Natural Variations : New Resources for the Functional Genomics in Rice

Abstract

Artificial mutations showing discrete phenotypes have often been used for analysis of the molecular basis of plant morphology and physiology. In contrast, although a wide range of phenotypic variation is ob served among strains or cultivars, these natural varia tions have not been well analyzed genetically because of their complex inheritance. However, recent progress in plant genome analysis has enabled us to perform comprehensive analysis of such naturally occurring variations. New tools (DNA markers) and information (sequences) have been used for the genetic dissection of complex traits into single genetic factors. Here, we demonstrate that naturally occurring variations are new resources for the functional analysis of rice genes. Flowering time (heading date) is a major determi nant of regional and seasonal adaptation of cultivated rice. Photoperiod and other environmental factors con trol heading date. A large amount of variation is observed in heading date and photoperiodic response among rice cultivars and strains, including wild rela tives. Quantitative trait locus (QTL) analyses of prog eny derived from several cross-combinations of culti vars has suggested that more than 15 loci are involved in heading date. In addition, new QTLs have been identified by genetic analysis of a cross between a cultivar and a wild relative. Nearly isogenic lines (NILs) have been developed by marker-assisted selec tion and used for the analysis of epistatic interaction among QTLs. Map-based cloning has been performed on several QTLs for photoperiodic response. We have demonstrated that Hd] is an ortholog of CO in Arabi dopsis and is involved in the promotion and inhibition of heading under short-day (SD) and long-day (LD) conditions, respectively. Hd6 is involved in inhibition under LD conditions and encodes the alpha-subunit of protein kinase CK2. Hd3a shows a high level of simi larity to Arabidopsis FT and functions as a flowering inducer. Ehdl is involved in promotion under SD conditions and encodes a B-type response regulator. Hd5 is involved in inhibition under LD conditions and encodes a putative subunit of a CCAAT-box-binding protein. Lhd4 is involved in inhibition under LD condi tions and encodes a protein with a CCT motif. We have identified several quantitative trait nucleotides (QTNs) that result in loss of function of these QTLs . However, some QTNs defining differences at the level of function remain to be clarified. Combining information from genetic and sequencing analyses has revealed that combinations of natural alleles with loss or gain of function at particular QTLs, such as Hd1, Hd5, Hd6, Ehd1, and Lhd4, seem to generate a wide range of continuous variation in photoperiodic flowering in rice. These genetic and molecular analyses have allowed us to propose a genetic pathway controlling photoperi odic flowering in rice. Identification and expression analyses of genes suggest that there are both con served and divergent features in photoperiodic control of flowering in rice, an SD plant, and A rabidopsis, an LD plant. Marker-based genetic approaches can also be used to analyze other complex traits besides flowering time. Several primary mapping populations, including back cross inbred lines (BILs) and doubled-haploid lines, have been developed to facilitate the analysis of com plex traits. Several QTLs have been identified with major effects on such characteristics as seed dormancy, cool-temperature tolerance at the booting stage, seed shattering habit, seed longevity, and UV-B tolerance. Marker-assisted genetic approach greatly facilitates the exploitation of agronomically and scientifically impor tant genes in natural variants. In order to mine scien tifically or agronomically important alleles, we have been developing novel mapping populations, such as chromosome segment substitution lines. In these lines, a particular chromosome segment from a donor line is substituted in the genetic background of the recurrent line. The substituted segments cover all chromosomes in a whole set of lines. Their potential power in genetic analysis has been demonstrated in plants. These mate rials allow detailed and reliable QTL analyses . Marker-assisted approaches have contributed to our understanding of a wide range of complex traits in rice, including plant morphology and environmental stress tolerance. These approaches have enabled us not only to identify genetically the genes involved in complex traits, but also to identify them at the molecular level . In addition, the advent of DNA marker technology has made it possible to establish new breeding strategies , such as marker-assisted breeding. That strategy re quires comprehensive dissection and understanding of complex traits of interest. Once we have identified tightly linked markers, NILs for target QTLs with