Poor development of spike differentiation triggered by lower photosynthesis and carbon partitioning reduces summer maize yield after waterlogging

Juan Hu, Baizhao Ren, Shuting Dong, Peng Liu, Bin Zhao, Jiwang Zhang

State Key Laboratory of Crop Biology and College of Agronomy, Shandong Agricultural University, Tai’an 271018, Shandong, China

Keywords:Summer maize Waterlogging Tassel and ear differentiation Yield Anthesis–silking interval

ABSTRACT Stable yield of staple grains must be ensured to satisfy food demands for daily dietary energy requirements against the backdrop of global climate change. Summer maize, a staple crop, suffers severe yield losses due to extreme rainfall events, threatening food security. A randomized block experiment with four treatments: control, no water stress (CK); waterlogging for 6 days at the third leaf, sixth leaf stage,and 10th day after tasseling,was conducted to investigate the mechanism of waterlogging-induced yield losses of summer maize.Waterlogging delayed plant growth and impaired tassel and ear differentiation,leading to high grain yield losses of Denghai 605(DH605).Waterlogging at third leaf(V3)stage reduced the photosynthesis of DH605, reducing total dry matter weight. Waterlogging at V3 stage reduced sucrose-cleaving enzymes activities in spike nodes and ears, reducing the carbon partitioned to ears (–53.1%), shanks (–46.5%), and ear nodes (–71.5%) but increasing the carbon partitioned to ear leaves(9.6%)and tassels(43.9%)in comparison with CK.The reductions in total carbon assimilate together with the reduced carbon partitioning to ears resulted in poor development of spikes (with respectively 15.2%and 20.6%reductions in total florets and fertilized florets)and lengthened the anthesis–silking interval by around 1 day, leading to high yield losses.

1. Introduction

By planting area and gross food production,maize is the largest food crop in China. With the world population growing and increased consumption of animal protein, the demand for maize in China is continually increasing [1–3]. However, the frequency and intensity of extreme weather events such as extreme heat,cold, floods, and droughts have also continually increased against the backdrop of global climate change, resulting in severe food security problems [4]. Improving maize resilience to extreme weather is a vital step towards meeting the food and fuel demand.

In China, the Huang-Huai-Hai region is the main production area of summer maize, accounting for respectively 35% and 37%of China’s maize yield and planting area [5]. However, maize production in the Huang-Huai-Hai Plain suffers increasingly severe losses from erratic rainfall patterns and frequent rainfall-induced waterlogging during the summer-maize growth cycle [6,7]. It is thus desirable to investigate summer maize response to waterlogging stress.

Maize is generally susceptible to flooding stress, which can occur at various growth stages. The response of summer maize yield to flooding depends on the growth stage and the duration of the flooding event. In previous studies [7,8], the greatest loss of grain yield in response to waterlogging occurred at the seedling stage, followed by the booting and flowering stages. Diverse responses to waterlogging have been observed. Waterlogging at early growth stages of summer maize triggered yield losses mainly by reducing the‘‘source”(the dry matter weight of summer maize plants) which limited the formation of the ‘‘sink” (grain), whereas waterlogging at reproductive stage restricted mainly the growth and development of the ‘‘sink”. In terms of yield components,waterlogging at the third and sixth leaf stages resulted in great decreases in kernels per ear, whereas waterlogging at the 10th day after flowering reduced mainly 1000-kernel weight [8].

Extension of the above results suggests that losses of grain yield may be attributed mainly to decreases in kernels per ear and 1000-kernel weight. However, these traits depend on a series of physiological processes involved in ear differentiation and grain development,including female spikelet differentiation,floret abortion, grain setting processes, and coordination of tassel and ears development (indicated by the anthesis–silking interval (ASI)),which are regulated by genotype and environmental factors [9–11]. Previous reports [12–14] have attributed variation in ear differentiation responses to management and environmental factors including drought, shading, planting density, nutrient rates, and soil moisture. However, the effects of these factors on maize ear differentiation have been highly variable.For example,some studies [15,16] have shown that the total number of florets per ear of maize was specific to hybrids and some reports [12–14] indicate that environmental factors,such as soil fertility,light,and temperature could also affect total number of florets.It is thus desirable to study waterlogging effects on spike differentiation of summer maize.

Leaf photosynthesis rate (Pn) influences maize yield, given that the development of harvested organs is driven by photosynthate accumulation.The direct correlation of grain yield with ear growth suggests a close relationship between Pnand ear differentiation.[17,18]. Water, temperature, light, and other abiotic stresses may trigger a decline in leaf photosynthetic rate,leading to undernourishment of ear and tassel.As a result,the differentiation and development of spikelet is inhibited,resulting in spikelet infertility and reduction in total florets [11,19,20]. Linkage between Pnand ear development may reflect a functional relationship between Pnand the plant ability to supply nutrition to ears.This function,however, has rarely been investigated in waterlogged summer maize.Sink organs of most plant species are supplied with carbon and energy in the form of sucrose,a major end product of photosynthesis. Sucrose must be cleaved by invertase and sucrose synthase to function in sink metabolism pathways. These sucrose-cleaving enzymes also play roles in the long-distance transport of sucrose,leading to the partitioning of assimilates among organs [21].

Previous studies have investigated the effects of waterlogging on photosynthesis rate and carbon partitioning to sink organs of summer maize. However, these studies have rarely investigated the relationship between plant production capacity,biomass partitioning, and spike differentiation processes. The objective of the present study was to investigate the effects of waterlogging on tassel and ear development of summer maize associated with photosynthetic capacity, sucrose metabolism, and carbon partitioning,with the aim of identifying the physiological mechanisms involved in waterlogging-induced yield losses.

2. Materials and methods

2.1. Experimental site and conditions

Field experiments were conducted in the 2017 and 2018 cropping seasons at an experimental farm of Shandong Agricultural University (36.09°N, 117.09°E) and State Key Laboratory of Crop Biology. The region has a temperate continental monsoon climate with a mean annual temperature of about 15 °C. The effective cumulative temperatures(temperature ＞10°C defined as effective temperature) of summer maize growth periods during 2017 and 2018 were respectively 1857.5 and 1836.2 °C d. The total annual precipitation was 569.6 and 752.7 mm in 2017 and 2018,of which the components occurring during the summer maize growth periods were 426.9 and 532.3 mm(Fig.S1).These weather parameters were collected from Tai’an meteorological station,China Meteorological Administration. The soil type was sandy loam, and soil pH was 8.05. The plowed soil (0–20 cm) before the experiment contained organic matter (C, 10.51 g kg-1), total N (0.85 g kg-1),rapidly available phosphorus(P,51.35 mg kg-1),and rapidly available potassium (K, 84.15 mg kg-1). N, P, and K fertilizers were applied as base fertilizer: 210 kg ha-1N (urea, 46% N), 75 kg ha-1P2O5(calcium superphosphate, 17% P2O5), and 150 kg ha-1K2O(muriate of potash, 60% K2O).

2.2. Plant materials and experimental design of trial 1

The summer maize hybrid Denghai 605 (DH605) was used as experimental material.The experiment was designed as a randomized block experiment with four treatments: control, no water stress (CK); and waterlogging for 6 days at the third leaf stage(V3-W), sixth leaf stage (V6-W), and 10th day after tasseling(10VT-W).Each treatment was replicated three times.The plot size was 4×4 m.Plots were separated by polyvinyl chloride partitions and the water to maintain the waterlogging condition for waterlogged pools was applied through water pipes. The water level was maintained at 2–3 cm above the soil surface in waterlogging treatments. The details are shown in Ren et al. [8]. Diseases and pests were well monitored and controlled.

2.2.1. Plant growth stages and tassel and ear development stages

Maize seed was sown on June 15, 2017 and June 8, 2018 to achieve a plant density of 67,500 plants ha-1. Plant growth rate was measured by leaf emergence, monitored by daily field observation. Basing on the positively correlated relationship between ear and tassel development and leaf number index, we sampled five representative plants from each plot at jointing growth stage(6th leaf stage, V6), booting stages (10th leaf stage, V10; 12th leaf stage, V12; 15th leaf stage, V15) and inflorescence emergence stage (tasseling stage, VT) respectively to determine the stages of tassel and ear development. In these plants, the growth status,height, leaf area, and other visible characteristics were similar to those of most plants in their treatment groups. Sampling dates were based on the growth stage of CK. The husks around the growth cone were stripped with a dissecting needle,and the development processes of ears and tassels were observed and photographed with a stereoscopic microscope. Ear and tassel length were also measured with the microscope with a micrometer(when ears and tassels were too small to be measured with a ruler) or with a ruler (when they were too large to be measured with the micrometer) [11]. The ASI was determined by daily observation as the interval between the dates when more than 50% of maize plants reached tasseling stage (D1) and silking (D2):

ASI (day) = D2 - D1

2.2.2. Ear and tassel characteristics

Ten representative plants of each plot were labeled before silking. After fertilization was complete, five representative plants were selected from each plot to measure the lengths of tassels and the numbers of tassel branches and flowers. Five ears from each plot were selected to measure ear floret differentiation on the fifth day after fertilization. Ovules with wilting silks indicated that fertilization had occurred, whereas those with turgid silks indicated that fertilization had not occurred and ovules without silks indicated degeneration.Floret fertility rate,floret setting rate,kernel abortion rate, and total set rate were calculated as follows[11]:

Floret fertility rate (%) = (fertilized floret number/total floret number) × 100

Floret set (%) = (number of normal kernels/number of fertilized florets) × 100

Kernel abortion rate(%)=(number of fertilized florets–number of normal kernels)/number of fertilized florets × 100

Total set rate (%) = (number of normal kernels/total floret number) × 100

2.2.3. Yield and yield components

Thirty ears from the middle three rows of each plot were sampled to measure yield and yield components. Maize kernels were determined with 14% moisture content.

Yield(kg ha-1)=ears(ears ha-1)×kernels(per ear)×1000-ker nel weight (g/1000 kernels) × 10-6

2.3. Plant materials and experimental design of trial 2

This experiment was conducted to investigate the physiological mechanism of waterlogging-induced yield losses in summer maize by investigating the photosynthetic capacity, sucrose metabolism,and carbon partitioning processes in 2017 and 2018. In this trial,the effects of waterlogging only at third leaf(V3)stage were investigated,as waterlogging at this stage triggers the most severe damage to summer maize [8]. The experimental material, planting density, management method, and other experimental details were as in trial 1.

2.3.1. Chlorophyll content

Ten complete and representative ear leaves were selected at VT stage respectively in 2017 and 2018.Their chlorophyll content was measured by a SPAD-502 chlorophyll meter (Minolta Camera CO.,Osaka, Japan). The reading value of SPAD-502 chlorophyll meter(SPAD) represents the relative leaf chlorophyll content.

2.3.2. Leaf net photosynthesis

In 2017, the Pnof ear leaves was measured in the middles of fully expanded leaves between 10:00 and 12:00 under saturating irradiance on days 1 (WL0), day 7 (WL7), day 21 (WL21), and 31(WL31) after waterlogging using a portable infrared gas analyzer(CIRAS II, PP System, Hansatech, U.K). In 2018, the Pnof ear leaves was measured on days 1 (WL0), 6 (WL6), 27 (WL27), and 32(WL32) after waterlogging. Measurement conditions were kept consistent:LED light source and photosynthetically available radiation (PAR)of 1600 μmol m-2. CO2concentration was maintained at a constant level of 360 μmol mol-1using a CO2injector with a high-pressure liquid CO2cartridge. Five plants per treatment (V3-W and CK) were randomly selected for measurement [8].

2.3.3. Dry matter weight

Five representative plants from each of the V3-W and CK treatments,on WL0,WL7, WL21,and WL31 in 2017 and on WL0,WL6,WL27, and WL32 in 2018 were sampled. Samples were dried at 80 °C in a forced-draft oven (DHG-9420A, Bilon Instruments Co.Ltd, Shanghai, China) to constant weight, and then weighed separately.

2.3.4.13C pulse labeling, sampling and analysis

At the 12-leaf stage,five plants were selected for13CO2labeling and five for reference from the V3-W and CK treatments in 2018.The 12th leaf was placed in the labeling chamber (a transparent plastic oven bag sealed at both ends) and 60 mL of13CO2air was pumped into the labeling chamber. After 1 h, the bags were removed from the plants. At VT stage, three13CO2labeled plants and three reference plants were separated into stems, leaves, ear leaves, spike nodes, shanks, ears, and tassels, respectively. These samples were dried, weighed, and ball-milled for analysis [22].

2.3.5. Assay of soluble sugar, sucrose, and starch contents

At VT stage, three plants were sampled from each treatment and separated into stems, leaves, ear leaves, spike nodes, shanks,ears, and tassels. All samples were dried, weighed, and ballmilled. Sugars were extracted following Hanft et al. [23]. 100-mg tissue samples were extracted directly in 6 mL boiling water for 20 min, the supernatant was collected, and the residues were extracted a second time in 6 mL boiling water for 20 min. The extract was then diluted to 50 mL constant volume with deionized water and named solution A.In addition,4 mL water and 2 mL 2 N HClO4were added to the residues,then were placed in boiling bath for 20 min. After this, the supernatant was collected, and the residues were extracted a second time in 5 mL water and 1 mL HClO4in boiling bath for 20 min. The extract was then diluted to 50 mL constant volume with deionized water and named solution B. For soluble sugar analysis, 100 μL solution A was reacted with 3 mL anthracenone solution for 10 min in a boiling water bath and after cooling, absorbance at 620 nm was measured with a spectrophotometer.For sucrose content analysis,20 μL solution A was reacted with 100 μL KOH (30%) for 10 min in a boiling water bath, cooled,and added to 3 mL anthracenone solution after cooling,and absorbance at 480 nm was measured.For starch content analysis,100 μL solution B was reacted with 3 mL anthracenone solution for 10 min in a boiling water bath and cooled,and absorbance at 620 nm was measured.

2.3.6. Activity assays for invertase and sucrose synthase

At VT stage, three plants from each treatment were sampled and separated into stems, leaves, ear leaves, spike nodes, shanks,ears, and tassels. These samples were frozen in liquid nitrogen and held at–80°C.Fresh samples of stems,leaves,ear leaves,spike nodes, shanks, ears, and tassels from different treatments were rapidly frozen in liquid nitrogen and ball-milled. Subsamples of 0.5 g were extracted in 4.5 mL phosphate acid buffer (pH 7.2–7.4)followed by centrifugation for 20 min at 3000 r min-1and collection of the supernatants for analysis. The activities of invertase and sucrose synthase were determined with a Plant Invertase ELISA kit and a Plant Sucrose synthase cleavage (SS-C) ELISA kit(mlbio, Shanghai, China) according to the manufacturer’s instructions. Sample dilution solution (40 μL), 10 μL collected supernatants, and 100 μL enzyme standards were added in sequence to each well of enzyme plate, then was incubated at 37 °C for 1 h and then washed five times. Color rendering solutions A and B(50 μL each) were added and mixed followed by dark incubation at 37 °C for 15 min and addition of 50 μL stop solution to stop the reaction. Enzyme activities were measured immediately at a wavelength of 450 nm.

2.4. Data analysis

Microsoft Excel 2013 (Microsoft, Redmond, WA, USA) and SigmaPlot 10.0 (Systat Software, Inc., Richmond, CA, USA) were used for data processing and plotting,IBM SPSS Statistics 21.0(IBM Corporation, Armonk, NY, USA) was employed for data statistics and analysis. Comparisons among groups were tested by one-way ANOVA and LSD tests.

3. Results

3.1. Yield and yield components

Large differences in yield and yield components were found in each year among the three waterlogging treatments (Table 1).Waterlogging at V3 stage caused the greatest yield loss, followed by V6-W and 10VT-W. On average, the grain yields of V3-W, V6-W, and 10VT-W were decreased by 27.2 (27.2 = (29.74 + 24.74)/2), 15.0 (15.0 = (14.66 + 15.26) /2), and 11.8% (11.8 =(11.8 + 11.79) /2), respectively, relative to CK. The waterlogginginduced decreases in kernels per ear (1.6%–23.2%) and 1000-kernel weight(1.1%–8.5%)were much larger than that in ear number per ha(0–4.9%)(Table 1).The decreases in kernels per ear and 1000-kernel weight in V3-W and V6-W contributed to respectivelyabout 70% and 20% of yield losses. In contrast, the yield loss of 10VT-W was caused mainly (75%) by the decrease in 1000-kernel weight (Fig. S2).

Table 1Grain yield and its components in the maize hybrid DH605.

3.2. Plant growth and spike differentiation stages

Waterlogging delayed plant growth and development. In 2018,the periods of V3 to 6th leaf stage(V6),V6 to 9th leaf stage(V9)of V3-W were 2 days longer than those of CK. In V6-W, the V9 and V12 stages were delayed by 2 and 3 days, respectively, relative to CK. In 10VT-W, the milk stage (R3) was delayed by 3 days.The ASI was increased by around 1 day in both V3-W and V6-W.The growth process of summer maize lagged mostly in V3-W, followed by V6-W and 10VT-W(Table S1).Accordingly,tassel and ear development lagged as indicated by the tassel and ear lengths,which were significantly shorter in V3-W and V6-W than in CK.The initiation of tassel and ear development was delayed by waterlogging at the V3 stage (V3-W). However, when waterlogging occurred at the V6 stage, the initiation of tassel development did not lag,evidently because the tassel had already entered the spikelet differentiation stage.Ear formation was delayed by 7 and 5 days in V3-W and V6-W, respectively, compared with CK. The stagnation of reproductive growth appeared to be more severe than that of vegetation growth. When waterlogged summer maize reached the same growth stage of CK, however, its ear and tassel differentiation still lagged. Thus waterlogging, especially at the V3 stage(V3-W), greatly impeded tassel and ear growth and development(Figs. 1, 2).

3.3. Tassel and ear characters

Numbers of tassel flowers, tassel length, tassel branches, and infertility of flowers were all affected by waterlogging.On average,the lengths of tassels of V3-W,V6-W,and 10VT-W were decreased by respectively 19.6%, 5.0%, and 3.2% relative to CK (Fig. S3). Relative to CK, the average total tassel florets and the mean branch numbers of V3-W and V6-W were much lower than that of CK(Table 2).Waterlogging at V3 stage caused the most severe damage to summer maize. With respect to ear development, waterlogging at V3 stage significantly reduced the florets per ear and floret setting rate in 2017,and decreased the florets per ear and floret fertility rate in 2018. Waterlogging at V6 stage reduced the total number of florets in 2017,and reduced the total number of florets and the floret fertility rate in 2018. In contrast, waterlogging at 10VT stage reduced the kernels per ear by reducing the floret setting rate (Table 3).The mean ear lengths of V3-W and V6-W were lower than that of CK. However, the mean ear length was not significantly affected by waterlogging at 10VT stage (Fig. S3). Waterlogging led to 20.6%, 9.3%, and 2.0% mean decreases of kernels per ears in V3-W,V6-W,and 10VT-W,respectively(Table 3).The mean losses of kernels per ears of 2017 and 2018 resulted in 75.2%,62.7%, and 17.1% reductions of grain yield in V3-W, V6-W, and 10VT-W, respectively, compared to that of CK (Fig. S2).

Table 2 The means of the total number of tassel flowers, aborted tassel flowers, and tassel branches of maize hybrid DH605.

Table 3 The means of total florets, fertilized florets and kernels per ear of maize hybrid DH605.

3.4. Partial correlation analysis of factors affecting yield composition

Kernels per ear was positively correlated with the total spikelet number, number of branches, and length of tassels and negatively correlated with the spikelet abortive rate. However, total spikelet number and number of branches were negatively correlated with 1000-kernel weight (Fig. 3A). In addition, total number of florets and number of fertilized florets were negatively correlated with 1000-kernel weight, but were positively correlated with kernel number per ear.Floret fertilization rate showed a positive correlation with kernels per ear, while floret fertilization rate showed a positive correlation with 1000-kernel weight (Fig. 3B). Tassel branches and total kernel setting rate were positively correlated,while sterile spikelet number was positively correlated with floret setting rate (Fig. 3C).

3.5. Photosynthetic characteristics and dry matter accumulation

Waterlogging at V3 stage significantly decreased SPAD by respectively 41.7%, 34.3%, 11.8%, and 12.0% on days 0, 7, 21, and 31 after waterlogging in 2017, and 45.8%, 27.7%, 12.2%, and 14.3%on days 0, 6, 27, and 32 after waterlogging in 2018 compared to that of CK. Waterlogging reduced the stomatal conductance and intercellular CO2concentration of V3-W by 20%–30% in 2017 and 2018, compared with those of CK. However, the transpiration rate was not significantly affected by waterlogging. The water use efficiency of V3-W was decreased by around 20% in 2017 and 2018,compared with that of CK (Fig.S4). As a result,leaf Pnvalues were significantly decreased. The Pnof V3-W treatment during the month after waterlogging was about 20%–40% lower than that of CK. Accordingly, the dry matter weight of waterlogged summer maize was decreased by waterlogging (Fig. 4).

Fig. 1. The growth stages of spike development in maize hybrid DH605. CK, control, no water stress; V3-W, waterlogging at the third leaf stage for 6 days; V6-W,waterlogging at the sixth leaf stage for 6 days.

3.6. The effects of waterlogging on 13C distribution

Waterlogging significantly reduced the dry matter weight accumulation of ear leaves, spike nodes, shanks, and ears and increased the dry matter weight of tassels. Accordingly, the carbon contents of ear leaves, spike nodes, shanks, and ears of waterlogged summer maize were respectively 26.5%, 46.0%,36.4%, and 52.6% lower than those of CK and the carbon content of tassels was 28.8% higher than that of CK. The13C allocation proportion of waterlogged summer maize was increased in ear leaves (9.6%) and tassels (43.9%), but was decreased in spike nodes (71.5%), shanks (46.5%), and ears (53.1%), compared with that of CK (Fig. 5).

Fig. 2. The growth stages of ear development in the maize hybrid DH605. CK, control, no water stress; V3-W, waterlogging at the third leaf stage for 6 days; V6-W,waterlogging at the sixth leaf stage for 6 days.

3.7. The effects of waterlogging on the content of soluble sugar,sucrose, and starch and the activities of sucrose-cleaving enzymes

The soluble sugar contents of ear leave and tassels for waterlogging treatment were increased by respectively 46.7% and 11.3%across years compared to that of CK. However, the soluble sugar contents of spike nodes and ears were significantly lower than those of CK across years.The content of starch in waterlogged summer maize was significantly decreased in spike node (–20.0%),shank (–25.0%), and ear (–20.3%) on average, compared with that of CK. The sucrose content of spike nodes was also significantly decreased after waterlogging, while was significantly increased in shanks and ears across years (Fig. 6). The sucrose synthase activities in ear leaf and shank were increased by 19.9% and 22.8% on average, respectively, but were decreased by respectively 21.2%,26.8%,and 29.6%in spike node,ear,and tassel of waterlogged summer maize, compared with that of CK. The activities of sucrose invertase in ear leaf and tassel were increased by 177.6% and41.2%, but decreased by respectively 78.2%, 64.2%, and 36.3% in spike node, shank, and ear across years, compared to that of CK(Fig. 7).

Fig.3. Correlation heat maps.According to the partial correlation coefficients of tassel traits with yield and yield components,ear traits with yield and yield components,and tassel traits with ear traits, thermal maps (heat maps) are shown. Red (1) represents a significant positive correlation and green (–1) a significant negative correlation. The darker the color of the color block,the stronger is the correlation.The data used for analysis are five biological replicates of each treatment.(A)Correlation heat map of tassel traits with yield and yield components. (B) Correlation heat map of ear traits with yield and yield components. (C) Correlation heat map of tassel traits with ear traits.

4. Discussion

4.1.Poor development and discordant of reproductive organs resulted in yield losses

This study indicated that the decrease of kernels per ear was the major contributor to waterlogging-induced grain yield losses. It has been reported that mature kernels per ear is determined by total florets, florets abortion, and post-fertilization seed abortion,especially by florets and post-fertilization seed abortion [24–26].In this study, waterlogging reduced the total number of florets,while the total abortion rate was maintained at the same level as CK or greater. As a result, the mature kernel seeds per ear was decreased, suggesting that total florets is a key factor determining kernels per ear of waterlogged summer maize. Waterlogging also reduced the length and number of branches and total spikelets of tassels.Previous studies[27–29]have suggested that abiotic stressors, including drought, shading, and high planting density, could hinder tassel exsertion and reduce the length,number of branches,and total spikelets of tassels.As a consequence,the total amount of pollen would be decreased, resulting in insufficient fertilization of female spikes and reduced kernel numbers. The ASI was lengthened in waterlogged maize. Discordance between silking period and anther dehiscence would further increase sterility [30,31].Therefore, the spike characteristics and the coordination of tassels and ears development were major determinants of the number of fertilized florets. The numbers of branches, spikelets, and kernels per ear, as well as spikelet length, were positively correlated with one another. Tassel branches and total kernel setting rate, sterile spikelets, and floret setting rate were also positively correlated.These correlations indicate that increasing the growth of sex organs coupled with coordinating tassel and ear development is necessary to achieve a high grain yield in summer maize.

Fig.4. Waterlogging effects on photosynthesis rate and dry matter weight of summer maize in 2017 and 2018.(A)The photosynthesis rate of waterlogged summer maize in 2017. (B)The dry matter weight of waterlogged summer maize in 2017. (C)The decrease rate of dry matter weight and photosynthesis rate of waterlogged summer maize,compared with that of CK in 2017.(D)The photosynthesis rate of waterlogged summer maize in 2018.(E)The dry matter weight of waterlogged summer maize in 2018.(F)The decrease rate of dry matter weight and photosynthesis rate of waterlogged summer maize,compared with that of CK in 2018.WL0,day 1 after waterlogging;WL6,day 6 after waterlogging;WL7,day 7 after waterlogging;WL21,day 21 after waterlogging;WL27,day 27 after waterlogging;WL31,day 31 after waterlogging;WL32,day 32 after waterlogging. Different letters on bars indicate significant differences among treatments at P ＜0.05 using LSD test. CK, control, no water stress; V3-W, waterlogging at the third leaf stage for 6 days.

4.2. Reduced assimilate supply impaired the growth of reproductive organs under waterlogging

Environmental stresses caused greater damage to reproductive than to vegetative organs, and these damages were always rapid and irreversible [19]. In the present study, waterlogging slowed leaf emergence; in particular, the interval from V3 to V10 stage was lengthened by 6 days in V3-W treatment,relative to CK.Moreover, the tassels and ears at V10 stage of waterlogged maize were much smaller and younger than those at V10 stage of CK. These results suggest that waterlogging slowed the growth rate of the reproductive organs much more than that of the vegetative organs.Among the treatments, waterlogging at V3 stage caused the most severe damage to summer maize.Poor supply of assimilates seems to be the major cause of impairment of ear and tassel growth[12].The primordial body of spikes, as an active growth zone, requires abundant photosynthetic products to drive its growth. Study [13]has also shown that floret differentiation rate is positively correlated with photosynthetic capacity in the month before tasseling,which is a key determinant of the total number of florets. We determined the plant photosynthesis rate and the dry matter weight accumulation during the month before tasseling of V3-W(this period also matched the month after waterlogging).The photosynthesis rate and dry matter accumulation rate of waterlogged summer maize were decreased significantly and maintained at a lower level than in CK. As a result, the total assimilates supplied to reproductive organs were decreased and this decrease may be the major contributor to the decreases in total florets of ear and tassel. Translocation of photo-assimilate was also impaired by waterlogging, reflecting a trade-off between yield and survival.Assimilates are accumulated in crops mainly as carbohydrates,consisting primarily of sucrose, fructan, and hexoses [32]. Photoassimilate is translocated mainly in the form of sucrose and used in the form of hexoses. In this study, at the end of waterlogging,the soluble sugar and sucrose of waterlogged summer maize leaves were significantly higher than those of CK(Fig.S5).Thus,waterlogging resulted in an increase in sucrose metabolism in leaves and an insufficient sucrose supply from the leaves.These processes would further enlarge the deficiency of carbohydrates to drive the ear and tassel growth.As a result,the development of ear and tassel could not match the leaf number index (Figs. S6, S7). Previous studies[33,34] have shown a clear relationship between the spike differentiation and leaf number index. The duration of the differentiation time of different differentiation stages was also disorganized. For example, the period of floret differentiation was shortened by 1 day in V3-W treatment, compared with that of CK (Fig. S8). As a result, tassel and ear characters were impaired in the form of decreases in spikelet length, number of branches,total spikelets of tassels, total number of florets, and numbers of fertilized florets of ears (Fig. 8).

Fig. 5. The waterlogging effects on dry matter accumulation and allocation in summer maize in 2018. (A) The dry matter weight of waterlogged summer maize. (B) The carbon content of waterlogged summer maize.(C)The 13carbon content of summer maize.(D)The proportion of 13carbon in different organs of waterlogged summer maize.Different letters on bars indicate significant differences among treatments at P ＜0.05 using LSD test.CK,control,no water stress;V3-W,waterlogging at the third leaf stage for 6 days.

Fig. 7. Waterlogging effects on the sucrose-cleaving enzymes in summer maize in 2017 and 2018. (A) The activity of sucrose synthase in 2017. (B) The activity of sucrose invertase in 2017.(C)The activity of sucrose synthase in 2018.(D)The activity of sucrose invertase in 2018.Symbols followed by different letters denote statistical difference at P ＜0.05 by LSD test. CK, control, no water stress; V3-W, waterlogging at the third leaf stage for 6 days.

4.3. Weak competition for assimilates between tassels and ears lengthened the ASI

Fig.6. Waterlogging effects on soluble sugar content,sucrose content,and starch content of summer maize in 2017 and 2018.(A)The content of soluble sugar in 2017. (B)The content of sucrose in 2017.(C)The content of starch in 2017(D)The content of soluble sugar in 2018.(E)The content of sucrose in 2018.(F)The content of starch in 2018.Symbols followed by different letters denote statistical difference at P ＜0.05 by LSD test.CK, control,no water stress;V3-W,waterlogging at the third leaf stage for 6 days.

Crops such as maize that depend on cross-pollination and synchronous flowering are susceptible to climate change,as some abiotic factors such as temperature and precipitation will be altered,inducing mismatched phenologies such as changes in flowering synchrony [31]. Lengthening of the interval between anthesis and silking is a response to environmental stresses [8]. The grain yield of maize decreased by 82%, while the ASI lengthened from 0 h to 28 h, indicating a strong negative correlation between ASI and grain yield [10]. Changes in ASI may indicate differences in partitioning of assimilate to the ear at flowering[35].In agreement with this hypothesis, many previous studies [36–42] have shown that impeding the transport of assimilates to tassels promoted ear growth, suggesting that competition for assimilates was present between tassels and ears. In the present study, waterlogging delayed silking, resulting in lengthening of ASI. Distributions of13C to leaf and tassel were increased, but were decreased in spike nodes, shanks, and ears of waterlogged summer maize, relative to CK.Thus,waterlogging impeded the translocation of assimilates out of leaves.Moreover,the difference in carbon partition between tassel and ear suggested ears are less competitive for assimilates than tassels in waterlogged summer maize. Therefore, differences in growth and development between tassel and ear were increased, leading to the increase in ASI.

Sucrose is the major photo-assimilate translocated from source to sink organs such as ears and tassels.Sucrose deficiency has been proposed [43–45] to be the major cause of growth suppression of maize kernels.In the present study,total soluble sugar and sucrose were significantly lower in spike node of waterlogged summer maize,indicating the inhibition of translocation of assimilates from ear leaf to spike node.Such inhibition may result in poor development of spike nodes, impeding the translocation of assimilates from source to sink. However, in this study, the sucrose in shank and ear was increased in waterlogged summer maize, compared with that of CK. This finding suggests that factors other than deficient sucrose content contribute to poor development of ears,rather.Indeed, previous studies [46,47] have indicated that depletion of sucrose in aborted kernels is the result, rather than the cause,of kernel abortion.The finding that the soluble sugar content of ears was decreased in waterlogged summer maize, compared with that of CK,suggests a poor ability of the ear to cleave sucrose into hexoses.Thus,poor ability to use sucrose may be a major contributor to impaired growth of the ear.

High invertase could facilitate apoplasmic phloem unloading of sucrose to developing organs such as ears[48].Upon translocation through the phloem to sinks, sucrose is degraded by either invertase or sucrose synthase into hexoses, which are then used in diverse ways to drive plant growth and development processes[48–52]. In this study, invertase and sucrose activities were decreased in spike nodes and ears,but were increased in ear leaves of waterlogged summer maize,suggesting reduced ability of spike nodes and ears to use sucrose.In turn,high sucrose content in ears may result in a negative feedback effect on the translocation of sucrose to ears. Thus, the carbon partition in ears was further impeded, in agreement with a lower13C proportion in ears, compared with that of CK. Thus, the low activities of sucrose-cleaving enzymes may be a major factor in the low availability of assimilates. Moreover, although the invertase activity of tassels was detected at a lower level, the sucrose synthase activity was at a higher level, relative to CK. The soluble sugar and sucrose content of tassels in waterlogged maize were maintained at the same level as that CK,suggesting that the increases of sucrose synthase activity counteracted the decreases of invertase activity in tassels. The difference in the activities of sucrose-cleaving enzymes between ear and tassel further pointed to a weak competition for assimilates of ears, compared with that of tassels. As a consequence,the ASI was lengthened under waterlogging (Fig. 8).

Fig.8. The mechanism of yield loss in summer maize induced by waterlogging.Waterlogging reduced the photosynthesis rate of summer maize,and affected the activity of sucrose-cleavage enzymes and sucrose content,thereby reducing the supply of assimilates to ears and tassels.As a result,the development of ear and tassel were impaired.Moreover, the competition by ears for assimilates was weaker than that by tassels, leading to a lengthened ASI. As a consequence, kernels per ear and kernel weight of waterlogged summer maize were reduced, resulting in severe yield loss.

Our findings are in agreement with those of former study [31]predicting that climate change will reduce investment in female function in some monoecious crops. We suggest that preventing the development of tassels by removing part of tassels in the field may be an effective management option for mitigating waterlogging-induced yield losses in current agricultural practice.

5. Conclusions

Waterlogging stress triggered extensive damage to the ear and tassel differentiation process, resulting in decreases in total and fertilized florets and an increase in ASI, contributing to grain yield loss in summer maize.Lower photosynthetic capacity of leaves and poor supply of assimilates were the major contributors to poor development of ears and tassels.Weak competition for assimilates of ears under waterlogging conditions contributed to lengthening ASI, aggravating the yield losses.

CRediT authorship contribution statement

Juan Hu:Data curation, Writing – original draft, Visualization,Investigation.Baizhao Ren:Writing - review & editing.Shuting Dong:Supervision.Peng Liu:Supervision.Bin Zhao:Supervision.Jiwang Zhang:Conceptualization, Writing - review & editing,Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the National Natural Science Foundation of China(31801296), China Agriculture Research System of MOF and MARA (CARS-02-18), the National Key Research and Development Program of China (2017YFD0300304), and the Postdoctoral Innovation Program of Shandong Province (202003039).

Appendix A. Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.08.001.