Cowpea (Vigna Unguiculata [L.] Walp.) Genotypes Response to Multiple Abiotic Stresses 2010 Journal of Photochemistry and Photobiology B Biology

Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146

Contents lists available at ScienceDirect

Journal of Photochemistry and Photobiology B: Biology

journal homepage: www.elsevier .com/locate / jphotobiol

Cowpea (Vigna unguiculata [L.] Walp.) genotypes response to multipleabiotic stresses

Shardendu K. Singh a, Vijaya Gopal Kakani a,1, Giridara-Kumar Surabhi a,2, K. Raja Reddy a,*

a Department of Plant and Soil Sciences, 117 Dorman Hall, Box 9555, Mississippi State University, Mississippi State, MS 39762, USA

a r t i c l e i n f o a b s t r a c t

Article history:Received 3 March 2010Received in revised form 6 May 2010Accepted 31 May 2010Available online 17 June 2010

Keywords:CO2

PollenResponse indexScreeningTemperatureUltraviolet-B

1011-1344/$ - see front matter � 2010 Elsevier B.V. Adoi:10.1016/j.jphotobiol.2010.05.013

* Corresponding author. Tel.: +1 662 325 9463; faxE-mail address: [email protected] (K.R. Red

1 Current address: Department of Plant and SoUniversity, Stillwater, OK 74078, USA.

2 Current address: Department of Biology, ColoradoCO 80523, USA.

The carbon dioxide concentration [CO2], temperature and ultraviolet B radiation (UVB) are concomitantfactors projected to change in the future environment, and their possible interactions are of significantinterest to agriculture. The objectives of this study were to evaluate interactive effects of atmospheric[CO2], temperature, and UVB radiation on growth, physiology and reproduction of cowpea genotypesand to identify genotypic tolerance to multiple stressors. Six cowpea (Vigna unguiculata [L.] Walp.) geno-types differing in their sites of origin were grown in sunlit, controlled environment chambers. The treat-ments consisted of two levels each of atmospheric [CO2] (360 and 720 lmol mol�1), UVB [0 and10 kJ m�2 d�1) and temperatures [30/22 and 38/30 �C] from 8 days after emergence to maturity. The ame-liorative effects of elevated [CO2] on increased UVB radiation and temperature effects were observed formost of the vegetative and photosynthetic traits but not for pollen production, pollen viability and yieldattributes. The combined stress response index (C-TSRI) derived from vegetative (V-TSRI) and reproduc-tive (R-TSRI) parameters revealed that the genotypes responded negatively with varying magnitude ofresponses to the stressors. Additionally, in response to multiple abiotic stresses, the vegetative traitsdiverged from that of reproductive traits, as deduced from the positive V-TSRI and negative R-TSRIobserved in most of the genotypes and poor correlation between these two processes. The UVB in com-bination with increased temperature caused the greatest damage to cowpea vegetative growth andreproductive potential. The damaging effects of high temperature on seed yield was not amelioratedby elevated [CO2]. The identified tolerant genotypes and groups of plant attributes could be used todevelop genotypes with multiple abiotic stress tolerance.

� 2010 Elsevier B.V. All rights reserved.

1. Introduction

The atmospheric carbon dioxide concentration [CO2] has in-creased globally by more than 100 lmol mol�1 (36%) over the last250 years with the highest recorded average growth rate of1.9 lmol mol�1 yr�1 over the last decade [1]. The current [CO2] ofapproximately 380 lmol mol�1 is estimated to reach between730 and 1020 lmol mol�1 by 2100 [1]. Changes projected in[CO2] and other greenhouse gases are expected to increase globalair temperature by 2.5–4.5 �C during the same period [1]. In addi-tion to these changes in climate, current and projected increase inground-level ultraviolet B (UVB) radiation is closely associatedwith stratospheric ozone (O3) column depletion as it attenuatesthe incoming solar UVB (280–320 nm) radiation [2–3]. Relative

ll rights reserved.

: +1 662 325 9461.dy).

il Sciences, Oklahoma State

State University, Fort Collins,

to the 1970s, the mid-latitudes O3 column losses for the 2002–2005 period were approximately 3% in the Northern and 6% inthe Southern hemisphere [3]. Current global distribution of meanerythemal daily doses of UVB radiation between the latitude40�N and 40�S during summer ranges from 2 to 9 kJ m�2 [4]. Thedaily dose of UVB radiation in USA for the period of June–August,2005 ranged between 0.02 and 8.75 kJ m�2 [5].

The interaction among the environmental stress factors such as[CO2], temperature, and UVB radiation evokes a variety of plant re-sponses. An increased in yield observed at elevated [CO2] [6] werenot observed when plants are grown in combination with hightemperature [7] or increased in UVB radiation [8–9]. Studies haveshown that the projected changes in climate will drastically reducecrop yields when they coincide with the reproductive stage ofplant growth [7,10]. Therefore, the interaction among the environ-mental factors will severely modify the magnitude and direction ofindividual climatic stress factor effects on plants leading to cascad-ing effects on terrestrial ecosystems [11–13]. Thus, an understand-ing of the effects of multiple environmental factors that simulateanticipated future climatic conditions will be useful to assess thegrowth and productivity of agronomic crops.

136 S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146

In nature, plants are routinely exposed to multiple abiotic stres-ses and recent studies demonstrate that plant responses to a singlefactor are much different than those responses under multiplestress conditions [14–16]. Hall and Ziska [17] recommended thatcrop breeders should consider the possible climate change whendeveloping a breeding strategy for yield improvement. Ahmedet al. [18] pointed that developing greater and sustained sinkcapacity will be needed for higher yields under stressful environ-ments. However, to date, the effects of multiple stress factors ongrowth and reproductive potential in many plants are lacking un-der realistic radiation environments. The quality and quantity oflight play an important role in determining plant responsivenessto a given environment [19–21]. Low light conditions have beenshown to reduce yield [19]. Moreover, UVB radiation defensemechanisms such as photo-repair system for DNA and biosynthesisof UVB-absorbing compounds require high light conditions similarto natural solar radiation regimes [22–23]. Many of the recentstudies evaluating the influence of combination of the abioticstresses have been carried out under lower solar radiation regimes[24] or unrealistically lower artificial light conditions (<300 lmolm�2 s�1 PAR) [9,14–15,25] compared to natural settings. Theinferences derived from these studies may not be reflective ofactual effect of those abiotic stress factors in natural environmentand hence limiting the portability of the results to fieldconditions.

In multiple abiotic stress scenarios, the interaction studies willhelp to elucidate whether interactions between atmospheric [CO2]and temperature can counteract the negative effects of UVB radia-tion and vice versa [16,26]. Premkumar and Kulandaivelu [27] re-ported that enhanced UVB radiation markedly alleviated theadverse effects of magnesium deficiency in cowpea whereas, inter-active effects of elevated UVB radiation and high temperaturecaused deleterious effects on soybean (Glycine max L.) growthand development [10,28]. Although, few studies have investigatedthe interactive effects of CO2 and temperature on crop plantsincluding cowpea [18], studies are limited that evaluated the ef-fects of a combination of [CO2], temperature, and UVB radiationand their interactions on crop growth and development, particu-larly on reproductive parameters [24,28–29]. Because of the ex-treme genetic diversity and wide range of climatic adaptation ofcowpea [30], it will be intuitive to study the relative responses ofthis species vegetative and reproductive plant attributes in accor-dance with the changing climate.

Recent studies dealing with multiple environmental factors onvarious plant processes from genes to canopies concluded croptolerance in many crops is needed to cope with changes pro-jected in climate [13,16,31]. Few crop genotypes have beenscreened by using various abiotic stress response indices derivedfrom the different stages of plant growth in response to single ormultiple abiotic stresses [10,32–33]. The earlier studies evaluat-ing the responsiveness of cowpea to abiotic stresses representedsmaller set of plant attributes usually measured either from partof plant organ and/or growth stage involving limited number ofgenotypes [18,27,34–35]. In this paper, we present the resultsof an experiment designed to explore the extent to which mostcommonly investigated plant responses including vegetativeand reproductive processes affected by a combination of multipleabiotic stress factors such as high [CO2], UVB radiation andtemperature.

We hypothesized that genotypic variability exists in cowpeatraits that provide tolerance to abiotic stresses and vegetativetraits differ from that of the reproductive traits on exposure tomultiple stresses. Furthermore, the magnitude and direction ofthe genotypes response to each of these abiotic stressors will bemodified under combination of multiple stress conditions. Theobjectives of this study were to determine whether doubling of

[CO2] will counteract the negative effects of UVB radiation andtemperature, and to evaluate interactive effects of [CO2], tempera-ture, and UVB radiation on growth, physiology and reproduction ofcowpea genotypes and to identify genotypic tolerance to multipleabiotic stressors.

2. Materials and methods

2.1. Research Facility and plant material

Eight sunlit, soil–plant-atmosphere-research (SPAR) units lo-cated at the R.R. Foil Plant Science Research Center (33�280N,88�470W), Mississippi State, Mississippi, USA, were used toconduct the current study. Each SPAR growth chamber has thecapability to precisely control the atmospheric [CO2], tempera-ture, UVB radiation, and desired nutrient and irrigation regimesat determined set points under near ambient levels of photosyn-thetically active radiation (PAR). Each SPAR chamber consists of asteel soil bin (1 m deep by 2 m long by 0.5 m wide) to accommo-date the root system, a Plexiglas chamber (2.5 m tall by 2 m longby 1.5 m wide) to accommodate aerial plant parts and a heatingand cooling system connected to air ducts that pass the condi-tioned air through plant canopy with sufficient velocity(4.7 km h�1) to cause leaf flutter, mimicking field conditions. Var-iable density shade cloths, designed to simulate canopy spectralproperties, placed around the edges of the plant canopy, were ad-justed regularly to match canopy height and to eliminate theneed for border plants. The Plexiglas chambers are completelyopaque to solar UVB radiation, but transmit 12% UV-A and>95% incoming PAR (wavelength 400–700 nm) [36]. During thisexperiment, the incoming daily solar radiation (285–2800 nm)outside of the SPAR units, measured with a pyranometer (Model4–8; The Eppley Laboratory Inc., Newport, RI, USA), ranged from1.5 to 24 MJ m�2 d�1 with an average of 18 ± 4 MJ m�2 d�1. TheSPAR units supported by an environmental monitoring and con-trol systems are networked to provide automatic acquisitionand storage of the data, monitored every 10 s throughout theday and night. Uniformity tests of SPAR units have indicated nostatistical difference in SPAR chambers for sorghum (Reddy, per-sonal communication) and wheat growth parameters [37]. Manydetails of the operations and controls of SPAR chambers havebeen described by Reddy et al. [38]. The relative humidity (RH)of each chamber was monitored with a humidity and tempera-ture sensor (HMV 70Y, Vaisala Inc., San Jose, CA, USA) installedin the returning path of airline ducts. The vapor pressure deficits(VPD) in the units were estimated from these measurements asper Murray [39].

Six contrasting genotypes of cowpea (Vigna unguiculata [L.]Walp.) representing differential sensitivity/tolerance to heat anddiverse sites of origin, ‘California blackeye (CB)-5’ and ‘CB-46’ (bothheat sensitive, University of California, Davis, USA), ‘CB-27’ (heattolerant, University of California, Riverside, USA), ‘MississippiPinkeye’ (MPE, heat sensitivity is not known, Mississippi StateUniversity, Mississippi, USA), ‘Prima’ (heat tolerant, Nigeria), and‘UCR-193’ (heat tolerant, India) [34,40–41], were evaluated in thepresent study. The genotypes were seeded in 15 cm diameter and15 cm deep plastic pots filled with fine sand on 26 July, 2005. Afteremergence (7 days after sowing), thirty pots having healthy plants(five pots for each genotype and three plants in each pot) weretransferred and arranged randomly in each SPAR chamber. Plantswere irrigated three times a day with full-strength Hoagland’snutrient solution delivered at 8:00, 12:00, and 17:00 h to ensureoptimum nutrient and water conditions for plant growth throughan automated and computer-controlled drip irrigation system.The excess solution was drained through the holes in the bottomof the pots and the SPAR soil bins.

S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146 137

2.2. Treatments

Eight treatments consisting of two levels of each of three envi-ronmental factors: CO2 [360 and 720 lmol mol�1 (+CO2)], temper-ature [(30/22 and 38/30 �C (+T)] and UVB (280–320 nm) radiationintensities [0 and 10 (+UVB) kJ m�2 d�1] were imposed from 8 daysafter emergence (DAE) to plant maturity, 53 DAE. The control treat-ment consisted of 360 lmol mol�1 CO2, 30/22 �C temperature and0 kJ m�2 d�1 UVB radiation and all SPAR chambers were main-tained at this condition until 8 DAE. The UVB radiation dose of10 kJ m�2 d�1 was designated to simulate 12% ozone depletion atthe experimental site. The daytime temperature was controlledin square-wave fashion and initiated at the sunrise and returnedto the night time temperature 1 h after sunset. The seasonal datafor daily mean temperatures and daytime [CO2] are presented inTable 1. The quality control of [CO2] and temperature in SPARchambers are described in detail by Reddy et al. [38].

The square-wave supplementation systems were used to pro-vide desired UVB radiation doses which were delivered from0.5 m above the plant canopy for 8 h, each day, from 8:00 to16:00 h by eight fluorescent UVB-313 lamps (Q-Panel Company,Cleveland, OH, USA) mounted horizontally on a metal frame insideeach SPAR chamber, driven by 40 W dimming ballasts. The UVBradiation delivered at the top of the plant canopy was monitoredat 10 different locations in each SPAR chamber daily at 10:00 hwith a UVX digital radiometer (UVP Inc., San Gabriel, CA, USA)which was calibrated against an Optronic Laboratory (Orlando FL,USA) Model 754 Spectroradiometer that is being used initially toquantify the lamp output. The lamp output was adjusted, asneeded, to maintain desired UVB level. To filter UV-C radiation(<280 nm), the lamps were wrapped with pre-solarized 0.07 mmcellulose diacetate (CA) film (JCS Industries Inc., La Mirada, CA,USA). The CA film was changed every 3 to 4-days to account forthe degradation of CA properties. The weighted total biologicallyeffective UVB radiation at the top of the plant canopy during theexperiment are presented in Table 1 which were calculated usinggeneralized plant response spectrum [42] as formulated by Greenet al. [43], normalized at 300 nm.

2.3. Vegetative growth measurements

One plant per pot (five plants per genotype) were harvested 10and 18 days after treatment (DAT) to determine plant height (PH),leaf area (LA), leaf number (LN), and dry matter (DM) of the leavesand stems. Leaf area was measured using Li-3100 leaf area meter(Li-Cor Inc., Lincoln, NE, USA), and specific leaf weight (SLW) wascalculated as leaf weight per unit of leaf area (g cm�2). The plantcomponents were oven dried for 72 h at 70 �C to obtain dryweights. The final remaining one plant per pot was harvested atthe maturity, 53 DAE.

Table 1The set treatments, atmospheric [CO2], ultraviolet B (UVB) radiation and day/night temperaradiation dosage, mean temperature, and daytime vapor pressure deficit (VPD) during the

Treatments Measured v

CO2 (lmol mol�1) UVB (kJ m�2 d�1) T (�C) CO2 (lmol

360 0 30/22 362.01 ± 0.0 38/30 361.32 ± 0.10 30/22 360.56 ± 0.10 38/30 360.11 ± 0.

720 0 30/22 722.28 ± 0.0 38/30 720.23 ± 0.10 30/22 721.61 ± 0.10 38/30 721.85 ± 0.

Each value represents the mean ± SE for one typical day for [CO2], and 10 August to 28

2.4. Photosynthesis and chlorophyll fluorescence measurements

Eighteen days after treatment, leaf net photosynthesis (Pnet),and electron transport rate (ETR) and fluorescence (Fv0/Fm0) weremeasured between 9:00 and 14:00 h on 3rd or 4th leaf from theterminal using an infrared gas analyzer built into a leaf cuvettein an open gas exchange system (Li-COR 6400) with an integratedfluorescence chamber head (Li-COR 6400–40 Leaf Chamber Fluo-rometer). The cuvette chamber conditions were set to provide pho-tosynthetic photon flux density of 1500 lmol m�2 s�1 and cuvetteblock temperature was maintained at the respective treatmentdaytime temperature using a computer-controlled Peltier modulemounted in the cuvette.

2.5. Leaf pigments, phenolics and cell membrane thermostabilitymeasurements

The total leaf chlorophyll, carotenoid and UVB radiation-absorbing compounds were extracted and determined (18 DAT)on five 0.38 cm�2 leaf disks by placing them in a vial containingeither 5 ml of dimethyl sulfoxide for pigments extraction or10 ml of a mixture of methanol, distilled water and hydrochloricacid in 79:20:1 ratio for phenolics extraction and incubated in darkfor 24 h. Thereafter, the concentration of the extract was deter-mined at 648, 662, and 470 nm for estimation of total chlorophylland carotenoid concentration and at 320 nm for phenolic com-pounds estimation by using Bio-Rad UV/VIS spectrophotometer(Bio-Rad Laboratories, Hercules, CA, USA). The equations of Lich-tenthaler [44] were used to estimate the chlorophyll and carote-noids concentrations, whereas the phenolic concentration wasestimated by using the method of Kakani et al. [45] and expressedas equivalent of p-coumaric acid.

The leaf cell membrane thermostability (CMT) in cowpea geno-types was assessed on 18 DAT according to the procedure de-scribed by Martineau et al. [46] with minor modifications. Inbrief, a sample for assay consist of a paired set namely; control(C) and treatment (T) sets, of five leaf disks each 1.3 cm�2, cut fromfive fully expanded 3rd or 4th leaves from the top of the axis se-lected randomly from each treatment with three replicate samples.Prior to assay, the paired set of leaf disks were placed in two sep-arate test tubes and washed thoroughly with four exchanges ofdeionized water, 10 ml each time, to remove electrolytes adheringto the cut surface of the leaf disks. After the final wash, both sets oftest tubes were filled with 10 ml of deionized water and sealedwith aluminum foil to avoid the evaporation. The T-set of the testtubes were incubated for 20 min at 50 �C in a temperaturecontrolled-water bath, while the C-set of test tubes were left atroom temperature (approx. 25 �C). Then, both sets of test tubeswere incubated at 10 �C for 24 h. Initial conductance readings ofboth sets (CEC1 and TEC1) were made by using an electrical

ture (T) conditions, and measured chamber [CO2] from a typical day, daily mean UVBexperimental period for each treatment.

ariables

mol�1) UVB (kJ m�2 d�1) Mean T (�C) VPD (kPa)

30 0.00 ± 0.00 25.97 ± 0.06 2.18 ± 0.0130 0.00 ± 0.00 33.73 ± 0.04 3.40 ± 0.0229 9.14 ± 0.12 25.98 ± 0.06 1.93 ± 0.0351 9.15 ± 0.09 33.69 ± 0.04 3.90 ± 0.01

63 0.00 ± 0.00 25.81 ± 0.05 2.37 ± 0.0154 0.00 ± 0.00 33.79 ± 0.03 3.21 ± 0.0240 9.20 ± 0.11 26.08 ± 0.06 2.28 ± 0.0168 9.10 ± 0.10 33.54 ± 0.04 3.54 ± 0.02

October 2005 for UVB, temperature, and VPD.

138 S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146

conductivity meter (Corning Checkmate II: Corning Inc., New York,USA) after bringing test tubes to room temperature. After that,tubes were again sealed with aluminum foil and autoclaved at120 �C and 0.15 MPa for 20 min to completely kill the leaf tissue.Autoclaved tubes were cooled to room temperature, contentsmixed thoroughly and final conductance (CEC2 and TEC2) mea-surements were recorded. The CMT was calculated as: CMT% ¼1�ðTEC1=TEC2Þ1�ðCEC1=CEC2Þ � 100, where, TEC and CEC are the measure of conduc-tance in treated and control test tubes, respectively, at initial(TEC1 and CEC1) and final (TEC2 and CEC2) conductance measure-ments.

2.6. Flower morphology, pollen production and pollen viabilitymeasurements

Number of days from sowing to the appearance of first openflower was recorded in all treatments. We found that cowpea an-thers dehisce between 05:00 and 08:00 h from a time-series obser-vations (data not shown), and therefore all flower and pollenparameters were measured during this time frame. Flower length(Fl length), percentage pollen viability (PV), and flower dry weight(Fl Dwt) were determined on 10 flowers randomly collected fromfive plants per genotype in each treatment. Flower length wasmeasured from the tip of the standard petal to the base of the ca-lyx. A 3% concentration of 2,3,5 Triphenyltetrazolium chloride(TTC) in 20% sucrose solution was found to be the best for cowpeapollen staining (data not shown). Sucrose was added to adjust theosmotic potential of the staining solution to prevent pollen grainsfrom bursting. Pollen grains were dusted gently by tapping with anartist brush on the microscope glass slides containing a drop of TTCsolution as described by Aslam et al. [47]. Preparations were storedat room temperature in dark, after 16 h, the total and stained pol-len grains were counted in two microscopic fields of 2.4 mm2 hav-ing >100 pollen grains from each field of view using a microscope(SMZ 800 microscope, Nikon Instruments, Kanagawa, Japan). Then,the same flowers were dried at 70 �C for 48 h in an oven to mea-sure flower dry weights.

2.7. Pod production and yield components

Cowpea plants were harvested when most of the pods were ma-ture and dry (53 DAE). The reproductive components such as totalnumber of pods plant�1, number of seeds pod�1, total seed wtplant�1, and weight of individual seeds (g seed�1, average of 100seeds) were determined on all plants in each genotype. Dryweights were measured after complete drying of pods and seedat room temperature. Shelling percentage was calculated as actualseed mass over pod mass multiplied by 100.

2.8. Analysis of variance (ANOVA)

The ANOVA was performed by using the general linear model‘‘PROC GLIMMIX” procedures of SAS [48] to test the significanceof atmospheric [CO2], temperature, UVB radiation, and genotypes,and their interactive effects on plant parameters studied. The leastsquare means (LSMEANS) comparisons were used to determinesignificance differences between treatments for each parameterusing PDIFF LINES option (P = 0.05).

2.9. Cumulative stress response index (CSRI) and total stress responseindex (TSRI)

The CSRI was calculated as the sum of stress response index(SRI) of individual plant-attribute response at a given treatmentcompared to the control, and is based on the response index con-

cept reported in the study of Dai et al. [32] which was calculatedas: SRI ¼ RVt�RVc

RVc� 100, where SRI = stress response index (that

could be measured at any treatment), RV = individual responsevariable (that could be any of 21 measured plant responses) undert = treatment and c = controlled conditions. For PH, LA, LN, SLW,and DM, the average of two measurements (10 and 18 DAT) wereused. All other growth parameters were from the final harvestdate. The CSRIs for vegetative (V-CSRI) and reproductive (R-CSRI)were calculated separately by the following equations:

V-CSRI ¼ PHt � PHc

PHcþ LAt � LAc

LAcþ LNt � LNc

LNcþ SLWt � SLWc

SLWc

�

þDMt � DMc

DMcþ Pnett � Pnetc

Pnetc

þ ETRt � ETRc

ETRc

þ Fv0=Fm0t � Fv0=Fm0

c

Fv0=Fm0c

þ Chlt � Chlc

Chlcþ Carot � Caroc

Caroc

þPhet � Phec

Phecþ CMTt � CMTc

CMTc

�� 100

R-CSRI¼ Fl lengtht�Fl lengthc

Fl lengthcþF Dwtt�F Dwtc

F Dwtc

�

þPPt�PPc

PPcþPVt�PVc

PVcþPod not�Pod noc

Pod noc

þseed wtt�seed wtc

seed wtcþg seed�1

t �g seed�1c

g seed�1c

þSeed Pod�1t �Seed Pod�1

c

Seed Pod�1c

þShellingt�Shellingc

Shellingc

!�100

where PH = plant height, LA = leaf area, LN = leaf number, SLW =specific leaf weight, DM = dry matter of plant shoot, Pnet = net pho-tosynthesis, ETR = electron transport rate, Fv0/Fm0 = the efficiency ofenergy harvesting by oxidized (open) PSII reactioncenters in thelight, Chl = total leaf chlorophyll, Caro = carotenoids, Phe = pheno-lics content, CMT = cell membrane thermostability, Fl length = flow-er length, Fl Dwt = flower dry weight, PP = pollen grains anther�1,PV = pollen viability, Pod no = pods plant�1, seed wt = seed weightplant�1, g seed�1 = individual seed weight (g seed�1), Seed Pod�1 =seed number pod�1, and Shelling = pod shelling percentage under,t = treatment and c = controlled conditions.

The TSRI, sum of the CSRIs over all the treatments, was evalu-ated for vegetative (V-TSRI) and reproductive (R-TSRI) responsesseparately and in combination (C-TSRI) for each genotype. Basedon the C-TSRI (sum of V-TSRI and R-TSRI) cowpea genotypes wereclassified as tolerant (P minimum C-TSRI + 2 standard deviation;SD), intermediate (P minimum C-TSRI + 1 SD and 6minimum C-TSRI + 2 SD) and sensitive (6 minimum C-TSRI + 1 SD) to multipleenvironmental factors individually and in combination.

2.10. Factor analysis

Factor analysis (FA) was used to summarize large number ofvariables by identifying the relationships among the groups of vari-ables, which when examined may suggest an underlying commonfactor that explains why these variables are correlated [49]. Factoranalysis was performed on the correlation matrix of 48 rows (sixgenotypes, eight treatments) and 21 columns (12 vegetative and9 reproductive response variables) using principal factor methodwith an iterative procedure of PROC FACTOR (SAS Institute Inc.,2004). The factors were rotated orthogonally by verimax optionand the numbers of underlying factors were determined by SBC(Schwarz’s Bayesian Criterion).

S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146 139

3. Results

3.1. Vegetative growth

Cowpea genotypes were very responsive to all treatments andtheir interactions for vegetative growth (Table 2, Fig. 1). Plantsgrown under + UVB conditions showed early (5 DAT) symptomsof minor yellowing of veinal and inter-veinal regions on leaveswhich developed into small chlorotic patches at a later stage. The[CO2] significantly interacted with UVB radiation and temperaturefor PH and LA (Table 2) resulting in an increase in PH either under+CO2 condition alone (59%) or in combination with +UVB (17%)and +T (26%), averaged over genotypes. However, plants grown un-der +CO2 + UVB + T condition were 35% shorter than control, whenaveraged across the genotypes. The damaging effects of individualstress factors (+UVB and +T) were less compared to the combinedeffects (+UVB + T) primarily due to significant UVB � T interactionexhibiting greater losses. Among the genotypes, CB-27 and UCR-193 showed the greatest decrease in PH and LA across the treat-ments (Fig. 1a and b). Specific leaf area exhibited a significantCO2 � UVB � T interaction, and on average, increased in all treat-ments except +UVB and +UVB + CO2 treatments (Fig. 1c). The maintreatment effects on LN were significant only for CO2 (Table 2). TheLN varied from 6 (MPE) to 8 (CB-27) under control condition and in-creased 1–2 leaves plant�1 under treatment conditions for all geno-types except CB-27 (data not shown).

The main effects of all treatments were highly significant forDM production (Table 2). Similar to the PH, +CO2 alone increasedthe DM by 68% as over the control when averaged across geno-types. However, this increment was less under +CO2 + UVB and+CO2 + T conditions (Fig. 1d). In contrast, without CO2 enrichment,temperature alone or in combination with +UVB significantly low-ered the DM production. Compared to the control, the highest de-crease in DM was observed in MPE (52%) at +UVB +T condition.

3.2. Leaf photosynthesis and chlorophyll fluorescence

Significant CO2 � UVB � T, CO2 � T and UVB � T interactionswere observed for Pnet (Table 2). Compared to the control, higherphotosynthetic rates were observed in all treatments except+UVB + T, which showed 12% lower rates (Fig. 2a), when averagedacross genotypes. Under +UVB condition, CB-27 showed a 17% de-crease whereas under +UVB + T condition, the decrease in photo-synthetic rate ranged from 11% (CB-27) to 25% (Prima) comparedto the control. The electron transport rate also showed a significant

Table 2Analysis of variance across the genotypes (G) and treatments of carbon dioxide [CO2], temand physiological attributes; plant height (PH), dry matter plant�1 (DM), leaf area (LA), leaffluorescence (Fv0/Fm0), electron transport rate (ETR), total chlorophyll (Chl), carotenoid (C

Source of Variation PH LA LN SLW DM

G ��� ��� �� ��� ���CO2 ��� ��� ��� ��� ���UVB ��� � NS ��� ���T ��� ��� NS ��� ���G � CO2 � � NS ��� ���G � UVB NS NS NS ��� �G � T ��� NS NS �� NSCO2 � UVB ��� � NS ��� NSCO2 � T ��� � ��� NS ��UVB � T �� � �� NS NSG � CO2 � UVB NS NS NS NS NSG � CO2 � T �� NS NS NS �G � UVB � T �� NS � NS NSCO2 � UVB � T NS NS NS �� NSG � CO2 � UVB � T NS NS NS NS NS

The significance levels ���, ��, �, and NS represent P 6 0.001, P 6 0.01, P 6 0.05 and P > 0

CO2 � UVB � T interaction and decreased significantly under +UVBcondition and in combination with either +CO2 or +T condition,when averaged over genotypes (Fig. 2b). However, ETR increasedin other treatments, exhibiting a similar trend to that of Pnet. TheFv0/Fm0 had significant UVB radiation � T interaction and showeda value close to the control or even higher under studied stressconditions (Fig. 2c).

3.3. Leaf pigments, phenolics and cell membrane thermostability

There was a CO2 � UVB � T interaction for both chlorophyll andcarotenoid concentrations in cowpea leaves (Table 2). High tem-perature caused substantial increase in chlorophyll and carotenoidconcentrations in most of the genotypes (Fig. 3a and b). In contrast,elevated UVB radiation caused a decrease in the concentration ofleaf chlorophyll and carotenoid contents either alone or in combi-nation with either +CO2 or +T. Compared to the control, the maxi-mum chlorophyll decrease of 20% was observed in CB-46 at +UVBcondition. The combined effect of +CO2 + UVB + T on chlorophylland carotenoid contents was positive for most of the genotypes,with Prima exhibiting 26 and 29% higher rates, respectively.

Significant CO2 � UVB � T interaction was also observed forphenolic concentrations in cowpea. Averaged over all the geno-types, UVB radiation increased the leaf phenolics either alone(17%) or in combination with +CO2 (27%),+T (2%) and with theirinteractions +CO2 + UVB + T (11%). Prima showed the highest in-crease across all treatments that ranged from 27 in +T to 91% in+CO2 + UVB + T condition. However, +UVB, +T showed a markeddecrease in phenolic concentrations with or without CO2 enrich-ment in all genotypes except Prima and UCR-193 (Fig. 3c). Cellmembrane thermostability was negatively affected for the plantsgrown only under +UVB condition with the maximum decrease ob-served in MPE (26%). Most of the genotypes exhibited improvedCMT when grown under +T condition either alone or in combina-tion with +CO2 (Fig. 3d).

3.4. Flower morphology, pollen production and pollen viability

All genotypes produced flowers in all treatments; however,flowers that were open were seen in all treatments except+UVB + T condition. Days to flowering varied among treatmentsand genotypes (29–46 DAS). Most of the genotypes grown in+CO2 and + T conditions flowered 1–3 d earlier. However, under+CO2 +T and +UVB conditions, the time to flower was delayed by1–3 d in all genotypes except CB-27. The greatest delay in

perature (T), ultraviolet B (UVB) radiation and their interaction on cowpea vegetativenumber plant�1 (LN), specific leaf weight (SLW), net photosynthesis (Pnet), chlorophyllaro), phenolics (Phe), and cell membrane thermostabilty (CMT).

Pnet ETR Fv0/Fm0 Chl Caro Phe CMT

��� ��� �� ��� ��� ��� ����� �� ��� ��� ��� NS ������ ��� NS ��� NS � ���� ��� � ��� ��� � ���� NS NS ��� ��� �� NS��� ��� NS NS � �� NS��� � NS ��� ��� NS ��NS NS � NS �� NS NS��� NS NS NS NS NS ������ NS �� NS NS NS NS��� NS NS � �� NS ������ ��� NS ��� ��� NS NS�� NS NS ��� ��� ��� NS��� � NS ��� �� �� ������ ��� NS NS �� ��� NS

.05, respectively.

Fig. 1. Temperature, carbon dioxide, and ultraviolet B (UVB) radiation effects eitheralone or in combination on (a) plant height, (b) leaf area, (c) specific leaf weight, and(d) plant dry matter (DM) of six cowpea genotypes measured at 18 days aftertreatment; control (360 lmol mol�1, 30/22 �C and 0 kJ UVB), +CO2

(760 lmol mol�1, 30/22 �C and 0 kJ UVB), +UVB (10 kJ UVB, 360 lmol mol�1, 30/22 �C), +T (38/30 �C, 360 lmol mol�1 and 0 kJ UVB), +CO2 + UVB (720 lmol mol�1,10 kJ UVB and 30/22 �C), +CO2 + T (720 lmol mol�1, 38/30 �C, and 0 kJ UVB),+UVB + T (10 kJ UVB, 38/30 �C, and 360 lmol mol�1), and +CO2 + UVB + T(720 lmol mol�1, 10 kJ UVB and 38/30 �C). The error bars show the standarddeviation from five replicates.

Fig. 2. Temperature, carbon dioxide, and ultraviolet B radiation effects either aloneor in combination on (a) net photosynthesis (Pnet), (b) electron transport rate (ETR),and (c) chlorophyll fluorescence (Fv0/Fm0) of six cowpea genotypes measured at18 days after treatment. The error bars show the standard deviation from threereplicates. Other details are as in Fig. 1.

140 S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146

flowering was recorded under +CO2 +UVB (2–6 d) and +CO2 +UVB + T (5–10 d) across the genotypes except CB-27.

All the treatments interacted significantly for flower length andflower dry weight in cowpea (Table 3). The +CO2 caused a small in-crease in flower length compared to the control. Temperature hadno effect on flower length either alone or in combination with +CO2

(Fig. 4a). The elevated CO2 and temperature interacted negativelywith UVB radiation for flower length. The highest decrease was ob-served at +UVB + T condition that ranged from 69% (MPE) to 82%

(CB-27). Averaged over genotypes, the flower dry weight waslower in all treatments compared to the control with the highestdecrease (79%) detected in +UVB + T condition (Fig. 4b). Additionof CO2 reduced the negative influence of +T and +UVB + T on flowerdry weight.

Pollen production and pollen viability were lower in all geno-types under all treatment conditions compared to the control(Fig. 4c and d), and significant interactions were observed amongtreatments (Table 3). High temperature caused significant decreasein pollen production either alone (31%) or in combination with+CO2 (34%) and +UVB (25%), averaged over genotypes. The highestdecrease in pollen production was observed in CB-27 (56%) fol-lowed by CB-5 (37%) at +CO2 + UVB + T condition (Fig. 4c). In thepresence of +UVB and/or +T, pollen viability showed greater de-crease when genotypes were grown under +CO2 compared withambient [CO2] in the presence of the same stressors. None of thegenotypes produced viable pollen grains under +UVB + T condition(Fig. 4d).

3.5. Pod production and yield components

Cowpea genotypes were highly influenced by high temperaturetreatments and failed to set pods under four treatments involving

Fig. 3. Temperature, carbon dioxide, and ultraviolet B radiation effects either aloneor in combination on (a) total chlorophyll, (b) carotenoid, (c) phenolic contents and(d) cell membrane thermostability (CMT) of six cowpea genotypes measured18 days after treatment. The error bars show the standard deviation from threereplicates. Other details are as in Fig. 1.

S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146 141

+T condition (Fig. 5). Therefore, the comparative statements in thissection do not include temperature and its interaction with otherenvironmental factors. Only +CO2 had small beneficial effect onpod number and yield components when averaged over all thegenotypes (Fig. 5a–d). Significant CO2 � UVB � T interaction forpod production and seed weight plant�1 were observed in cowpeagenotypes (Table 3). For instance, compared to the control, higherpod numbers (13%), seed weight (26%), and seeds pod�1 (10%)observed in the plants grown under +CO2 condition were not ob-served in the plants grown under +CO2 + UVB condition. Moreover,the addition of CO2 exacerbated the deleterious effect of +UVB onpod production (Fig. 5a). The greatest decrease in pod number

was observed in CB-27 (47%) followed by UCR-193 (46%) under+CO2 + UVB condition when compared to the control. Whereas,the seed weight was highly influenced by +UVB alone which pro-duced the lowest seed weight in UCR-193 (55%) followed by CB-27 (36%) (Fig. 5b). Similar to the seed weight, the seeds pod�1

was also substantially reduced in UCR-193 (17%) and CB-27(12%) at +UVB condition.

The individual seed weight (g seed�1) increased by 10–14% inCB-5 and CB-46 at elevated [CO2], while 8–9% decrease was ob-served in CB-27 and Prima (Fig. 5c). Compared to the control, the+UVB condition caused the highest decrease (6–30%) in the indi-vidual seed weight, averaged over genotypes. At +CO2 condition,the shelling percentage increased across the genotypes with high-est increase in MPE (40%). Among the cowpea genotypes, the +UVBlowered the shelling percentage by 20–30% whereas this decreasewas less at +CO2 + UVB condition (Fig. 5d).

3.6. Stress response index

The cumulative stress response index (CSRI) representing theoverall stress response of plant attributes for a given treatmentas compared to the control showed varying degree of sensitivityof cowpea genotypes to different stress conditions (Table 4). Mostof the genotypes exhibited positive CSRI for vegetative parameters(V-CSRI, Table 4). Only one negative V-CSRI was evident for Prima,MPE and UCR-193 whereas, CB-27 showed the highest numbers ofnegative V-CSRIs. The negative V-CSRI was mostly associated with+UVB and +UVB + T conditions with the highest negative value of�221 (CB-27) at +UVB. The V-TSRI, sum of V-CSRI over all the treat-ment conditions, varied greatly from �18 (CB-27) to +1619 (UCR-193).

In contrast to V-CSRI, the R-CSRI representing the cumulativeresponses of reproductive parameters for a given treatment condi-tion were mostly negative in all the genotypes, from �2260 (MPE)to �2746 (CB-27) (Table 4). Positive R-CSRI was only observed un-der +CO2 condition for all genotypes except in UCR-193. MPEexhibited positive CSRI for both vegetative and reproductiveparameters under + UVB condition. Highest negative values wereobserved in +UVB + T condition across all genotypes and environ-ments. There was no significant correlation (r2 = 0.04, P > 0.05) be-tween V-TSRI and R-TSRI.

The combined cumulative stress response index (C-CSRI), repre-senting the combined stress responses over vegetative and repro-ductive plant attributes (V-CSRI + R-CSRI), was mostly negativeand highly varied among the genotypes. However, positive C-CSRIswere observed under +CO2 and +CO2 + UVB conditions in all thegenotypes except CB-27. Highest negative C-CSRI was recoded at+UVB + T condition for all genotypes. The C-TSRI, representingthe sum of C-CSRI over all treatment conditions, was all negativeand varied from �1088 (UCR-193) to �2761 (CB-27) (Table 4).

The environmental stress response index (ESRI) representingthe damaging effect of a given environmental factor either aloneor in combination with other factors, on overall performance ofcowpea, was calculated separately for vegetative (V-ESRI) andreproductive (R-ESRI) parameters (Table 4). The ESRIs were rankedfrom 1–7 (1 being the most negative and 7 being the positive orleast negative). Similar to the CSRIs, the V-ESRI was mostly positivewhereas, R-ESRI and C-ESRI were mostly negative. The +UVB + Twas ranked 1 and the +CO2 as 7 in all the cases.

3.7. Factor analysis: grouping the plant attributes

Factor analysis revealed that the 21 measured variables can begrouped into four groups and thus underlying factors influencingcowpea responsiveness to multiple environmental conditions(Table 5). Marked patterns in the loadings of variables under each

Table 3Analysis of variance across the genotypes (G) and treatments of carbon dioxide [CO2], temperature (T), ultraviolet B (UVB) radiation and their interaction on cowpea reproductiveattributes; flower length (Fl length), flower dry weight (Fl Dwt), pollen production anther�1 (PP), pollen viability percentage (PV), pod number plant�1 (Pod No.), seed weightplant�1 (seed wt), individual seed weight (g seed�1), seeds number pod�1 (Seed pod�1), and shelling percentage.

Source of Variation Fl length Fl Dwt PP PV Pod No. Seed wt Seeds pod�1 g seed�1 Shelling

G ��� ��� ��� ��� ��� ��� ��� ��� ���CO2 ��� ��� ��� ��� NS NS NS NS NSUVB ��� ��� �� ��� �� ��� ��� �� �T ��� ��� ��� ��� ��� ��� ��� ��� ���G � CO2 ��� �� NS ��� ��� � NS ��� ���G � UVB ��� � ��� ��� � NS NS NS NSG � T ��� �� ��� ��� ��� ��� ��� ��� ���CO2 � UVB ��� ��� NS ��� NS �� NS NS �CO2 � T ��� ��� NS ��� NS NS NS NS NSUVB � T ��� ��� ��� ��� NS ��� �� �� NSG � CO2 � UVB ��� ��� NS ��� NS NS NS ��� NSG � CO2 � T ��� NS NS ��� ��� � NS ��� ���G � UVB � T ��� NS �� ��� NS NS NS NS NSCO2 � UVB � T ��� ��� � ��� � �� NS NS �G � CO2 � UVB � T ��� NS � ��� NS NS NS ��� NS

The significance levels ���, ��, �, and NS represent P 6 0.001, P 6 0.01, P 6 0.05 and P > 0.05, respectively.

142 S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146

factor helped to propose the common underlying group. First fac-tor had the largest eigenvalue and higher communalities for mostof the variables. Plant attributes largely loaded on the Factor 1were pollen production, pollen viability, pod number, total andindividual seed weights, seed number, and shelling percentage (Ta-ble 5). These parameters were the traits that are known to contrib-ute for crop yield. Therefore, this group was named as theunderlying factor ‘‘yield attributes”. Second factor had the higherloading for the traits contributing to vegetative growth along withCMT and photosynthesis; therefore it was named as ‘‘Growth attri-butes”. Third factor consisted of higher loadings for SLW, chloro-phyll, carotenoid, and phenolics and grouped as an underlyingfactor ‘‘Leaf attributes”. The two variables highly loaded in thefourth factor were flower length and flower dry weight suggestingan underlying factor ‘‘Flower attributes”.

4. Discussion

Cowpea genotypes varied significantly in their vegetative andreproductive performance under multiple abiotic stress conditions.The co-existence of two or more important climatic factors, [CO2],UVB radiation, and temperature, modified the magnitude anddirection of individual stress factor response, thus supporting ourhypothesis. For instance, the +CO2 compensated the negative ef-fects of +UVB and +T singly or in combination for most of the veg-etative and physiological traits including plant height, leaf area, netphotosynthesis, and dry matter production. The +CO2+ UVB + Ttreatment negated some damaging effects of +UVB + T on flowerlength and weights and pollen production and viability, but thisrecovery was not up to the level of control. Moreover, under+UVB + T condition, the flower development was severely inhib-ited, but large number of non viable pollen was observed showingadditive effect of these two stress factors which resulted in greaterlosses. Treatments in combination with +T caused complete repro-ductive failure in all cultivars suggesting that high temperaturemight have affected pollen germination as in other studies [10]and thus zero seed yield in all cowpea cultivars. The current studyalso supported the hypothesis that the vegetative and reproductiveprocesses operate differently under multiple abiotic stress condi-tions, as deduced from the opposite response and lack of correla-tion between these two processes.

Substantial reductions in PH, LA, and DM observed in the cur-rent study have also been reported in several tropical legumes ex-posed to UVB or temperature [50–51]. Stimulation ofphotosynthesis in cowpea caused by +CO2 alone or in combination

with either +UVB or +T in the current study is in agreement withthe observed response in other C3 crops such as canola (Brassica na-pus L.), soybean (Glycine maxi (L.) Merr.) and sunflower (Helianthusannuus L.) [2,29,52]. However, it contrasted with the results ob-tained in a previous study, which reported decrease in photosyn-thesis at higher temperature (day/night, 33/22 �C) alone or incombination with high [CO2] compared to the control temperature(day/night, 33/30 �C) in cowpea [18]. This dissimilarity might havebeen due to the temperature treatment differences, as only night-time temperature varied in that study. Interestingly, compared tothe control, the average photosynthetic rate was much higher un-der +CO2 + T (92%) condition than in either +CO2 (69%) or +T (35%)condition. The lower photosynthetic rate observed for single fac-tors (e.g.+CO2 or +T) compared to their interaction might be ex-plained by the feedback inhibition of photosynthesis due tofaster accumulation of starch in leaves under +CO2 conditionwhereas limited supply of carbohydrate under +T condition dueto increase in respiration [2,18]. In contrast, significant decreasein photosynthesis rate was observed under +UVB + T conditioncompared to the control. However, addition of [CO2](+CO2 + UVB + T condition) compensated the negative effect of+UVB + T. ETR and Fv0/Fm0 also exhibited a pattern similar to thatof photosynthesis under +UVB + T and +CO2 + UVB + T conditions,respectively,

The leaf chlorophyll concentration followed the trend similar tophotosynthesis, however, varying degrees of UVB and temperatureinduced stimulation in the synthesis of carotenoid and phenoliccompounds were observed. The carotenoid and phenolic com-pounds have been considered as protective response against thesestress conditions [27,52]. In general, UVB radiation increased thephenolic compounds while + T alone or in combination with+CO2 caused marked decrease in phenolic compounds. Elevated[CO2] and temperature tend to shorten the time between plantingand flowering while UVB alone or in combination with other treat-ments caused delay in the flowering.

Contrary to the trends in vegetative growth and photosynthesis,+CO2 did not counteract the negative effects of UVB radiation andtemperature on plant reproductive processes. A slight increase inyield components observed at +CO2 in this study is a common ben-eficial effect of CO2 enrichment of increasing carbon availabilityleading to greater yield when other conditions are normal [6,50].However, elevated [CO2] failed to counteract the negative effectsof UVB radiation in most of the genotypes and even recorded lowerpod numbers, seed weight, and shelling percentage. UVB radiationinduced decrease in seed yield has also been reported in other

Fig. 4. Temperature, carbon dioxide, and ultraviolet B radiation effects either aloneor in combination on (a) flower length, (b) flower dry weight, (c) pollen production,and (d) pollen viability (pollen viability was not found at +UVB + T) of six cowpeagenotypes measured between 30 and 40 days after emergence. The error bars showthe standard deviation from ten flower length and pollen viability, and five (pollenproduction) replicates. Other details are as in Fig. 1.

Fig. 5. Temperature, carbon dioxide, and ultraviolet B radiation effects either aloneor in combination on (a) pod number plant�1, (b) total seed weight plant�1, (c)individual seed weight, and (d) shelling percentage of six cowpea genotypesmeasured at 53 days after emergence. These parameters were not found at +T,+CO2 + T, +UVB + T, and +CO2 + UVB + T treatments. The error bars show thestandard deviation from five replicates. Other details are as in Fig. 1.

S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146 143

tropical legumes [33]. Rajendiran and Ramanujam [53] reportedsmaller and fewer seeds per pod along with decrease in pod num-ber (25%), seed weight (45%), and shelling percentage (7%) in Vignaradiata exposed to UVB radiation. This appeared to be due to smal-ler flowers with lower dry weight and reduced pollen viability.Additionally, the increase in the allocation of carbon resources to-wards the repair mechanisms and biosynthesis of UVB absorbingcompounds at the expense of reproductive structures might havealso contributed for the decreased flower characteristics and seedyield attributes.

The substantial decrease in flower size and viable pollen pro-duction caused by UVB radiation and/or temperature in the currentstudy are in accordance with the previous studies including

cowpea [7,34,50]. Fully developed flowers were observed underall treatment conditions except +UVB + T in which flowers pro-duced were small and did not open as in other treatments. Surpris-ingly, the flowers produced under +UVB + T condition showeddeveloped anthers with substantial amount of nonviable pollengrains (Fig. 4c), indicating that pollen germination is being affectedby these stress conditions.

The stress response indices (CSRI, TSRI and ESRI, Table 4) usedto assess the quantitative effects of multiple abiotic stressors inthe current study is equally effective as in other crops with high in-tra-specific variability [28,32–33]. Generally, positive values ofvegetative parameters (V-CSRI and V-TSRI) compared to the nega-tive values for reproductive attributes (R-CSRI and R-TSRI) clearly

Table 4Cumulative stress response index (CSRI), sum of relative individual plant attribute stress responses index (SRI) at a given treatment; and total stress response index (TSRI), sum ofCSRI over all the treatments of six cowpea genotypes in response to elevated carbon dioxide (720 lmol mol�1,+CO2), high temperature (38/30 �C,+T), and increased UVB radiation(10 kJ m�2 d�1,+UVB) and their interactions. TSRIs were separated into vegetative (V-TSRI), reproductive (R-TSRI), and added together to obtain combined TSRIs� (C-TSRI). TheCSRI is the sum of relative responses with treatments in comparison to control, i.e. 360 lmol mol�1 (CO2), 30/22 �C temperature (T) and 0 kJ m�2 d�1 (UVB) observed forvegetative (V-CSRI: plant height, dry matter, leaf area, leaf number, specific leaf wt, net photosynthesis, chlorophyll fluorescence, electron transport rate, chlorophyll, carotenoid,phenolics, cell membrane thremostability) and reproductive (R-CSRI: flower length, flower dry wt, pollen production, pollen viability, pod number, seed wt, individual seed wt,number of seeds pod�1, shelling percentage) parameters studied. A combined CSRI (C-CSRI) is the sum of V-CSRI and R-CSRI. ESRI (environmental stress response index),calculated separately for vegetative (V-ESRI), reproductive (R-ESRI), and combined (C-ESRI) parameters, indicates the damaging effect of a given stress on overall cowpeaperformance of genotypes with ranks indicated in parentheses.

Stressor Genotypes

Prima CB-5 CB-27 CB-46 MPE UCR-193

Vegetative cumulative stress response index (V-CSRI) V-ESRI+CO2 +419 +358 +111 +356 +235 +429 +1908 (7)+UVB +19 �72 �221 �45 +168 �18 �170 (2)+T +114 +170 +49 +131 +147 +197 +809 (3)+CO2 + UVB +353 +157 �1 +98 +92 +279 +978 (4)+CO2 + T +247 +330 +157 +340 +212 +409 +1696 (6)+UVB + T �120 �44 �104 �89 �33 +12 �378 (1)+CO2 + UVB + T +79 +176 �9 +170 +260 +311 +987 (5)V-TSRI� +1111 +1075 �18 +961 +1081 +1619 -

Reproductive cumulative stress response index (R-CSRI) R-ESRI+CO2 +12 +169 +32 +191 +1 �12 +393 (7)+UVB �32 �20 �96 �86 +104 �136 �266 (6)+T �574 �559 �520 �571 �491 �556 �3271 (4)+CO2 + UVB �115 �62 �120 �87 �49 �3 �437 (5)+CO2 + T �591 �583 �594 �587 �544 �579 �3478 (3)+UVB + T �766 �798 �800 �794 �755 �770 �4684 (1)+CO2 + UVB + T �680 �637 �644 �590 �526 �651 �3728 (2)R-TSRI� �2746 �2490 �2742 �2524 �2260 �2706 -

Combined cumulative stress response index (C-CSRI) C-ESRI+CO2 +431 +527 +142 +547 +236 +418 +2302 (7)+UVB �13 �91 �317 �132 +272 �154 �436 (5)+T �460 �389 �471 �440 �344 �359 �2462 (3)+CO2 + UVB +238 +95 �121 +10 +43 +275 +541 (6)+CO2 + T �344 �253 �437 �247 �332 �170 �1782 (4)+UVB + T �886 �842 �904 �883 �787 �759 �5062 (1)+CO2 + UVB + T �602 �461 �653 �420 �266 �340 �2741 (2)C-TSRI� �1636 �1414 �2761 �1565 �1178 �1088 -

Table 5Rotated factor loadings of 21 measured plant attributes representing group-wise responsiveness of cowpea to multiple environmental stresses.

Response variable Factor 1 (yield attribute) Factor 2 (growth attribute) Factor 3 (leaf attribute) Factor 4 (flower attribute) Communality

Plant height 0.31 0.80* 0.07 0.29 0.82Leaf area 0.76 0.75* �0.05 0.16 0.94Leaf number 0.01 0.64* 0.21 �0.01 0.46Specific leaf weight �0.58 0.32 0.66* 0.04 0.87Dry matter 0.37 0.84* 0.36 0.14 0.99Net photosynthesis �0.31 0.53* 0.17 0.31 0.50Chlorophyll fluorescence 0.00 0.57* 0.00 0.26 0.39Electron transport rate �0.51 0.22 0.26 0.13 0.39Chlorophyll �0.19 0.24 0.92* 0.13 0.95Carotenoid �0.11 0.18 0.91* �0.04 0.87Phenolic 0.50 0.29 0.59* �0.15 0.52Cell membrane theromstability �0.42 0.51* 0.28 �0.33 0.55Flower length 0.15 0.27 0.03 0.95* 1.00Flower dry weight 0.22 0.24 0.02 0.89* 0.90Pollen production 0.82* 0.00 0.10 0.09 0.69Pollen viability 0.72* 0.19 0.04 0.55 0.72Pod number 0.86* 0.17 �0.11 0.20 0.82Seed weight plant�1 0.88* 0.12 �0.12 0.22 0.85Seed weight seed�1 0.90* 0.00 �0.23 0.26 0.94Seed number pod�1 0.82* 0.06 �0.36 0.30 0.89Shelling percentage 0.81* 0.06 �0.35 0.29 0.87Eigenvalues� 182 82 33 15 –

* Indicates the variables with large factor loadings in the corresponding column.� Indicates the eigenvalues of the correlation matrix.

144 S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146

show high negative impact of abiotic stresses on cowpea reproduc-tive potential. As expected, the data showed high degree of geno-typic variation for both vegetative and reproductive traits andoverall stress effect was negative in all genotypes as deduced from

the C-TSRI. However, the magnitudes of genotypic responses werehighly modified by different stresses either alone or in combina-tion. This modified degree of response mechanisms might havebeen caused due to the differences in co-activation of different

S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146 145

response pathways by simultaneous exposure of plants to differentabiotic stresses leading to a synergistic (for example most of thevegetative growth and photosynthetic parameters of cowpea inthis study) or antagonistic (reproductive processes and yield attri-butes) effects [13]. There was no significant (r2 = 0.04, P > 0.05) cor-relation between V-TSRI and R-TSRI suggesting that the genotypeswhich performed well for vegetative parameters did not perform inthe same way for reproductive growth in the presence of the samestress condition. Vast amount of energy and resources are requiredfor plants to acclimate to abiotic stress conditions, hence, nutrientdeprivation including carbon could pose a serious problem toplants attempting to cope with heat or UVB radiation stress [13].This increase in allocation of carbon and other resources towardsrepair mechanisms and biosynthesis of protective compounds suchas carotenoids and/or phenolic compounds at the expense of repro-ductive structures along with carbon-independent processes suchas pollen vitality might have caused high sensitivity of reproduc-tive traits. The combined response of vegetative and reproductivetraits to multiple abiotic stresses (C-TSRI) facilitated the relativeclassification of cowpea genotypes into three groups as tolerant(UCR-193, MPE and CB-5), intermediate (CB-46 and Prima), andsensitive (CB-27) to multiple abiotic stresses.

The magnitude of genotypic variability of a species offers anopportunity for a plant breeder to design and develop specificplant type to suit an agro-ecological environment. The effective-ness of selection for a trait depends on its genetic control underdifferent environmental condition which is expressed as herita-bility of the trait [17,54]. The genetic association of a trait withhigher level of physiological and/or developmental attributesthat facilitate adaptation for a stress condition are very usefulfor plant breeding purposes and to develop improved lines ofa crop species [55]. By categorizing the interactions across plantattributes, it is evident from the result of this study that thestress protective response ‘‘Leaf attributes” identified by factoranalysis exhibit parallel increasing or less decreasing responsepatterns along with ‘‘growth and yield attributes” for at leastin three cowpea genotypes (Prima, MPE and UCR-193). Simi-larly, the inheritance studies have demonstrated that heat toler-ance during reproductive development requires a higherheritable recessive gene for flower production [54]. In thecurrent study, the appearance of flower and comparable pollenproduction observed even under +UVB + T condition haveremarkable potential for trait-based selection criterion thatmay be used in other species to enhance stress tolerance viagenetic manipulation.

Plant adaptation to abiotic stresses will dependent upon theactivation of molecular networks involved in stress perception, sig-nal transduction and expression of specific stress related genes andmetabolites, which ultimately result in morphological and physio-logical development [56]. The linkage between stress-associatedmolecular mechanisms and physiological response is still a majorgap in our understanding of crop tolerance to different stress con-ditions [57]. Most of the studies published so for involving combi-nation of stress factors have used either short-term stresstreatments and/or low radiation environments, rather than evalu-ating stress response over plant life cycle under reasonable radia-tion environment [9,14–15,24–25,28]. Therefore, due to theemergent nature of yield from physiological processes, and thephysiological processes are the outcome of various molecular net-works in response to different stresses, the results from these stud-ies may not be transferable under natural environment and willlack the association with actual crop yield. A comprehensive port-folio of molecular and physiological basis of stress tolerance thatcombined the traditional and molecular breeding (genetic engi-neering) will help to improve crop tolerance and yield across abi-otic stress conditions.

In conclusion, the current study revealed that regardless of CO2

enrichment, a combined effect of UVB radiation and temperaturepossibly will pose a serious problem for cowpea and most likelyfor many summer-grown crops in future climatic conditions. Allcowpea genotypes responded in the same direction while the mag-nitude of these responses to multiple stress conditions variedwidely among genotypes. Elevated [CO2] did not negate the dam-aging effects of UVB radiation and/or high temperature on repro-ductive traits, particularly viable pollen production and seedyield. The identified tolerant cowpea genotypes and groups ofplant attributes could be used for selection and development ofgenotypes tolerance to multiple abiotic stresses by trait-basedplant breeding or genetic engineering programs. The cowpea veg-etative and reproductive attributes in response to abiotic stresseswere not correlated indicating the tolerance mechanisms in boththese processes operate differently. In addition, cumulative envi-ronmental stress response indices of vegetative (E-ESRI) and repro-ductive (R-ESRI) parameters yielded poor correlation indicating thefactors that may positively contribute for vegetative traits may notgo hand-in-hand with reproductive traits. Therefore, developingcultivars for the future climate is daunting challenge addressingmany facets of crop growth and development under multiple envi-ronmental stress conditions.

Acknowledgements

This research was funded in part by the Department of Energy,and USDA-UVB Monitoring Program at Colorado State University,CO. We also thank Drs. Harry Hodges for his comments and sugges-tions and Mr. David Brand for technical support. We thank Dr. JeffEhlers, Department of Botany and Plant Sciences, University of Cal-ifornia Riverside, CA, USA for providing seed. This article is a con-tribution from the Department of Plant and Soil Sciences,Mississippi State University, Mississippi Agricultural and ForestryExperiment Station, paper no. J11412.

References

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

[2] S.P. Long, Modification of the response of photosynthetic productivity to risingtemperature by atmospheric CO2 concentrations: has its importance beenunderestimated?, Plant Cell Environ 14 (1991) 729–739.

[3] WMO, Scientific Assessment of Ozone Depletion: 2006, in: A.L.N. Ajavon, D.L.Albritton, R.T. Watson (Eds.), Global Ozone Research and Monitoring Project,Geneva, Switzerland, 2007, p. 572.

[4] R.L. McKenzie, P.J. Aucamp, A.F. Bais, L.O. Björn, M. Ilyas, Changes inbiologically active ultraviolet radiation reaching the Earth’s surface,Photochem. Photobiol. Sci. 6 (2007) 218–231.

[5] USDA, UV-B-Monitoring and Research Program (<http://uvb.nrel.colostate.edu/UVB/index.jsf>), in, Colorado State University, CO USA, 2005.

[6] B.A. Kimball, K. Kobayashi, M. Bindi, Responses of agricultural crops to free-airCO2 enrichment, Adv. Agron. 77 (2002) 293–368.

[7] P.V.V. Prasad, K.J. Boote, L.H. Allen Jr., J.M.G. Thomas, Super-optimaltemperatures are detrimental to peanut (Arachis hypogaea L.) reproductiveprocesses and yield at both ambient and elevated carbon dioxide, GlobalChange Biol. 9 (2003) 1775.

[8] A.H. Teramura, J.H. Sullivan, L.H. Ziska, Interaction of elevated ultraviolet-Bradiation and CO2 on productivity and photosynthetic characteristics in wheat,Rice, and soybean, Plant Physiol. 94 (1990) 470–475.

[9] M.M. Qaderi, D.M. Reid, Growth and physiological responses of canola (Brassicanapus) to UV-B and CO2 under controlled environment conditions, Physiol.Plant. 125 (2005) 247–259.

[10] S. Koti, K.R. Reddy, V.R. Reddy, V.G. Kakani, D.L. Zhao, Interactive effects ofcarbon dioxide, temperature, and ultraviolet-B radiation on soybean (Glycinemax L.) flower and pollen morphology, pollen production, germination, andtube lengths, J. Exp. Bot. 56 (2005) 725–736.

[11] D.B. Lobell, G.P. Asner, Climate and management contributions to recenttrends in US agricultural yields, Science 299 (2003) 1032.

146 S.K. Singh et al. / Journal of Photochemistry and Photobiology B: Biology 100 (2010) 135–146

[12] S.P. Long, E.A. Ainsworth, A.D.B. Leakey, J. Nosberger, D.R. Ort, Food forthought: Lower-than-expected crop yield stimulation with rising CO2

concentrations, Science 312 (2006) 1918–1921.[13] R. Mittler, Abiotic stress, the field environment and stress combination, Trends

Plant Sci. 11 (2006) 15–19.[14] L. Rizhsky, H. Liang, R. Mittler, The combined effect of drought stress and heat

shock on gene expression in tobacco, Plant Physiol. 130 (2002) 1143–1151.[15] L. Rizhsky, H. Liang, J. Shuman, V. Shulaev, S. Davletova, R. Mittler, When

defense pathways collide. The response of arabidopsis to a combination ofdrought and heat stress, Plant Physiol. 134 (2004) 1683–1696.

[16] M.M. Caldwell, J.F. Bornman, C.L. Ballaré, S.D. Flint, G. Kulandaivelu, Terrestrialecosystems, increased solar ultraviolet radiation, and interactions with otherclimate change factors, Photochem. Photobiol. Sci. 6 (2007) 252–266.

[17] A.E. Hall, L.H. Ziska, Crop breeding strategies for the 21st century, in: K.R.Reddy, H.F. Hodges (Eds.), Climate Change and Global Crop Productivity, CABInternational, Oxon, 2000, pp. 407–423.

[18] F.E. Ahmed, A.E. Hall, M.A. Madore, Interactive effects of high temperature andelevated carbon dioxide concentration on cowpea (Vigna unguiculata (L.)Walp.), Plant Cell Environ. 16 (1993) 835–842.

[19] R.J. Summerfield, P.A. Huxley, P.J. Dart, A.P. Hughes, Some effects ofenvironmental stress on seed yield of cowpea (Vigna unguiculata (L.) walp.)cv. Prima, Plant Soil 44 (1976) 527–546.

[20] D.J. Allen, S. Nogues, N.R. Baker, Ozone depletion and increased UV-Bradiation: is there a real threat to photosynthesis?, J Exp. Bot. 49 (1998)1775–1788.

[21] E. Goto, Effects of light quality on growth of crop plants under artificiallighting, Environ. Control. Biol. 41 (2003) 121–132.

[22] M.M. Caldwell, S.D. Flint, P.S. Searles, Spectral balance and UV-B sensitivity ofsoybean: a field experiment, Plant Cell Environ. 17 (1994) 267–276.

[23] P. Adamse, S.J. Britz, C.R. Caldwell, Amelioration of UV-B damage under highirradiance I: role of blue light photoreceptors, Photochem. Photobiol. 60 (1994)110–115.

[24] R. Tegelberg, R. Julkunen-Tiitto, M. Vartiainen, R. Paunonen, M. Rousi, S.Kellomäki, Exposures to elevated CO2, elevated temperature and enhancedUV-B radiation modify activities of polyphenol oxidase and guaiacolperoxidase and concentrations of chlorophylls, polyamines and solubleproteins in the leaves of Betula pendula seedlings, Environ. Exp. Bot. 62(2008) 308–315.

[25] M.M. Qaderi, D.M. Reid, E.C. Yeung, Morphological and physiological responsesof canola (Brassica napus) siliquas and seeds to UVB and CO2 under controlledenvironment conditions, Environ. Exp. Bot. 60 (2007) 428–437.

[26] V.C. Runeckles, S.V. Krupa, The impact of UV-B radiation and ozone onterrestrial vegetation, Environ. Pollut. 83 (1994) 191–213.

[27] A. Premkumar, G. Kulandaivelu, Influence of increased solar UV-B radiation onmagnesium deficient cowpea seedlings: changes in growth and foliarconstituents, Plant Sci. 161 (2001) 1–8.

[28] S. Koti, K.R. Reddy, V.G. Kakani, D. Zhao, W. Gao, Effects of carbon dioxide,temperature and ultraviolet-B radiation and their interactions on soybean(Glycine max L.) growth and development, Environ. Exp. Bot. 60 (2007) 1–10.

[29] U. Mark, M. Tevini, Effects of solar ultraviolet-B radiation, temperature andCO2 on growth and physiology of sunflower and maize seedlings, Plant Ecol.128 (1997) 224–234.

[30] B.B. Singh, Cowpea (Vigna unguiculata (L.) Walp.), in: R.J. Singh, P.P. Jauhar(Eds.), Genetic resources, Chromosome Engineering, and Crop Improvement,CRC Press, Boca Raton, FL, USA, 2004, pp. 117–161.

[31] A.E. Hall, Comparative ecophysiology of cowpea, common bean, and peanut,in: T.N. Henry, B. Abraham (Eds.), Physiology and Biotechnology Integrationfor Plant Breeding, Marcel Dekker, New York, 2004, pp. 271–385.

[32] Q. Dai, S. Peng, A.Q. Chavez, B.S. Vergara, Intraspecific response of 188 ricecultivars to enhanced UV-B radiation, Environ. Exp. Bot. 34 (1994) 433–442.

[33] M. Saile-Mark, M. Tevini, Effects of solar UV-B radiation on growth, floweringand yield of central and southern European bush bean cultivars (Phaseolusvulgaris L.), Plant Ecol. 128 (1997) 115–125.

[34] M.O.A. Warrag, A.E. Hall, Reproductive responses of cowpea to heat stress:genotypic differences in tolerance to heat at flowering, Crop Sci. 23 (1983)1088–1092.

[35] C.F. Musil, S.B.M. Chimphango, F.D. Dakora, Effects of elevated ultraviolet-Bradiation on native and cultivated plants of southern Africa, Ann. Bot. 90(2002) 127–137.

[36] D. Zhao, K.R. Reddy, V.G. Kakani, J.J. Read, J.H. Sullivan, Growth andphysiological responses of cotton (Gossypium hirsutum L.) to elevated carbondioxide and ultraviolet-B radiation under controlled environmentalconditions, Plant Cell Environ. 26 (2003) 771–782.

[37] D.H. Fleisher, D.J. Timlin, Y. Yang, V.R. Reddy, K.R. Reddy, Uniformity of soil–plant-atmosphere-research chambers, Trans. ASABE 52 (2009) 1721–1731.

[38] K.R. Reddy, H.F. Hodges, J.J. Read, J.M. McKinion, J.T. Baker, L. Tarpley, V.R.Reddy, Soil–plant-atmosphere-research (SPAR) facility: a tool for plantresearch and modeling, Biotronics 30 (2001) 27–50.

[39] F.W. Murray, On the computation of saturation vapor pressure, J. Appl.Meteorol. 6 (1967) 203–204.

[40] W.W. Hare, ‘Mississippi Pinkeye’ cowpea, HortScience 26 (1991) 76–77.[41] J. Fang, C.C. Chao, P. Roberts, J. Ehlers, Genetic diversity of cowpea [Vigna

unguiculata (L.) Walp] in four West African and USA breeding programs asdetermined by AFLP analysis, Genet. Resour. Crop Evol. 54 (2007) 1197–1209.

[42] M.M. Caldwell, Solar ultraviolet radiation and the growth and development ofhigher plants, in: A.C. Giese (Ed.), Photophysiology, Academic Press, New York,1971, pp. 131–177.

[43] A. Green, T. Sawada, E. Shettle, The middle ultraviolet radiation reaching theground, Photochem. Photobiol. 19 (1974) 251–259.

[44] H.K. Lichtenthaler, Chlorophylls and carotenoids: pigments of photosynthesis,Methods Enzymol. 148 (1987) 350–352.

[45] V.G. Kakani, K.R. Reddy, D. Zhao, W. Gao, Senescence and hyperspectralreflectance of cotton leaves exposed to ultraviolet-B radiation and carbondioxide, Physiol. Plant. 121 (2004) 250–257.

[46] J.R. Martineau, J.H. Williams, J.E. Specht, Tolerance in soybean. II. Evaluation ofsegregating populations for membrane thermostability, Crop Sci. 19 (1979)79–81.

[47] M. Aslam, M.S. Brown, R.J. Kohel, Evaluation of seven tetrazolium salts as vitalpollen stains in cotton Gossypium hirsutum L, Crop Sci. 4 (1964) 508–510.

[48] SAS Institute Inc, SAS/STAT User’s Guide, Version 9.1.3, Cary, NC, in, 2004.[49] D.E. Johnson, Applied Multivariate Methods for Data Analysts, Duxbury Press,

New York, 1998.[50] P.V.V. Prasad, K.J. Boote, L.H. Allen, J.M.G. Thomas, Effects of elevated

temperature and carbon dioxide on seed-set and yield of kidney bean(Phaseolus vulgaris L.), Global Change Biol. 8 (2002) 710–721.

[51] A. Singh, Growth, physiological, and biochemical responses of three tropicallegumes to enhanced UV-B radiation, Can. J. Bot. 74 (1996) 135–139.

[52] M.M. Qaderi, L.V. Kurepin, D.M. Reid, Growth and physiological responses ofcanola (Brassica napus) to three components of global climate change:temperature, carbon dioxide and drought, Physiol. Plant. 128 (2006)710–721.

[53] K. Rajendiran, M.P. Ramanujam, Improvement of biomass partitioning,flowering and yield by triadimefon in UV-B stressed Vigna radiata (L.)Wilczek, Biol. Plant. 48 (2004) 145–148.

[54] S. Thiaw, A.E. Hall, Comparison of selection for either leaf-electrolyte-leakageor pod set in enhancing heat tolerance and grain yield of cowpea, Field CropsRes. 86 (2004) 239–253.

[55] B.B. Singh, B. Sharma, Restructuring cowpea for higher yield, Indian J. Genet.Plant Breed. 56 (1996) 389–405.

[56] B. Vinocur, A. Altman, Recent advances in engineering plant tolerance toabiotic stress: achievements and limitations, Curr. Opin. Biotechnol. 16 (2005)123–132.

[57] T.R. Sinclair, L.C. Purcell, Is a physiological perspective relevant in a‘genocentric’ age?, J Exp. Bot. 56 (2005) 2777–2782.