Title: Identification of genetic loci underlying the kernel fissure-resistance exhibited by 'cypress' and 'saber' Authors
Submitted to: Rice Technical Working Group Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: February 18, 2014
Publication Date: N/A
Technical Abstract: The economic value of broken rice is about half that of whole milled rice, so one goal of producers, millers, and rice breeders is to reduce broken grains that result from the dehusking and milling processes One of the primary causes of rice breakage is fissuring, or cracking, of the rice before it enters the mill. A common cause of rice fissuring is the exposure of drying, mature kernels to humid field or postharvest conditions that cause the kernels to reabsorb moisture. Fissures may be caused by rain or dew in not-yet-harvested fields. A few rice varieties produce grain more resistant to fissuring than others, and breeders would like to incorporate these genes into improved rice varieties. Identifying breeding progeny containing the desired genes is difficult, however, for traits such as kernel fissuring that are highly sensitive to variable environmental effects. Identification of molecular gene-tags to support marker-aided breeding selections is especially beneficial for environmentally sensitive traits. Marker-assisted selection (MAS) is based on the principle that when markers linked to a gene affecting a desired trait are selected, the physically linked trait is also in selected individuals. The present study took advantage of the fact that the reverse also holds true, and was accomplished by selecting for fissure resistance (FisR) versus fissure susceptibility (FisS) among the progeny of two populations, then identifying molecular marker alleles that were present in different proportions between the FisR and FisS subgroups. We identified marker-gene linkages in two populations, because identification of the same genes in multiple populations and environments increases confidence in those genes, and enhances our knowledge as to which genes will be most effective under a variety of genetic and field conditions, i.e., the genes most useful to breeders and the industry. The first study population was a set of 300 F2s from a cross between ‘Cypress’ (FisR) and ‘LaGrue’ (FisS). Cypress is also semidwarf whereas LaGrue grows to a taller plant height. We expected to find segregation for plant height among the Cypress x LaGrue progeny, but what was not anticipated was that all of the most FisR F2 and F3 progeny were of semidwarf height, while the FisS progeny were generally tall in height, suggesting the presence of a FisR gene near the sd1 gene on chromosome 1. Linkage between Cypress’ FisR and sd1 was confirmed by non-random distribution of several molecular markers along chromosome 1 among the FisR and FisS Cypress x LaGrue progeny. This raised the question of whether FisR was due to a different gene physically linked to the sd1 gene on chromosome 1, or if FisR was a secondary effect of the semidwarf gene itself, for example, caused by the distance between the grains on the plant and flood water during grain-fill. Non-random marker distribution among the Cypress x LaGrue progeny detected a FisR locus on chromosome 8 as well as on chromosome 1. The second QTL mapping population was derived from a cross between the variety ‘Cybonnet’, which presumably inherited its hull-related FisR from Cypress, and ‘Saber’, a variety which was previously shown to have a FisR mechanism that was not hull-related, and thus different from that in Cypress and Cybonnet. This population consisted of 280 recombinant inbred lines (CbSa-RILs), which, being pure-breeding, allowed us to use replication across years and locations to obtain better estimates of FisR of each of the progeny lines (2 replications each for TX2007, AR2007, and TX2009). Furthermore, because Cybonnet and Saber are both semidwarf in height, study of this population allowed us to evaluate the FisR genes in a genetic environment (population) where differences in plant height did not exist. The 280 CbSa-RILs were molecularly characterized using a SNP chip designed to identify polymorphisms between japonica genotypes. Of the 384 SNPs, 28 proved noninformative due to both parents being either null at that locus while 212 were polymorphic. There were 144 monomorphic SNPs suggesting several genomic regions to be identical by descent between Cybonnet and Saber, which are known to share ancestors. The QTL regions previously identified on chromosomes 1 and 8 were further saturated with the addition of 20 SSR loci. QTL mapping among the 280 CbSa-RILs confirmed the existence of three FisR alleles originating from Cypress (from CBNT? Do you know these are in CPRS?), two on chromosome 1, and one on chromosome 8. The fact that the CbSa-RILs did not segregate for plant height but showed linkage between markers linked to sd1 and FisR clarified that a FisR gene is linked to but not allelic with the sd1 locus. The FisR QTLs contributed by Cybonnet were all of higher confidence (LOD score) and larger phenotypic effect than those originating from Saber which mapped to chromosomes 5, 10, and 12. Grain shape QTLs were also located on these same chromosomes. However, shape does not appear to be a large driving factor of Saber’s FisR in that the single largest QTL for grain shape, on chromosome 3, was not associated with FisR. The markers we identified as linked to the FisR genes from Cybonnet and Saber can be used by breeders to improve the incorporation and stacking of these FisR alleles into improved U.S. rice varieties via MAS.