Take home messages

• Wheat is most susceptible to heat stress during grain-filling
• Multiple Environment Analysis (MET) can identify important genotype and environmental drivers for complex outcomes such as wheat yield
• Biomass at maturity (r = 0.64), grain protein content (r = -0.58), harvest index (r = 0.48), and grain-filling duration (r = 0.27) were the key measurable traits contributing to yield across different Australian environments
• Night-time temperature (average seasonal minimum and daily minimum during vegetative development) explained most of the underlying genetic variance for wheat yield with strong negative associations between yield and elevated night-time temperatures evident in NSW and Victoria
• Wheat genotypes varied strongly for their grain-filling performance under heat stress conditions, however, two important traits were found (increased grain-filling rate and stem biomass remobilisation for both grain weight and yield maintenance
• The water-soluble carbohydrate composition, distribution, and quantities during grain-filling in wheat stems were also a key indicator of heat tolerance
• In a controlled environment study, low night-time temperatures were shown to ameliorate high day-time temperature stress in wheat.

Introduction

Heat stress can cause significant yield and quality reductions in wheat and is common during the sensitive grain-filling phases of wheat in many environments across Australia. Thermotolerance in wheat has been improved through breeding approaches and adoption of specific traits (i.e. biomass, phenology, harvest index, grain number) which drive yield. However, greater advancements of analytical methods and plant genetics are required for improving current understanding and driving new discoveries for yield improvement in wheat. Greater sources of heat tolerance are required and on-going due to the current detrimental impacts of heat stress and the expected increase in stress (i.e. heat) event intensity, duration, and frequency in future climate change scenarios.

Heat stress is known to reduce the duration of grain-filling in wheat, resulting in smaller grain size which reduces yield and quality. Multiple compensatory mechanisms exist to assist with reducing the duration of grain-filling and hence exposure to heat stress, mostly through increasing grain-filling rates, although the drivers for these beneficial traits are poorly understood. Despite grain-filling rates increasing under heat stress, it still cannot compensate for the reduced duration of grain-filling associated with heat stress. Grain-filling is a complex development phase, particularly in relation to carbon utilisation, as photosynthetic capacity, senescence, sink signalling, and remobilisation of stored reserves all attempt to contribute carbon and nitrogen to the grain in final stages of development. A lack of understanding remains about the key mechanisms limiting grain-filling in wheat under heat stress, and different conclusions within the scientific community have been drawn for their trade-offs, particularly between grain-filling rate and stay-green traits.

Accumulated water-soluble carbohydrates (WSC) in the stem are very beneficial for wheat thermotolerance and assist in mitigating yield and quality loses caused by heat stress, particularly during grain-filling. Under ideal conditions wheat is known to be an over producer of resources for yield, however, under stress conditions mechanisms for yield and quality reductions are poorly understood. WSC stored in the stems act as a reserve pool the plant can utilise in times of stress or demand, alleviating the dependence on photosynthesis to produce carbon. The WSC profile of wheat is very dynamic, both spatially (i.e. stem internodes) and temporally (i.e. at different growth stages). At anthesis, the WSC composition has been reported as 80% sucrose, 2% glucose and 2% fructose, while later in grain-filling (20–30 days after anthesis; DAA) the composition is primarily fructans (>80%) (Zhang et al., 2015). Sucrose is a disaccharide and has been identified as a key molecule for signalling (stress response, vegetative and reproductive development) and provides building blocks (i.e. protein, cellulose, and starch) for plant tissue (Ruan 2014). Fructans on the other hand are large linear or branched polymers comprised of fructose monomers and a terminal glucose, and can bind to membranes under stress improving membrane stability under abiotic stress (Kerepesi et al., 1998; Vágújfalvi et al., 1999), or may be utilised as a form of polymer trapping (Wang and Nobel 1998). Nevertheless, the composition and distribution of WSC in wheat stems are often overlooked, particularly under heat stress conditions, due to the impact drought has on photosynthesis.

Time of sowing (TOS) experiments across Australia

The aim of these experiments was to replicate core genotypes across different temperature environments and regions of Australia to highlight the contributions and drivers of measurable traits and environmental variables for yield. 32 diverse genotypes (elite pre-breeding material and commercially available; data not shown) were grown over three years (2017–2019) in three key wheat production regions of Australia (NSW, Victoria, and Western Australia) and each trial utilised multiple times of sowing (TOS) treatments (TOS 1 – optimum, TOS 2 – one month late, TOS 3 – two months late) creating 17 unique environments (Table 1) to identify the key genotypes, environmental, and trait associations between each year, region, and TOS interaction. Genotypes varied significantly for crucial traits, including yield, biomass, thousand kernel weight (TKW), phenology, test weight, grain protein, screenings, plant height and harvest index. Each environment was controlled for drought stress through supplementary irrigation if necessary to allow for the impact of heat stress alone to be examined.

Table 1. Location, year and sowing dates (TOS) of replicated field experiments.

Using a combination of forefront statistical methods, including multiple environment trait (MET), and multiple-environment, multi-trait MET analysis alongside proven broad-sense heritability and environmental covariate correlations, allowed for the discovery of site-specific interactions of environmental conditions and physiological traits driving yield performance. This study identified significant genotype x environment (G x E) variations between environments, primarily due to regions and years over TOS effects, while the most important traits for yield across all environments included maturity biomass (r = 0.64), grain protein content (r = -0.58), harvest index (r = 0.48), and grain-filling duration (r = 0.27). Unexpectedly, broad-sense heritability was largely influenced by region, and some traits (anthesis and maturity days after sowing, grain-filling duration, protein and screening percentages) were significantly correlated with changes in average seasonal temperatures.

Environmental covariate analysis observed vegetative minimum temperature explained the most underlying genetic variance from the MET-analysis and highlighted the strong negative associations for elevated night-time temperatures within the NSW and Victorian environments in relation to yield. Significantly greater understanding and improvements for important traits such as yield and dynamic drivers (maturity biomass, harvest index, grain-filling duration, and minimum temperatures) in a range of environments (regions and/or TOS) were attained.

Drivers of grain-filling performance under heat stress conditions

More detailed measurements were taken on 14 selected genotypes at Narrabri in 2019, to assess mechanisms which drive heat tolerance during grain-filling. Measurements were taken of phenology (heading, anthesis, physiological maturity), morphology (viable tiller number, biomass at anthesis and physiological maturity, and plant height), yield and grain components (TKW, screenings, test weight, and protein content). Additionally, main stems were collected weekly from anthesis to physiological maturity to assess responses throughout grain-filling (grain weight, grain number, individual grain weight, stem biomass) between the two TOS.

Key differences were evident between the genotypes and their grain-filling performance between the two environments (data not shown). The rate of grain-filling in the stress environment (TOS 3) was overall +10.45% faster than the optimum environment (TOS 1), however, three genotypes reduced or had zero change in rate from the optimum to the stressed environment highlighting susceptibility. The strongest correlations for final main stem grain weight were found in response to rate (g-1 week-1) across both TOS 1 (R2 = 0.93) and TOS 3 (R2 = 0.85) (Figure 1).

Figure 1. Linear regression between final grain weight of the main stem at physiological maturity and the rate (g-1 week-1) estimated from the change of grain weight from anthesis to physiological maturity for all genotypes (14) in both TOS 1 (n =4) and TOS 3 (n = 4).

Figure 2. Linear regression analysis for stem biomass remobilisation rate (g-1 week-1) and final grain weight at maturity (n = 4) across both TOS 1 (R² = 0.23) and TOS 3 (R² = 0.33) for all genotypes.

The average rate (g-1 week-1) for stem biomass remobilisation was slightly faster in TOS 1 at
-0.20 (± 0.002 SE) compared to TOS 3 at -0.19 (± 0.003 SE). In TOS 3, five genotypes did not increase stem remobilisation rate which highlighted susceptibility, whereas the tolerant genotypes increased stem remobilisation rate under stress. Stem remobilisation rate (g-1 week-1) was moderately associated (R² = 0.47) with final grain weight under stressed conditions (TOS 3), compared to a low association at the ideal sowing time in TOS 1 (R² = 0.18) (Figure 2).

Susceptible and tolerant genotypes varied for grain-filling performance, namely comparing grain size and grain number between different environments. Tolerant genotypes employed a combination of two methods to mitigate stress, such as the increased rates of grain-filling (i.e. grain weight over time), and stem biomass remobilisation (i.e. stem biomass over time) for main spike grain weight and yield maintenance respectively. In contrast, susceptible genotypes were characterised by their reduced remobilisation efficiency of stem biomass (rate and mass), which limited both grain-filling and yield maintenance under heat stress. The delayed TOS treatments greatly limited the grain-filling performance of all genotypes. However, the tolerant genotypes were able to increase grain-filling rate, accumulate and remobilise stem biomass, and maintain grain number throughout the grain-filling period and across ideal (TOS 1) and stressed environments (TOS 3).

Accumulation and reallocation of water-soluble carbohydrates in wheat stems during grain-filling

Significant genetic x environment (G x E) interactions were found (data not shown), where WSC accumulation quantity and duration, composition (fructans, sucrose, fructose, and glucose), and remobilisation (quantity and efficiency) during the grain-filling phase were affected. Heat tolerant genotypes were able to increase pre-anthesis accumulated WSC (mainly fructans and sucrose) within the stressed environment, in either the lower large storage capacity internodes or higher in the peduncle. In contrast, susceptible genotypes significantly reduced accumulation quantity during pre- and post-anthesis phases, especially in the lower internodes (>60% and >50% respectively). The ability to increase pre-anthesis WSC and maintain post-anthesis WSC accumulation (quantity and duration) were major indicators of thermotolerance and were associated with greater grain-filling of the main stem under stress conditions.

Priming, low night-time temperatures, and high reserve pool content improves grain-filling and yield in response to terminal heat stress

Plants which experience non-lethal abiotic stress during vegetative growth (aka ‘priming’) have been shown to have improved responsiveness to subsequent heat stress during highly susceptible reproductive phases by upregulating defensive metabolites, promoting carbon assimilation and translocation. Measuring these beneficial responses can assist in the identification of thermotolerant mechanisms and could be targeted to improve heat tolerance in wheat.

Six genotypes varying for thermotolerance and susceptibility were grown in controlled environment conditions (24/12°C) and exposed to early non-lethal priming (35/28°C for 48 hours at the seven-leaf stage) and varying heat stress treatments (4 or 7 days at 38/5°C at 21 days after anthesis, DAA) during peak fructan accumulation (Table 2). Detailed measurements of numerous traits were used to determine key hypothesised adaptiveness and alterations to sink and source relations under heat stress during peak grain-filling.

Priming incurred a slight improvement across measured traits; however, genotypes varied greatly in their responsiveness to priming and subsequent heat stress, particularly in yield and grain components, photosynthetic attributes, and WSC composition and accumulation. Interestingly, genotype 20 improved yield under heat stress irrespective of priming treatments, which may have resulted from high WSC content (data not shown) in comparison to other genotypes during the stress event.

Table 2. Summary of the factorial treatment combinations used in glasshouse experiment, which including two priming (N/P) and three different heat treatment durations (4D and 7D) and control.

Low night-time temperatures appeared to ameliorate and improve overall performance (i.e. yield) under high day-time temperature stress during grain-filling. Despite being exposed to high day-time temperature stress (38 °C) over consecutive days (4 or 7) improving plant performance highlights the significance of night-time temperatures. Other studies have shown considerably larger yield losses (-19%) under lower temperature stress conditions for the same duration (1 week) imposed at 25 DAA (Castro et al., 2007). Similarly, key grain components (grain length, grain width, and grain thickness) were significantly reduced due to heat stress during early grain-filling (11–14 DAA) from only a three-day exposure (38/20°C) (Rangan et al., 2019). The heat stress treatment used in this study had low night-time temperatures (38/5°C) compared to other studies (Castro et al., 2007; Rangan et al., 2019). Consequently, the low night-time temperatures may have significantly reduced the carbon loss through night-time respiration (Impa et al., 2019) and mitigated the effect of day-time heat stress and respiration advantages of the tolerant genotypes (Posch et al., 2019). Greater considerations are required for the carbon availability (stem WSC content) and night-time temperatures when comparing and assessing thermotolerance during grain-filling in future experiments.

Conclusion

These PhD studies highlight the importance of conducting large-scale field trials to ensure identification of relevant traits and research questions associated with heat stress in wheat before further investigation in subsequent controlled environment experiments. Key traits and sources of tolerance were identified in these studies, however, further investigation into the associated genetics and mechanisms for carbon accumulation and remobilisation such as sink signalling, photosynthesis (sucrose export from source tissue), remobilisation (composition change – polymer to monomer), and utilisation (carbon-use efficiency) needs to be conducted. These studies also highlighted the importance of night-time temperatures in heat stress tolerance, which seem to be much more significant than day-time temperatures and are currently poorly understood in wheat.

References

Castro M, Peterson CJ, Rizza MD, Dellavalle PDí, Vázquez D, IbáÑez V, Ross A (2007) Influence of Heat Stress on Wheat Grain Characteristics and Protein Molecular Weight Distribution. Wheat Production in Stressed Environments. (Eds HT Buck, JE Nisi, N Salomon) pp. 365–371. (Dorhrecht: Springer Netherlands: Mar Del Plata, Argentina).
(doi: 10.1007/1-4020-5497-1_45)

Impa SM, Sunoj VSJ, Krassovskaya I, Bheemanahalli R, Obata T, Jagadish SVK (2019) Carbon balance and source-sink metabolic changes in winter wheat exposed to high night-time temperature. Plant Cell and Environment 42,1233–1246.
(doi: 10.1111/pce.13488)

Joudi M, Ahmadi A, Mohamadi V, Abbasi A, Vergauwen R, Mohammadi H, Van den Ende W (2012) Comparison of fructan dynamics in two wheat cultivars with different capacities of accumulation and remobilization under drought stress. Physiologia Plantarum 144, 1–12.
(doi: 10.1111/j.1399-3054.2011.01517.x)

Kerepesi I, Galiba G, Bányai É (1998) Osmotic and Salt Stresses Induced Differential Alteration in Water-Soluble Carbohydrate Content in Wheat Seedlings. Journal of Agricultural and Food Chemistry 46, 5347–5354.
(doi: 10.1021/jf980455w)

Méndez-Espinoza AM, Garriga M, Ben Mariem S, Soba D, Aranjuelo I, Del Pozo A (2021) Carbohydrate and amino acid dynamics during grain growth in four temperate cereals under well-watered and water-limited regimes. Agronomy 11.
(doi: 10.3390/agronomy11081516)

Posch BC, Kariyawasam BC, Bramley H, Coast O, Richards RA, Reynolds MP, Trethowan R, Atkin OK (2019) Exploring high temperature responses of photosynthesis and respiration to improve heat tolerance in wheat. Journal of Experimental Botany 70, 5051–5069.
(doi: 10.1093/jxb/erz257)

Rangan P, Furtado A, Henry R (2019) Differential response of wheat genotypes to heat stress during grain filling. Experimental Agriculture 55, 818–827.
(doi: 10.1017/S0014479718000406)

Ruan Y (2014) Sucrose Metabolism: Gateway to Diverse Carbon Use and Sugar Signaling. Annual Review of Plant Biology 65, 33–67.
(doi: 10.1146/annurev-arplant-050213-040251)

Vágújfalvi A, Kerepesi I, Galiba G, Tischner T, Sutka J (1999) Frost hardiness depending on carbohydrate changes during cold acclimation in wheat. Plant Science 144, 85–92.
(doi: 10.1016/S0168-9452(99)00058-8)

Wang N, Nobel PS (1998) Phloem Transport of Fructans in the Crassulacean Acid Metabolism Species Agave deserti. Plant Physiology 116, 709–714.
(doi: 10.1104/pp.116.2.709)

Zhang J, Chen W, Dell B, Vergauwen R, Zhang X, Mayer JE, Van den Ende W (2015) Wheat genotypic variation in dynamic fluxes of WSC components in different stem segments under drought during grain filling. Frontiers in Plant Science 6, 1–11.
(doi: 10.3389/fpls.2015.00624)

Acknowledgements

The research undertaken as part of this project is made possible by the significant contributions of growers through both trial cooperation and the support of the GRDC, the author would like to thank them for their continued support.

Special thank you to all the collaborators, fellow PhD students, and staff who contributed to this project including Australian Grain Technologies, Birchip Cropping Group, Statistics for Australian Grain Industry, Australian National University, and University of Newcastle.

Contact details

Dr Mitchell Clifton
Department of Primary Industries and Regional Development
4 Marsden Park Road, Calala, NSW, 2340
Email: mitch.clifton@dpi.nsw.gov.au
X: @mclifton_DPI

Date published
February 2025