Linkage mapping using mutant genes in rice

Because rice is an important food for nearly half the world's population, a considerable amount of data concerning rice genetics and breeding has been accumulated by many workers [204, 213, 222, 310].

Construction of linkage maps has also progressed steadily. However, owing to confusion due to inadequate knowledge of genetical analysis of complex characters such as anthocyanin coloration, linkage studies were delayed for a considerable time in comparison with other major food crops such as maize and barley [346].

However the number of marker genes has reached over 900 by the efforts of rice geneticists in the world and 571 of them have been assigned to the 12 chromosomes corresponding to the haploid chromosome number of rice.

Depending on the gene analysis of various morphological, physiological and biochemical characters, the marker genes were classified into 31 genetic traits as shown in Table 1. Many of these were identified as naturally occurring variations of spontaneous origin in varietal populations; others were induced through mutagenic treatments. The morphological markers affecting various plant parts such as roots, stems, leaves, panicles, florets and kernels are easiest to identify and have been thoroughly investigated. Genes for disease and insect resistance can be identified only on the basis of tests using pathogens and pests. Besides the usual gene analyses, a number of cloned genes and cDNAs from callus, endosperm, roots and other tissues were identified in a large-scale sequence analysis [259, 277, 477, 665].

On account of new developments in rice genetics, the need for examining gene symbols and linkage groups has been progressively intensified. For this reason, Nagao [345, 346], and Kadam and Ramiah [183] proposed a unified system of gene symbols. Following the international rule for gene symbolization of all organisms, which was accepted by the Tenth International Congress of Genetics, the working group of the International Rice Commission (IRC) provided guidelines for gene symbolization and a list of suggested gene symbols [44]. Finally, the Rice Genetics Cooperative (RGC), which was established to promote cooperation in rice genetics in 1985, adopted new rules for gene symbolization. Since then, one of the standing committees of RGC coordinates and monitors gene symbols. New gene symbols and revised linkage maps are published annually in Rice Genetics NewsLetter (RGN).

Table 1 lists the marker genes alphabetically along with classification of traits, linkage groups (chromosomes), genetic nature, mutagenesis and the references except for the genes not allotted to the linkage groups.

In spite of efforts of coordinators in RGC, there are still some problems on gene symbolization. In addition there are many unauthorized genes which were marked with asterisks in the table. Above all prompt registration with the convener is expected in accordance with rule 7 mentioned in RGN Vol. 3 (pp. 4-5). Thus the confusion related to the numbering and name of genes will be avoided.

According to Nagao [346], the first reliable linkage relation was established between C (brown apiculus color) and wx (glutinous endosperm) with recombination values of 20~22" [34, 562, 663]. Later, Nagao and Takahashi [348] were the first to construct 12 linkage groups. Cytogenetical approaches using reciprocal translocations and primary trisomics have been developed to investigate relationships between linkage groups and individual chromosomes [160, 161, 162, 484, 694]. Iwata and Omura [165], and Khush et al. [206] published independently new revised linkage maps based on their trisomic analyses.

On the other hand, Misro et al. [322] presented linkage maps based on the data of indica rice. However, it is difficult to identify different linkage groups because of a discrepancy in genic suppositions on anthocyanin coloration and related characters.

Rice chromosomes were difficult to distinguish from each other because of their small size and lack of characteristics, such as banding patterns and centromere positions. To avoid confusion among researchers, a rule was set up by RGC, indicating that chromosomes were assigned Arabic numerals in descending order of their pachytene length (or centromere position in case of ambiguity of length). After the joint observations and intensive discussions on the extra chromosome of primary trisomic plants, a unified system of numbering rice chromosomes was accepted during the Second International Rice Genetics Symposium in 1990. The numbers of linkage groups, trisomics, translocations and karyotypes followed this agreed system of the chromosome numbering [RGN Volumes, 7(p.13) and 12(p.9), Ref. 204, 214].

The construction of current linkage maps (1998) is presented as shown in Fig. 1 and Table 2. In the linkage maps, the Kosambi mapping function was used to correct the recombination values between genes, and the recombination values in the table are expressed as cM or percentage. Thus 571 genes were allotted to the twelve linkage groups and 209 marker genes were positioned on twelve chromosomes.

On the other hand, saturated molecular linkage maps are already constructed in several groups [31, 37, 90, 259, 302, 343, 451] and the integration of the different linkage maps has been progressed [228, 655, 657, 700]. Recently the positions of centromeres and arm locations of RFLP markers on the molecular linkage maps were detected through gene dosage analysis using the secondary and telotrisomics. Depending on the information the orientation of the maps and allocation of genes to short and long arms were resolved [210, 374, 534, 535]. Present linkage maps (Fig. 1) followed this orientation.

We need to intensify cooperative efforts to locate the unlocated genes to respective chromosomes and map those which have been assigned to the linkage groups. Some of the linkage groups (chromosomes 8, 9, 10, and 12) are still sparsely populated with marker genes. Efforts should be made to find additional markers for these linkage groups.

New linkage maps that integrate conventional and biochemical markers such as RFLP, RAPD, AFLP, microsatellite and cloned genes are highly valuable for future work in genetics and breeding. Continued efforts are needed to integrate all existing markers into one unified, densely populated map.

The utility of biochemical and molecular markers is based on finding tight linkages between RFLP markers and the target genes. Major genes tagged with molecular markers hitherto in rice linkage groups are listed in Table 2.

Once a gene is tagged with molecular markers, it can easily be transferred to other varieties by indirect selection with the molecular markers in the progeny from an appropriate cross. Tagging genes for blast resistance, bacterial blight resistance and insect resistance may ultimately lead to marker-aided selection and the cloning of those genes via chromosome walking (marker-based gene cloning) [314, 326, 596].

The rice gene Xa-21 which confers resistance to Xanthomonas campestris pv. oryzae race 6 was already cloned by the map-based cloning [440, 441, 640, 647]. The transgenic plants showed high levels of resistance to pathogens [541].

Mapping of quantitative trait loci (QTLs) in various characters provide precious information for establishing breeding programs, and the extensive cooperation among various institutions is desirable for this purpose, using common materials such as recombinant inbreds (RI), doubled haploids (DH) and backcross families (BC) under different environmental conditions [266, 279, 557, 595, 678]. Chromosomal regions associated with marker segregation distortion were estimated by comparative linkage mapping between different population types (BC, F₂, DH and RI )[660].

Genetic linkage maps constructed on the basis of orthologous loci were compared with each other in rice, wheat and maize [4, 48, 260, 626]. It was revealed that gene content and orders are highly conserved between different species within the grass family. Some chromosome rearrangements involving inversions and transpositions have arisen during or after speciation [46, 330, 331, 332]. Comparative maps may provide an opportunity to identify not only major gene loci but also QTLs of agricultural importance such as disease and insect resistance, flowering time, shattering and other seed characters [46, 265, 419].

Construction of a rice physical map covered by YAC or BAC clones have reached over half of the genome length [261], and also high resolution map of chromosome 4 was presented [673]. Physical maps are an important resource for most molecular research facilitating positional cloning of conventional genes, sequencing of genomic DNAs and analysis of chromosome and genome structure in detail.

In situ hybridization is another effective method for localizing specific DNA or RNA sequences on chromosomes [67, 68, 177]. Considerable variation was detected between genetic and physical maps of several chromosomes [86, 542]. Rice 5S and 17S rRNA genes (rDNA) have been mapped simultaneously by multicolor fluorescence in situ hybridization (McFISH) [384]. FISH is also effectively used for visualization of cloned genes and molecular markers. As an example, physical loci of the rice blast genes, Pib and Pita-2 and a rice A genome-specific tandem repeat sequence (Trs A) were clearly visible on the chromosomes [365, 385].

Thus, genome research in rice is going to enter into a new era. By applying the basic information on new linkage mapping, many efficient and reliable means for rice breeding will be developed in near future.

The author is greatly indebted to Dr. I. Takamure from the Faculty of Agriculture, Hokkaido University for his assistance.