Characterisation of morphological, physiological and biochemical traits for heat tolerance in aerobic rice

The Australian rice industry, established in the cooler environment of NSW and Northern Victoria, is gradually expanding into North Australia. In Northern Australia, rice frequently experiences heat stress (>34°C) particularly at anthesis causing spikelet sterility, reduced yield, and grain quality. Wide genetic variation exists in the rice germplasm for heat tolerance which can be potentially harnessed for addressing the issues of spikelet sterility. This research evaluated contrasting rice genotypes under field conditions, semi-controlled (glasshouse), controlled (growth cabinet) environments at elevated temperatures for characterisation of rice morphological, physiological, biochemical, and reproductive traits associated with spikelet sterility. Under semi-controlled environment in the glasshouse, the spikelet sterility varied significantly between 27 genotypes and ranged from 0.8% to 74.3%. Hayayuki recorded the lowest spikelet sterility, with no negative effects from heat stress imposed at flowering. Swarna and KDML-105 had significantly higher spikelet sterility in the GH compared to all other genotypes, however the heat treatment did not significantly increase spikelet sterility. This may be due to the high temperatures that the spikelets were exposed to in the GH during flowering. Many genotypes had a significant increase in spikelet sterility when exposed to high temperatures at flowering under aerobic conditions, however Nagina 22 and Vandana showed the largest increase in spikelet sterility. Time of planting experiments for diverse genotypes were conducted to evaluate high temperature effects during microsporogenesis and at anthesis in pot and field experiment in the tropics. Spikelet sterility of up to 75% was recorded for the highly heat tolerant genotype Nagina 22 (N22) in November and December planting, whilst the heat susceptible genotype, Moroberekan recorded significantly lower spikelet sterility of 33 – 40% for the same planting dates. N22 flowered during February, 69 days after sowing, coinciding with high temperatures, whilst Moroberekan flowered in April coinciding with lower temperatures. Reducing heat induced spikelet sterility, flowering outside the hotter months (December to February) avoids heat stress. This can be achieved through selecting genotypes with a longer maturity date or adjust planting dates to synchronise anthesis when critically high or low temperatures are unlikely to occur and when plant available water is non-limiting. The effects of high temperature were also assessed through changes in the anther lipidome fatty acid composition. Six selected rice genotypes grown under controlled environment growth chambers were evaluated for spikelet sterility and fatty acid compositions of the anthers from plants grown throughout the crop period at normal unstressed temperature (28/21°C day/night), or transient exposure at high temperatures (38/28°C day/night) for a short period during flowering only, or prolonged exposure at high temperatures from flowering to iv seed maturity and harvest. In this experiment, N22, was the only genotype conferring heat tolerance across transient and prolonged heat stress. All other genotypes (Hayayuki, TeQing, Sasanishiki, Lemont and Moroberekan) showed severe spikelet sterility due to prolonged heat stress. N22 consistently demonstrated significantly higher level of saturated and unsaturated fatty acids, palmitic and stearic acid than the other genotypes. The fatty acid content and the composition of the anther lipidome may be linked with heat tolerance expressed as low spikelet sterility in rice at higher temperatures during flowering. A diverse population of 309 japonica genotypes were characterised in a genome wide association study in Northern Australia for genetic mapping of spikelet sterility and determination of QTLs associated with spikelet sterility in the tropical rice. Under tropical aerobic conditions, spikelet sterility ranged from 0.3 to 78.1%, with 16.8% of observations having a spikelet sterility of 5% or less, 22.3% of observations had a spikelet sterility of 5 to 10% and 60.9% of observations had a spikelet sterility greater than 10%. The marker traits association using the MLM (Q+K) workflow analysis within Tassel 5.0, generated the Manhattan plot. Using a LOD score of ≥3.5 as the significant threshold for the SNPs for percentage spikelet sterility, eight SNPs were identifiable for spikelet sterility in a tropical aerobic environment. These data suggest that the spikelet sterility in tropical rice are multigenic controlled and opportunity exist for selection of heat tolerance for rice genotype for breeding and developing new lines of crop that provide advantage for the adaptation in the tropics due to low spikelet sterility. This can contribute for developing new rice varieties needed for greater tropical adaptation, hence useful for the tropical rainfed rice production environment. An integrated approach for managing spikelet sterility through targeted breeding for spikelet sterility and the agronomic approach (time of planting) is suggested.