Yuling Guo,Guanmin Huang,Qing Guo,Chuanxi Peng,Yingru Liu,Mingcai Zhang,Zhaohu Li,Yuyi Zhou,Liusheng Duan

State Key Laboratory of Plant Physiology and Biochemistry,College of Agronomy and Biotechnology,China Agricultural University,Beijing 100193,China

Keywords:Maize Drought resistance Coronatine Root development Grain yields

ABSTRACT Drought stress caused by insufficient irrigation or precipitation impairs agricultural production worldwide.In this study,a two-year field experiment was conducted to investigate the effect of coronatine(COR),a functional analog of jasmonic acid(JA),on maize drought resistance.The experiment included two water treatments(rainfed and irrigation),four COR concentrations(mock,0 μmol L?1;A1,0.1 μmol L?1;A2,1 μmol L?1;A3,10 μmol L?1)and two maize genotypes(Fumin 985(FM985),a drought-resistant cultivar and Xianyu 335(XY335),a drought-sensitive cultivar).Spraying 1 μmol L?1 COR at seedling stage increased surface root density and size,including root dry matter by 12.6%,projected root area by 19.0%,average root density by 51.9%,and thus root bleeding sap by 28.2%under drought conditions.COR application also increased leaf area and SPAD values,a result attributed to improvement of the root system and increases in abscisic acid(ABA),JA,and salicylic acid(SA)contents.The improvement of leaves and roots laid the foundation for increasing plant height and dry matter accumulation.COR application reduced anthesis and silking interval,increasing kernel number per ear.COR treatment at 1 μmol L?1 increased the yield of XY335 and FM985 by 7.9% and 11.0%,respectively.Correlation and path analysis showed that grain yields were correlated with root dry weight and projected root area,increasing maize drought resistance mainly via leaf area index and dry matter accumulation.Overall,COR increased maize drought resistance mainly by increasing root dry weight and root area,with 1 μmol L?1 COR as an optimal concentration.

1.Introduction

Maize(Zea mays)is one of the three major food crops,used for feed and biofuel materials[1].The total planting area of maize around the world is 197 million hectares and the total yield is 1.15 billion tons of grain[2].The continued growth of population and consumption and climate change have brought great challenges to agricultural production.With increasing global population,the food demand by 2050 will increase by 60%,which requires more irrigation water(World Water Assessment Programme,United Nations Educational,Scientific and Cultural Organization,New York,NY,USA).Drought is the main abiotic stress limiting the production of maize.China has 41.3 Mha of maize planting area and produces 38.9%of the total production of all food crops[3].However,China is a country severely deficient in fresh water resources.The average annual water consumption for each person is 2194 m3,28% of the global average.In China,about 60%of arable land is dryland under water deficit,with 40%of its distribution occurring in semi-arid areas affected by drought[4].Drought and extreme heat caused a 9%–10% reduction in national cereal production from 1964 to 2007,and reductions were 8%–11%higher in developing than in developed countries[5].Increasing the drought resistance of maize in semi-arid areas will thus increase food security.

Leaves function in plant growth and development and are sensitive to external environmental changes.For example,leaves lose water and curl under drought conditions.The chlorophyll content of leaves decreases with the increase of osmotic stress,leading to leaf drying and death in severe cases.Drought stress reduced the chlorophyll content of leaves and caused irreversible damage to them[6].Maintaining high levels of chlorophyll content in leaves under drought conditions is an effective way to improve maize drought resistance.Drought stress could also inhibit maize leaf development and reduced leaf area.An experimental study[7]employing an unmanned aerial vehicle showed that leaf staygreen decreased by 13.7% under drought conditions in response to drying and death of the leaves.Results of a study[8]have shown that drought reduced single-plant leaf area by about 50%relative to a control treatment.The maize ear leaf is a source organ for yield production.Ear leaf morphological traits influence maize plant configuration and drought resistance[9].Drought stress can also change the physiological and biochemical characteristics of maize leaves.Accumulation of peroxides in cells increased under drought conditions,causing cell membrane damage,manifested as an increase in electrical conductivity[10].There is a positive relationship between ABA content and crop drought resistance.ABA concentration increased under osmotic stress and induced stomatal closure to increase drought resistance[11].As typical stressresponsive hormones,SA and JA are induced by drought stress and positively regulate the drought resistance of plants[12].

The root system is the bridge between soil and plant and absorbs moisture from the soil for the plant.Plants can increase drought resistance by inhibiting the growth of aboveground parts and increasing the growth of belowground parts to promote the absorption of water[13,14].Under drought-stress conditions,a decrease in root length and biomass will eventually lead to a decrease in root volume and surface area,limiting shoot growth[15].Drought-resistant maize genotypes have vigorous root systems with higher root biomass,surface area,and volume[16].Root bleeding sap is affected by root pressure and represents root activity[17],so that root bleeding sap reflects root behavior,especially water absorption[18].The flow rate of root-bleeding sap decreased by 75.69%under drought treatment compared with a control[19].

Crop drought resistance is a complex trait involving several metabolic and morphological adaptive pathways[20].Grain yield is the primary trait reflecting drought resistance.The yield loss of drought-sensitive varieties is higher than that of droughtresistant varieties under drought conditions.The mean loss in crop yield was about 30% caused by drought stress[21],and extreme drought could even prevent any yield.Secondary traits also figure in assessing drought resistance of maize genotypes.Dry matter accumulation and plant height in shoot are sensitive to drought stress.Under drought conditions,dry matter accumulation decreased by respectively 17%,19% and 44% at the V7,V8,and V9 stages[22].In drought resistance breeding,secondary traits include anthesis-to-silking interval(ASI).ASI refers to the time interval between anthesis and silking,which can be extended by drought during reproductive growth stage,resulting in seed abortion and yield losses[23].

One method of increasing crop drought resistance is the application of plant growth regulators.Application of 2-(3,4-dichlorophenoxy)triethylamine(DCPTA)ameliorated the effects of drought on nitrogen metabolism in maize,increasing yield under drought conditions[19].Jasmonic acid(JA)is an endogenous growth regulator in higher plants[24]that plays multiple roles in plant resistance to both biological and abiotic stresses.The content of JA in plants increased rapidly under drought conditions[25].The JA signaling pathway can regulate a variety of transcription factors to participate in improving drought resistance via synergistic or antagonistic effects with other hormones.JAZ(jasmonate ZIMdomain)formed a complex with ABA receptor PYL to activate MYC2 transcriptional activity[26].Spraying JA maintained a high relative water content in maize leaves,increasing drought resistance[27].JA-treated seedlings had more branched roots than the control treatment[28].

Coronatine(COR)is a host-nonspecific phytotoxin produced by several pathovars of Pseudomonas syringae[29]that functions in resistance to abiotic stress.COR is a functional analog of JA that has similar function but also its own unique functions[30].COR increased drought resistance in tobacco,inhibited mesocotyl elongation in maize seedlings,and promoted lateral-root elongation in cotton[29,31,32].But despite studies of the roles of plant growth regulators in drought resistance,mechanisms by which COR increases plant drought stress tolerance under field conditions have not been reported.

The objectives of the present study were to(1)investigate the effects of COR on leaf development,root growth,grain yield,and other agronomic traits associated with drought in the field environment,(2)identify the most effective indicators for increasing final grain yield,and(3)identify the optimum COR concentration to provide practical support for agricultural production.

2.Materials and methods

2.1.Experimental site

Two-year field experiments were conducted in 2018 and 2019 at Qian’an Experimental Station of Academy of Agricultural Sciences in Jilin province,China(42°37′182′′N,125°48′106′′E;elevation 300 m above mean sea level).It is a typical site characterized by a temperate continental monsoon climate.Spring maize is the main crop in this area.The mean yearly sunshine duration and frost-free periods are 2866.6 h and 146 d,respectively.The mean annual precipitation was 418 mm from 1957 to 2018.The mean yearly evaporation is 1872 mm,which is 4.48 times the precipitation[33].The main soil type in this area is chernozem.The upper layer of the test site contained 17.8 g kg-1of organic matter,0.94 g kg-1of total nitrogen,8.65 mg kg-1of available phosphorus,and 106 mg kg-1of available potassium.The mean soil bulk density in the 0–40 cm layer was 1.2 g cm-3and the mean field water capacity was 24.7%.

2.2.Experimental design

The experiment was laid out in a split-plot design with two factors.The main-plot factor was irrigation and the subplot factors were cultivar and plant growth regulator treatment.There were two water treatments:well-watered(irrigation to ensure that the relative water content of the soil exceeded 60%;W)and drought(rainfed;D)treatments.Two maize genotypes Fumin 985(FM985,drought-resistant)and Xianyu 335(XY335,droughtsensitive)were used as test materials,both widely cultivated in Northeast China.The experimental plant growth regulator was COR and distilled water was used as a control(mock).Based on preliminary screening in the laboratory,the COR concentrations were 0.1 μmol L-1(A1),1 μmol L-1(A2),and 10 μmol L-1(A3),respectively.COR was sprayed at the V4 stage,applied with a 5 L precision sprayer at a 450 L ha-1.There were 16 treatments with four replicates.In accord with local planting practice,maize planting density was set to 75,000 plants ha-1with a row spacing of 75–45–75 cm(Fig.1A).Each treatment plot area was 33.6 m2(7 m length×4.8 m width).There was a wide buffer zone(2.4 m)between irrigated and rainfed areas.Plastic mulching with drip irrigation was applied shortly after planting.A drip tape was shared between the two short rows.Fertilizer was applied following high-yield practice:110 kg ha-1N,67.5 kg ha-1P2O5,and 85 kg ha-1K2O were used before sowing,followed by 150 kg ha-1N,21.5 kg ha-1P2O5,and 26 kg ha-1K2O applied at the V8 stage via drip irrigation.Other management measures followed local field practice.The sowing data,harvest date,and other growth stages are shown in Table S1.No insect and weed damage occurred during the growing season.

Fig.1.Control of soil moisture content by precipitation and drip irrigation system.(A)Schematic diagram of planting and drip irrigation system.Drip irrigation systems provided timely and adequate irrigation to the soil.(B)Annual precipitation(bar)and temperature(line)in Qian’an county in recent 20 years.(C)Mean monthly precipitation(bar)and temperature(line)in 2018 and 2019 and the 20-year mean in Qian’an.(D)Soil moisture at maize growth stages under irrigation and rainfed conditions.SW,sowing;SD,seedling;JT,jointing;TS,tasseling;MK,milk;PM,physiological maturity;HV,harvesting.

2.3.Environmental variables and maize trait measurements

2.3.1.Soil water content(SWC)

SWC was measured with a soil auger at different maize growth periods with an interval of about 20 days at sowing(SW),seedling(SD),jointing(JT),tasseling(TS),milk(MK),physiological maturity(PM),and harvesting(HV)stages.Soil samples were collected from the 0–80 cm soil layer at 20-cm depth intervals with 3–5 replicates.Soil samples were immediately weighed(W1)and then dried at 105 °C to constant weight(W2).The field capacity(Fc)of the soil at 40 cm was measured before sowing[34].SWC was calculated as follows:

where n is the number of sample layers and i indexes the first(0–20 cm),second(20–40 cm),third(40–60 cm),and fourth(60–80 cm)soil layers.

2.3.2.Measurement of chlorophyll content,leaf area index(LAI),and electrical conductivity(EC)

The chlorophyll contents(SPAD values)of the V9 and V13(ear leaf)stages were measured with a SPAD-502(Minolta Camera Co.,Ltd.,Osaka,Japan)at the jointing and grain-filling stages,respectively.Maize plants with the same growth status were selected,and each treatment had 6–8 replicates.The values measured at the base,middle,and tip of the leaf were averaged to represent the SPAD value of the leaf.

In 2018,LAI values were measured with a LAI-2200C canopy analyzer(LI-COR Biosciences,Lincoln,NE,USA).Five equidistant points were selected between the two rows of large ridges(75 cm),and LAI values were measured 20 cm above the ground.To measure changes specifically in leaves,in 2019,green leaf area(LA)per plant and LAI were measured using the following formula[35]:

where L is leaf length;W is maximum leaf width;n indexes leaf number,and i=1,2,3...;PD is plant density(75,000 plants ha-1).

Electrical conductivity(EC)of leaves was measured using an electrical conductivity meter(Model DDSJ-308A,Shanghai REX Instrument Factory,Shanghai,China).Samples of about 0.5 g of tissue were taken from the middle position of the ear leaf using a hole punch.The samples were immediately soaked in 6 mL distilled water in a 10 mL tube.After 10 h in the dark,the initial EC was measured(C1).The tubes were then boiled in a water bath for 15 min and the EC was measured(C2)after cooling to room temperature.EC was calculated as.

2.3.3.Endogenous hormone content

Owing to the difficulty of sampling and the slow progress of soil drought under field conditions,the samples used for plant hormone analysis were planted in a greenhouse with a controlled climate.Seedlings of XY335 and FM985 were grown in a box(10×10×8 cm)filled with soil.Soil moisture was controlled by weighing at 9:00 AM every day.COR was sprayed at the V3 stage,followed by seven days of drought treatment,with well-watered treatment as control.For each treatment,50 mg leaves with four replicates were sampled for determination of abscisic acid(ABA),indole-3-acetic acid(IAA),salicylic acid(SA),and JA by highperformance liquid chromatography(HPLC).Working solutions of internal standards and extraction solvent were added to the samples in sequence and centrifuged at 14,000 r min-1for 5 min at 4 °C.Then 1.2 mL of the solvent from the lower phase was extracted and concentrated with a nitrogen evaporator.For analysis,samples were redissolved in methanol.The last step of the LCMS was injection into a Q Exactive Mass Spectrometer(Thermo Fisher Scientific,Waltham,MA,USA)at the Biological Mass Spectrometry Laboratory of China Agricultural University.

2.3.4.Root sampling

Under field conditions,maize roots were deeply embedded in the soil,and it was difficult to separate the complete root system from the soil.Consequently,a rigorous method was used to extract the roots below the soil surface.Maize surface roots were distributed mainly within 15 cm from the center of the maize plant under field conditions.Accordingly,a column of soil with radius 15 cm and height 25 cm centered on each maize plant was withdrawn.The soil was first washed away and the roots left behind were photographed for measurement.Analysis of the images was performed at http://dirt.cyverse.org/?q=welcome(DIRT)[36].Briefly,the root images were initially processed,including cutting the excess and correcting the orientation,and uploaded to the DIRT site.The appropriate parameters were selected,including the Scale Marker setting of 25.00 mm and the Masking Threshold setting of 5.00.The system then automatically analyzed the images and output statistics.The indicators analyzed included projected root area(number of pixels belonging to the root,as in general image analysis of roots),average root density(ratio of foreground to background pixels within the root shape)and maximum width of root system(maximum width of root system measured horizontally from the first to the last foreground pixel).After root system phenotype acquisition,roots were cut from the stems and placed in an oven for drying at 85°C to a constant weight.The dry weight of the surface roots was measured with a precision balance.

Root-bleeding sap collection followed Guan et al.[17]with modification.Briefly,maize stems were horizontally cut 10 cm above the soil surface.The wound was quickly covered with a wad of absorbent cotton.Preliminary experiments had shown that the amount of absorbent cotton was sufficient to absorb all the bleeding sap during the measurement period.The cotton was wrapped in plastic film and tied tightly to the stem.The cotton was placed at 8:00 AM and collected at 2:00 PM.Net rootbleeding sap was determined as the weight change over the collection period.

2.3.5.Asi

The arrivals of the anthesis and silking stages were defined when respectively the male and female tassels reached 2 cm in length.The arrival of 50% of the plot population at the stage was the time used for calculation.All plots were observed and recorded at 8:00 AM,2:00 PM,and 8:00 PM every day.Each treatment had four repetitions.

2.3.6.Grain yield and yield components

At the R6 stage,double rows of maize ears with 7 m length(8.4 m2)were manually harvested in each plot.The numbers of plants,ears,and earless plants were recorded.Plants with no ears or＜10 kernels per ear were considered earless plants.Ten representative ears were selected in each plot by a weighing method.Ear length,ear diameter,bare tip length,kernel number per row and 1000-kernel weight(TKW)were determined from the representative ears.TKW was determined by measuring 300 kernels(4 replicates)after drying at 85°C to constant weight.Grain number per ear was obtained by multiplying the number of rows per ear by the number of kernels per row.The final grain yield was standardized to 14% moisture.

2.4.Statistical analysis

Data recording and processing were conducted with Excel 2016(Microsoft Corp.,Redmond,WA,USA).Multiple comparisons of root parameters,leaf parameters,plant hormones,aboveground biomass,yield and yield-related parameters were tested by the LSD test at the P＜0.05 level with SPSS 22.0(SPSS Inc.,Chicago,IL,USA).Tests for normality and homogeneity of variance were performed before the multiple comparison test.The interaction effects between water,genotypes and plant growth regulators were analyzed using GLM in SPSS.Pearson’s correlations between parameters were calculated with SPSS.Path analysis of projected root area,root dry weight,leaf area index,dry matter accumulation,and yield was performed with SPSS.OriginPro 2021 software(OriginLab Co.,Northampton,MA,USA)was used for drawing figures.

3.Results

3.1.Meteorological drought revealed by precipitation and soil moisture

Precipitation in the west of Jilin province in the 20 years from 2000 to 2019 was relatively low(Fig.1B).The year 2001 received the lowest precipitation of 220.3 mm,while 2012 received the highest annual precipitation of 686 mm.There were nine years with precipitation below 400 mm and 16 years with precipitation below 500 mm,in line with a proverb‘‘Nine years of drought in ten years”in the region.The temperature change remained relatively constant in the 20 years,with the highest mean temperature of 11.9°C,lowest mean temperature of 1.2°C,and annual mean temperature of 6.2°C.The mean precipitation during the maize growing season(from May to October)was 420.9 mm in 2018,39.4 mm higher than the 20-year mean(20-yravg;Fig.1C).The precipitation in 2019,a wet year,increased by 143 mm compared with 20-yravg,and the increase in rainfall was concentrated in July.Rainfall events dropped 45.4 mm on July 5 and 35.2 mm on July 23,causing severe surface runoff.Fig.1D shows the mean soil water content(SWC)of the soil layer at 0–80 cm at several growth stages.Irrigation provided 60 mm in 2018 and 70 mm in 2019 to maintain a drought-free environment in the control area.In the rainfed plots,the mean SWC of the 0–80 cm soil layer was 50%–59%in 2018 and 48%–60% in 2019,causing moderate(40%＜SWC≤50%)and mild(50%＜SWC≤60%)drought conditions(GB/T 20481–2006,https://zwgk.cma.gov.cn/zfxxgk/gknr/flfgbz/bz/202102/t20210210_2719989.html).Drought occurred mainly in the seedling and grainfilling stages in the two years.

3.2.Root development

Treatment with 1 μmol L-1COR markedly increased surface root density and size under drought conditions (Figs.2A, S1).In XY335,1 μmol L-1COR treatment increased(P＜0.01)rootbleeding sap by respectively 27.5%and 14.0%compared to the control in 2018 and 2019,and 10 μmol L-1COR treatment resulted in respective increases(P＜0.001)of 43.5%and 21.6%in root-bleeding sap(Fig.2B).In the drought-resistant cultivar FM985,1 and 10 μmol L-1COR treatments increased(P＜0.05)root-bleeding sap by respectively 79.2% and 58.0%,in 2019.All the surface roots were cut from the stems and dried to constant weight.The surface root dry matter of FM985 was 8.0% greater than that of XY335 on average during the two-year experiment(Fig.2C;Table S2).Under drought conditions,root dry matter of XY335 and FM985 with 1 μmol L-1COR treatment increased(P＜0.01)by respectively 16.0%and 13.1%(2018)and 12.9%and 8.4%(2019),compared with the control.There was an interaction(P＜0.001)between maize genotypes and COR treatments,such that COR increased surface root dry matter.

There was no significant interaction effect between maize genotype and COR treatment with respect to projected root area(PRA),average root density(ARD)and maximum width of root system(MW;Fig.2D–F;Table S2).However,COR increased these three root indicators(P＜0.01).In 2018,PRA and ARD of XY335 treated with 1 μmol L-1COR increased(P＜0.05)by 22.2% and 44.7%,respectively,and ARD and MW of FM985 increased by 85.1%(P＜0.001)and 42.1%(P＜0.05),respectively,compared with control.In 2019,1 μmol L-1COR increased(P＜0.05)the ARD of XY335 and FM985 by 21.0% and 56.7%,respectively,compared with control.

3.3.Leaf characteristics

Under drought conditions,LAI of XY335 and FM985 decreased by respectively 4.8% and 4.5%(2018)and 25.2% and 13.8%(2019)in comparison with well-watered plants(Fig.3A).In 2018,COR had no significant effect on LAI under well-watered conditions,but it alleviated the inhibitory effect of drought on LAI of both maize genotypes.Compared with the control,1 μmol L-1COR increased the LAI of XY335 and FM985 by 11.5%(P＜0.01)and 6.8%,respectively,under drought conditions.In agreement with the trend of LAI in 2018,leaf LAI markedly decreased under drought conditions in 2019.LAI showed a trend of first increasing and then decreasing under drought conditions with the increase of COR concentration.Treatment with 1 μmol L-1COR increased the LAI of XY335 and FM985 by 19.7%(P＜0.01)and 3.7%respectively,compared with the control.

Fig.2.Root phenotype of maize surface roots in the top 25 cm of soil in 2018 and root system parameters under rainfed conditions in 2018 and 2019.(A)Images are processed by and exported from http://dirt.cyverse.org/?q=welcome.White disk diameters represent 25 mm.Representative photographs are from three individual plants of each treatment.(B)The root bleeding sap of two maize genotypes collected for 6 h.(C)All surface roots were cut from the stems and dried to constant weight.The traits of the surface root system were analyzed and quantified as projected root area(D),average root density(E)and maximum width of root system(F).Vertical bars indicate means±standard deviation with three replicates.Different letters within the same moisture condition indicate differences at P＜0.05 by LSD test.Projected root area,number of pixels belonging to the root(as in general image analysis of roots),Average root density,ratio of foreground to background pixels within the root shape.Maximum width of root system,maximum width of root system measured horizontally from the first to the last foreground pixel.A1,0.1 μmol L-1 COR;A2,1 μmol L-1 COR;A3,10 μmol L-1 COR.XY335,Xianyu 335;FM985,Fumin 985.

Fig.3.Effect of COR on leaf area index(LAI),chlorophyll content(SPAD),and electrical conductivity(EC)of two maize genotypes in 2018 and 2019.(A)COR increased leaf growth and leaf area index under drought conditions.Vertical bars represent means±standard deviation with six replicates.COR alleviated drought inhibition of ear leaf growth(B),and increased the chlorophyll content of leaves at two growth stages(C,D).Each treatment for leaf area had six replicates and each treatment for SPAD had eight replicates.(E)Effect of COR on EC under drought conditions.Vertical bars represent means±standard deviation with four replicates.Different letters within the same moisture condition indicate differences at P＜0.05 by LSD test.*,P＜0.05;**,P＜0.01;***,P＜0.001.ns,not significant.A1,0.1 μmol L-1 COR;A2,μmol L-1 COR;A3,10 μmol L-1 COR.XY335,Xianyu 335;FM985,Fumin 985.

Green leaf area(LA)per plant was measured to explain the changes specifically in leaves in 2019.Leaves closest to the maize ear contribute most to yield.Drought stress caused a marked decrease in LA.Under drought conditions,the LAs of the 12th,13th,and 14th leaves were reduced by 14.8%,13.7% and 11.3%(XY335),1.9%,3.2%and 2.0%(FM985),respectively,compared with well-watered plants(Fig.3B).Compared with control,1 μmol L-1COR increased the LAs of the 12th,13th,and 14th leaves by 16.5%(P＜0.01),12.3%(P＜0.05),12.1%(P＜0.05;XY335),and 5.4%(P＜0.001),5.2%(P＜0.05)and 4.0%(FM985),respectively.

Compared with control under well-watered conditions,SPAD decreased under drought conditions.The SPAD values of the 9th leaf(at jointing stage)of XY335 and FM985 under drought conditions were reduced by 6.1% and 8.2%(2018),3.4% and 4.4%(2019)(Fig.3C),respectively,compared with well-watered plants,and those of the ear leaf decreased by 4.8% and 3.5%(2018),6.3% and 5.1%(2019)(Fig.3D).COR increased the SPAD values of leaves under drought conditions.Compared with control under drought conditions,1 μmol L-1COR increased the SPAD values of the 9th leaf(at jointing stage)of XY335 and FM985 by 4.9%(P＜0.05)and 7.5%(P＜0.05;2018),4.4%(P＜0.001)and 4.5%(P＜0.01;2019),respectively,and those of the ear leaf increased by 6.7%(P＜0.001)and 4.4%(P＜0.05;2018),and 6.2%(P＜0.001)and 5.8%(P＜0.001;2019),respectively.The lower the EC value,the less was the drought damage to leaves.Under drought conditions,1 μmol L-1COR reduced the EC of XY335 and FM985 by 13.8%(P＜0.01)and 6.1%(P＜0.05;2018),11.9%(P＜0.05)and 6.1%(2019),respectively,compared with control(Fig.3E).

3.4.Changes in endogenous plant hormones

Under well-watered conditions,1 μmol L-1COR increased(P＜0.01)the ABA contents of XY335 and FM985 by 26.0 % and 32.1%,respectively(Fig.4A).In agreement with the trend in ABA content,COR application also significantly increased the contents of JA and SA under well-watered conditions(Fig.4B,C).Plants can adapt to drought stress by altering the homeostasis of endogenous hormones.Drought stress significantly increased the contents of ABA,JA and SA by 156.4%,63.8% and 25.8%,respectively,compared with control under well-watered conditions.COR application also significantly increased the contents of ABA,JA,and SA under drought conditions.Compared with the drought control,1 μmol L-1COR increased(P＜0.01)the ABA contents of XY335 and FM985 by 28.2% and 19.8%,JA contents by 41.1% and 27.3%,and SA contents by 21.8%and 16.8%,respectively.Drought stress inhibited plant growth and caused a dramatic decrease in IAA content(Fig.4D).COR application reduced the IAA content in XY335 and FM985 under both drought and well-watered conditions.Under drought conditions,1 μmol L-1COR reduced(P＜0.001)the IAA contents of XY335 and FM985 by 56.4% and 49.5%,respectively.

3.5.Aboveground growth

Drought inhibited aboveground growth,resulting in the decrease of dry matter accumulation and plant height.Under drought conditions,the plant height of XY335 and FM985 decreased by 2.2% and 4.0%(2018),4.6% and 7.3%(2019),respectively,compared with irrigated plants(Fig.5A).COR mitigated the inhibitory effect of drought.After 1 μmol L-1COR treatment,the plant heights of XY335 and FM985 were increased by 3.0%(P＜0.05)and 1.2%(2018),3.4%(P＜0.01)and 4.8%(P＜0.05)(2019),respectively,compared with drought control.

Fig.4.Effect of COR on endogenous hormone contents in 2019.The contents of four endogenous hormones responsive to drought stress were determined by HPLC:abscisic acid(A),jasmonic acid(B),salicylic acid(C),and indole-3-acetic acid(D).Vertical bars represent means±standard deviation with four replicates.Different letters within the same moisture condition indicate differences at P＜0.05 by LSD test.*,P＜0.05;**,P＜0.01;***,P＜0.001.ns,not significant.A1,0.1 μmol L-1 COR;A2,1 μmol L-1 COR;A3,10 μmol L-1 COR.XY335,Xianyu 335;FM985,Fumin 985.

Fig.5.Effect of COR on plant height(A),dry weight accumulation(B),and ASI(C)of two maize genotypes in 2018 and 2019.Vertical bars represent means±standard deviation with eight replicates for plant height,three replicates for dry matter,and at least three replicates for ASI.Different letters within the same moisture condition indicate differences at P＜0.05 by LSD test.*,P＜0.05;**,P＜0.01;***,P＜0.001.ns,not significant.A1,0.1 μmol L-1 COR;A2,1 μmol L-1 COR;A3,10 μmol L-1 COR.XY335,Xianyu 335;FM985,Fumin 985.

Aboveground dry matter accumulation showed a trend similar to that of plant height(Fig.5B).Drought stress reduced the dry matter accumulation of XY335 and FM985 by 4.1% and 3.5%(2018),10.8% and 2.5%(2019),respectively,compared with the irrigated control.Under drought conditions,0.1 and 1 μmol L-1COR increaseed dry matter accumulation in both maize genotypes.Treatment with 1 μmol L-1COR increased(P＜0.05)the dry matter accumulation of XY335 by 15.9%and 11.3%compared with control in 2018 and 2019,respectively,indicating that 1 μmol L-1COR increased drought resistance of maize by promoting aboveground growth.

3.6.Asi

Drought increased anthesis-to-silking interval(ASI),and asynchronicity in ASI leads to yield losses.As shown in Fig.5C,the ASI of a variety was usually about 24 h(1 d)under adequate moisture conditions.However,under drought conditions,ASI of XY335 and FM985 increased by 28 h(127.3%)and 18 h(52.9%)(2018),30 h(83.3%)and 24 h(66.7%)(2019),respectively,compared with the well-watered control.COR significantly reduced ASI under drought conditions.Treatment with 1 μmol L-1COR reduced ASI of XY335 and FM985 by 16.0%(P＜0.05)and 30.8%(P＜0.05)(2018),54.5%(P＜0.01)and 40.0%(P＜0.05)(2019),respectively.

3.7.Grain yields and yield components

Irrigation and COR treatments affected yield(P＜0.001)(Table 1).The mean yields of XY335 and FM985 under irrigation increased(P＜0.001)by 3.4% and 9.6%(2018),5.1% and 3.8%(2019)respectively,compared with the mean yield under drought conditions.COR increased(P＜0.001)the mean yield of XY335 and FM985 by 4.4% and 7.4%(2018),5.0% and 3.6%,respectively,compared with the control.Among the three concentrations of COR treatments,1 μmol L-1COR showed the greatest effect.Under drought conditions,1 μmol L-1COR increased the yield of XY335 and FM985 by 8.4%(P＜0.05)and 17.9%(P＜0.001)(2018),7.4%(P＜0.001)and 5.4%(P＜0.001)(2019),respectively,compared with the drought control.

?

Yield components and maize ear traits were affected by irrigation,genotype,and plant growth regulator(PGR)treatment.In 2018,irrigation and maize genotype showed an effect(P＜0.001)on ear length,ear diameter,bare tip length,kernel number per row,grain number per ear,and TKW.PGRs significantly increased ear length and kernel numbers per row.No interaction was observed between irrigation and PGR.However,kernel number per row was affected by maize genotype(P＜0.001)and PGR(P＜0.01),and their two-way interactions(P＜0.01).Yield components and maize ear traits in 2019 showed a trend similar to those in 2018.In general,irrigation and maize genotype influenced(P＜0.01)yield components and maize ear traits except for ear diameter and TKW.PGRs increased(P＜0.01)ear length,ear diameter,kernel number per row,and decreased(P＜0.05)bare tip length.Specifically,1 μmol L-1COR decreased the bare tip length of XY335 and FM985 by 14.0%(P＜0.01)and 24.0%(P＜0.05)(2018),10.3%(P＜0.01)and 7.3%(2019),respectively,thereby increasing the kernel numbers per row of XY335 and FM985 by 8.0%(P＜0.01)and 5.5%(P＜0.001)(2018),4.7%(P＜0.01)and 3.9%(P＜0.01)(2019)respectively,compared with control under drought conditions,which resulting in an increment in yield.

Fig.6.Correlation and path coefficients between grain yields and PRA,RDW,LAI,DW in 2018 and 2019.Red lines represent correlation coefficients(left)and green lines represent path coefficients(right).The direction of the arrow indicates a causal relationship between indexes.Values on solid lines are direct path coefficients.Values on dotted lines are indirect path coefficients.The data used for the regression analysis are the mean values of the two maize genotypes.*,P＜0.05;**,P＜0.01;***,P＜0.001.ns,not significant.PRA,projected root area;RDW,root dry weight;LAI,leaf area index;DW,dry matter accumulation.

3.8.Correlation and path analysis

Regression analysis(Fig.6A)shows the relationship between grain yields and root system parameters.Root dry weight(RDW)and projected root area(PRA)increased final grain yields.The coefficients of determination were 0.566(P＜0.001)and 0.446(P＜0.01),respectively,indicating that root system influenced in drought resistance.There were significant positive correlations among RDW,PRA,leaf area index(LAI)and dry matter accumulation(DW).The highest determination coefficients were observed between PRA and LAI(0.791),followed by RDW and DW(0.679),indicating that LAI and DW were affected mainly by the change in the root system.

Path analysis was performed to evaluate the magnitude and significance of causal connections between grain yield and PRA,RDW,LAI,and DW(Fig.6B).There were significant positive correlations between these four indicators and grain yield,as well as the four indicators.LAI showed the greatest and most positive direct effect on grain yield,indicating that leaves were the main factors contributing to greater drought resistance in this study.However,the increase in LAI was caused mainly by the change in root system,as indicated by PRA and RDW.RDW and PRA had little direct effect on yield(based on direct path coefficient values of 0.230 and-0.007,respectively),but increased maize drought resistance mainly by promoting LAI(based on indirect path coefficient values of 0.296 and 0.392,respectively).DW also contributed to drought resistance,and RDW and PRA increased grain yields by promoting DW(based on indirect path coefficient values of 0.230 and 0.162,respectively).These results suggested that PRA and RDW increased drought resistance of maize mainly through LAI and DW.

4.Discussion

4.1.COR promoted root and leaf development under drought conditions

Drought stress is a complex trait controlled by multiple factors.There are many ways to improve the drought resistance of maize,and the most effective way is to promote the development of the root system[37].The maize root system absorbs water from the soil and plays an important role in improving drought resistance[38].COR is a structural and functional analog of JA that has effects similar to those of JA[30].High concentrations of COR and JA inhibited root elongation but promoted root branching[28].Pretreatment of chickpea with COR promoted root growth by increasing the activity of root antioxidant enzymes under drought and heat stress conditions[39].Foliar-applied COR in soybean increased root dry weight by 37%[40].In cotton,COR was likely associated with IAA transport to increase lateral root growth[31].In the present study,1 μmol L-1COR treatment increased root dry weight,projected root area(PRA),and average root density(ARD)of surface roots in the top 25 cm of soil under drought conditions(Figs.2,S1;Table S2).The improvement of the root system under drought conditions helped plants absorb more water from the soil[38].Methyl jasmonate treated rice plants showed reduced stomatal opening and increased root bleeding rate[41].As in previous studies,1 μmol L-1COR treatment increased the root bleeding sap and thus improved maize drought resistance(Fig.2B).

As a source organ,leaves are the main place for photosynthesis[42].Drought stress has multiple effects on leaf development,one of which is inhibiting leaf growth[43].In agreement with previous studies,this study revealed that the LAI of XY335 and FM985 decreased by 17.9%and 10.4%under drought conditions.COR protected the leaves of maize seedlings and increased leaf fresh weight and relative leaf water content under drought conditions[44].In the present study,1 μmol L-1COR increased the LAI of two genotypes(Fig.3A).The 12th,13th(ear leaf)and 14th leaves are key leaves for grain yield formation[45].Treatment with 1 μmol L-1COR increased the LA of 12th,13th,and 14th leaves(Fig.3B).Drought stress caused accumulation of peroxide in plant leaves,resulting in cell membrane damage and thus affecting normal leaf development[46].Damage to the cell membrane resulted in increasing EC in plants[47].COR alleviated leaf damage from drought stress and increased chlorophyll content and photosynthetic ability[48–50].In this study,1 μmol L-1COR increased chlorophyll content and decreased EC in leaves of two maize genotypes under drought conditions(Fig.3C,D,E).

4.2.Effect of COR on plant hormones,plant height,dry matter accumulation,and ASI

Dry matter accumulation in plants decreases with increase in drought intensity and duration[8,51].In this study,drought stress reduced the dry matter accumulation of XY335 and FM985 by 7.7%and 3.0%compared with the irrigated control(Fig.5B).COR alleviated inhibition maize growth by drought stress[44,52].After 1 μmol L-1COR treatment,dry matter accumulation of XY335 and FM985 increased under drought conditions.Drought stress suppressed plant height and caused a dwarf phenotype[53].In our study,1 μmol L-1COR alleviated inhibition by drought stress and increased the plant height of two maize genotypes(Fig.5A).

Drought stress changed phytohormonal homeostasis in plants under osmotic stress[54].Generally,stress-responsive hormones such as ABA,JA,and SA are induced by drought stress and positively regulate the drought resistance of plants[55].In our study,1 μmol L-1COR increased the contents of ABA,JA and SA in two maize genotypes under both irrigation and drought conditions(Fig.4).Plants can escape the effects of drought stress by slowing their growth[13].Thus,reducing the concentration of growthpromoting phytohormones such as gibberellin and IAA in leaves under drought conditions help to improve plant drought resistance[56,57].On the whole,the contents of ABA,JA and SA increased,while the contents of IAA,which promotes plant growth,decreased under drought conditions,indicating that COR may increase drought resistance by inhibiting shoot growth and promoting root growth in the early stage.

ASI is the key indicator reflecting drought resistance of maize genotypes[23].Generally,drought stress had little effect on tasseling,but it could delay the date of silking,and asynchronicity between anthesis and silking caused grain sterility[58].In our study,1 μmol L-1COR reduced ASI of XY335 and FM985 by 37.9% and 35.7%,respectively(Fig.5C).

4.3.COR increased grain yields under drought conditions

Yield is the ultimate goal of agricultural production.Increased dry matter accumulation and shortening of ASI contributed to increases in grain yield under drought conditions[59].In this study,COR increased maize grain yield under well-watered conditions(Table 1).Under drought conditions,1 μmol L-1COR increased the yield of XY335 and FM985 by 7.9%and 11.0%,respectively.This increase in grain yield was attributed mainly to a decrease in bare tip length and an increase of kernel numbers per row,in agreement with the findings of Dong et al.[60].Treatment with 1 μmol L-1COR reduced bare tip length by 12.8% and increased kernel number per row by 5.5%.This difference could be explained by a reduction in ASI,increasing grain pollination[61].Increase in grain yield was associated with increased kernel numbers and TKW[62].In this study,COR increased kernel number and TKW by 3.6% and 3.7%,respectively,under drought conditions.

Grain yield is determined largely by RDW and PRA(Fig.6).The plant root is an important organ for the absorption of soil water[63].Chu et al.[64]reported that larger root biomass improved drought resistance of plants.In our study,COR increased RDW and PRA.Regression analysis showed that root indexes were significantly positively correlated with LAI and DW,indicating that improvement of the root system increased LAI and DW under drought conditions.Among the indicators,LAI was the most highly correlated with grain yield,and the correlation coefficient was 0.784,indicating that the leaf is an organ greatly affected by drought stress,in agreement with Feng et al.[65].In this study,COR increased the LAI of maize genotypes under drought conditions,in agreement with the previously reported[49]increase by COR of leaf growth under drought stress conditions.DM was significantly(r=0.773,P＜0.001)correlated with grain yield under drought conditions.High aboveground biomass at maturity increased grain yield[61].Our results showed that 1 μmol L-1COR increased the DW of two maize genotypes under drought conditions.The explanatory power of the error term of grain yield variation that could not be explained by these four variables was 0.52.Accordingly,further path analysis was performed to assign path coefficients between grain yield and 11 biological traits(Table S3).The results showed that RDW and PRA could also increase drought resistance of maize by increasing chlorophyll content.Significant correlations were observed between grain yields and chlorophyll content,LAI,and dry matter accumulation(Fig.S2),indicating that COR could strengthen seedlings by increasing leaf growth under normal water conditions.Collectively,PRA and RDW increased drought resistance of maize mainly through LAI and DW.

5.Conclusions

Frequent meteorological drought was verified by analysis of precipitation over the past 20 years and soil moisture in Northeast China.COR application at the seedling stage increased surface root density and size,including root dry matter accumulation,projected root area,average root density,and root bleeding sap.LAI and SPAD were severely inhibited by drought stress.Treatment with 1 μmol L-1COR increased leaf area and SPAD values and reduced EC under drought conditions,a finding attributed to the improvement of the root system.Treatment with 1 μmol L-1COR increased ABA,JA,and SA contents and reduced the contents of endogenous IAA under drought conditions,increasing maize drought resistance.COR application reduced ASI,increasing final grain yields under drought conditions.In summary,application of 1 μmol L-1COR increased maize drought resistance of maize mainly by increasing root dry weight and root area.

CRediT authorship contribution statement

Yuling Guo:Writing–original draft.Guanmin Huang:Methodology.Qing Guo:Resources.Chuanxi Peng:Investigation.Yingru Liu:Investigation.Mingcai Zhang:Supervision.Zhaohu Li:Validation.Yuyi Zhou:Funding acquisition.Liusheng Duan:Writing–review & editing.

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 research was funded by National Key Research and Development Program of China(2017YFD0300405-2).

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2022.05.005.