Alyne Oliveir Lvinsky*，Pulo Césr Mglh?esRoniel Gerlo ávilb，Mrin Melo Dinizc，Thigo Corrê e Souz

aEmbrapa Milho e Sorgo，Sete Lagoas，MG，CEP：35701-970，BrazilbUniversidade Federal de Lavras，Lavras，MG，CEP 37200-000，BrazilcUniversidade Federal de S?o Jo?o Del Rey，Sete Lagoas，MG，CEP 35701-970，BrazildUniversidade Federal de Alfenas，Alfenas，MG，CEP 37130-000，Brazil

?

Partitioningbetweenprimaryandsecondarymetabolism of carbon allocated to roots in four maize genotypes under water deficit and its effects on productivity

Alyne Oliveira Lavinskya，*，Paulo César Magalh?esa，Roniel Geraldo ávilaa，b，Mariana Melo Diniza，c，Thiago Corrêa de Souzad

aEmbrapa Milho e Sorgo，Sete Lagoas，MG，CEP：35701-970，Brazil
bUniversidade Federal de Lavras，Lavras，MG，CEP 37200-000，Brazil
cUniversidade Federal de S?o Jo?o Del Rey，Sete Lagoas，MG，CEP 35701-970，BrazildUniversidade Federal de Alfenas，Alfenas，MG，CEP 37130-000，Brazil

A R T I C L E I N F O

Article history：

Receivedinrevisedform26March2015 Accepted 1 June 2015

Available online 6 June 2015

Starch

Sugars

Lignin

Phenols

WinRHIZO

A B S T R A C T

Plantsmayrespondtodroughtbyalteringbiomassallocationtoshootsandrootsorbychanging the metabolic activities in these organs.To determine how drought changes the partitioning of carbon allocated to growth and secondary metabolism in maize roots and how it affects photosynthesis（A）and productivity in maize，we evaluated leaf gas exchange，yield componentes，root morphology，and primary and secondary metabolites including total soluble sugars（TSS），starch（S），phenolics（PHE），and lignin（LIG）.Data were collected from pot-grownplantsoffourmaizegenotypes:BRS1010and2B710（sensitivegenotypes）andDKB390 and BRS1055（tolerant genotypes）under two soil water tensions:field capacity（FC，?18 kPa）and waterdeficit（WD，?138 kPa）.WDwasappliedatthepre-floweringstagefor12 daysandthenthe water supply was restored and maintained at optimum levels until the end of the cycle.For genotypeBRS1055underFC，thegreatestAdidnotresultingreatergrainbiomass（DGB）because the accumulated photoassimilates had already filled the cells，and thus the excessive TSS synthesized in leaves was allocated to roots in large amounts.However，the sharp decrease in A caused by WD imposition in this genotype did not affect the influx pressure of leaf TSS，which was due largely to conversion of primary metabolites to PHE compounds to increase the length of fine roots.In leaves of DKB390 under WD，both S and TSS were reduced，whereas PHE were increased to prevent excessive water loss and xylem cavitation.Under WD，both BRS1010 and 2B710 genotypesdisplayed reduced allocation ofbiomass to shootsand roots and LIGcontentin leaves，aswellaslowerAandDGBvalues.InBRS1010thisresponsewascoupledtoSdecreasein leaves and TSS increase in roots，whereas in 2B710 there was a concomitant S increase in roots. ?2015 Crop Science Society of China and Institute of Crop Science，CAAS.Production and

hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license （http://creativecommons.org/licenses/by-nc-nd/4.0/）.

1.Introduction

The threat of global climate change is causing concern in agriculture，given that climatic factors essential for crop development will be severely affected，reducing the productionand quality of food［1］.Given that most crops are produced in tropical regions characterized by low water availability and the occurrence of frequent and severe drought［2，3］，research to identify the priority adaptive needs for investment in agriculture has become relevanttoreducethe severityof thepredicted impactsonproduction ofcrops［4］such asmaize（ZeamaysL.）in tropical regions.

Maize is an important cereal for human food and animal feed，and the severe WD experienced in Brazil has reduced the production and supply of this grain，affecting mainly small farmers and agro-industries in the northeast that use maize for animal feeding［5］.The main objectives of selection for increasing productivity in maize under drought include staygreen，anthesis—silking interval，and number of ears per plant，which are advantageous characteristics under WD. Thus，studies for phenotypic characterization and evaluation of the productivity of maize genotypes under WD are initial steps in elucidating mechanisms involved in the physiological differential behavior of some maize genotypes under this stress，and in providing indirect selection criteria for drought tolerance［6］.In this context，the search for new descriptors that reflect productivity，such as the characterization of roots is very important［7］.

According to the theory of functional balance［8］，plants increase the allocation of biomass to shoots if carbon gain is affected by limited resources aboveground，such as light and CO2.Similarly，plants increase biomass allocation to roots in the presence of low levels of belowground resources，such as water and nutrients［9］.By altering their structure to increase the angle of root growth，plants using drought-avoidance strategies can exploit water in deeper layers of the soil，allowing marked improvement in grain yield［10］.Although the flexibility of biomass allocation to this organ of capture and storage of resources constitutes a key point in the determination of physiological changes in plants in environments with low soil water availability［11］，little is known about this relationship［12］.The relationship between carbon allocation and water economy has been studied mainly in aboveground organs，with particular emphasis on leaves. Moreover，plants are likely to respond to drought not only by altering their allocation of biomass to shoots and roots，but also by changing the metabolic activities of these organs［13］.

Primary metabolites in plants，such as sugars，starch，amino acids，and fatty acids，are synthesized mainly in the photosynthetic tissues of shoots，where the concentrations of these metabolites are higher than in roots［13］.Under drought conditions，however，the plasticity of the plants allows a shift to increased synthesis or allocation of several primary metabolites to roots and decreased allocation to shoots［13］. This shift may increase levels of plant secondary metabolites such as PHE and LIG，which are synthesized from primary metabolites and often referred to as compounds associated with stress response［13］.According to the hypothesis of compromise between growth and defense［14—16］，the plant faces an energy competition between synthesizing defense compounds（which incur high metabolic costs for synthesis and storage）or investing their energy into other functions such as growth，maintenance and reproduction.In a pioneer study［16］with 41 tropical species，the energetic compromise between growth and defense was evident.Species with higher growth rates showed leaves with lower concentrations of compounds associated with chemical defense，providing a greater amount of energy to growth.However，these species were more susceptible to defoliation by herbivores，suggesting that a negative relationship between growth rate and cost of synthesis holds.

The goal of this study was to determine how drought changes the partitioning of carbon allocated to growth and secondary metabolism in maize roots and how it affects A and productivity in four maize genotypes under WD.

2.Material and methods

2.1.Growth conditions，plant material，and experimental design

The experiment was conducted in a greenhouse at the National Maize and Sorghum Research Center（19°28′S，44°15″08″W，732 m.a.s.l.），and the plant material consisted of four maize genotypes with contrasting drought tolerance: two tolerant（DKB 390 and BRS1055）and two sensitive（BRS 1010 and 2B710）.The choice of genotypes was based on results of previous field experiments performed by researchers from the breeding program of the National Maize and Sorghum Research Center，which，over the years，has accumulated experience in maize phenotyping for drought tolerance［17］.

Plants were grown in plastic pots containing 20 kg of typical dystrophic Red Latosol soil.The water content in the soil was monitored daily between 9:00 a.m.and 3:00 p.m.，with a moisture sensor（GB Reader N1535；（Measurement Engineering，Australia））installed at the center of each pot with the aid of a screw auger at a depth of 20 cm.These sensors detect the water content in the soil based on electrical resistance and are coupled to digital meters.Water replacement by irrigation was based on the data obtained with the sensor and water was added to reach FC during the period preceding the imposition of the treatments.The water replacement calculations were performed with a spreadsheet and based on a soil water retention curve.In parallel，all necessaryculturalandphytosanitarytreatmentswere performed.

At the pre-flowering growth stage，half of each initial treatment was subjected to WD the other half continued to receive daily irrigation in order to maintain soil moisture close to FC，with a soil water tension of?18 kPa.WD was imposed by daily provision of 50%of the total available water until the soil water tension reached at least?138 kPa.After twelve days under this condition，the leaf water potential was determined with a Scholander-type pressure pump at midday（Ψmd）［18］. Leaf gas exchange and chlorophyll a fluorescence were measured in leaves corresponding to the ear insertion with aninfraredgasanalyzerequippedwithafluorometer （LI-6400-40；LI-COR，USA）［18］.Samples of leaves and roots were collected and stored in liquid nitrogen for determination of the contents of TSS，S，PHE，and LIG.The water supply was then restored and maintained at optimum levels until the end of the cycle.At harvest，the agronomic parameters associated with productivity were analyzed according to the methodology detailed in the“Agronomic parameters”section.Theexperimental unit was the pot containing two plants，with six replications per treatment.

For the statistical analysis，the results were submitted to variance analysis and the means were compared by the Scott—Knott test at 5%probability.

Fig.1-Leaf water potential at midday（Ψmd）in four maize genotypes with contrasting characteristics for drought tolerance grown under different water levels in the soil.Drought-sensitive genotypes：BRS1010 and 2B710；drought-tolerant genotypes：DKB390 and BRS1055；FC：field capacity；WD：water deficit.Means followed by the same letter do not differ statistically.Lowercase letters denote comparisons between genotypes within the same level of water in the soil and uppercase letters indicate comparisons between the water levels in the soil within the same genotype.Scott-Knott test at 5%probability was applied.

2.2.Biochemical analysis

Tissues were macerated in 80%ethanol（v/v），incubated at 70°Cfor90 min，andsubjectedtotwocentrifugations （15，000×g，10 min）.In the insoluble fraction，S was analyzed enzymatically［19］.In the soluble fraction，TSS were determined［20］.The contents of total PHE was quantified by the Prussian Blue method［21］，and LIG by the Klason method［22］.

2.3.Agronomic parameters Leaf number，ear number，leaf area，plant height，ear insertion height，and diameter and length of the ear were quantified ［18］，in six plants per treatment.Leaf area was measured with an area meter（LI-COR）.The plants were then partitioned into roots，stems，leaves，tassel，ears（cob，straw，and grains），and dried in an oven with forced air circulation at 70°C for 72 h. Based on the dry weights of the different parts，the harvest index，number of grains per ear，and weight of 100 grains were estimated［18］.

A group of 50 kernels was subjected to morphometric characterization by measurement of the dimensions of the kernels（length，widthandthickness）usingadigitalpachymeter，with three replications per treatment.These kernels were then soaked overnight in ethylenediamine（10%，w/v）and longitudinally cut with a knife to evaluate possible changes in embryo size，depending on the treatment.Photographs were obtained using a stereoscopic microscope and the ImageJ program was used to calculate the ratio between the areas of the endosperm and the embryo.

For the evaluation of the root system the computerized system WinRHIZO（WinRHIZO Pro，Regent Inc.Instr.，Canada）was used to measure length，diameter，volume and surface area of roots by diameter class，as follows:very fine roots （? less than 0.5 mm），fine roots（＞0.5 ?＜2.0 mm）and thick roots（?＞2.0 mm）［7］.For this analysis，roots were collected from three replicates per treatment.

3.Results

All genotypes showed similar tolerance to leaf dehydration，with reduced leaf water potential at midday（Ψmd）under WD compared to plants maintained in soil moisture at FC （Fig.1）.

Under FC，the A and stomatal conductance of water vapor （gs）were different among genotypes，with higher values for BRS 1055，intermediate for 2B710 and DKB 390，and lower for BRS1010.Even under this condition，the transpiration rates（E）were higher for genotype 2B710，intermediate for BRS 1055，and smaller for BRS 1010 and DKB 390，and the internal concentration of carbon（Ci）unchanged（Table 1）.In turn，irrespective of genotype，there was a significant reduction in A，gsand maximum efficiency of photosystem II（Fv/Fm）in plants exposed to stress caused by WD compared to those for which the soil moisture was maintained close to FC，whereas Civalues increased（Table 1）.Plants of the tolerant genotypes （DKB390 and BRS1055）showed average values of A and gs70.33%and 64.66%higher under WD，respectively，than the susceptible genotypes（2B710 and BRS1010）grown under the same condition.In addition，sensitive genotypes showed lower transpiration rates（E）compared to tolerant ones. Under WD，BRS 1055 showed larger values of Fv/Fm，than the other genotypes，and the BRS1010 genotype showed an A value 7.81 times lower than that of 2B710 but an E only 2.18 times lower，explaining the values of A/E for 2B710 similar to those for DKB 390 and BRS 1055.

The genotype BRS 1055 under FC showed an increase of 21.0%in A，with a 7.6%increase in total biomass（TDB）and 9.8%decrease in DGB compared to DKB 390 under the same condition（Table 2）.Given that these two genotypes showed similar tolerance to WD，it was expected that higher A values would result in higher grain productivity.It is possible that in BRS 1055 under FC the largest A did not result in greater DGB because the cells that accumulate photoassimilates were already completely filled.In addition，the metabolic costsassociated with the highest root dry biomass（DRB）in BRS 1055 compared to the other genotypes offset the gains from A，thus reducing productivity.

Under WD，genotypes DKB 390 and BRS 1055 showed similar values of A and TDB，but the DGB was 28%higher in the DKB 390，resulting in a higher harvest index（HI）than that of BRS 1055（Table 2）.At first sight，the lower HI values in BRS 1055 might be interpreted as a low tolerance for WD. However，compared to the control（under FC）a decrease in HI of only 9.3%occurred for BRS 1055 under WD.The sensitive genotypes BRS 1010 and 2B710 under WD showed reductions of 22.0%and 24.0%in HI，respectively.In fact，plants from BRS 1010 and 2B710 showed reduced leaf area （LA），area of embryo relative to the endosperm （EM:E）and DGB under WD compared to FC，indicating the occurrence of a low photoassimilate flow to the grain in these two maize genotypes under WD，compared to genotypes DKB 390 and BRS 1055.Plants of BRS 1055 have a constitutively higher source capacity per se，associated with a root system specialized for water absorption.Thus large reductions in A had little effect on the availability of photoassimilates in the source under WD，and thus the flow of photoassimilates from source to sink organs（such as in grain filling）was similar to that in plants under FC.

Morphometric analysis of root images previously scanned with WinRHIZO showed no difference between genotypes under FC with respect to length，surface area，and volume in different diameter classes of（Table 3）.However，under WD the BRS1055 genotype showed higher fine root length（FRL），surface area of fine roots（SAFR），total surface area of roots （TSAR），and volume of fine roots（VFR）than did the other genotypes（Table 3）.In addition，plants of this genotype under WD showed greatly increased FRL，surface area of very fine roots（SAVFR），SAFR，TSAR，and FRV compared to those grown under FC，whereas in the genotypes BRS 1010 and DKB 390 these variables did not change.For the sensitive genotype 2B710 under WD，SAVFR，SAFR，and TSAR decreased in comparison with plants of the same genotype under FC （Table 3）.

Plants of genotype 2B710 showed increased root S levels under WD compared to plants of the same genotype under FC，so that under WD this genotype showed significantly higher levels of S relative to the others under the same water condition（Fig.2）.Regardless of the water content in the soil，BRS 1055 had higher levels of TSS in roots than did the other genotypes（Fig.2）.Only plants of BRS 1010 showed increased TSS root levels under WD relative to plants under FC（Fig.2）. There was no significant variation in LIG levels in maize roots with the imposition of WD（Fig.2）.In turn，PHE levels in roots were greatly increased only in BRS1055 exposed to the stress caused by WD（Fig.2）.The concentration of PHE in BRS 1055 under WD was 2.57 times higher than that recorded in plants of this genotype under FC（Fig.2）.In leaves of DKB390 under WD，both S and TSS were reduced，whereas PHE was increased.Under WD，both BRS1010 and 2B710 genotypes showed reduced LIG content in leaves and S decreased in leaves of BRS1010.

Table 1-Gas exchange parameters obtained in four maize genotypes with contrasting characteristics for drought tolerance grown under different water levels in the soil.

4.Discussion

Higher maize yield is not always associated to higher values of A ［23］.In the present study we found that BRS1055 genotype showed higher A values than DKB390，BRS1010，and 2B710 under FC，but did not use the additional photosynthate to increase DGB.It is possible that the largest A did not result in greater DGB because the cells that accumulate photoassimilates were already completely filled，and as a result the excess of TSS synthesized in leaves was allocated and used in large amounts in roots.Grain size and the potentialtoaccumulatephotoassimilatesinmaizeare probably determined by the number and size of endosperm cells［24］.In addition，the metabolic costs associated with the highest DRB in BRS 1055 compared to the other genotypes may contribute to this response.

In maize，stomatal closure is one of the first events that occurs with declining leaf water status，and invariably coincides with reduced A and E［18］.The ameliorative effectof A on tolerant maize genotypes exposed to early-term drought is believed to occur largely via counteracting oxidative stress by modulating ABA-mediated antioxidant enzymes such as catalase（CAT）［25］.However，with long-term drought exposure an A decline becomes even more evident with failure of the ABA-mediated antioxidant defense system，mainly when other mechanisms for counteracting oxidative stress are lacking.In the present study，after long-term drought exposure all genotypes showed severely reduced A values.In fact，a four-to six-day episode reduces maize A to near zero and can reduce yield［26］，as seen mainly in the sensitive genotype BRS1010.

Table 2-Agronomic production in four maize genotypes with contrasting characteristics for drought tolerance grown under different water levels in the soil.

In WD plants of BRS1055，the sharp decrease in A did not affect the influx pressure of leaf TSS through the plasmodesmata in the phloem into kernel cells；the values of DGB and EM:E remained unchanged compared to FC.Under WD，the S，PHE，and LIG values in leaves of this genotype did not change in comparison with FC，and the values of these compounds were lower in leaves than in roots.In plants of BRS1055 under WD only PHE in roots were increased in comparison with FC，suggesting that large TSS primary metabolites in the roots were transglycosylated and used in PHE compounds of secondarymetabolismotherthanLIG，suchasquinic，chlorogenic，cinnamic，ferulic，coumaric，caffeic，and salicylic acids and/or tocopherol（also associated with mechanisms of protection against damage caused by exposure to WD）.In fact，WD plants of BRS1055 showed increases in FRL without change in DRB.The largest amount of water absorbed by the root system is due to fine and very fine roots［27］，and only plants with conservative characteristics in arid environments show improvement in water absorption efficiency for producing very fine roots［28］.

Plants of the DKB390 genotype showed decreased A values under WD compared to FC conditions，and despite their lack of mechanisms for increasing the absorption of water by roots，their yield was maintained.Plants of DKB390 show a smaller distance between the vascular bundles in the leaf blade under WD than under FC and high phenylalanine ammonia lyase（PAL）activity［18］.In the present study，we also showed that leaves of DKB390 under WD showed increased PHE，whereas S and TSS decreased and LIG did not change.We observed neither oxidative damage nor CAT activity increase in WD plants of this genotype（data not shown），suggesting diversion of the H2O2accumulated in the leaf to biosynthesis of PHE compounds other than LIG，to prevent excessive water loss and xylem cavitation［29］.It may be that PAL activity increase［18］is more highly expressed than LIG formation，suggesting formation of PHE compounds ofsecondary metabolism otherthanLIG.HIwasalso correlated with root features such as increases in the formation of aerenchyma［18］.In field studies with maize under WD conditions，genotypes with a higher proportion of aerenchyma in their roots presented better performance （higher root growth and higher biomass of the above-ground part）［30］.A higher proportion of aerenchyma can allow enhanced soil exploration and water intake，since these structures decrease the metabolic cost of growing roots，due to a decrease in the number of cells that are undergoingcellular respiration.However，WD plants from DKB390 genotype showed no significant variation in growth，surface area and root volume and DRB，leading us to believe that the increase in the amount of aerenchyma reported in［18］was used only for the maintenance of the physiological processes in the root，which although beneficial，does not improve water capture in this genotype.

Drought reduced the amount of carbon allocation to shoots and roots，as well as DGB and LIG content in leaves in plants of BRS1010 and 2B710 in comparison with droughttolerant lines.For example，roots of genotype 2B710 under WD showed substantial reductions in SAVFR，SAFR，and TSAR.Considering that the root cap acts in geotropism and protects the root meristem from damage by mucilage secretion at the edges of the cells［31］，it is possible that under WD the area of these roots was reduced，leaving the meristem unprotected.Asaresult，there wasasmallerreplacementofthe damagedcellsoftherootcap bythe quiescentcenter，which led to a restriction in SAVFR，SAFR，and TSAR without affecting DRB.

Table 3-Morphometric characterization of roots in four maize genotypes with contrasting characteristics for drought tolerance grown under different water levels in soil.

LIG occupies the spaces in the cell wall between cellulose，hemicellulose，and pectin，especially in xylem tracheids and vessel elements，creating crosslinkages between polysaccharides that support the cell wall［32］.The mechanical strength of cell walls in leaves of these maize drought-sensitive lines was low，as evidenced by reduced LIG deposition.In fact，a previous study found that WD decreased the level of lignifying enzymes and consequently the LIG level in maize leaves［33］. We also observed an S decrease in leaves and a TSS increase in roots of BRS1010，whereas in 2B710 there was only a concomitant S increase in roots.In leaves，osmotically active compounds have been shown to originate mainly in S degradation，and the same mechanism was suggested to underlie the osmotic adjustment in roots of plants under WD［34］.Our data indicate that drought increases the partitioning of recent assimilates to osmotic adjustment in roots and that S breakdown in leaves contributes to a drought-induced increase of root TSS in BRS1010.The increased S in 2B710 plants under WD relative to FC suggests that either drought was not causing carbon starvation，or if plants were starving，they would not use S to address any carbon imbalance.

Few studies have examined how root traits are associated with A and yield in plants grown under WD［10］.The main hindrance to these studies is the difficulty of root phenotyping in field-grown plants.Several methods have been developed to study root growth under both field and controlled environmental conditions.Under controlled environments，root scanning based on WinRHIZO is one of the most efficient methods，allowing image analysis and examination of root morphological traits［7，35］.A reasonable compromise to avoid the difficulty of studying roots in the field is also offered by growing plants in large pots with soil.Pot experiments allow precise measurement of the amount of water provided to each plant and estimation of the capacity of roots to penetrate a wax layer of high mechanical impedance mimicking a soil hardpan，often the main constraint to access of roots to soil moisture in deeper soil layers［36］.

The results of the present study provide clear evidence that genotypes DKB 390 and BRS1055 are more tolerant to WD than BRS 1010 and 2B710.However，these two tolerant genotypes have different mechanisms for overcoming the stress generated by WD.DKB 390 uses mainly physiological mechanisms at the shoot level for the maintenance of productivity，by minimizing water loss and escaping xylem cavitation，a response that reduces dependence on metabolic adjustments in the root system to increase the absorption of water.BRS 1055 drought tolerance，in contrast，is associated with increase in water uptake or transport.

Fig.2-Starch content（S），total soluble sugars（TSS），lignin（LIG），and phenols（PHE）in roots and leaves of four maize genotypes with contrasting characteristics for drought tolerance grown under two water levels in the soil.Drought-sensitive genotypes：BRS1010 and 2B710；drought-tolerant genotypes：DKB390 and BRS1055；FC：field capacity；WD：water deficit.Means followed by the same letter do not differ statistically.Lowercase letters denote comparisons between genotypes within the same soil water level and uppercase letters indicate comparisons between soil water levels within the same genotype.The Scott-Knott test at 5%probability was applied.

Acknowledgments

This research was supported by the Foundation for Research Assistance of Minas Gerais State，Brazil（FAPEMIG，Grant BPD-00477-13）granted to AOL.

R E F E R E N C E S

［1］IPCC，Summary for policymakers，in:S.Solomon，D.Qin，M. Manning，Z.Chen，M.Marquis，K.B.Averyt，M.Tignor，H.L. Miller（Eds.），Climate Change 2007:The Physical Science Basis，Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate ChangeCambridge University Press，United Kingdom and New York 2007，pp.1—18.

［2］W.E.Easterling，P.K.Aggarwal，P.Batima，L.M.Brander，L. Erda，S.M.Howden，F(xiàn)ood，fiber and forest products，in:M.L. Parry，O.F.Canziani，J.P.Palutikof，P.J.Van der Linden，C.E. Hanson（Eds.），Climate Change 2007:Impacts，Adaptation and Vulnerability，Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate ChangeCambridge University Press，United Kingdom 2007，pp.273—313.

［3］T.N.Tubiello，G.Fischer，Reducing climate change impacts on agriculture:global and regional effects of mitigation 2000—2080，Technol.Forecast.Soc.74（2007）1030—1056.

［4］D.B.Lobell，M.B.Burke，C.Tebaldi，M.D.Mastrandrea，W.P.Falcon，R.L.Naylor，Prioritizing climate change adaptation needs for food security in 2030，Science 319 （2008）607—610.

［5］A.P.Gutierrez，N.L.Engle，E.D.Nys，C.Molejón，E.S.Martins，Drought preparedness in Brazil，Weather Clim.Extreme 3 （2014）95—106.

［6］F.O.M.Dur?es，P.C.Magalh?es，A.C.Oliveira，índice de colheita genético e as possibilidades da genética fisiológica para melhoramento do rendimento do milho，Rev.Bras. Milho Sorgo 1（2002）33—40.

［7］P.C.Magalh?es，T.C.Souza，F(xiàn).R.Cant?o，Early evaluation of root morphology of maize genotypes under phosphorus deficiency，Plant Soil Environ.57（2011）135—138.

［8］R.Brouwer，Distribution of dry matter in the plant，Neth.J. Agric.Sci.10（1962）399—408.

［9］H.Poorter，O.Nagel，The role of biomass allocation in the growth responses of plants to different levels of light，CO2and water:a quantitative review，Aust.J.Plant Physiol.27 （2000）595—607.

［10］Y.Uga，K.Sugimoto，S.Satoshi，S.Ogawa，J.Rane，M.Ishitani，N.Hara，Y.Kitomi，Y.Inukai，K.Ono，N.Kanno，H.Inoue，H. Takehisa，R.Motoyama，Y.Nagamura，J.Wu，T.Matsumoto，T.Takai，K.Okuno，M.Yano，Control of root system architecture by DEEPER ROOTING 1 increases rice yield under drought conditions，Nat.Genet.45（2013）1097—1105.

［11］W.J.Davies，J.Zhang，Root signals and the regulation of growth and development of plants in drying soil，Annu.Rev. Plant Physiol.Plant Mol.Biol.42（1991）55—76.

［12］F.Fort，C.Jouany，P.Cruz，Root and leaf functional trait relations in Poaceae species:implications of differing resource-acquisition strategies，J.Plant Ecol.2（2012）1—9.

［13］S.A.Martin，Disorders of primary metabolites in response to drought may increase the synthesis of natural products for medicinal purposes:South American herbs—a case study，Rec.Pat.Biotechnol.8（2014）36—46.

［14］D.F.Rhoades，R.G.Cates，Toward a general theory of plants antiherbivore chemistry，in:J.W.Wallace，R.L.Mansell（Eds.），Biochemical Interactions Between Plants and Insects:Recent Advances in Phytochemistry，Plenum Press，New York 1976，pp.168—213.

［15］P.D.Coley，J.P.Briant，F(xiàn).S.Chapin，Resource availability and plant antiherbivore defense，Science 230（1985）895—899.

［16］P.D.Coley，Effects of plant growth rate and leaf lifetime on the amount and type of ant—herbivore defense，Oecologia 74 （1988）531—536.

［17］L.J.Guimar?es，F(xiàn).F.Mendes，S.N.Parentoni，P.E.O.Guimar?es，C.A.P.Pacheco，J.R.A.Machado，W.F.Meirelles，A.R.Silva，Desempenho de Híbridos de Milho quanto à Tolerancia à Seca，XXIX Congresso Nacional de Milho e Sorgo，águas de Lindóia—26 a 30 de Agosto de，2012.

［18］T.C.Souza，E.M.Castro，P.C.Magalh?es，L.O.Lino，L.E.T.Alves，P.Albuquerque，Morphophysiology，morphoanatomy，and grain yield under field conditions for two maize hybrids with contrasting response to drought stress，Acta Physiol.Plant.35 （2013）3201—3211.

［19］R.N.Trethewey，P.Geigenberger，K.Riedel，M.R.Hajirezaei，U. Sonnewald，M.Stitt，J.W.Riesmeier，L.Willmitzer，Combined expression of glucokinase and invertase in potato tubers leads to a dramatic reduction in starch accumulation and a stimulation of glycolysis，Plant J.15（1998）109—118.

［20］M.Stitt，R.M.C.Lilley，R.Gerhardt，H.W.Heldt，Metabolite levels in specific cells and subcellular compartments of plant leaves，Methods Enzymol.174（1989）518—552.

［21］R.Budini，D.Tonelli，S.Girotti，Analysis of total phenols using the Prussian blue method，J.Agric.Food Chem.28（1980）1236—1238.

［22］J.L.Gomide，B.J.Demuner，Determina??o do teor de lignina em material lenhoso método Klason modificado，Papel.47 （1986）36—38.

［23］S.P.Long，X.G.Zhu，S.L.Naidu，D.R.Ort，Can improvement in photosynthesis increase crop yields？Plant Cell Environ.29 （2006）315—330.

［24］P.C.Magalh?es，R.Jones，Aumento de fotoassimilados na taxa de crescimento e peso final dos gr?os de milho，Pesq. Agrop.Brasileira 25（1990）1747—1754.

［25］T.C.Souza，P.C.Magalh?es，E.M.Castro，N.P.Carneiro，F(xiàn).A. Padilha，C.C.Gomes-Jr，ABA application to maize hybrids contrasting for drought tolerance:changes in water parameters and in antioxidant enzyme activity，Plant Growth Regul. 73（2014）205—217.

［26］J.Boyer，M.E.Westgate，Grain yields with limited water，J.Exp. Bot.55（2004）2385—2394.

［27］E.F.Coelho，F(xiàn).D.C.Oliveira，E.C.E.Araújo，L.F.L.Vasconcelos，Distribui??o de raízes de laranja‘Pêra'sob sequeiro e irriga??o por microaspers?o em solo arenoso，Pesq.Agrop. Brasileira 37（2002）603—611.

［28］G.Liu，G.T.Frescher，X.Pan，J.H.C.Cornelissen，Y.Li，M.Dong，Coordinated variation in leaf and root traits across multiple spatial scales in Chinese semi-arid and arid ecosystems，New Phytol.188（2010）543—553.

［29］M.T.Tyree，J.S.Sperry，Do woody plants operate near the point of catastrophic xylem dysfunction caused by dynamic water stress？Plant Physiol.88（1988）574—580.

［30］J.Zhu，K.M.Brown，P.J.Lynch，Root cortical aerenchyma improves the drought tolerance of maize（Zea mays L.），Plant Cell Environ.33（2010）740—749.

［31］M.Iijima，T.Higuchi，P.H.Barlow，Contribution of root cap mucilage and presence of an intact root cap in maize（Zea mays）to the reduction of soil mechanical impedance，Ann.Bot.94 （2004）473—477.

［32］N.Bellaloui，A.Mengistu，L.H.S.Zobiole，H.K.Abbas，A. Kassen，Soybean seed phenolics，sugars，and minerals are altered by charcoal rot infection in MG III soybean cultivars，F(xiàn)ood Nutr.Sci.5（2014）1843—1859.

［33］D.Vincent，C.Lapierre，B.Pollet，G.Cornic，L.Negroni，M.Zivy，Water deficits affect caffeate O-methyltransferase，lignification，andrelatedenzymesinmaizeleaves.Aproteomicinvestigation，Plant Physiol.137（2005）949—960.

［34］R.Hasibeder，L.Fuchslueger，A.Richter，M.Bahn，Summer drought alters carbon allocation to roots and root respiration in mountain grassland，New Phytol.3（2015）1117—1127.

［35］C.Wijewardana，M.Hock，B.Henry，K.R.Reddy，Screening corn hybrids for cold tolerance using morphological traits for early-season seeding，Crop Sci.55（2015）851—867.

［36］R.Tuberosa，Phenotyping for drought tolerance of crops in the genomics era，F(xiàn)ront.Physiol.3（2012）347.

*Corresponding author.

E-mail address:alynelavinsky@gmail.com（A.O.Lavinsky）.

Peer review under responsibility of Crop Science Society of China and Institute of Crop Science，CAAS.

http://dx.doi.org/10.1016/j.cj.2015.04.008

2214-5141/?2015 Crop Science Society of China and Institute of Crop Science，CAAS.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http://creativecommons.org/licenses/by-nc-nd/4.0/）.

22 January 2015