyping with sample tracking to minimize errors. Moreover, web-based SNP data analysis tools have been deployed through the IRRI Galaxy workbench to speed up SNP data analysis. Future efforts will focus on large-scale deployment of GBS across breeding materials to enable QC genotyping, tracking of donor introgressions, and integration of genome-wide prediction into the variety development pipelines. The large-scale application of high-density markers will help transform IRRI’s rice breeding programs and increase the rate of genetic gain towards developing high-yielding, stress-tolerant varieties for target environments and market segments

The main objectives of IRRI’s variety development should meet the needs of customers/farmers from diverse rice sectors in each target region. The dynamic market change asks rapid variety development with highly valued QTLs/genes. Molecular breeding implemented through the efficient crossing, high throughput genotyping and rapid generation advancement will provide packages to breeders to develop new varieties quickly and more economically. The more efficient and cost-effective marker-assisted backcrossing service will provide the more opportunity for the success in molecular breeding platform. To make MABC system more successful, the development of molecular marker system for the high-throughput SNP genotyping is must. Currently Genotyping Service Lab (GSL) of IRRI provides high-throughput SNP genotyping service using BeadXpress and Fluidigm system. Meanwhile, the linked SNP markers for the specific traits are being developed. For abiotic stress tolerances, the markers for submergence, drought, heat, anaerobic germination, salinity, and phosphorus deficiency for Fluidigm system are being developed and tested in variety diversity panel and segregating populations. In MABC, due to the high number of crossings, the labor- and space-saving crossing system is being developed. As a result of an integrated MABC platform will speed up the development of pre-breeding line which are containing single or multiple QTLs/genes.

This study aims to identify major quantitative trait loci (QTL) for yield components under low-input systems in tropical regions in rice. A total of 156 highly advanced recombinant inbred lines (RILs) have been developed from a cross between two temperate rice varieties, Dasanbyeo (Tongil-type indica) and TR22183 (japonica). Both parental lines and RILs were tested under two different regimes of irrigation and phosphorus (P) application levels. During the wet season of 2012, under mild drought conditions, TR22183 showed more vigorous root growth at 15 days after sowing than Dasanbyeo. The early root establishment of TR22183 may have contributed to the enhanced P uptake from the top soil in the early growth stage. The linkage map for DT-RILs was constructed with 312 single nucleotide polymorphism markers aided by 384-plex platform of BeadXpress high-throughput genotyping system. For the vegetative growth, major QTLs on chromsome 6 for plant height and tiller numbers were identified. For the grain yield related traits, major QTLs on chromsome 2 were closely linked to each other. On the other hand, panicle length QTLs were identified on chromsomes 2 and 9. We are currently analyzing phenotypic data and the experiment is being repeated in dry season and temperate region.

Hybrid sterility is one of the major barrier to the application of wide crosses in plant breeding and is commonly encountered in crosses between indica and japonica rice varieties. Ten mapping populations comprised of two reciprocal F2 and eight BC1F1 populations generated from the cross between Ilpumbyeo (japonica) and Dasanbyeo (indica) were used to identify QTLs and to interpret the gametophytic factors involved in hybrid fertility or sterility between two subspecies. Frame maps were constructed using a total of 107 and 144 STS markers covering 12 rice chromosomes in two reciprocal F2 and eight BC1F1 populations, respectively. A total of 15 main-effect QTLs and 17 significant digenic- epistatic interactions controlling spikelet fertility (SF) were resolved in the the entire genome map of F2 BC1F1 populations . Among detected QTLs responsible for hybrid ferility, four QTLs, qSF5.1 and and qS F5.2 on chromosome 5, qSF6.2 on chromosome 6, and qSF12.2 on chromosome 12 were identified as major QTLs since they were located at corresponding position in at least three mapping populations. Loci qSF5.1, qSF6.1 and qSF6.2 were responsible for both female and male sterility, whereas qSF3.1, qSF7 and qSF 12.2 affected the spikelet fertility only through embryosac factors, and qSF9.1 did through pollen factors. Five new QTLs identified in this study will be helpful for understanding the hybrid sterility and for breeding programs via inter-subspecific hybridization.

Hybrid sterility is one of the major barrier to the application of wide crosses in plant breeding and is commonly encountered in crosses between indica and japonica rice varieties. Ten mapping populations comprised of two reciprocal F2 and eight BC1F1 populations generated from the cross between Ilpumbyeo (japonica) and Dasanbyeo (indica) were used to identify QTLs and to interpret the gametophytic factors involved in hybrid fertility or sterility between two subspecies. Frame maps were constructed using a total of 107 and 144 STS markers covering 12 rice chromosomes in two reciprocal F2 and eight BC1F1 populations, respectively. A total of 15 main-effect QTLs and 17 significant digenic- epistatic interactions controlling spikelet fertility (SF) were resolved in the the entire genome map of F2 BC1F1 populations . Among detected QTLs responsible for hybrid ferility, four QTLs, qSF5.1 and and qS F5.2 on chromosome 5, qSF6.2 on chromosome 6, and qSF12.2 on chromosome 12 were identified as major QTLs since they were located at corresponding position in at least three mapping populations. Loci qSF5.1, qSF6.1 and qSF6.2 were responsible for both female and male sterility, whereas qSF3.1, qSF7 and qSF 12.2 affected the spikelet fertility only through embryosac factors, and qSF9.1 did through pollen factors. Five new QTLs identified in this study will be helpful for understanding the hybrid sterility and for breeding programs via inter-subspecific hybridization.