Ll Qiang, WANG Zheng-rui, Ll Ding, WEl Jian-wei, QlAO Wen-chen, MENG Xiang-hai, SUN Shu-luan,Ll Hui-min, ZHAO Ming-hui, CHEN Xiu-min, ZHAO Feng-wu

Dryland Farming Institute, Hebei Academy of Agricultural and Forestry Sciences/Key Laboratory of Crop Drought Tolerance Research of Hebei Province, Hengshui 053000, P.R.China

1. lntroduction

Extreme weather events in recent years have become frequent due to global warming. Heat waves and increased temperatures have become especially damaging, causing major decreases in crop yields (Goodinget al. 2003;Hoffmannet al. 2006; Figueiredoet al. 2015). Wheat(Triticum aestivumL.) is one of the most important cereal crops in the world. As climates warm, heat stress during the post-anthesis period (terminal heat) has negatively affected wheat production. This increased temperature not only hastens the phenological stages of wheat development but also reduces the duration of the grain filling stages,thereby lowering grain yield and quality (Farooqet al. 2011).Wheat yield losses at this rate severely threaten national food security. According to Wiegand and Cuellar (1981),1°C increases from mean daily air temperatures between 15.8–27.7°C during wheat development would shorten the grain filling period by 3.1 days and decrease the weight per grain by as much as 2.8 mg. Two different cultivars in two cropping systems were used in the aforementioned study and the effect of temperature was quantified by a regression equation. Here, we measured the effect of temperature on yield and kernel weight in 48 winter wheat cultivars. The effect of temperature variation on these two wheat quality characteristics were easily measured and still informative.Research on heat tolerance remains one of the most limited understandings in wheat genetics and breeding. This is mainly due to the lack of effective methods to quantify heat stress and heat tolerance in wheat cultivars. Currently,heat stress quantification methods have been utilized under different stress temperatures (i.e., 42°C, Tewari and Tripathy 1998; 45°C, Sharkey 2005; 40°C, Allakhverdievet al. 2008).In this study, we used stress intensity for temporary analysis to offer more detailed research on temperature ranges for heat stress.

Many plant physiological processes are often measured as a gauge for heat tolerance phenotypes. One process is tetrazolium triphenyl chloride reduction, and is related to early heading high temperature index and stay-green,three phenotypes that may be used as heat tolerance indicators (Towill and Mazur 1974; Xuet al. 2000; Rane and Nagarajan 2004; Tewoldeet al. 2006). However, there is no direct evidence elucidating the correlation between these processes and wheat cultivar yield. Other characteristics like photosynthetic rate, leaf chlorophyll content, canopy temperature depression, membrane stability, and flagleaf stomata conductance may be correlated with field performance, especially with grain filling under heat stress(Reynoldset al. 1994, 1998; Amaniet al. 1996; Blumet al.2001). These characteristics may be used to select or screen wheat germplasm resources for heat tolerance. However,their numerical values cannot completely reflect the heat tolerance among different wheat cultivars. The ultimate indicator of cultivar-specific heat tolerance is manifested in the absolute yield and relative yield, that is, both yield potential and yield stability. An ideal cultivar-specific heat tolerance would exhibit a durable, consistent yield with the minimal yield reduction (YR) under heat stress conditions.

The aim of this study was to use various wheat cultivars to evaluate a new method for heat tolerance quantification and compare it to three established methods. We designed a portable greenhouse with temperature control for our experimental purposes. Using this design, we proposed a new measurement for heat tolerance that we designated the heat tolerance index (HTI). The results indicated that HTI calculated by yield and YR significantly correlated with cultivar-specific wheat yield and YR under heat stress.The new quantification method combining yield potential and yield stability could be a useful selection method for characterizing cultivar performance under heat stress.

2. Materials and methods

2.1. Cultivars

A total of 48 winter wheat (Triticum aestivumL.) cultivars(Table 1) in the North China Plain were evaluated for heat tolerance. These cultivars were mechanically sown by WINTERSTEIGER (Australia) at the research station in Hengshui City, Hebei Province (37°44′N, 115°42′E; elevation 20 m above mean sea level). A commonly used control wheat cultivar in Hebei Province regional experiments,Heng 4399, was selected as the control cultivar in this study.

The experiment was conducted on sandy loam soil in field conditions. A split block design with three replicates and two treatments was employed during the 2009–2010 and 2010–2011 crop seasons. Each cultivar was grown in a plot size of 11.16 m2(nine rows of 8 m length with 15.5 cm space between rows). After making an adjustment for seed size, the seed rate was maintained at a uniform population of 300 plants m–2(3 million plants per hectare), according to1 000-grain weight and germination percentage. Standard agronomic practices recommended for normal fertility(340 kg N ha–1:172.5 kg P2O5ha–1:40 kg K2O ha–1) were followed. Full rates of K2O and P2O5were applied at the time of sowing. Nitrogen was supplied in split applications:170 kg N ha–1at sowing, and 170 kg N ha–1at the first irrigation. Care was taken to avoid moisture and biotic stress by ensuring timely irrigation and pesticide control.

Table 1 The 48 winter wheat cultivars evaluated in this study

2.2. Heat treatments

Fifteen days after the anthesis for 80% of these wheat cultivars, the temperature-controlled phytotron greenhouse was transferred to cover the 48 cultivars. These were covered for 20 days, increasing the maximum temperature from 28.3 to 44.0°C and from 31.5 to 43.3°C during 2009–2010 and 2010–2011, respectively (Fig. 1). According to the 30-year meteorological data for this area, the temperature inside the portable greenhouse was set to (5±0.5)°C higher than the outside temperature to more realistically simulate the heat stress of this region. Untreated controls were also tested for each cultivar.

2.3. Heat tolerance assessment

Four methods of heat tolerance assessment were used in this study. One method of heat tolerance uses a subjective visual rating system assessing the stay-green trait (Wanouset al. 1991). The stay-green trait has historically been used for determining stress tolerance (Borrelet al. 2000; Joshiet al. 2007a). The stay-green trait of wheat flag leaves was measured in this study. The flag leaf was chosen for assessment as it is considered to be the greatest contributor to grain yield because of its close location to the wheat spikes. It also remains green longer than the other leaves.Our wheat field ratings were noted as green leaf retention(GLR). GLR was calculated by estimating the percentage of flag leaf area that was green. Ten randomly selected plants were used to assess flag leaf GLR in each plot at the 5th day after heat treatment (Fig. 2).

The second method for assessing heat tolerance was based on the capacity for grain filling under heat stress(Fokaret al. 1998). This was calculated as the difference between the 1 000-kernel weight under non-stress and stress treatments. The kernel weight reduction (KWR)formula is as follows: KWR=1–kw/kwp, where, KWR is the reduction rate of 1 000-kernel weight, kw is 1 000-kernel weight under heat stress, andkwpis 1 000-kernel weight under non-stress.

The third method used to assess heat tolerance in wheat was the ‘susceptibility index’ (S). Fischer and Maurer (1978) proposed the calculation of a simple S that provides a measure of stress tolerance based on yield loss under stress as compared to optimum conditions. For each genotype, S estimates the rate of change in yield between the two environments relative to the mean change for all genotypes, that is:S=(1–Y/Yp)/(1–X/Xp), whereYis yield under stress,Ypis yield without stress andXandXprepresent average yield over all cultivars under stress and non-stress conditions, respectively. The term (1–X/Xp) is defined as ‘stress intensity’ (δi). This method is commonly used for estimating stress tolerance (Blumet al. 1989; Lazaret al. 1995; Ahmadet al. 2003; Villegaset al. 2007; Masonet al. 2010).

The fourth method used for assessing heat tolerance includes a new parameter entitled the HTI. HTI is based on both yield potentiality and stability under heat stress,and was determined using a modified formula given for the S (Fischer and Maurer 1978). In order to indicate the yield potential under heat stress, the absolute yield under heat stress (Y) was amended into the formula. In order to remove the effect of variation in yield potential under non-stress, the relative yield (HRC=Y/YP, where,HRCmeans heat related yield) was calculated. The relative yield has been used in many studies for the understanding of stress tolerance (e.g.,Chinoy 1947; Blum 1973; Fischer and Maurer 1978). In addition, control cultivar corresponding data was used in the formula to remove other non-relevant variable disturbances like heat intensity.

Fig. 1 The maximum temperature inside and outside of the greenhouse in the 2009–2010 (A) and 2010–2011 (B) seasons. An approximate 5°C differences between the inside and outside temperatures were controlled for 20 days. Bars mean SE.

Fig. 2 Green leaf retention (GLR) ratings in wheat flag leaves post-heat stress treatment. On the scale of 0 to 10 used for GLR,0=100% green, 1=90% green, 2=80% green, 3=70% green, 4=60% green, 5=50% green, 6=40% green, 7=30% green, 8=20%green, 9=10% green, and 0=no green. Using this scale, the genotypes were classified into four categories: resistant (>0 and ≤1; I),moderately resistant (≥2 and ≤5; II), moderately susceptible (≥6 and ≤8; III), and susceptible (≥9 and ≤10; IV).

The cultivars were then classified by HTI as resistant(≥1.20), moderately resistant (1.00 to ≤1.19), moderately susceptible (0.80 to ≤0.99), and susceptible (≤0.79)(Table 2).

Analysis of variance, multiple comparisons (Duncan test), and correlation analysis were performed using SAS software.

3. Results

3.1. Meteorological data

The maximum temperature inside and outside of the greenhouse during the 2009–2010 and 2010–2011 crop growing seasons are presented in Fig. 1. The highest average temperature recorded was 35°C in 2005, and the highest temperature during the grain filling stage was 40.5°C, also in 2005. These maximum temperatures have a difference of 5.5°C. We used this temperature difference in the controlled greenhouse experiments by exposing the treatment plants to a heat stress treatment which 5°C higher than the control treatment.

The range in the average daily maximum air temperatures per year recorded since 1990 were the lowest in 1990 at 32.2°C and the highest in 2016 at 34.1°C (Fig. 3). A rising trend in the average daily maximum temperature indicated an increase in environmental heat pressure during the wheat grain filling stage in this region.

Table 2 The opinion scale of heat tolerance index (HTI) about different wheat cultivars

3.2. Heat treatment

The drastic increase in heat treatment applied at 7 DAA (days after anthesis) in 2010 due to a spike in outside temperature created a heat shock at 44°C in the greenhouse. This temperature was a chronic high temperature by 15 DAA in 2011. Although these temperatures varied greatly from year to year, our heat treatment experimental design is advantageous for heat tolerance studies for two reasons.First, unlike other studies where heat treatments are carried out from seedling stage or kept constant at some high temperatures (Lobellet al. 2007), here we provided more natural temperature fluctuations at the grain feeding stage that can be used to create a database for modeling and predicting the impact on grain output caused by future extreme weather, up to 5°C higher. Second, this method saves time by isolating treatments at the crucial developmental grain feeding stage for yields, allowing a short experimental time window with simple temperature effect quantification on grain yield and other characteristics.

3.3. Wheat cultivars evaluation

Using at-test for pairwise comparisons, a significant difference was found between the mean of the 48 cultivars compared to the control cultivar for YR, GLR, and KWR over the 2-year study (Table 3).

The five cultivars Nongda 318, Nongda 212, Nongda 189,Nongda 3492, and Yannong 19 of the 48 cultivars tested had higher yields and lower reduction under heat stress than the mean value of all cultivars across the two crop seasons.These result indicated heat tolerance may have a genetic basis and breeding for this trait is possible.

Mean yield in the 48 cultivars not subjected to heat stress was 6 744.3 kg h–1. The mean yield of the 48 cultivars was 5 762.2 kg ha–1when they were treated with heat stress in 2010.δifor yield in this experimental year was 0.15. Significant (P＜0.01) differences were detected when comparing each of the 48 cultivars to each other in yields under heat stress. These yields ranged from 4 375.5 to 7 041.0 kg ha–1. Mean yield in the untreated controls was 7 133.0 kg ha–1while the mean yield was 5 717.2 kg ha–1under heat stress in 2011.δifor yield in this experimental year was 0.20. Significant (P＜0.01) differences were detected for yields under heat stress among cultivars and ranged in yields from 4 042.5 to 7 467.0 kg ha–1. The mean YR percentage of all cultivars increased from 14.1% in 2010 to 19.23% in 2011 (δ2010=0.15＜δ2011=0.20). The mean KWR percentage decreased from 13.3% in 2010 to 5.65% in 2011(δ2010=0.13>δ2011=0.06).

Mean 1 000-kernel weight in the untreated controls was 43.6 g and was 37.8 g under heat stress in 2010. The kernel weightδiin the 2010 experiment was 0.13. Significant(P＜0.01) differences were detected for kernel weight under heat stress among cultivars and ranged in yields from 31.9 to 44.7 g. Mean 1 000-kernel weight in the untreated controls were 45.3 g and was 48 g under heat stress in 2011. The kernel weightδiin the 2011 experiment was 0.06. Significant(P＜0.01) differences were detected for kernel weight under heat stress among cultivars and ranged in yields from 37.5 to 52.5 g.

Fig. 3 The average maximum air temperature per year from 1990 to 2016.

Table 3 Comparison of mean yield under heat stress (Y), yield reduction (YR), green leaf retention (GLR), kernel weight reduction(KWR), susceptibility index (S), and heat tolerance index (HTI) of 48 wheat cultivars for the 2-year study

The results showed that the yields of the 48 wheat cultivars decreased from 26.25 to 1.67 kg ha–1with an average of 9.82 kg ha–1in 2010 and from 26.75 to 0.92 kg ha–1with an average of 14.16 kg ha–1in 2011 (δ2010=0.15 andδ2011=0.20) under a 1°C increase each day compared to normal growth temperatures during wheat grain filling stages. Weight per kernel of the 48 wheat cultivars was decreased from 0.097 to 0.02 mg with an average of 0.058 mg in 2010 and from 0.08 to 0.001 mg with an average of 0.027 mg in 2011 (δ2010=0.13 andδ2011=0.06) under a 1°C increase each day compared to normal growth temperatures during wheat grain filling stages.

The KWR mean percentage ranged from 3.9 to 17.6%with an average of 9.5%. The YR ranged from 2.9 to 29.7%with an average of 16.7%. The YR mean percentage ranged from 2.79 to 34.2% with an average of 14.1% in 2010, while the YR mean percentage ranged from 1.73 to 32.9% with an average of 19.2% in 2011. The KWR mean percentage ranged from 4.72 to 22.1% with an average of 13.3% in 2010, while the KWR mean percentage ranged from 0.21 to 16.7% with an average of 5.65% in 2011.

There were 28 and 22 cultivars which had above-average yield under heat stress in 2010 and 2011, respectively. The cultivar Heng 6632 produced the highest mean yield,7 016.25 kg ha–1, under heat stress while the YR mean rate was 17.3%. There were 27 and 20 cultivars which had below-average YR under heat stress in 2010 and 2011,respectively. The cultivar 08CA95 had the lowest mean YR, 2.9%, while the mean yield under heat stress was 5 727 kg ha–1. The cultivars Nongda 3492 and Nongda 189 had higher yields under heat stress and lower YR.

The cultivar Nongda 3492 showed the largest variation in yield under heat stress with a difference of 1 425 kg ha–1between the two crop seasons. The cultivar Hengguan 35 showed the least variation in yield under heat stress with a difference of 12 kg ha–1between the two crop seasons.The cultivar DH155 showed the largest variation on the YR rate with a difference of 26.4% between the two crop seasons. The cultivar 55319 showed the smallest variation on the YR rate with a difference of 0.13% between the two crop seasons.

3.4. Heat tolerance assessment

The average GLR values of the 48 cultivars were 69.5 and 64.1 in 2010 and 2011, respectively. The GLR reduced from 12.6 to 2% with an average of 6.1% in 2010 and reduced from 11.8 to 2.0% with an average of 7.2% in 2011 under a 1°C increase each day compared to normal growth temperatures during wheat grain filling stages. The cultivar with the maximum mean GLR was 08CA95 and had a value was of 89.5. The cultivar with the minimum mean GLR was 56 487 and had a value of 40.5. The cultivar Jingdong 8, which had a yield under heat stress of 5 184 kg ha–1and a YR of 19.1%, showed a maximum value of 90 GLR in 2011. The cultivar 05CA306, which had a yield under heat stress of 4 917 kg ha–1and a YR of 2.38%,showed a minimum value of 40 GLR in 2011. The cultivar with the largest variation between the two years was 55319.There was a significant correlation across cultivars between mean GLR and KWR differences in the two crop seasons in the heat treated cultivars compared to the untreated controls (R=0.3126,P=0.0305). Additionally, a significant correlation across cultivars under heat stress was found between GLR and yield (R=0.2926,P=0.0436), and GLR and YR (R=–0.3314,P=0.0214) in the 2010 crop season.No significant correlations were found in 2011.

The mean KWR percentages in the 48 cultivars were 13.3 and 5.65% in 2010 and 2011, respectively. Cultivars differed significantly (P＜0.05) in this respect, ranging from a 3.9% reduction in Nongda 3492 to a 17.6% reduction in Shi B07-4056. The cultivar Shi 6207, which had a yield under heat stress of 5 100 kg ha–1and a YR of 20.73%,showed a minimum KWR rate of 0.21%. The largest and smallest variation occurred in cultivar 05CA349 and 67257,that reached 17.53 and 0.38%, respectively. These results showed that KWR differed significantly (P＜0.01) in the two crop seasons suggesting that the kernel weight was greatly affected by stress intensity. There was no significant correlation across cultivars among the rates of KWR, yield and YR rates under heat stress except for the correlation between KWR and YR under heat stress in 2010 (r=0.3126P=0.0305). The KWR rate of the CA0629 cultivar was found to be 12.8% while its YR rate was 3.78%. The percent change suggested that the heat tolerance of this cultivar could be related to withstanding less reduction in kernel number under heat stress compared to other cultivars (Fokaret al. 1998). Nevertheless, the KWR rate of the Heng 6632 cultivar was 7.5% while its YR rate was 13.35%. This cultivar may express its relative heat tolerance in reducing kernel weight loss under heat stress, but with somewhat lower tolerance reflected in kernel number per ear compared with the CA0629 cultivar.

The maximum S was 2.35 in the Hengguan 33 cultivar(Y=5 041.5 kg ha–1, YR=34.24%) and 1.65 in the Shi 8 cultivar (Y=5 451 kg ha–1, YR=32.91%) in 2010 and 2011,respectively. The minimum S was 0.19 in the 08CA95 cultivar (Y=5 791.5 kg ha–1, YR=2.79%) and 0.09 in the Heng 07-5205 cultivar (Y=5 200.5 kg ha–1, YR=1.73%) in 2010 and 2011, respectively. A significant correlation was found across cultivars among the S, yield and YR under heat stress except for the correlation between S and Y in 2011 (r=–0.2336,P=0.11).

The maximum HTI was 1.17 in the Nongda 212 cultivar(Y=6 499.5 kg ha–1, YR=3.11%) and 1.46 in the Nongda 3492 cultivar (Y=7 467 kg ha–1, YR=4.38%) in 2010 and 2011,respectively. The minimum HTI was 0.58 in the Shi 4185 cultivar (Y=4 666.5 kg ha–1, YR=32.53%) and 0.55 in the Han 6228 cultivar (Y=4 042.5 kg ha–1, YR=32.44%) in 2010 and 2011, respectively. The HTI was greater in cultivars with higher yield and lower YR under heat stress. Thet-test showed there was no significant difference for HTI between the two years. In this study, we found a significant(P＜0.01) correlation existed across cultivars and yield,YR under heat stress and HTI (Table 4). The significant positive correlations (r=0.8657 andr=0.8418 in 2010 and 2011, respectively;P＜0.01) were found between HTI and mean yield under heat stress across the 48 wheat cultivars(Fig. 4). This result indicated that HTI can describe the yield potential under heat stress. The significant negative correlations (r=–0.8344 andr=–0.7158 in 2010 and 2011,respectively;P＜0.01) were found between HTI and YR rate under heat stress across the 48 wheat cultivars (Fig. 4).This result indicated that HTI can describe the yield stability under heat stress.

4. Discussion

4.1. Wheat cultivars evaluation

As expected, high temperature had an impact on crop yield. However, the impact caused by heat shock or chronic high temperature was different. We hypothesize that heat shock may inhibit crucial enzyme activities such as PEPCase, NADP-ME, FBPase, PPDK, and Rubisco(Demirevska-Kepovaet al. 2005), or damage PSII (Sharkey 2005), or diminish chlorophyll biosynthesis (Tewari and Tripathy 1998) to cause severe KWR. Meanwhile, chronic high temperatures had even more of an effect on grain yield. This result was not consistent with a former study(Wardlawet al. 2002) and perhaps this is because of our higher heat shock temperatures. The 1 000-kernel weight of the cultivars Shixin 733, Shi 6207, and Jingdong 8 were the most affected by heat shock. Future studies on these three cultivars could reveal the effect of heat shock on physiological and biochemical characteristics by setting different temperature gradients during post-anthesis stages. Conversely, the 1 000-kernel weight of the cultivars Nongda 212, 08CA95, and CA0629 were affected the least by heat shock. A functional protein analysis duringexposure to high temperatures could elucidate proteins involved in heat tolerance phenotypes.

Table 4 Correlations among yield under heat stress (Y) and yield reduction under heat stress (YR%) as a function of green leaf retention (GLR), kernel weight reduction (KWR),susceptibility index (S), and heat tolerance index (HTI) in 2009–2010

Fig. 4 The correlations found across the 48 wheat cultivars among yield, yield reduction (YR) and heat tolerance index (HTI). The 48 dots on the graph represent average values for each cultivar.

The results indicated that the heat shock at 7–8 days after heat treatment affected grain filling, but otherwise might be an effective ‘genetic switch’ to induce gene expression to control kernel number and efficient panicle of unit-acreage to achieve less YR. Moreover, the cultivar 08CA95 had a more typical compensation featureviathe other yield components under heat shock (YR 2.79%vs. KWR 12.18%in 2010; YR 3.08%vs. KWR 7.25% in 2011). The different compensation mechanisms of different wheat cultivars under heat stress are worth studying in the future to assess whether it’s sufficient to compensate for the loss in dry matter accumulation due to reduction in grain filling duration.

4.2. Heat tolerance assessment

The stay-green trait is the ability of plants to remain green for a longer time, thereby contributing photosynthates for a longer period towards grain development (Thomas and Howarth 2000). The stay-green trait has also been suggested as a useful trait for determining heat tolerance(Joshiet al. 2007a). GLR is a subjective visual rating system of stay-green that can be performed rapidly in the field (Wanouset al. 1991). In this study, GLR showed low accuracy and poor repeatability in the two years of experimental data (Table 3). Correlation analysis revealed that there was not always a correlation between GLR and yield and YR under heat stress.

Some morphophysiological traits are associated with heat tolerance according to the capacity for grain filling under heat stress (Fokaret al. 1998). In this study, the reduction rate of 1 000-kernel weight (RKW) was added as a reference.The RKW of the Nongda 3432 cultivar was 42.0 and 45.3 g in the heat- and non-stress treatments, respectively. The yield of the Nongda 3432 cultivar in the two treatments is 5 374.5 and 5 667 kg ha–1, respectively. The RKW reduction rate was 7.3%. These results indicated that this cultivar was suitable for heat tolerance as a genetic resource, but was not intrinsically a good cultivar for heat tolerance due to the low yield performance under heat stress. Meanwhile, the HTI of this cultivar was 0.9458. This value considers both relative yield and absolute yield under heat stress. It was lower than the control cultivar Heng 4399, indicating that this cultivar was not suitable for actual field production under heat stress. The KWR value may be useful in screening heat resistant germplasm but is not useful as a sole criterion for choosing heat resistant cultivars.

The smaller the S value the better the heat tolerance(Fischer and Maurer 1978). The YR of 08CA95 and Heng 07-5205 cultivars with the minimum S were low, meanwhile the yield under heat stress was also low, this indicating that they were not good cultivars for heat tolerance.Correlation analysis also revealed that there was not always a correlation between S and Y. This index only describes the yield stability under heat stress.

Result showed that the larger the HTI was , the higher the yield and the lower the YR under heat stress were.The mean yield of a cultivar under heat stress could be used to assess the yield potential of a cultivar under heat stress. The significant positive correlation between HTI and mean yield under heat stress across the 48 wheat cultivars(Fig. 4) provides support to the conclusion that HTI can indicate the yield potential of wheat cultivars under heat stress. The YR rate under heat stress may characterize the yield stability between two environments, although they cannot account for differences in yield potential among cultivars (Clarkeet al. 1992). These significant (P＜0.01)correlations existed across cultivars among yield, YR under heat stress and HTI (Table 4). This index can reflect the yield potential and the stability in heat stress of different wheat varieties simultaneously. Heat tolerance of different wheat cultivars can be much better defined using the HTI parameter.

4.3. Evaluation and application of HTl

The 10 cultivars Nongda 212, Nongda 318, Nongda 189,Nongda 211, Nongda 3492, Nongda 413, Heng 6632,Heng 07-5205, 08CA95, and Yannong 19 had a higher HTI than the others in the two crop seasons, which included the former five cultivars (Nongda 318, Nongda 212, Nongda 189, Nongda 3492, and Yannong 19).

Heat stress imposed at the anthesis stage reduced wheat yield in all cultivars. Here we investigated the relationship of the yield, YR, and YR rate between the 10 wheat cultivars above the average value of the 48 wheat cultivars(Table 5). The average yield under heat stress of these 10 wheat cultivars was 6 461 kg ha–1, more than 12.6% of the average yield of the 48 wheat cultivars which was 5 740 kg ha–1in the two crop seasons. This result indicated that the 10 wheat cultivars showed better yield performance under heat stress than the other cultivars. Two cultivars Heng 07-5205 and 08CA95 showed lower yield than the mean value of all 48 cultivars. However, the YR of Heng 07-5205 was the smallest of all cultivars during the 2010 season.Meanwhile, the mean YR of 08CA95 was the smallest when both crop seasons were taken into account. According to this research, HTI is a comprehensive index that can not only reflect yield under heat stress but also YR and yield potential.

There were high variations in YR and YR rate within cultivars. The YR under heat stress ranged from 91.5 to 2 625 kg ha–1, and the YR rate under heat stress ranged from 1.73 to 34.24%. The average YR under heat stress of the 10 wheat cultivars was 10.8%, less than 35% of the average YR of the 48 wheat cultivars which was 16.69%.Of the 10 cultivars, the YR of Heng 6632 reached 17.34%.This cultivar had a large YR under heat stress, but the mean yield under heat stress was 7 016.25 kg ha–1, ranking number one of the 48 cultivars in the two years. This was an excellent wheat cultivar for heat tolerance. This agreed with the result of HTI test. Stability in yield for each cultivar may be estimated by the yield difference between stress and non-stress environments (Blumet al. 1989).

To distinguish between the heat tolerance of wheat cultivars, the 48 wheat cultivars were classified according to the grading scheme based on HTI (Table 2). The results showed that the HTI of the 10 wheat cultivars reached above level 2, indicating adequate heat tolerance, and should be considered for wheat breeding.

The HTI approach better estimated heat tolerance compared to the other methods for four reasons. First, the heat treatment was easier to quantify because of usingδi.Second, HTI had greater repeatability over different years and different stress intensities. Third, the correlations among HTI, yield and YR under heat stress was always significant in different years. Finally, HTI linked the yield under heat stress to the YR and reflected a heat tolerance with yield potential and stability in heat stress of different wheat cultivars simultaneously.

Table 5 The evaluation on the yield and yield component of 48 wheat varieties1)

5. Conclusion

The different cultivar responses to heat stress indicated that the heat tolerance ability of wheat cultivars varied.HTI was able to represent these varied responses and significantly correlate with the yield and YR of wheat under heat stress. Accordingly, this index can reflect the yield potential and the stability to heat stress of different wheat varieties simultaneously. Cultivars with a high HTI (>1)can be considered to be heat resistant, as they exhibited higher yield and smaller YRs under heat stress compared with the other cultivars. HTI was a reliable evaluation for heat-tolerance of wheat cultivars. In this paper, 10 wheat cultivars showed high HTI (high yield potential and stability), offering a genetic resource and foundation for further testing on their heat conferring characteristics.

Acknowledgements

Work reported here was partially supported by the Generation Challenge Program, CIMMYT (International Maize and Wheat Improvement Center) (GCP, G7010.02.01), the earmarked fund for China Agriculture Research System(CARS-3-2-3), and the National Key Technology R&D Program of China (2016YFD0100502, 2016YFD0300407).

Ahmad R, Qadir S, Ahmad N, Shah K H. 2003. Yield potential and stability of nine wheat varieties under water stress conditions.International Journal of Agriculture & Biology,5, 7–9.

Allakhverdiev S I, Kreslavski V D, Klimov V V, Los D A,Carpentier R, Mohanty P. 2008. Heat stress: An overview of molecular responses in photosynthesis.Photosynthesis Research, 98, 541–550.

Amani I, Fischer R A, Reynolds M P. 1996. Canopy temperature depression associated with yield of irrigated spring wheat cultivars in a hot climate.Journal of Agronomy and Crop Science, 176, 119–129.

Blum A. 1973. Components analysis of yield responses to drought of sorghum hybrids.Experimental Agriculture, 9,159–167.

Blum A, Klueva N, Nguyen H T. 2001. Wheat cellular thermotolerance is related to yield under heat stress.Euphytica, 117, 117–123.

Blum A, Shipler L, Golan G, Mayer J. 1989. Yield stability and canopy temperature of wheat genotypes under drought stress.Field Crops Research, 22, 289–296.

Borrell A K, Hammer G L, Douglas A C L. 2000. Does maintaining green leaf area in sorghum improve yield under drought? I. Leaf growth and senescence.Crop Science,40, 1026–1037.

Chinoy J J. 1947. Correlation between yield of wheat and temperature during ripening of grain.Nature(London),159, 442–444.

Clarke J M, Depaw R M, Tounley-Smith T F. 1992. Evaluation of methods for quantification of drought tolerance in wheat.Crop Science, 32, 723–727.

Demirevska-Kepova K, Holzer R, Simova-Stoilova L, Feller U. 2005. Heat stress effects on ribulose-1,5-bisphosphate carboxylase/oxygenase, Rubisco binding protein and Rubisco activase in wheat leaves.Biologia Plantarum, 49,521–525.

Farooq M, Bramley H, Palta J A, Siddique K H M. 2011. Heat stress in wheat during reproductive and grain filling phases.Critical Reviews in Plant Sciences, 30, 491–507.

Figueiredo I C R D, Pinto C A B P, Ribeiro G H M R. 2015.Efficiency of selection in early generations of potato families with a view toward heat tolerance.Crop Breeding and Applied Biotechnology, 15, 210–217.

Fischer R A, Maurer R. 1978. Drought tolerance in spring wheat cultivars. I: Grain yield response.Australian Journal of Agricultural Research, 29, 897–912.

Fokar M, Blum A, Nguyen H T. 1998. Heat tolerance in spring wheat. II. Gram filling.Euphytica, 104, 9–15.

Gooding M J, Ellis R H, Shewry P R, Schofield J D. 2003. Effects of restricted water availability and increased temperature on the grain filling, drying and quality of winter wheat.Journal of Cereal Science, 37, 295–309.

Hoffmann S, Debreczeni K, Hoffmann B, Nagy E. 2006. Grain yield and baking quality of wheat as affected by crop year and plant nutrition.Cereal Research Communications, 34,473–476.

Joshi A K, Chand R, Arun B, Singh R P, Ortiz R. 2007a.Breeding crops for reduced-tillage management in the intensive, rice-wheat systems of South Asia.Euphytica,153, 135–151.

Joshi A K, Kumari M, Singh V P, Reddy C M, Kumar S, Rane J, Chand R. 2007b. Staygreen trait: Variation, inheritance and its association with spot blotch tolerance in spring wheat(Triticum aestivumL.).Euphytica, 153, 59–71.

Kumari M Y, Pudake R N, Singh V P, Joshi A K. 2013.Association of staygreen trait with canopy temperature depression and yield traits under terminal heat stress in wheat (Triticum aestivumL.).Euphytica, 190, 87–97.

Lazar M D, Salisbury C D, Worrall W D. 1995. Variation in drought susceptibility among closely related wheat lines.Field Crops Research, 41, 147–153.

Lobell D, Ortiz-Monasterio I. 2007. Impacts of day versus night temperature on spring wheat yields: A comparison of empirical and CERES model predictions in three locations.Agronomy Journal, 99, 469–477.

Mason R E, Mondal S, Beecher F W. 2010. QTL associated with heat susceptibility index in wheat (Triticum aestivumL.)under short-term reproductive stage heat stress.Euphytica,174, 423–436.

Rane J, Nagarajan S. 2004. High temperature index - for field evaluation of heat tolerance in wheat varieties.Agricultural Systems, 79, 243–255.

Reynolds M P, Balota M, Delgado M I B, Amani I, Fischer R A.1994. Physiological and morphological traits associated with spring wheat yield under hot, irrigated conditions.Australian Journal of Plant Physiology, 21, 717–730.

Reynolds M P, Singh R P, Ibrahim A, Ageeb O A A, Larque-Saavedra A, Quick J S. 1998. Evaluating hysiological traits to complement empirical selection for wheat in warm environments.Euphytica, 100, 85–94.

Sharkey T D. 2005. Effect of moderate heat stress on photosynthesis: Importance of thylakoid reactions, rubisco deactivation, reactive oxygen species and thermotolerance provided by isoprene.Plant Cell Environment, 28, 269–277.

Tewari A K, Tripathy B C. 1998. Temperature-stress-induced impairment of chlorophyll biosynthetic reactions in cucumber and wheat.Plant Physiology, 117, 851–858.

Tewolde H, Fernandez C J, Erickson C A. 2006. Wheat cultivars adapted to post-heading high temperature stress.Journal of Agronomy and Crop Science, 192, 111–120.

Thomas H, Howarth C J. 2000. Five ways to staygreen.Journal of Experimental Botany, 51, 329–337.

Towill L E, Mazur P. 1974. Studies on the reduction of 2,3,5-triphenyl tetrazolium chloride as a viability assay for plant tissue culture.Canadian Journal of Botany, 53,1097–1102.

Villegas D, Garcia L F, Rharrabti Y, Martos V, Royo C. 2007.Morphological traits above the flag leaf node as indicators of drought susceptibility index in durum wheat.Journal of Agronomy and Crop Science, 193, 103–116.

Wanous M K, Miller F R, Rosenow D T. 1991. Evaluation of visual rating scales for green leaf retention in sorghum.Crop Science, 31, 1691–1694.

Wardlaw I F, Blumenthal C, Larroque O, Wrigley C W. 2002.Contrasting effects of chronic heat stress and heat shock on kernel weight and flour quality in wheat.Functional Plant Biology, 29, 25–34.

Wiegand C L, Cuellar J A. 1981. Duration of grain filling and kernel weight of wheat as effected by temperature.Crop Science, 21, 95–101.

Xu W, Rosenow D T, Nguyen H T. 2000. Stay-green trait in grain sorghum: Relationship between visual rating and leaf chlorophyll concentration.Plant Breeding, 119, 365–367.