Genome-Wide Responses to Selection and Genetic Networks Underlying Submergence Tolerance in Rice

Simultaneous improvement and genetic dissection of drought and submergence tolerances in rice (Oryza sativa L.) by selective introgression

Introduction

Drought and submergence are contrasting abiotic stresses that often occur in the same rice crop season and cause complete crop failure in many rain-fed lowland areas of Asia.

Methods

To develop rice varieties with good tolerances to drought and submergence, 260 introgression lines (ILs) selected for drought tolerance (DT) from nine BC2 populations were screened for submergence tolerance (ST), resulting in 124 ILs with significantly improved ST.

Results

Genetic characterization of the 260 ILs with DNA markers identified 59 DT quantitative trait loci (QTLs) and 68 ST QTLs with an average 55% of the identified QTLs associated with both DT and ST. Approximately 50% of the DT QTLs showed 'epigenetic' segregation with very high donor introgression and/or loss of heterozygosity (LOH). Detailed comparison of the ST QTLs identified in ILs selected only for ST with ST QTLs detected in the DT-ST selected ILs of the same populations revealed three groups of QTLs underlying the relationship between DT and ST in rice: a) QTLs with pleiotropic effects on both DT and ST; b) QTLs with opposite effects on DT and ST; and c) QTLs with independent effects on DT and ST. Combined evidence identified most likely candidate genes for eight major QTLs affecting both DT and ST. Moreover, group b QTLs were involved in the Sub1regulated pathway that were negatively associated with most group aQTLs.

Discussion

These results were consistent with the current knowledge that DT and ST in rice are controlled by complex cross-talks between or among different phytohormone-mediated signaling pathways. Again, the results demonstrated that the strategy of selective introgression was powerful and efficient for simultaneous improvement and genetic dissection of multiple complex traits, including DT and ST.

A Genetic Network Underlying Rhizome Development in Oryza longistaminata

The rhizome is an important organ through which many perennial plants are able to propagate vegetatively. Its ecological role has been thoroughly studied on many grass species while the underlying genetic basis is mainly investigated using a rhizomatous wild rice species-Oryza longistaminata. Previous studies have revealed that the rhizome trait in O. longistaminata is jointly controlled by multiple loci, yet how these loci interact with each other remains elusive. Here, an F₂ population derived from Oryza sativa (RD23) and O. longistaminata was used to map loci that affect rhizome-related traits. We identified 13 major-effect loci that may jointly control rhizomatousness in O. longistaminata and a total of 51 quantitative trait loci (QTLs) were identified to affect rhizome abundance. Notably, some of these loci were found to have effects on more than one rhizome-related trait. For each trait, a genetic network was constructed according to the genetic expectations of the identified loci. Furthermore, to gain an overview of the genetic regulation on rhizome development, a comprehensive network integrating all these individual networks was assembled. This network consists of three subnetworks that control different aspects of rhizome expression. Judging from the nodes' role in the network and their corresponding traits, we speculated that qRHZ-3-1, qRHZ-4, qRHI-2, and qRHI-5 are the key loci for rhizome development. Functional verification using rhizome-free recombinant inbred lines (RILs) suggested that qRHI-2 and qRHI-5, two multi-trait controlling loci that appeared to be critical in our network analyses, are likely both needed for rhizome formation. Our results provide more insights into the genetic basis of rhizome development and may facilitate identification of key rhizome-related genes.

Mapping of genomic regions associated with arsenic toxicity stress in a backcross breeding populations of rice (Oryza sativa L.)

BACKGROUND

Arsenic (As) is an unwanted toxic mineral that threatens the major rice-growing regions in the world, especially in South Asia. Rice production in Bangladesh and India depends on As-contaminated groundwater sources for irrigating paddy fields, resulting in elevated amounts of As in the topsoil. Arsenic accumulating in rice plants has a significant negative effect on human and animal health. Here, we present a quantitative trait locus (QTL) mapping study to identify candidate genes conferring As toxicity tolerance and accumulation in rice (Oryza sativa L.) seedlings. An early backcross breeding population consisting of 194 lines derived from a cross between WTR1 (indica) and Hao-an-nong (japonica) was grown in hydroponics for 25 days, from the seventh day exposed to an environmentally relevant concentration of 10 ppm As.

RESULTS

Arsenic toxicity leads to significantly negative plant responses, including reduced biomass, stunted plant growth, reduced leaf chlorophyll content, and increased shoot As concentration ranging from 9 to 20 mg kg- 1. Marker-trait association was determined for seven As-related traits using 704 single nucleotide polymorphism (SNP) markers generated from a 6 K SNP-array. One QTL was mapped on chromosome 1 for relative chlorophyll content, two QTLs for As content in roots were mapped on chromosome 8, and six QTLs for As content in shoots were mapped on chromosomes 2, 5, 6, and 9. Using the whole-genome sequence of the parents, we narrowed down the number of candidate genes associated with the QTL intervals based on the existence of a non-synonymous mutation in genes between the parental lines. Also, by using publicly available gene expression profiles for As stress, we further narrowed down the number of candidate genes in the QTL intervals by comparing the expression profiles of genes under As stress and control conditions. Twenty-five genes showing transcription regulation were considered as candidate gene nominees for As toxicity-related traits.

CONCLUSIONS

Our study provides insight into the genetic basis of As tolerance and uptake in the early seedling stage of rice. Comparing our findings with the previously reported QTLs for As toxicity stress in rice, we identified some novel and co-localized QTLs associated with As stress. Also, the mapped QTLs harbor gene models of known function associated with stress responses, metal homeostasis, and transporter activity in rice. Overall, our findings will assist breeders with initial marker information to develop suitable varieties for As-contaminated ecosystems.

Genetic Dissection and Simultaneous Improvement of Drought and Low Nitrogen Tolerances by Designed QTL Pyramiding in Rice

Drought and low nitrogen are the most common abiotic stresses limiting rice productivity in the rainfed rice areas of Asia and Africa. Development and adoption of green super rice (GSR) varieties with greatly improved drought tolerance (DT) and low nitrogen tolerance (LNT) are the most efficient way to resolve this problem. In this study, using three sets of trait-specific introgression lines (ILs) in a Xian (indica) variety Huanghuazhan (HHZ) background, we identified nine DT-QTL and seven LNT-QTL by a segregation distortion approach and a genome-wide association study, respectively. Based on performances of DT and LNT and genotypes at the detected QTL, two ILs M79 and M387 with DT and LNT were selected for cross-making to validate the identified QTL and to develop DT and LNT rice lines by pyramiding two DT-QTL (qDT3.9 and qDT6.3) and two LNT-QTL (qGY1 and qSF8). Using four pairs of kompetitive allele specific PCR (KASP) SNP markers, we selected 66 F2 individuals with different combinations of the target DT- and LNT-QTL favorable alleles and they showed expected improvement in DT and/or LNT, which were further validated by the significant improvement in DT and/or LNT of their F3 progeny testing. Based on evaluation of pyramiding lines in F3 lines under drought, low nitrogen (LN) and normal conditions, four promising pyramiding lines having different QTL favorable alleles were selected, which showed significantly improved tolerances to drought and/or LN than HHZ and their IL parents. Our results demonstrated that trait-specific ILs could effectively connect QTL mapping and QTL pyramiding breeding, and designed QTL pyramiding (DQP) using ILs could be more effective in molecular rice breeding for complex quantitative traits.

Simultaneous Improvement and Genetic Dissection of Salt Tolerance of Rice (Oryza sativa L.) by Designed QTL Pyramiding

Breeding of multi-stress tolerant rice varieties with higher grain yields is the best option to enhance the rice productivity of abiotic stresses prone areas. It also poses the greatest challenge to plant breeders to breed rice varieties for such stress prone conditions. Here, we carried out a designed QTL pyramiding experiment to develop high yielding "Green Super Rice" varieties with significantly improved tolerance to salt stress and grain yield. Using the F4 population derived from a cross between two selected introgression lines, we were able to develop six mostly homozygous promising high yielding lines with significantly improved salt tolerance and grain yield under optimal and/or saline conditions in 3 years. Simultaneous mapping using the same breeding population and tunable genotyping-by-sequencing technology, we identified three QTL affecting salt injury score and leaf chlorophyll content. By analyzing 32M SNP data of the grandparents and graphical genotypes of the parents, we discovered 87 positional candidate genes for salt tolerant QTL. According to their functional annotation, we inferred the most likely candidate genes. We demonstrated that designed QTL pyramiding is a powerful strategy for simultaneous improvement and genetic dissection of complex traits in rice.

Characterization of Salt-Induced Epigenetic Segregation by Genome-Wide Loss of Heterozygosity and its Association with Salt Tolerance in Rice (Oryza sativa L.)

In a breeding effort to develop salt tolerant (ST) rice varieties by designed QTL pyramiding, large numbers of progenies derived from four crosses between salt- or drought- tolerant BC2F5 IR64 introgression lines, were subjected to severe salt stress, resulting in 422 ST plants. The progeny testing of the selected F3 lines under more severe salt stress resulted in identification of 16 promising homozygous lines with high levels of ST. Genetic characterization of the 422 ST F3 progeny and 318 random F2 plants from the same four crosses using 105 segregating SSR markers lead to three interesting discoveries: (1) salt stress can induce genome-wide epigenetic segregation (ES) characterized by complete loss of heterozygosity (LOH) and nearly complete loss of an allele (LOA) in the F3 progenies of four rice populations in a single generation; (2) ∼25% of the stress-induced ES loci were transgenerational and inherited from their salt- and drought- selected parents; and (3) the salt-induced LOH and LOA loci (regions) appeared to contain genes/alleles associated with ST and/or drought tolerance. 32 genomic regions that showed one or more types of salt-induced ES in the random and salt-selected progenies from these crosses. The same or different types of ES were detected with two large genomic regions on chromosomes 1 and 6 where more and the strongest ES were found across different populations. 14 genomic regions were found where the salt-induced ES regions were overlapping with QTL affecting ST related traits. The discovery of the three types of salt-induced ES showed several interesting characteristics and had important implications in evolution and future breeding for developing stress-resilient rice and crops.