SHAO Rui-xin,YU Kang-ke,Ll Hong-wei,JlA Shuang-jie,YANG Qing-hua,ZHAO Xia,ZHAO Ya-li,LlU Tian-xue

1 Collaborative Center Innovation of Henan Food Crops/National Key Laboratory of Wheat and Maize Crop Science/College of Agronomy,Henan Agricultural University,Zhengzhou 450046,P.R.China

2 Institute of Grain Crops,Henan Academy of Agricultural Sciences,Zhengzhou 450002,P.R.China

Abstract Compared to other crops,maize production demands relatively high temperatures.However,temperatures exceeding 35°C lead to adverse effects on maize yield.High temperatures (≥35°C) are consistently experienced by summer maize during its reproductive growth stage in the North China Plain,which is likely to cause irreversible crop damage.This study investigated the effects of elevating temperature (ET) treatment on the yield component of summer maize,beginning at the 9th unfolding leaf stage and ending at the tasseling stage.Results demonstrated that continuous ET led to a decrease in the elongation rate and activity of silks and an elongated interval between anthesis and silking stages,and eventually decreased grain number at ear tip and reduced yield.Although continuous ET before tasseling damaged the anther structure,reduced pollen activity,delayed the start of the pollen shedding stage,and shortened the pollen shedding time,it was inferred,based on phenotypical and physiological traits,that continuous ET after the 9th unfolding leaf stage influenced ears and therefore may have more significant impacts.Overall,when maize plants were exposed to ET treatment in the ear reproductive development stage,the growth of ears and tassels was blocked,which increased the occurrence of barren ear tips and led to large yield losses.

Keywords:summer maize,North China Plain,elevating temperatures,reproductive stage,barren ear tip,yield

1.lntroduction

Average global temperatures are rising according to climate records.Several current climate models (Rafteryet al2017;Rogeljet al2018) have indicated that by 2100 the mean annual global surface temperature is projected to increase by 1 to 3.5°C,which may lead to changes in spatial and temporal patterns of precipitation.Climate change will influence agricultural production by affecting crop yields and productivity in a variety of ways (Zandalinaset al2018).Compared to other staple crops,maize is most vulnerable to climate variability (Kamaliet al.2018).Global warming has already revealed negative impacts on maize yield in many regions,and has been correlated with a global maize yield reduction of 4% over a 29-year period from 1980 to 2008.The North China Plain (NCP) is one of the major agricultural production areas in China,where a winter wheat and summer maize (Zea maysL.) rotation is the primary cropping system.NCP may be one of the most vulnerable regions to climate change due to increasing temperatures and water-related stresses (Tao and Zhang2010;Moet al.2013;Wang Det al.2020;Wang Y Xet al.2020).

Sustained temperature increases over the growing season impact the growth time (from sowing to maturity)of crops (Heim 2017;Van Oortet al2018),which is an important climate change-related yield reducing factor (Tao and Zhang2010).Summer maize in NCP experiences high temperature stress (HTS,≥35°C) from July to August during its spike differentiation and pollination.Reproductive organs,in comparison to vegetative organs,have a considerably lower temperature threshold for damage caused by heat stress.Yield reduction due to high temperature during the reproductive stage has been associated with a decrease in kernel number and weight.However,if HTS occurs in different growth stages,the corresponding damages to reproductive organs and losses of kernel yield in maize vary (Lizasoet al.2018).HTS during pollination leads to a reduction in kernel number per ear by decreasing the number of ovules that differentiate,are fertilized,and develop into kernels (Schoperet al1986).HTS after pollination causes the reduction by increasing kernel abortion (Cheikh and Jones 1994;Yanget al.2020).Exposure to temperatures over 36°C significantly decreases the fertilization rate of pollinated spikelets,and the subsequent failure to fertilize an ovule is primarily attributed to the abnormal development of male reproductive tissues that are more sensitive to heat stress than female ones (Dupuis and Dumas 1990).HTS during the early stage of kernel development disrupts endosperm development and may lead to abortion or premature growth cessation (Cheikh and Jones 1994).Although HTS-dependent variation is greater during grain filling than before anthesis in the vegetative period (Shimet al2017),the asynchronous timing of anthesis and silking is also considered as an important factor in decreasing fertilized ovules.

However,few studies have explored the effects of HTS before tasseling at the anthesis and silking stages.Maize flower styles and stigmas,commonly known as ‘silks’,elongate rapidly during flowering to emerge from bracts (or husks) and collect pollen from the apical male inflorescences(or tassels) (Tardieuet al2015;Turcet al2017).They exhibit a mono-dimensional development similar to that of monocotyledon leaves,although there are notable differences (Turcet al2016).Tissue elongation rates and cell division are reduced by adverse environmental factors,which results in delayed silk emergence.The silk emergence date plays a pivotal role in controlling reproductive success under adverse environmental conditions (Tardieuet al2000;Mickelbartet al2015).

Therefore,the primary objective of this study was to elucidate the effects of elevating temperature (ET),above 35°C for 20 (2013)/13 (2014) days (d) from vegetative stage–9th leaf collar fully unfolded (V9 stage) to vegetative tasseling (VT) stage,on the growth and development of reproductive organs,such as tassels and ears.Since yield components indicated that ET increasingly led to barren ear tips,the second objective was to investigate the underlying mechanisms of the effects of ET on ear tip development in field-grown summer maize.

2.Methods and materials

2.1.Experimental location and materials

Experiments were conducted at the Comprehensive Test Site of the Zhengzhou Agricultural Meteorological Observation Station of Henan Meteorological Research Institute (34°16´–34°58´N,112°42´–114°14´E).The maize cultivars used for experiments were Zhengdan 958 (ZD958 from Henan Academy of Agricultural Sciences,Zheng58/Chang7-2) and Xianyu 335 (XY335 from American Pioneer Company,PH6WC/PH4CV),which occupy extremely large planting areas in China.Seeds were sown at a density of 67 500 plants ha–1with basal compound fertilizer (750 kg ha–1;15% N,15% P2O5,15% K2O) and top-dressed with urea (245 kg ha–1;46% N) at the 13-leaf stage.A pesticide regiment was applied to control insects and diseases.No indications of water or nutrient stresses were observed during the growth period.Experiments were carried out in 2013 and 2014.

2.2.Experimental treatments and growing conditions

The high temperature processing facility established by Zhanget al.(2008) was used in this study,which included the use of an angle iron framed growth chamber that was 4 m (length)×3 m (width)×4 m (height).The chamber was fixed in a field,covered with a resin film with a light transmittance of about 95%,and was 80% sealed with a 20% gap for gas exchange.Each maize variety included two treatments:a control group (CK,no ET treatment) and an ET group (elevating temperatures from V9 to VT stages),with each treatment repeating six times.When the 9th unfolding leaf was present,plants were treated with the covered resin film that can elevate temperatures during 08:30–17:30 every day,until the end of the VT stage.The field water capacity for all treatments was kept at a normal state (75±5%).

2.3.Sampling and measurements

Growth dynamics of tassels and earsUnder the ET treatment,four plants with the same growth rate per treatment were selected every 3 days for observing the number of differentiated tassel florets,measuring the length and accumulated dry matter of the main spike and branches of the tassel,and measuring the tassel volume using a previously described drainage method (Morenoet al1996).Silk length was recorded after the first silk emerged,and the rates of silk emergence (%) were calculated based on the silk number from husks and the total silk number.Rate of silking per ear (%)=100×Silking number from husks/Total silking number.

Flowering dynamics and amount of pollen shedThe events of tassel emergence,pollen shedding,silk emergence,blossoming,and end of shedding were recorded.The tasseling date was recorded when about 2–3 cm of tassel was visible,the pollen shedding date was recorded when the first anther started shedding pollen,the silking date was recorded when the silk emerged about 1–2 cm,the blossoming date was recorded when the maximum shedding number was reached,and the end of shedding date was recorded when pollen was no longer shed.The anthesis–silking interval (ASI) was calculated as the difference between the shedding and silking dates.The duration of the pollen shedding was calculated as the time period between the shedding and the end of shedding dates.

The amount of pollen and the fresh weight of the pollen shed per tassel from three plants were measured for each temperature treatment through the collection of daily shed pollen according to the previously described method by Wanget al(2019).The percentage of shedding pollen was calculated according to Yu (2016).

Pollen viability assessment and additional microscopyThe vitality of mixed pollen from open anthers collected between 09:00–10:00 was assessed by staining it with a 0.5% triphenyl-tetrazolium-chloride (TTC) solution and subsequently using microscopic inspection (Nollet al.2010).The remaining pollen was used for hand-pollinating to test its viability.About 10–15 min after staining,the pollen color was observed and photographed with a stereomicroscope(Olympus BX51,Japan),and additional pictures were obtained with an Olympus microscope (MicroPublisher 3.3 RTV Digital Iimaging System).Only pollen grains that were stained red were regarded as viable.Pollen viability was expressed as the percentage of viable pollen grains of the total pollen grains in the microscopic field.

Pollen was viewed using a Nissan S3400N scanning electron microscope (Hitachi High Technologies Co.,Tokyo,Japan) and a transmission electron microscopy (TEM;JEM-1230,Tokyo,Japan).Samples of fresh pollen collected from the open anthers between 09:00–10:00 were immediately fixed and stored in 2.50% glutaraldehyde.For morphological analysis,pollen was washed with a 100-mmol L–1sodium phosphoric acid buffer,fixed with 1% osmium tetroxide,dehydrated in graded ethanol (30–100%),and then critical point dried (Leica EM CPD300,Leica Vienna).The pollen was subsequently sputter coated with gold palladium and scanned with a SEM.To analyze pollen anatomy,the stored pollen was fixed with a 1% osmium tetroxide,dehydrated in a graded acetone (30–100%),embedded in an epoxy resin,cut into ultrathin sections using an ultramicrotome,stained with uranyl acetate and lead citrate,and finally observed with a TEM.

Pollen germination analysisApproximately 10 mg of homogenous pollen grains from each replicates were added to a microcentrifuge tube with 500 μL of a pollen germination solution (0.35 mol L–1sucrose,0.03% CaCl2and 0.01%H3BO3) (Broglia and Brunori 1994).The tube contents were mixed well by gentle pipetting,and then the mixture was evenly spread in a Petri dish containing pollen growth medium (0.35 mol L–1sucrose,0.03% CaCl2,0.01% H3BO3,0.7% agarose) modified from Aylor (2004).Prepared Petri dishes were sealed with parafilm,incubated at 25°C for 3 h,and then immediately observed under a microscope(Olympus,BX51) to count germinated pollen grains in 10 fields per Petri dish.Approximately 200 pollen grains from each sample were examined,with three different plants as technical replicates,and the experiment was repeated three times in total.For each treatment,the germination rate was assessed as the total number of germinated pollen grains divided by the total number of germinated and ungerminated pollen grains.

Silk length and fertilizated activitySilk vigor from middle part of ear was measured using a TTC staining method(Phillips and Hayman1970).Stain intensity,fluff number and length of the top and middle parts were correlated with the fertilized activity.After 2–6 days of silking,ears attached to the ET treated plants were bagged and manually pollinated from 2 days after silking for four successive days (between 09:00 and 11:00) to ensure the maximum pollination with fresh pollen collected from the CK treatment plants.At 24 h after bagging,bags were retrieved from ears and husk leaves were carefully removed.According to the silk color and growth status,the rate of successful fertilization was calculated as follows:Fertilization rate per ear (%)=Number of fertilized silks/Total number of silks×100.

MDA content and autioxidant enzyme activitiesOn the 12th d after the ET treatment,three plants with similar growth states were selected.Their tassels and ears were collected,and their superoxidative dismutase (SOD),peroxidase(POD),catalase (CAT) activities and malonic dialdehyde(MDA) content were measured.These contents were measured according to the standards of Songet al.(2012).Measurement was repeated three times in each treatment.The relative change (%)=(Treatment–Control)/Control×100.

Yield components and barren ear tip occurrenceTwenty plants from each treatment were harvested at physiological maturity to measure grain yield,kernel weight,number of kernels,and ear length and diameter.Grains were threshed by hand,dried at 80°C to a constant weight,and then adjusted to have a 14% water content.The occurrence of barren ear tip in maize plants was measured.

2.4.Statistical analyses

The ultrastructure of anther,pollen and starch granules was assessed based on data collected in 2014,with additional indices being a two-year average for data collected in 2013 and 2014.Figures were drawn using SigmaPlot 10.0 Software (Systat Software Incorporation,Chicago,IL),where error bars represent the standard deviation and each data point in each figure represents the mean±SD of at least three experimental replicates.Significant differences among treatments were assessed using SPSS 11.5 Software (International Business Machines Corporation,Armonk,NY).

3.Results

3.1.Growth of tassels and ears

The changes in temperature and precipitation during the maize growth stages are presented in Fig.1,and the temperature fluctuation between the CK and ET treatments from the V9 to VT stages was shown in Fig.2.ET did not significantly affect the length and branch number of tassels(Fig.3-A),however,it led to a significant decrease in tassel volume in the ZD958 plants (Fig.3-B).For the XY335 plants,from 6th d of the ET treatment,the length and volume of tassels under the ET treatment were found to be significantly lower than those under the CK treatment,but at the tasseling stage there were no considerable differences between the two treatments.The dry matter of tassels under ET and CK also confirmed that ET obviously suppressed the growth of tassels (Fig.3-C).The differences between ET and CK were significant from 10 d after the ET treatment (Fig.3-C).For example,at the 12th d after the ET treatment,the dry weight under ET decreased by 74.10 and 62.80% in the ZD958 and XY335 plants,respectively.

Fig.1 Mean,maximum and minimum temperatures (Tmax,Tmin and Tmean) and precipitation (P) from 1 Jun.to 30 Sep.in 2013 (A)and 2014 (B).

Fig.2 Daily maximum temperature records from vegetative stage–9th leaf collar unfolded (V9) to vegetative tasseling (VT) stage during elevating temperature (ET) in 2013 and 2014.CK,the control.

Fig.3 Effects of elevating temperature (ET) from vegetative stage–9th leaf collar fully unfolded (V9) to vegetative tasseling(VT) stage on tassels’ length (A),volume (B) and dry matter(C) in Zhengdan 958 (ZD958) and Xianyu 335 (XY335).Each value represents the mean±SD (n=4).＊ indicates significant difference between the control (CK) and ET at P＜0.05.

The ear length and diameter of the XY335 plants were found to be influenced by high temperatures from the V9 to VT stages.Moreover,XY335 showed a greater and stastitsically significant difference in ear length between ET and CK than ZD958 (Fig.4-A).Conversely,for ZD958,there were no obvious differences between the treatments before the VT stage.In regards to the ear diameter,the two genotypes showed very similar growth dynamics,both of which rose obviously at the 10th d after the V9 stage,but XY335 started to decrease earlier and more significantly than ZD958 (Fig.4-B).

3.2.Physiological traits of tassels and ears

To determine the physiological resistance of ears and tassels to the ET treatment,the MDA content and CAT,POD and SOD activities were measured at the tasseling stage(Fig.5).Increases in MDA content after the ET treatment were evident,especially in XY335,but the degrees to which they increased varied between the two maize genotypes and different inflorescences,with the increase rate in ears being higher than that in tassels (Fig.5-A).It was found that the relative changes of CAT,POD and SOD activities in the two genotypes were surprisingly consistent (Fig.5-B–D).In tassels,the increase rates of CAT,POD and SOD activities in XY335 were higher than those in ZD958.However,a different trend was observed regarding ears,where the CAT,POD and SOD activities in ZD958 increased after the ET treatment,but those in XY335 decreased significantly.

3.3.Pollen,anther and anthesis-silking intervals

Fig.4 Effects of elevating temperature (ET) from vegetative stage–9th leaf collar fully unfolded (V9) to vegetative tasseling (VT)stage on the growth dynamic parameters of ears.A is the ear length in the Zhengdan 958 (ZD958) (left) and Xianyu 335 (XY335)(right).B is the ear diameter in the ZD958 (left) and XY335 (right).Each value represents the mean±SD (n=4).＊ and ＊＊ indicate significant differences between the control (CK) and ET at P＜0.05 and P＜0.01,respectively.

Fig.5 The effect of elevating temperature (ET) from vegetative stage–9th leaf collar fully unfolded (V9) to vegetative tasseling (VT)stage on the relative change of malonic dialdehyde (MDA) content (A),catalase (CAT) activity (B),peroxidase (POD) activity (C),and superoxidative dismutase (SOD) activity (D) of tassels and ears in Zhengdan 958 (ZD958) and Xianyu 335 (XY335),respectively.Each value represents the mean±SD (n=3).＊ indicates significant differences between the control (CK) and ET at P＜0.05.

Anther cross sections from the two maize genotypes were observed at the end of floret differentiation.The percutaneous and mesenchymal cell deformities were loosely arranged,the vascular bundles appeared smaller and thinner under ET condition than that of CK,and the pollen grains had a crumpled appearance.These characteristics were more obvious in XY335 than in ZD958 (Fig.6-A1–A4).After performing consistent staining on pollen grains from each treatment,the CK pollen grains had a deeper staining color and a regular shape as compared to those under ET.The enriched pollen in ZD958 and XY335 under CK accounted for 94.00 and 91.43% of the pollen grains in the field of view,respectively (Fig.6-B1 and B3).After the ET treatment,the degree of staining of pollen grains varied from no visible red or light red to deep red color.The enriched pollen in ZD958 and XY335 accounted for 85.00 and 78.00%of the pollen grains in the field of view,respectively.As shown in Fig.6-C1–C4,the pollen germination rate was lower and the pollen tube length was shorter within the same incubation time after the ET treatment than before the treatment,but the difference between the genotypes was not significant.The average growth rates of the pollen tubes in ZD958 and XY335 were 46.24 and 45.00 μm h–1,respectively.After the ET treatment,the growth rates were 38.50 and 37.00 μm h–1,which were decreased by 16.74 and 17.78%,respectively.The microstructure of the pollen grains indicated that the reticulation on the surface of the XY335 pollen grains became coarse and the gaps increased after ET stress,and the pollen germination pores and the vicinity of these pores were severely deformed,some of which were obviously obstructed (Fig.6-C1–C4,D1–D4).The pollen grain surface in ZD958 was slightly thickened or not significantly changed after the ET treatment,and the germination pores were slightly depressed but to a much lower degree than those of the pollen grains in XY335.As shown in Fig.6-E1–E4,the starch grains in the pollen of XY335 were significantly decreased,and that of ZD958 primarily displayed a notable decrease in volume.

Fig.6 Scanning electron microscope and microscope images of anther,pollen grains and starch granule in pollens of elevating temperature (ET) treated maize plants.A1–A4,B1–B4,C1–C4,D1–D4,and E1–E4 represent the anther’s structure (×200),the viability after formaldehyde-acetic acid-ethanol fixative (FAA) staining (×250),germination efficiency (×250),surface ultrastructure(×1 000),and starch granule (×4 000) of pollens,respectively.And the four lines represent the different treatments including the control (CK) and ET in Zhengdan 958 (ZD958),CK and ET in Xianyu 335 (XY335),respectively.

It was evident that both ZD958 and XY335 showed a delay in the date of pollen shedding after the ET treatment,with the shedding dates of ZD958 and XY335 being delayed for 3 and 5 days,respectively (Fig.7-C).The pollen shedding rate of ZD958 reached a maximum on the 2nd d after pollen shedding and then gradually decreased,while that of XY335 peaked on the 3rd d (Fig.7-A and B).Trends of pollen shedding rates were consistent between the two genotypes (Fig.7-C).The time and duration of pollen shedding within the same variety at its peak were not affected by the ET treatment.The peak of pollen shedding in ZD958 was 3 days after pollen release,while that in XY335 was 4 days after pollen release (Fig.7-A and B).

3.4.Elongation,electrical conductivity and hydraulic potential of the silks

Rapid growth usually occurs during maize silk elongation.However,the ET treatment was found to inhibit the silk elongation,causing a delay in the rapid elongation stage (Fig.8).The silk growth rate is a characteristic distinguishing the parts of ear,and the extent to which ET stress influenced these rates was also distinct.For the ZD958 and XY335 plants under the ET treatment,the apical silks shed husks decreased by 25.79 and 39.33%,respectively,and the basal silks decreased by 33.06 and 40.34%,respectively.The growth rate of silks decreased after the ET treatment,and the silk length of the ET plants was shorter than that of the CK plants.At the end of pollen shedding,it was found that the length of top silks decreased by 30.15 and 32.61%,the length of central silks decreased by 17.53 and 23.00%,and the length of basal silks decreased by 21.09 and 25.95%,respectively,for the ZD958 and XY335 plants.

In regards to electrical conductivity,XY335 was found to be more sensitive to the ET treatment than ZD958.Furthermore,the silks at different positions on the XY335 plants showed significant differences due to the ET treatment,where the effects on the basal silks were more significant than those on the top silks (Fig.9-A).After the ET treatment,the water potential of silks was decreased,and there were significant differences between the genotypes or among different silk positions on the plants.XY335 showed significant variations.The top silks of the XY335 and ZD958 plants under ET decreased by 25.00 and 6.45%,and their basal silks decreased by 41.11 and 8.30%,compared with XY335 and ZD958 respectively,under CK (Fig.9-B).

3.5.Reduction of apical kernels and yield components

Fig.7 The fresh weight of pollen per day in Zhengdan 958 (ZD958) (A) and Xianyu 335 (XY335) (B) after pollen shedding,and percentage of pollen shedding (C).Each value represents the mean±SD (n=3).＊ and ＊＊ indicate significant differences between the control (CK) and elevating temperature (ET) at P＜0.05 and P＜0.01,respectively.

Fig.8 Effects of elevating temperature (ET) from vegetative stage–9th leaf collar fully unfolded (V9) to vegetative tasseling (VT)stage on the silk length in different positions of ears in Zhengdan 958 (ZD958) (A) and Xianyu 335 (XY335) (B).Each value represents the mean±SD (n=4).＊ indicates significant difference between the control (CK) and ET at P＜0.05.

The bald tip length of ZD958 ears was very short under normal conditions,and its percentage of bald tip in the ear was low (Fig.10-A).After the ET treatment,although the length and percentage of the bald tip were increased,the increase was not significant compared to the control.However,when the XY335 plants were exposed to ET stress,the percentage of barren tip showed a significant increase of 118.60% (Fig.10-B).In 2013,the reduction of grain yield was significant in the ET treated plants,particularly in XY335,which was reduced by 39.19%.In 2014,the reduction of grain yield due to ET stress was more significant in the two genotypes than in 2013.Correspondingly,the 100-kernel weight,kernel number,ear length,and ear diameter parameters were decreased under the ET treatment (Table 1).

Fig.9 Effects of elevating temperature (ET) from vegetative stage–9th leaf collar fully unfolded (V9) to vegetative tasseling (VT)stage on silk conductivity (A) and water potential (B) in different position of ears.ZD958,Zhengdan 958;XY335,Xianyu 335.CK,the control.Each value represents the mean±SD (n=4).＊ indicates significant differences between the control (CK) and ET at P＜0.05.

Fig.10 Effects of elevating temperature (ET) from vegetative stage–9th leaf collar fully unfolded (V9) to vegetative tasseling (VT)stage on the length of bald tip and its percentage in Zhengdan 958 (ZD958) (A) and Xianyu 335 (XY335) (B).Each value represents the mean±SD (n=4).＊ and ＊＊ indicate significant differences between the control (CK) and ET at P＜0.05 and P＜0.01,respectively.

4.Discussion

With the projected acceleration of global warming,high temperatures will have an impact on the whole yielddetermining processes and yield of maize.In NCP,the temperature during the reproductive development in maize is often higher than the optimal temperature for the maximum differentiation of spikes (25–32°C).In this study,the maximum temperature above 35°C lasted for 20/13 d before tasseling in field summer maize during the ET treatment,and the phenological development,flowering characteristics,yield-related traits,and grain yield of maize were investigated.Generally,elevating temperatures above the ambient temperature range tended to increase the percentage of barren ear tips and decrease productivity by exerting negative effects on almost all the yield-determining traits.

4.1.Correlation between maize yield and the ETinduced weak development of tassels and ears

Tassel and ear differentiation is the basis of three elements influencing yield in maize,and the development of tassels and ears is mainly after the jointing stage.Extreme climatic conditions after this stage would cause irreversible damage to both the tassels and ears of maize (Hatfieldand Prueger 2015).The damage becomes serious especially after the V9 stage,when tassels have mostly formed,while ears remain at a critical stage of differentiation (Loussaertet al.2017).The ear differentiation stage is an important periodin which spike length,spike thickness and the number of potential florets are established (Fig.8).At this stage,the ET treatment caused some irreversible damages to the resulting phenotypes and physiological conditions of the ears.It was obvious that the ET treatment also had a corresponding influence on the tassel development and eventual pollen development.Results indicated that the ET treatment led to a decrease in the weight and volume of tassels (Fig.3),which resulted in a decrease in pollen(Fig.7).These results were consistent with those of Wanget al.(2019).Furthermore,anther dehiscence was inhibited by the ET treatment,and shedding time was delayed and shortened (Figs.6 and 7).These findings were consistent with those of Habbenet al(2014).The results of this study indicated that the hypoplasia of both maize tassels and ears directly caused a drastic decrease in the final yield of maize.

Table 1 Grain yield,kernel weight and number,ear length and diameter in the control (CK) and elevating temperature (ET)treatments before tasseling period in 2013 and 2014

4.2.A more significantly adverse impact of ET on ears than on tassels due to physiological traits

SOD,POD and CAT are the main components of the enzymatic active oxygen removal system,which plays an important role in the reduction of reactive oxygen accumulation in plants by resisting membrane lipid peroxidation and alleviating membrane damage (Cicchinoet al2013;Choudhuryet al.2017).In this study,SOD,POD and CAT activities in both tassels and ears of the ZD958 plants were increased under the ET treatment,and SOD activity exhibited the greatest increase (Fig.5).However,these activities decreased in the ears of the XY335 plants after the ET treatment.These results indicated that the influence of high temperatures varied greatly between the two genotypes,and ET before tasseling caused a more significant disadvantage to ears than to tassels,especially in the XY335 plants.These results were consistent with those of Jajoo and Allakhverdiev (2017).MDA is the final product of membrane lipid peroxidation and is one of the important indicators and indices of membrane system damage.Results showed that MDA accumulation in both the tassels and ears increased,which indicated that ET stress induces cell membrane damage.Together these findings revealed that high temperature resistant maize may temporarily remove the intracellular free radicals of tassels when ET is encountered,but there is severe damage to the cell membranes of both tassels and ears (Slamaet al2017).

4.3.Slowed growth and activity of silk as the early determinant of kernel reduction

Although the difference in kernel number among genotypes is mainly casued by different characteristics of genotypes,the decrease in kernel number within each genotype is primarily due to seed failure under adverse environmental conditions (Borráset al.2018).The ear of maize is a coalesced inflorescence,which is composed of spikelet pairs arranged in rings and sequentially initiated at the apex of the ear (Turc and Tardieu 2018).The ear silks are the primary reproductive organs of maize.The first silks to emerge originate from the basal spikelets and the last are from the apical spikelets (Shenet al2018).High temperatures make it very difficult for the silks to protrude from husks,which delays the development of silks at the top of female spikes,showing a link between silk growth and floret abortion (Ouryet al2016).This is an original example of coupling between expansive growth and developmental processes.In this study,the ET treatment led to a significant decrease in the elongation rate of silks,especially those in the apical position of ear.Since the development of the top and basal silks was relatively late,their silk length directly determined whether they can be pollinated properly,and thus the kernels at the top and basal positions were aborted due to poor polination.

Furthermore,a change in the activity of the silks (electron conductivity and water potential) at different positions also reflected the severity to which the ET treatment destroyed the silks.Abortion occurs in the youngest ovaries whose silks do not emerge at least 2 d before the arrest of silk growth,irrespective of plant water status,ovary volume,and growth rate or sugar status (Prasadet al2005;Ouryet al2016).This occurrence linked to silk growth accounted for 90% of drought-related abortion in a previous series of experiments with contrasting water deficits and genotypes(Ouryet al2016).This was similar to the control in other species in which abortion follows the reciprocal of the pollination/fertilization gradient for ovaries within a fruit(Ouryet al2016),for fruits within an inflorescence,and for inflorescences within a plant (Slaferet al2015).

4.4.Kernel reduction caused by anther quality and delayed-pollination

The threshold for damage caused by high temperatures in reproductive organs is considerably lower in maize than in vegetative organs (Barnabáset al2008).Maize is a cross-pollinated plant,therefore,anther development has a significant impact on the formation of late grains.Furthermore,pollen formation is one of the most heatsensitive developmental stages in cereals (Saini 1997;Stone 2000).The ET treatment inhibited the maturity and dehiscence of anthers,shortened the dispersing time of tassels,reduced the amount of dispersing pollen,and had an irreversible effect on the microstructure of anthers (Fig.6),which reduced their vitality and provided condition for kernel abortion in later stage.

The ASI has received considerable attention from maize breeders and physiologists.This is a simple visual observation that has been found to correlate with kernel number and yield under stressful conditions(Musvosviet al2018).The earliest studies describing the negative correlations between ASI and yield under adverse environmental stress identified that the lack of pollen for late appearing silks was the cause of reduced kernel numbers (Aslamet al2015;Shinet al.2015).In this study,the ET treatment significantly increased the ASI and decreased pollen weight (Table 2),which resulted in poor ear pollination (Wanget al2019) and reduction in grain number per ear.

The results of this study showed that from the floret differentiation stage to the tasseling stage,maize yield was reduced due to ET stress,which differed among genotypes.Grain weight differences were mainly due to material accumulation and transport capacity,and the decrease of leaf photosynthetic capacity was the primary reason for grain weight reduction under adverse environmental conditions(Andradeet al.1999;Hernándezet al.2014).

5.Conclusion

Continuous ET from the stage V9 to VT significantly affected the growth and development of both tassels and ears.ET damaged the anther structure so that pollen shedding was delayed and its duration was shortened.ET inhibited the growth of silk and reduced the activies of silk in ears.Furthermore,the influence of continuous ET on the development of ears was found to be more significant than that on tassel development.However,ET elongated ASI and decreased the quantity and quality of pollen shedding,which were the main reasons for kernel abortion at ear tip and the reduction of maize yield under ET.

Table 2 Time interval between tassels emergence (TE) and first silk emergence (SE),anthesis–silking interval (ASI),duration of pollen shedding,and fresh weight of pollen

Acknowledgements

This research was fnancially supported by the National Key Research and Development Program of China(2018YFD0300704),the Special Funds for Public Welfare Industry (Agriculture) Research,China (201203029),the Open Foundation of State Key Laboratory of Crop Biology in China (2019KF03),and the Open Foundation of Chinese Academy of Agricultural Sciences,China/Key Laboratory of Crop Water Use and Regulation,Ministry of Agriculture and Rural Affairs,China (FIRI2019-02-0103).

Declaration of competing interest

The authors declare that they have no conflict of interest.