review evolutionary context for understanding and manipulating plant responses to past...

on June 3, 2018http://rstb.royalsocietypublishing.org/Downloaded from

Phil. Trans. R. Soc. B (2012) 367, 613629

doi:10.1098/rstb.2011.0248

Review

* Autho

One conevolution

Evolutionary context for understandingand manipulating plant responses to past,

present and future atmospheric [CO2]Andrew D. B. Leakey1,* and Jennifer A. Lau2

1Department of Plant Biology and Institute for Genomic Biology, University of Illinois,Urbana-Champaign, 1201 W. Gregory Drive, Urbana, IL 61801, USA

2W. K. Kellogg Biological Station and Department of Plant Biology, Michigan State University,3700 E Gull Lake Drive, Hickory Corners, MI 49060, USA

Variation in atmospheric [CO2] is a prominent feature of the environmental history over which vas-cular plants have evolved. Periods of falling and low [CO2] in the palaeo-record appear to havecreated selective pressure for important adaptations in modern plants. Today, rising [CO2] is akey component of anthropogenic global environmental change that will impact plants and the eco-system goods and services they deliver. Currently, there is limited evidence that natural plantpopulations have evolved in response to contemporary increases in [CO2] in ways that increaseplant productivity or fitness, and no evidence for incidental breeding of crop varieties to achievegreater yield enhancement from rising [CO2]. Evolutionary responses to elevated [CO2] havebeen studied by applying selection in controlled environments, quantitative genetics and trait-based approaches. Findings to date suggest that adaptive changes in plant traits in response tofuture [CO2] will not be consistently observed across species or environments and will not belarge in magnitude compared with physiological and ecological responses to future [CO2]. Thislack of evidence for strong evolutionary effects of elevated [CO2] is surprising, given the large effectsof elevated [CO2] on plant phenotypes. New studies under more stressful, complex environmentalconditions associated with climate change may revise this view. Efforts are underway to engineerplants to: (i) overcome the limitations to photosynthesis from todays [CO2] and (ii) benefit maxi-mally from future, greater [CO2]. Targets range in scale from manipulating the function of a singleenzyme (e.g. Rubisco) to adding metabolic pathways from bacteria as well as engineering the struc-tural and functional components necessary for C4 photosynthesis into C3 leaves. Successfullyimproving plant performance will depend on combining the knowledge of the evolutionary context,cellular basis and physiological integration of plant responses to varying [CO2].

Keywords: adaptation; climate change; evolution; yield; fitness

1. INTRODUCTION: WHY TAKE ANEVOLUTIONARY PERSPECTIVE?Assimilation of CO2 from the atmosphere into biomassby higher plants is fundamental to: (i) providing food,fuel and fibre for human consumption; (ii) supplyingenergy to terrestrial ecosystems; and (iii) regulatingthe concentration of atmospheric CO2 ([CO2]) andclimate. In almost all higher plants, photosyntheticCO2 fixation (A) and stomatal conductance (gs) areinstantaneously sensitive to variation in [CO2] overthe range of past to present and/or predicted future[CO2]. The rise in [CO2] starting during the IndustrialRevolution and continuing today is notable for howquickly it is altering plant function. Changes in A andgs caused by increasing [CO2] initiate a set of cellularand physiological responses, which typically increase

r for correspondence ([email protected]).

tribution of 12 to a Theme Issue Atmospheric CO2 and theof photosynthetic eukaryotes: from enzymes to ecosystems.

613

growth and can increase reproductive output. Genoty-pic variation in almost all elements of these responsescreates the potential for ecological and evolutionaryconsequences over a wide range of timescales.

Plant and ecosystem responses to varying [CO2] arecurrently best understood at the physiological andecological levels and on timescales of one generationor less. Investigating plant responses to varying atmos-pheric [CO2] in an evolutionary context is importantbecause variations in [CO2] over geological timescalesare believed to have played important roles in the evol-ution of ecologically and economically important traitsin extant species. In addition, if plants evolve in responseto twenty-first century [CO2], changes in future ecosys-tem structure, function and services will extend beyondwhat can be predicted from knowledge of physiologicaland ecological responses to elevated [CO2]. From apractical perspective, there is the possibility that despitemajor breeding successes, present elite crop varietiesmay not be adapted for optimal performance under pre-sent and future [CO2]. Accordingly, improving plant

This journal is q 2012 The Royal Society

mailto:[email protected]://rstb.royalsocietypublishing.org/

614 A. D. B. Leakey & J. A. Lau Review. Plant adaptation to atmospheric [CO2]


productivity in high [CO2] environments may be anopen opportunity for biotechnology or breeding toimprove crop performance now and in the future. Thispaper builds on previous reviews [14] to synthesizethe present knowledge on each of these topics. It dis-cusses how an evolutionary perspective could advanceefforts to understand and manage plant and ecosystemresponses to rising [CO2] by addressing the followingfive questions:

(1) Have plants evolved in response to varying [CO2]on a geological timescale?

(2) Have plants evolved by natural or artificial selec-tion in response to contemporary increases in[CO2]?

(3) Will rising [CO2] drive natural selection in thefuture, and if so how?

(4) What traits are favoured under high [CO2]?(5) How does evolutionary history impact and

inform efforts to engineer crops for improvedperformance in present and future [CO2]?

Detailed evaluation of the approaches available forintegrating evolutionary biology with physiology andecology are reviewed authoritatively elsewhere [57].

2. QUESTION 1: HAVE PLANTS EVOLVEDIN RESPONSE TO VARYING [CO2] ON AGEOLOGICAL TIMESCALE?Variation in [CO2] is proposed to have played a key rolein driving the evolution of plants since they colonizedthe land 400 Ma [812]. Estimates of the palaeo-[CO2] record have been generated by modelling theweathering and burial of CaMg silicates and organiccarbon [13], as well as a range of proxies includingstomatal characteristics of fossil leaves [14] and thestable isotope composition of pedogenic carbonates,marine organic matter and fossil bryophytes [1518].There is significant uncertainty associated with eachindividual methodology and variation across methods[13,16, 19,20]. Nevertheless, the following generaltrend emerges. The earliest land plants became estab-lished at high [CO2] (15003000 ppm) before aperiod of low [CO2] (less than or equal to 1000 ppm),which started 350 Ma and lasted 50100 Myr(figure 1a). Between 250100 Ma [CO2] appears tohave been maintained at approximately 1000 ppm.With the exception of a period in the Eocene 4050 Ma, all proxies and models indicate [CO2] of lessthan 1000 ppm for the last 100 Myr (figure 1a).

Periods of falling or low [CO2] have been linked tothe evolution of a number of important plant traits aswell as diversification of the vascular flora (figure 1b)[8,27]. The development of megaphyll leaves in mul-tiple independent lineages was coincident with thetransition from high to low [CO2] during the Devo-nian and Carboniferous 400350 Ma [25,28]. As[CO2] dropped, rates of A would have becomeincreasingly limited by the resistance to diffusion ofCO2 from the atmosphere to the site of fixation byRubisco. Changes in stomatal density and stomatalsize (a combination of pore depth and pore cross-sectional area determined from measurements of theentire guard cell complex, which better predicts gs

Phil. Trans. R. Soc. B (2012)

than pore cross-sectional area alone) observed infossil leaves have been used to drive modelspredicting increases in maximum gs at this time[14,23,29], which would have relieved resistance toCO2 diffusion through the epidermis and maintainedrates of A [30]. The rise in gs is proposed to haveincreased the capacity for evaporative cooling ofleaves, allowing greater leaf areas to develop forintercepting radiation without causing overheating[25,28]. These changes in the structure and functionof leaves coincided with the first major increase in thenumber of vascular plant species [24]. Although mostdata indicate that periods of low [CO2] are associatedwith novel adaptations and diversification, Willis &McElwain [12] found that over geological timescales,periods of high [CO2] corresponded with greater orig-ination rates of fossil species. These two observationsappear contradictory on first examination. However,origination rates are defined as the rate at which newspecies appear in the fossil record, and this coulddiffer from the rate of change of overall species richness.If this is the case, a combination of decreasing rates ofspecies gain in low [CO2] with even greater decreasesin the rate of species loss could be the basis of theobserved patterns.

The second major period of falling and low [CO2](from 100 Ma until the present day) also overlappedwith increases in stomatal density and decreases in indi-vidual stomatal size that suggest plants were developinggreater maximum gs to counteract the CO2-limitationof photosynthesis (figure 1b) [14,23,29]. This waspreceded by a rapid and significant increase of vein den-sity in angiosperm leaves that started 150 Ma [26].Greater leaf hydraulic conductance resulting fromgreater vein density would have allowed plants topotentially achieve greater A by delivering more waterto the leaf in order to sustain greater stomatal con-ductance. Therefore, increasing vein density has beenproposed to have been a major fitness advantage forangiosperms and contributed to their subsequent radi-ation (figure 1a,b) [26,3133]. The magnitude of thebenefit to A from greater hydraulic conductance sup-porting greater gs is negatively correlated with [CO2].Modelling these relationships suggests that increases inhydraulic conductance on the scale observed would sup-port several fold greater A at [CO2] of 280 ppm, but beof more modest benefit (13%) at [CO2] of 1000 ppm[26]. The interdependence of the reported variations in[CO2], vein density and stomatal characteristics is hardto determine. The change in vein density appears tohave significantly predated the decrease in [CO2]during the Cretaceous, as well as changes in stomatalcharacteristics (figure 1a,b) [34]. Uncertainties in thepalaeo-[CO2] record and dating of samples in fossilstudies could contribute to this disparity. This leavesopen the possibility that additional datawill demonstratethat the three events were coordinated, as understandingof physiology in extant species would lead us to expect.On the other hand, the majority of papers in this fieldopenly acknowledge that other environmental factorsalso varied during these key periods of plant evolutionand could have been the selective agent for adaptivetraits. A further possibility related to this scenario isthat greater hydraulic capacity was an exaptation [35]

http://rstb.royalsocietypublishing.org/

PalaeozoicO5000(a)

(b)

4000

3000

2000

1000

[CO

2] (

ppm

)re

lativ

e tr

ait v

alue

, cla

denu

mbe

r or

spe

cies

num

ber

1.0

0.8

0.6

0.4

0.2

0450 400 350 300 250

time (Myr)

200 100

GEOCARBSULFliverwortspalaeosolsmarine boronphytoplanktonstomatal indices

stomatal conductancevascular speciesstomatal densityleaf sizeC4 grass cladesvein density

B/Ca

150 50 0

0

S D C P Pa NTr J KMesozoic Cen.

Figure 1. Comparative time courses over most of the Phanerozoic of: (a) estimated atmospheric [CO2] predicted from a geo-chemical model of the carbon cycle (GEOCARBSULF, adapted from Berner [13]) and multiple proxies of [CO2] (stomatal

indices and isotope analysis of liverworts, palaeosols, marine boron, phytoplankton and B/Ca, updated from the compilation ofRoyer [21]), with a dashed line at 1000 ppm indicating the atmospheric [CO2] above which photosynthesis is saturated in mostmodern plants [22]; (b) estimated maximum stomatal conductance (adapted from Franks & Beerling [23]), estimated vascularspecies richness (adapted from Knoll & Niklas [24]), stomatal density (redrawn from Royer et al. [20]), Devonian andCarboniferous leaf size (adapted from Osborne et al. [25]), C4 grass clade richness (adapted from Edwards et al. [9]) andangiosperm vein density (adapted from Brodribb & Feild [26]), all of which are expressed relative to the maximum value inthe individual records of each parameter from the cited studies.

Review. Plant adaptation to atmospheric [CO2] A. D. B. Leakey & J. A. Lau 615


that developed under some other selective forcebut then resulted in further fitness gains when [CO2]subsequently decreased.

An important constraint on the role of varying [CO2]in driving plant evolution that has been rarely discussed isthe nonlinearity of [CO2] effects on plant performanceand the concentration at which [CO2] is saturating.A survey of a wide variety of modern plant species,including angiosperms and gymnosperms, woodyand herbaceous species, indicates that A is almostuniversally saturated at intercellular [CO2] of less than700 ppm, which corresponds to atmospheric [CO2] of1000 ppm [22]. Changes in leaf water use and energybalance associated with altered gs are also minimalabove 1000 ppm [36]. If this saturating [CO2] was main-tained across the course of plant evolutionary history, itwould set a threshold above which variations in [CO2]would have no consequence for plant physiology or fit-ness (figure 1). For example, in this scenario, the initialdecline in [CO2] during the Devonian from peakvalues of 15003000 ppm would have no direct effecton plants until dropping below 1000 ppm 350 Ma(figure 1a). However, the saturating [CO2] for earlyplants may have been much higher than for modern


species. This is partly because the resistance for CO2 topass through the epidermis was much greater, owing totheir large stomatal sizes and lower stomatal densities[23,29]. In addition, it is possible that cell wall and chlor-oplast envelope structures of ancient species reducedmesophyll conductance relative to modern species [37].Both of these factors would increase the saturating atmos-pheric [CO2] for photosynthesis [38,39] and, therefore,the threshold for atmospheric [CO2] effects on plant fit-ness. This alternative scenario is consistent with the factthat stomatal conductance and leaf size both started toincrease 390 Ma, when [CO2] estimates from a varietyof proxies ranged from 1500 to 3000 ppm. Establish-ing that the saturating [CO2] for early plants is thathigh will require further experimental and modellinganalysis. Further investigation of changes in mesophyllconductance and saturating [CO2] in plants from thelate Mesozoic and Cenozoic might also clarify the rolethat changes in [CO2] played in triggering the evolu-tion of high water use and photosynthetic capacityin angiosperms.

Plants with C4 photosynthesis appear to have emergedduring the most recent period of low [CO2] (less than1000 ppm), before becoming ecologically important in




many ecosystems 38 Ma [9]. Molecular clock datacalibrated using fossils suggest that the origins of all theknown C4 grass clades occurred between 3 and32 Ma, and it has been proposed that this was driven bya mid-Oligocene drop in [CO2] that favoured C4 speciesover C3 species because of their greater photosyntheticefficiency at low [CO2], particularly when combinedwith high temperatures and drought stress (figure 1a,b)[4042].Carbon isotope analysis of pollen grains suggeststhat a significant fraction (more than 25%) of grass specieswere C4 1533 Ma [43]. Isotopic evidence fromorganic matter in abyssal sediments of the AtlanticOcean were initially proposed to suggest that C4 speciesmight have originated earlier, as much as 90 Ma, butcould alternatively be explained by expansion of marinearchaea at that time [44,45]. Crassulacean acid metab-olism (CAM) photosynthesis is another adaptation thatinvolves a CO2 concentrating mechanism to maximizewater-use efficiency, and it is estimated to have alsoevolved in the orchids during the period of relatively low[CO2] (65 Ma) [46], while the cacti diverged fromtheir closest relatives 35 Ma [47].

In summary, there are several lines of evidence that fall-ing and low [CO2] creates selective pressure for two majorclasses of adaptation. First, adaptations to acquire and usewater in exchange for [CO2] (smaller stomata, greater sto-matal density, megaphyll leaves and greater vein density),which were presumably restricted to plants existing inmesic environments. Second, adaptations for CO2 con-centrating mechanisms that increase photosyntheticefficiency and maximize water-use efficiency (C4 andCAM photosynthesis), which were presumably favouredin hot and dry environments. The potential strength oflow [CO2] as an agent of selection on plant traits hasbeen highlighted by work on plants growing 855 Ka,which included a glacial period with very low [CO2](180220 ppm). Isotopic evidence indicates that theratio of intercellular [CO2] to atmospheric [CO2] wassimilar to that observed in modern plants, and thereforeglacial trees were operating close to the photosyntheticCO2 compensation point where carbon starvation isexperienced [48]. In contrast, periods of rising or high[CO2] have not widely been proposed to drive majorevents in plant evolution. The relative rarity of major evol-utionary events during periods of high [CO2] is unlikely toresult from the absence of genetic variation in plant sensi-tivity to high [CO2], or heritability of key traits controllingplant response to [CO2] (see Question 3) [49]. Alterna-tively, it might reflect that fitness and selection weremore strongly driven by genetic variation in plantresponses to other environmental influences when[CO2] was high and imposing little or no limitation onphotosynthesis and productivity.

3. QUESTION 2: HAVE PLANTS EVOLVEDBY NATURAL OR ARTIFICIAL SELECTIONIN RESPONSE TO CONTEMPORARYINCREASES IN [CO2]?Anthropogenically driven global environmental changesince the mid-twentieth century has been detectableagainst the background variability in climate andatmospheric composition [50]. In addition, biologicalresponses to global environmental change are detectable


in both natural and agricultural ecosystems [5154].Included in these biological responses are evolutionarychanges across a range of taxa in response to air pollu-tants, drought and temperature [5558]. However,there is still no unequivocal evidence that plants haveevolved in response to contemporary increases in[CO2]. Stomatal density and stomatal size of ninediverse Floridian plant species have changed over thelast 150 years, causing a decrease in maximal gs as[CO2] has risen from 290 to 390 ppm [51]. However,these changes were interpreted as being driven by anacclimation response, not genetic changes. This is con-sistent with numerous other studies on diverse taxa,finding that physiological adjustments play moresignificant roles than evolutionary responses to recentenvironmental change [59].

It is possible that crop breeding programmes will haveincidentally selected for genotypes with improvedresponsiveness to elevated [CO2]. This possibility hasnot been intensively studied, but the available evidencesuggests that it is not the case. In fact, the opposite scen-ario where [CO2] responsiveness has been selectedagainst may have occurred. Two experimental compari-sons of wheat genotypes released at different dates overthe nineteenth and twentieth centuries both showed thatstimulation of yield by elevated [CO2] predicted formid- to late twenty-first century compared with ambientor pre-industrial [CO2] was greater in genotypes withearlier release dates (figure 2) [60,61]). There is alsono evidence for a greater CO2-fertilization effecton yield of more recently released soybean genotypesin the US Department of Agriculture germplasm collec-tion (R. Nelson & E. A. Ainsworth 2011, unpublisheddata). These findings have suggested that optimizingthe performance of crops under [CO2] today and inthe decades to come will not happen incidentally.Accordingly, high yield under the elevated [CO2] pre-dicted in the future needs to be included as a target incrop breeding and biotechnology programmes [1,62].

4. QUESTION 3: WILL RISING [CO2] DRIVENATURAL SELECTION IN THE FUTURE,AND IF SO HOW?While palaeoeological studies have implied evolutionaryresponses to past changes in atmospheric [CO2], recentquantitative genetic and selection experiments havetested whether predicted future elevated [CO2] con-centrations will cause further evolutionary change[6371]. The evolutionary effects of rising atmospheric[CO2] are likely to be fundamentally different fromevolutionary effects of other types of anthropogenicenvironmental change because the rise in [CO2]occurs almost uniformly across the globe. Whereas evol-ution in response to other types of global change, suchas global warming, may be facilitated by spatial variation(e.g. populations from warmer regions may possessgenes that facilitate adaptation to warming climates),[CO2] does not vary substantially across speciesranges, and, therefore, little genetic differentiation in[CO2] responsiveness among populations across aspecies range is expected. Still, because the traitsthat mediate [CO2] responsiveness are influenced by awide variety of abiotic environmental conditions that


35(a) (b) (c) (d)

400 600 800 400 600 800 400 600 800 400 600 800

25

15

yiel

d (g

pla

nt1

)

growth [CO2] (ppm)

10

5

0

20

30

Figure 2. Seed yield as a function of growth at [CO2] ranging from subambient (293 ppm) to ambient (385 ppm) and elevated[CO2] (715 ppm) for four different Spring wheat lines released in (a) 1903 (Marquis), (b) 1921 (Thatcher), (c) 1965 (Chris)and (d) 1996 (Oxen). Treatment means are adapted from Ziska et al. [60].



vary both spatially and temporally (e.g. drought andtemperature), genetic variation for the physiologi-cal traits underlying [CO2] response may exist in bothnatural [72] and crop species [73,74]. In this section,we review quantitative genetic studies that test howelevated [CO2] will affect predicted future evolutio-nary trajectories, what traits are likely to change inresponse to further increases in [CO2], and constraintson adaptation to future, elevated [CO2]. In addition,we highlight some of the challenges to predictingevolutionary responses to future increases in [CO2].

Evolutionary responses depend on: (i) selection, orwhether elevated [CO2] alters the relationship betweenplant traits and plant fitness; and (ii) heritability andgenetic covariance, or whether trait and fitnessresponses to [CO2] are passed to subsequent gener-ations. Two approaches have been employed tounderstand how varying [CO2] will influence plant evol-ution. The first approach uses selection in controlledenvironments experiments (sensu [75] e.g. [68,71]).Replicated plant populations are grown for multiplegenerations under ambient [CO2] or elevated [CO2]predicted for mid- to late twenty-first century. Offspringfrom populations that had evolved under ambient[CO2] conditions versus elevated [CO2] conditions arethen compared, ideally in both ambient [CO2] and elev-ated [CO2] environments. Any divergence betweenpopulations can be attributed to genetic changes inplant traits in response to the [CO2] environment, pro-vided that maternal environmental effects are controlledfor. Increased fitness of populations that had evolvedunder elevated [CO2] conditions compared with popu-lations evolved under ambient [CO2] conditions whengrown in elevated [CO2] environments is evidencefor adaptation to elevated [CO2]. The second approachemploys quantitative genetics to compare predictedevolution in ambient [CO2] versus elevated [CO2]environments [6365,69]. This approach involves esti-mating components of the evolutionary process(selection, heritability and/or genetic covariances) onplant populations grown in ambient [CO2] or elevated


[CO2]. The advantage of the selection in a controlledenvironment approach is that it specifically tests forwhether an evolutionary response occurs; however,the mechanisms underlying the response cannot beidentified. The advantage of the quantitative geneticapproach is that it identifies how the mechanismsof evolutionary change (altered patterns of natural selec-tion, heritabilities or genetic covariances between traits)are affected by [CO2] and also can identify specifictraits underlying adaptation to elevated [CO2] (seeQuestion 4). Predicting the effects of [CO2] on long-term evolutionary change with this method, however,is complicated by assumptions that heritabilities andcovariances remain constant over time [76].

Most studies that employ the selection in controlledenvironments approach find limited evidence thatplants adapt to elevated [CO2], even though geneticchanges in plant traits are often observed. Potvin &Tousignant [68] simultaneously manipulated [CO2]concentration and temperature to simulate futureenvironmental conditions by increasing [CO2] concen-trations from 370 to 650 ppm and temperature from208C to 23.68C over seven generations. They detectedlittle evidence that populations of Brassica junceaadapted to simulated future environments. Althoughhalf of the 14 traits measured had diverged betweenpopulations that had evolved under present versusfuture environmental conditions, only one measuredtrait showed an adaptive response, and no fitnessmeasures showed a pattern of local adaptation. Simi-larly, Ward et al. [71] isolated the effects of [CO2] byartificially selecting on fecundity in replicate Arabidopsisthaliana populations grown under subambient [CO2](200 ppm) and elevated [CO2] (700 ppm) for five gen-erations. They found that subambient populations hadadapted to low [CO2] and produced more seed thanlines selected under elevated [CO2] when grown at200 ppm [CO2]; however, elevated [CO2] populationshad not adapted to elevated [CO2]populations thathad evolved under 200 ppm and 700 ppm did notdiffer significantly in seed production in elevated




[CO2] environments. Moreover, elevated [CO2] selec-tion lines actually increased biomass less in responseto elevated [CO2] compared with control lines thatwere not artificially selected, suggesting that the biomassincreases commonly observed in response to elevated[CO2] in single generation studies will not be increasedby evolutionary change. Finally, Collins & Bell [77]used the model organism Chlamydomonas to investigateevolutionary responses over 1000 generations in eitherconstant [CO2] environments (430 ppm) or steadilyincreasing [CO2] (4301050 ppm). As with the earliermentioned studies, no Chlamydomonas populationsshowed evidence for adaptation to elevated [CO2]even though changes in photosynthesis and respirationrates occurred. These changes reduced the fitness ofpopulations evolved under elevated [CO2] when theywere grown in lower [CO2] environments but did notaffect fitness when they were grown in elevated [CO2].Similar results were observed in natural Chlamydomonaspopulations found in CO2 springs [78]. Collins & Bellattribute their findings to the fixation of conditio-nally neutral mutations in the carbon concentrationmechanismthese changes had no effect when [CO2]was saturating, but reduced growth and fitness when[CO2] was limiting. Together, these studies suggestthat genetic changes may occur in response to eleva-ted [CO2], but that these changes do not necessarilyresult in increased fitness or productivity in elevated[CO2] environments.

Similar to the results observed from selection incontrolled environment studies, quantitative geneticexperiments also find little evidence that elevated[CO2] has large effects on predicted evolutionary tra-jectories. For example, Lau et al. [64] failed to detectevidence that elevated [CO2] alters patterns of natu-ral selection on plant traits, heritabilities or geneticcovariances, despite employing a statistically powerful,well-replicated experiment on a highly variable popu-lation. Both Steinger et al. [69] and Bazzaz et al. [63]found evidence that elevated [CO2] alters patterns ofnatural selection and/or heritabilities; however, effectsof [CO2] on evolutionary processes typically weresmall in magnitude. All of these studies focus pri-marily on down-stream traits, such as phenologyand growth. Selection acting on physiological traits israrely measured, in part because of the difficulty ofmeasuring physiological traits on the hundreds orthousands of individuals necessary for rigorous quanti-tative genetics analyses. This is unfortunate given that:(i) physiological traits might be expected to respondmost strongly to elevated [CO2]; (ii) several studieshave shown that [CO2] alters phenotypic integrationand the trade-offs between plant traits [70,79]; and(iii) results from selection in controlled environmentstudies suggest genetic changes in physiological traitsin response to variation in [CO2] [77,80,81].

Together, the selection in controlled environmentstudies and quantitative genetic studies conducted todate indicate that adaptive evolutionary responsesto elevated [CO2] will be weak relative to ecologicaland physiological responses. The lack of evidence forstrong evolutionary responses is surprising, given thelarge effects of elevated [CO2] on plant phenotypes.However, it is consistent with mixed results from


studies of populations growing along gradients of[CO2] at natural CO2 springs. Many studies fail tofind evidence for adaptation to elevated [CO2] or evi-dence for genetic changes towards increasedproductivity, even though reduced allocation to photo-synthetic apparatus is sometimes observed inpopulations with an evolutionary history of high[CO2] [78,82,83] (but see [84]). One recent examplein which adaptation to elevated [CO2] was observeddocumented genetic divergence between Plantagoasiatica populations growing near or far from naturalCO2 springs [83]. When reared in common environ-ments, genotypes collected from locations near thesprings (greater [CO2]) had lower photosyntheticcapacity and gs compared with genotypes far awayfrom the springs (lesser [CO2]), but also had greatershoot-to-root ratios and achieved greater productivity.Interestingly, the study populations included in theseexperiments had experienced [CO2] concentrationsranging from 380 to 5338 mmol mol21. Genetic differ-ences in phenotypic traits were observed betweenpopulations that experienced [CO2] between 380and 1044 ppm. In contrast, there was no additionaldifferentiation between the populations experiencing[CO2] of 1044 and 5338 ppm. This is consistentwith the idea presented in Question 1 that [CO2] willcease to be an agent of selection above the [CO2](approx. 1000 ppm) at which physiological responsesare saturated. Moreover, the physiological and growthstimulation effects of elevated [CO2] begin to attenuateat [CO2] even lower than 1000 ppm, potentiallyexplaining the minimal evolutionary effects of elevated[CO2] but larger evolutionary responses to subambient[CO2] [3,83].

Although most quantitative genetic studies have beenconducted in relatively simplistic growth chamber andgreenhouse environments where both abiotic andbiotic stressors are absent, some studies suggest thatthe evolutionary effects of [CO2] may be heightened inthe presence of competitors or herbivores [63,65].The effects of [CO2] in more complex communitiesmay result through two processes. First, if elevated[CO2] alters the intensity or likelihood of biotic inter-actions and biotic interactions are strong agents ofnatural selection, then elevated [CO2] may alter evol-ution when those interactors are present, even ifelevated [CO2] has minimal direct effects on evolution-ary processes. For example, if a plant is grown in thepresence of competitors and elevated [CO2] altersthe outcome of competition because species varyin the magnitude of their growth response to [CO2],then the strength of competition as a selective agentmay be altered. Lau et al. [65] provide empirical evi-dence in support of this mechanism; elevated [CO2]reduces the fitness effects of competition on A. thaliana.Because competition is a strong agent of selection onA. thaliana size traits, elevated [CO2] minimizes theselective effects of competition, and differences in pat-terns of natural selection are observed betweenpopulations grown in ambient [CO2] versus elevated[CO2] environments when competitors are present,even though [CO2] has no direct effects on predictedevolutionary trajectories in the absence of competitors.Second, elevated [CO2] may alter evolutionary process




if elevated [CO2] affects the expression of traits thatmediate interactions with other species. For example,Vannette & Hunter [85] find that five genotypes ofAsclepias syrica respond differently to elevated [CO2]in terms of the expression of defensive chemicalsbut not growth or reproductive traits. If herbivoresnegatively impact plant fitness, then the effects ofelevated [CO2] on defence trait expression could trans-late into differential effects on plant fitness whenherbivores are abundant, even though elevated [CO2]is unlikely to affect evolutionary processes whenherbivores are absent.

In sum, the available evidence to date suggests thatevolutionary responses to elevated [CO2] will not beconsistently observed or large in magnitude relativeto ecological and physiological responses. This isdespite substantial evidence indicating that subambi-ent [CO2] concentrations are an important selectiveagent [11,71], potentially responsible for large evol-utionary changes in a wide variety of plant traitsand even the diversification of vascular plants (seeQuestion 1). Most studies to date, however, havebeen conducted in relatively simplistic environmentalconditions, where biotic and abiotic stress is minimal.Given that some evolutionary effects have beenobserved or are predicted when plants experiencecompetition [63,65,69] or herbivory [85,86], evol-utionary effects may be more likely in more stressfulbiotic environments. Similarly, evolutionary effects ofelevated [CO2] also may be more likely when plantsexperience abiotic stress, such as drought or nutrientlimitation. Under stressful environments, it is possiblethat the genetic changes in physiological traitsobserved in numerous studies may change frombeing conditionally neutral to beneficial, therebyresulting in differential effects on growth and fitness.Few studies, however, have investigated evolutionaryconsequences of rising atmospheric [CO2] in subopti-mal environmental conditions. Given that temperatureand potentially drought stress will increase simul-taneously with [CO2], such studies are needed toidentify evolutionary effects and traits under selectionin future environments.

5. QUESTION 4: WHAT TRAITS ARE FAVOUREDUNDER HIGH [CO2]?Physiological, palaeoecological and quantitative gen-etics experiments suggest that leaf and photosynthetictraits are responsive to [CO2] and, therefore, may playa key role in mediating adaptive evolutionary responsesto elevated [CO2]. Moreover, both inter- and intraspeci-fic comparisons reveal variations in [CO2] response(reviewed by Poorter & Navas [87], see table 1 in Lauet al. [64]). Still, we have a rather limited understandingof which plant traits are most likely to produce higheryields or increased fitness in the elevated [CO2] environ-ments predicted for the future, as well as which specificcombinations of traits are necessary for strong growthenhancement responses to elevated [CO2] [88]. Therecent advent of trait-based approaches and associatedmulti-species trait datasets, combined with intraspecificcomparisons and genetic or phenotypic manipulationsof traits, may provide improved methods to more


thoroughly understand the traits underlying [CO2]responsiveness and to predict which genotypes andspecies will be favoured in an elevated [CO2] world.

To date, most studies attempting to identify traitsunderlying variation in [CO2] response have usedinterspecific comparisons and have focused on differ-ences among broad functional groups (e.g. C3 forbs,C3 grasses, C4 grasses and legumes). Such studiestypically find that C3 species have greater responsesto elevated [CO2] than C4 species and that legumeshave relatively higher CO2 responses than non-legumes, provided that other nutrients are not limiting[89,90]. Fewer studies have focused on more specificgrowth, phenological or physiological traits. Traitsunderlying adaptation to elevated [CO2] may be ident-ified by growing species under ambient or elevated[CO2] conditions and correlating phenotypic traitswith the growth stimulation effects of elevated [CO2]or fitness/yield in elevated [CO2] environments.In one such study, Atkin et al. [91] compared thegrowth stimulation effects of elevated [CO2] on tenAcacia species that varied in relative growth rate(RGR), and found that fast-growing species (greaterRGR) responded more to elevated [CO2] than slow-growing species. However, later studies on the sameset of Acacia species found no relationship betweenRGR or leaf traits (specific foliage area) and CO2response [92]. Furthermore, studies on other systemshave found the opposite pattern [93]. Such empiricalstudies have the advantage of manipulating [CO2] onmultiple species grown in common environmentalconditions, but are limited by the small number oftaxa that can be considered in any one experiment.In addition, manipulating [CO2] in a single environ-ment may be problematic given that environmentalconditions affect [CO2] responses and these environ-mental effects may vary across genotypes. In thestudy by Atkin et al., for example, the [CO2] manipu-lation took place under optimal nutrient and waterconditions, an environment that favours the fast-growing species. Slow-growing species typically inhabitmore stressful environments, and a very differentfinding may have resulted if the experimental environ-ment more closely matched environments to whichslow-growing species were adapted.

More recently, meta-analyses have been conductedon data from hundreds of existing empirical studiesto look for broad patterns in [CO2] response acrosstaxonomic scales. Poorter & Navas [87] conducted ameta-analysis on 350 different experiments that exam-ined the growth stimulation effects of elevated [CO2]on 350 different plant species. Surprisingly, only18 per cent of the variation in growth response wasexplained by species, possibly because CO2 stimulationeffects also depend on ontogeny, environment andintraspecific variation. Still, the meta-analysis confirmedexpectations: C3 plants exhibited the strongest growthresponse to elevated [CO2]; C4 plants showed the smal-lest growth response; and CAM plants demonstratedintermediate responses. These functional group classifi-cations based on photosynthetic mechanisms explainedonly 10 per cent of the variation among species in CO2response. This may reflect the importance of ontogeny,environment and genotypes as mentioned earlier, but




could also indicate that other traits play strong rolesin mediating productivity responses to [CO2]. Forexample, among C3 forbs, fast growers respondedmore strongly to elevated [CO2] than slow growers anddifferences were also observed between dicots andmonocots as well as legumes versus non-legumes [87].Other meta-analyses and reviews have focused onother traits. Kerstiens [94] found that biomass responsesto elevated [CO2] were greater for shade-tolerant speciesthan shade-intolerant species. Similarly, Niinements[39] observed that plants with more robust leaves (i.e.evergreen schlerophylls) responded more positively toelevated [CO2] than species with less robust leaves,emphasizing the importance of mesophyll conductanceand its role in influencing [CO2] supply to chloroplasts.

Similar to the interspecific comparisons describedearlier, intraspecific variation in plant traits also canbe correlated with growth responses to elevated[CO2] or high fitness/yield in elevated [CO2] environ-ments. A study measuring RGR on 29 Picea glaucagenotypes found no association between RGR andstimulation of productivity at elevated [CO2] [95].Moreover, P. glauca growth at elevated [CO2] wastightly correlated with growth at ambient [CO2]. Inother words, the most productive genotypes in ambi-ent [CO2] were also the most productive genotypes/highest yielders in elevated [CO2]. This result wasused to argue that the association between RGR andstimulation of productivity by elevated [CO2] observedin prior interspecific comparisons was due to traitscorrelated with both RGR and enhancement of pro-ductivity by elevated [CO2], rather than a directrelationship between RGR and enhancement of pro-ductivity by elevated [CO2]. Similarly, Liu et al. [96]observed that four provenances of Populus tremuloidesexhibited different responses to elevated [CO2] interms of gs and transpiration rate, but these dif-ferences did not translate into differences among theprovenances in biomass response to elevated [CO2].

Conclusions reached from both intra- and inter-specific approaches illustrate an important challengeto identifying traits associated with adaptation to elev-ated [CO2] environments. To pinpoint exactly whattraits are responsible for high growth responses to elev-ated [CO2] or high fitness/yield in elevated [CO2]environments, all relevant traits must be measured. Ifall relevant traits are included in the regressionmodel, then the particular traits responsible for adap-tation to elevated [CO2] can be identified. In anintraspecific context, this is essentially what is donein the phenotypic selection analysis approach devel-oped by Lande & Arnold [97]. This approach usesmultiple regression to account for correlations amongtraits. As a result, it can differentiate between traitsthat are directly associated with fitness versustraits that are indirectly associated with fitness due tocorrelations with other phenotypic traits. It should benoted, however, that there are several statistical andbiological challenges to be dealt with when applyingthis approach (summarized by Mitchell-Olds & Shaw[98]), including issues with identifying traits directlyunder selection due to correlations between measuredtraits and unmeasured traits that are under selection.In the selection analyses in ambient [CO2] versus


elevated [CO2] environments conducted to date, how-ever, there is little evidence suggesting that differenttraits are associated with high fitness in ambient[CO2] versus elevated [CO2] [64]. It is important tonote, however, that most such studies focus ongrowth or phenological traits rather than on physio-logical traits. Similarly, comparisons amongrecombinant inbred lines (as suggested by Zhanget al. [95]), mutant-wild-type comparisons [99] andexperimental manipulations of phenotypic traits[100] also may effectively identify traits involved inCO2 response because focal traits segregate indepen-dently of genetic background. Alternatively,modelling approaches [101] can be used to predictwhich traits are important to fitness/yields in futureenvironments, identifying focal traits for further inves-tigation in empirical studies.

Similar approaches can be applied to interspeci-fic comparisons [102]. While previous interspecificcomparisons were largely limited to coarse-scale com-parisons among functional groups or to studies onfew species in a single clade, the advent of new traitdatabases, which include physiological, as well as mor-phological and phenological traits, such as TRY [103],may allow for robust multi-trait analyses on the hun-dreds of species for which [CO2] responses have beenmeasured. Although such trait-based approaches havenot yet been used to investigate traits underlying[CO2] response, they have been employed to predictchanges in community composition in responseto other anthropogenic environmental changes (e.g.habitat fragmentation [104]).

A challenge to all studies focused on finding traits orgenes underlying adaptation to elevated [CO2] is thatmultiple traits are probably involved. For example,interspecific variation in photosynthetic efficiency isnot due to one trait, but is instead a result of severalphysiological traits each with relatively small effects[105]. Moreover, these traits are often correlated andmay have synergistic effects [72]. For example, hypothe-tically, having low Amass and high Nmass is expected toyield very unfit genotypes because the high cost of main-taining Rubisco would exceed carbon gains [72]. Manysuch correlations are probably maintained by selection,suggesting that the correlations may change in responseto novel environmental conditions. Although it wouldrequire large sample sizes and many trait measurements,in the quantitative genetics framework, correlatio-nal selection studies could be used to identify how theadaptive value of one trait depends on other traitvalues. Similar approaches could be applied to inter-specific trait datasets. Alternatively, cluster analyses ofmorpho-physiological diversity may be used to identifysuites of correlated traits underlying adaptation [102].Cluster analyses have been used to identify traits andpotential diversity for improved agronomic breeding tonovel stressors, such as drought tolerance [106].

The trait-based approaches described earlier may bepowerful tools for identifying individual traits and suitesof correlated traits underlying high fitness in future[CO2]. Ultimately, suites of traits with synergisticinteractions are likely to contribute to high fitness innovel environments. As a result, large datasets consist-ing of many traits for each of many genotypes or




species are necessary. Prior palaeoecological studieson evolutionary responses to variation in [CO2] con-centration have identified several potential traits thatmay be involved in [CO2] responses, and modellingapproaches may identify many more. Improvements inboth the number of traits and number of taxa includedin existing trait datasets, as well as identification oflikely adaptive traits through modelling or knowledgeof physiological responses, may allow for rapid advancesin identifying the physiological, phenological and mor-phological traits that may lead to increased growthand fitness in future, elevated [CO2] environments.Moreover, crosses between distinct genotypes thatdiffer in entire suites of traits may yield novel genotypesand trait combinations that could contribute to fitnessand yield improvement in ambient [CO2] versuselevated [CO2].

6. QUESTION 5: HOW DOES EVOLUTIONARYHISTORY IMPACT AND INFORM EFFORTSTO ENGINEER CROPS FOR IMPROVEDPERFORMANCE IN PRESENT ANDFUTURE [CO2]?Circumventing the limitations of current low [CO2] tophotosynthesis in many of our main crop plants is akey target for biotechnological efforts to improvecrop productivity [107110]. In addition, there isemerging recognition that intervention to adapt cropsby breeding or biotechnology for optimal performancein elevated [CO2] is needed [111]. In response, targetsfor selection are beginning to be identified [1], andapproaches that account for the challenges of geneticand trait-based approaches described earlier (Ques-tions 3 and 4) have been proposed [62]. Here, wereview present targets for crop improvement relatedto present and future [CO2].

The enzyme responsible for photosynthetic fixationof CO2 in all plants, Rubisco (RibUlose-1,5-BISpho-sphate Carboxylase Oxygenase), is fundamentallyinefficient. This stems from a relatively low CO2 affi-nity and low carboxylation reaction catalytic rate,which plants compensate for by synthesizing verylarge quantities of the enzyme, at the expense of avery large fraction of their leaf nitrogen. In addition,a significant fraction of reactions catalysed by Rubiscoresult in oxygenation rather than carboxylation ofRuBP (RibUlose-1,5-BisPhosphate). Recycling of thetoxic 2PG (2-PhosphoGlycolate) that is one of theproducts of the oxygenation reaction is achieved bythe photorespiratory pathway, but at the expense ofenergy, carbon and nitrogen [109]. Efforts to circum-vent these inefficiencies fall into three categories. First,engineering of the Rubisco to improve its enzymaticperformance. Second, engineering of CO2 concentrat-ing mechanisms to saturate the carboxylation reactionand suppress the oxygenation reaction. Third, modifi-cations to the photorespiratory pathway that reducelosses of carbon, nitrogen and energy.

There has been considerable selective pressure forthe evolution of more efficient Rubisco throughout thehistory of land plants. During the periods of sub-saturating [CO2], whichat a minimuminclude50 Myr in the Carboniferous/Permian and the last


30 Myr (figure 1a), modifications leading to greater[CO2] specificity, greater catalytic rate or impairedoxygenation without deleterious side-effects couldhave increased photosynthetic efficiency [110]. Duringperiods of saturating [CO2], modifications leading togreater catalytic rate without deleterious side-effectscould have increased photosynthetic efficiency. In eachcase, greater photosynthetic efficiency would resultin greater carbon gain for the same investment inresources, or equivalent carbon gain but with greaterwater-use efficiency and nitrogen-use efficiency. Never-theless, it appears that evolution of Rubisco has beenconstrained in a fundamental manner as it remains thelimiting step in metabolism for most modern C3plants in many growing conditions.

The factors that have constrained the evolution andengineering of improved Rubisco up until now arebecoming better understood [110]. There are threedifferent clades of Rubisco [112]. Clade 1 includesall vascular plants along with cyanobacteria andsome algae and proteobacteria. Clade 2 is found inchemoautotrophs, dinoflagellate algae and some pro-teobacteria. Clade 3 is exclusive to the archaea. Theyall probably share a common ancestor, which was amethanogenic archaea [112]. Even with considerablevariation in amino acid sequence and biological func-tion, key active site residues are conserved across thethree clades [113]. This has resulted in a shared acti-vation process and catalytic chemistry that suggeststhat there are considerable constraints upon modifi-cation of enzyme function [114]. One fundamentalissue may be that CO2 directly binds to the RuBPenediol, rather than forming a Michaelis complexwith Rubisco, which reduces the capacity for discri-mination against O2 binding [115]. An importanttrade-off exists in which Rubiscos with greater speci-ficity for CO2 relative to O2 have lower catalytic rates(figure 3) [116,117]. Modelling the effects of variationin Rubisco specificity on canopy photosynthesis whileaccounting for the constraints of this relationship indi-cated that the Rubisco specificity and catalytic rate ofmodern C3 plants is optimal for [CO2] of approxi-mately 200 ppm [116]. This may reflect adaptationof Rubisco for optimal carbon gain at the average[CO2] of the last 400 000 years. If so, Rubisco evol-ution has not kept pace with anthropogenic [CO2]rise, and will become increasingly maladapted overthe course of this century. In addition, the evolution-ary changes that have occurred could be consideredfine-scale tuning of Rubisco without solving its morefundamental inefficiencies. Transgenic Rubiscos havebeen produced that break the trade-off between speci-ficity and catalytic rate, but typically the results havebeen reduced rather than improved enzyme perform-ance [116]. However, the modelling analysis doesemphasize that among the natural diversity of Rubisco,there are enzymes with more favourable characteristicsthan currently found in C3 plants. Improved under-standing of the expression and assembly of Rubiscois necessary to allow the expression of non-nativeRubisco in higher plants. The process is complicatedbecause Rubisco in higher plants is a complex offour dimers of a large subunit encoded in the plastidgenome plus eight small subunits encoded in the nuclear


100

100 200 300

CO2 concentration (mmol mol1)

C2C1

2

1

400 500 600 700

90

80

70

60

50

optim

al s

peci

fici

ty f

acto

r,

Figure 3. Assuming a fixed number of Rubisco active sites

per unit leaf area and the dependence of catalytic rate peractive site kcc on specificity described for different photo-synthetic organisms by Zhu et al. [116], the line shows, forany given atmospheric [CO2], the specificity (t) that willgive the highest light-saturated rate of leaf photosynthetic

CO2 uptake (Asat). The average t for terrestrial C3 cropplants (92.5) is indicated (t1) together with the interpolatedatmospheric [CO2] at which it would yield the maximumAsat (C1). Point t2 is the specificity that would yield the high-est Asat at the current [CO2] of the atmosphere (C2). At C2,decrease in t from present average (t1) to the optimum forcurrent [CO2] (t2) can increase light-saturated leaf photo-synthetic carbon uptake by 12%. Reproduced withpermission from Zhu et al. [116].



genome [110]. Progress has been made in using plas-tome transformation to replace tobacco Rubisco withbacterial and archaeal Rubiscos [118,119]. However,attempts to express more efficient red algal andsunflower Rubiscos in tobacco have failed owing todifferences in their requirements for protein foldingand assembly [110]. An additional molecular constraintto the modification of Rubisco is the need for sugar-phosphate inhibitors bound to its active site to beremoved during an interaction with Rubisco activase.The identity of residues required for successful inter-action with Rubisco activase has been proposed [120],but awaits experimental confirmation [110]. Theabsence of regulation by Rubisco activase and colocaliza-tion of genes for both Rubisco subunits and chaperoneproteins on the plastome may explain why evolution ofRubisco appears to have progressed further in redalgae than in higher plants [121].

While Rubisco engineering attempts to overcomethe limitations of natural evolution, an alternative strat-egy is derived from knowledge of successful naturalevolutionary responses in which a CO2 concentratingmechanism overcomes the CO2-limitation of photosyn-thesis. Higher plants with C4 and CAM photosynthesis,as well as cyanobacteria with carboxysomes and algaewith pyrenoids, all achieve efficient photosynthesis byconcentrating CO2 around Rubisco in order to stimu-late carboxylation and inhibit oxygenation. Effortshave begun to engineer these traits in C3 plants as ameans to increase productivity and yield.

C4 photosynthesis is thought to have evolved asan adaptation to limit photorespiration at times ofsub-saturating [CO2] [42]. Successful conversion ofrice from a C3 plant to a C4 plant would likely increase


A, water-use efficiency and nitrogen-use efficiency,especially in hotter and drier environments [107].Achieving this goal will be challenging because C4photosynthesis is a highly polygenic trait. Accordingly,it will require not just the expression of genes encodingthe enzymes of the CO2 concentrating mechanism, butalso engineering Kranz anatomy and transporters sup-porting flux between mesophyll and bundle sheathcompartments [122]. This is reflected in a proposedevolutionary scheme involving a series of steps in theevolution from C3 to C4 photosynthesis: (i) generalpreconditioning, i.e. gene duplication; (ii) anatomicalpreconditioning, i.e. close veins; (iii) enhancement ofbundle sheath organelles; (iv) addition of photore-spiratory pump, including localization of glycinedecarboxylase to the bundle sheath; (v) enhancementof Phosphoenol pyruvate carboxylase activity;(vi) integration of components; and (vii) optimizationof components [42]. Linked to this knowledge, signi-ficant effort has recently focused on determiningthe transcriptional control of C4 leaf structural andmetabolic development [123]. The fact that C4 photo-synthesis has independently evolved in many geneticbackgrounds on different occasions increases the like-lihood that successful engineering can be achieved[107]. Specific evidence for this assertion includesthat: (i) independent lineages of C4 species sharecommon mechanisms controlling the localization ofkey enzymes for C4 photosynthesis in bundle sheathcells; and (ii) specific localization of enzymes in C4leaves to bundle sheath versus mesophyll cells can beachieved by modification of trans-factors without achange in existing cis-regulation of C3 species [124].

An effort is also beginning to engineer tobaccoplants with carboxysomes from cyanobacteria (http://www.nsf.gov/news/news_summ.jsp?cntn_id=119017).This could also enhance carbon gain, water-use effi-ciency and nitrogen-use efficiency. Carboxysomes arestructures where photosynthetic enzymes are localized,creating high [CO2]. Engineering carboxysomes intochloroplasts of crop plants would effectively create anevolutionary flashback to when cyanobacteria weresymbiotically recruited into host cells as the precursorsof chloroplasts. While the strategy requires conglom-eration of traits from now distantly related species,it benefits from being limited to manipulation ofmetabolism within individual cells.

An alternative in preventing photorespiratory lossesby the development of CO2 concentrating mecha-nisms is direct manipulation of the photorespiratorypathway in order to prevent or reduce the typicallosses of carbon, nitrogen and energy once the oxygen-tion reaction has produced 2PG [109]. Twoindependent approaches in achieving this goal haveinserted multi-enzyme pathways into plant chloro-lasts. The first fully oxidizes glycolate in thechloroplast into CO2 [125]. The second inserts a bac-terial pathway into the chloroplast that convertsglycolate into glycerate (producing some CO2 as aby-product), which can then be phosphorylated andre-incorporated into the Calvin cycle [126]. Bothmodifications are beneficial because they: (i) releaseCO2 in the chloroplast that can be refixed by Rubisco,potentially at higher concentrations; (ii) avoid release

http://www.nsf.gov/news/news_summ.jsp?cntn_id=119017http://www.nsf.gov/news/news_summ.jsp?cntn_id=119017http://www.nsf.gov/news/news_summ.jsp?cntn_id=119017http://rstb.royalsocietypublishing.org/



of ammonia and the energetic cost usually associatedwith its reassimilation; and (iii) produce additionalreducing equivalents in the chloroplast [109].Expression of the bacterial pathway in Arabidopsis ledto a 30 per cent stimulation of biomass production[127]. One potential problem that remains to betested is whether excess reducing equivalents will beproduced under high light conditions, as photorespira-tion can play an important role as an alternativeelectron sink during periods of stress that impairphotosynthetic quenching [109,128].

All of the approaches to enhancing photosynthesisdescribed above tackle the limitation to photosynthe-sis by current [CO2]. They address the urgent needto boost crop production in the face of growing foodinsecurity. However, looking further into the future,the continuing rise of [CO2] will gradually diminishthis limitation to photosynthesis and optimization ofcrop productivity will present a modified set of chal-lenges. The speed of anthropogenically driven [CO2]rise means forward thinking is particularly necessaryto optimize crops to their growth [CO2]. Based onthe evidence reviewed earlier, natural selection forimproved performance in elevated [CO2] is weak andthere is unlikely to have been incidental breedingfor improved performance at elevated [CO2] to date(Questions 2 and 3). However, understanding ofplant cellular, physiological and agronomic responsesto elevated [CO2] has allowed preliminary identifi-cation of targets for biotechnological improvement [1].

Future, elevated [CO2] will favour replacement ofRubisco in C3 crops with Rubisco that has lowerspecificity and greater catalytic rate (derived from C4species or algae) even more so than under presentconditions [116]. Some organisms are capable ofexpressing different Rubiscos whose characteristicsare tailored to variation in growth conditions [127].Engineering such a regulatory system into cropscould provide additional benefits, such as expressingdifferent Rubiscos in sun and shade leaves [116].

At elevated [CO2], A becomes limited by the capacityfor regeneration of RuBp [129]. Modelling suggeststhat allocation of greater nitrogen resources to enzy-mes involved in RuBp regeneration in the Calvincycle will stimulate photosynthesis at elevated [CO2][130]. Transgenic tobacco overexpressing one of theseenzymes, sedoheptulose-1,7-bisphosphatase, achievesenhanced photosynthesis and productivity [131].

C3 plants grown at elevated [CO2] consistentlyaccumulate substantially larger pools of carbohydratesin leaves and other tissues, even when grown withunlimited rooting volume in the field [132]. The pri-mary molecular response of soybean to growth atelevated [CO2] is transcriptional reprogramming of therespiratory pathway, allowing greater use of theadditional available assimilate [133]. Even though soy-bean undergoes this metabolic rewiring and is able tomatch greater C fixation with enhanced N assimilation[134], there is still significant accumulation of leafstarch for much of the growing season [133,135]. Thisimplies that greater enhancement of productivity at elev-ated [CO2] might be achieved by increasing utilizationof photoassimilate. Despite improved understanding ofhow carbohydrate status drives productivity [136,137],


further work is needed to determine how C utilizationis controlled by interactions among sink metabolism,photoassimilate transport capacity and energy demandfor photoassimilate export from source leaves [133,138141]. Given the projected increases in tempera-ture, drought and disease stress that crops willexperience due to global environmental change, greaterallocation of carbon resources to metabolites associatedwith stress tolerance could have multiple advantages [1].For example, greater production of osmolytes, such aspinotol, mannitol and raffinose, can provide protectionfrom dehydration under high temperature and droughtstress, while antioxidant metabolites, such as ascorbate,reduce oxidative damage from elevated ozone anddrought stress [142144]. Greater carbon resourcesare typically assumed to allow plants to invest greatresources in defence [145]. However, the changes inhormone signalling and secondary metabolism ofsoybean grown under elevated [CO2] provide an inter-esting exception to this rule. When grown at elevated[CO2] in the field, the inducible defence response ofsoybean to damage by Japanese beetle, induction of aprotease inhibitor that hinders the beetles digestive pro-cess, is impaired [146]. Changes in sugarhormoneinteractions are thought to underpin the responseand may provide another target for enhancing cropproduction in elevated [CO2].

Growth at elevated [CO2] increases nitrogen-useefficiency by stimulating A per unit leaf N and byallowing photosynthetic acclimation in which less Nis allocated to Rubisco, leaving greater N resourcesavailable for other processes including growth[129,147]. Despite these physiological changes, Navailability strongly limits the response of productivityto elevated [CO2] [132]. Legumes are able to achievelarge increases in yield and maintain tissue C : N ratiosunder elevated [CO2] because they can allocateadditional carbon to N-fixing nodules, providedother nutrients are not limiting [148150]. Engineer-ing other C3 crops with the capacity to fix N throughsymbiotic relationships with nodule-forming or endo-phytic microbes would allow them to benefit morefrom rising [CO2] and would be favoured by con-ditions of greater C availability. The biofuel crop,Miscanthus giganteus, shows the potential of suchan approach, as it was recently shown to achieve itsextremely high productivity by combining C4 photo-synthesis with N fixation by endophytic microbes[151]. A potentially deleterious coupling between inhi-bition of photorespiration at elevated [CO2] andimpairment of leaf N assimilation in Arabidopsis hasalso recently been proposed [152]. If this is the casein crop species under field conditions, and theresponse is not counteracted by greater root N assim-ilation, elucidation of the mechanism of responsecould yield a further target for biotechnologicalimprovement to optimize the coupling of carbon andnitrogen metabolism and maximize productivity.

7. CONCLUSIONThere are several lines of evidence that periods of fallingand low [CO2] in the palaeo-record created selectivepressure for two major classes of adaptation: (i)




adaptations to acquire and use water in exchange for[CO2], which were presumably restricted to plants exist-ing in mesic environments and (ii) adaptations for CO2concentrating mechanisms that increase photosyntheticefficiency and maximize water-use efficiency, whichwere presumably favoured in hot and dry environments.Nevertheless, while contemporary global environmentalchange is impacting many elements of plant biology,there is still no unequivocal evidence for plant adaptationto contemporary increases in [CO2]. This includes noevidence for incidental breeding of crop varieties toachieve greater yield enhancement from future [CO2].The studies of evolution in response to elevated [CO2]conducted to date applying selection in controlledenvironments, quantitative genetics and trait-basedapproaches suggest that the evolutionary responses ofnatural plant populations to future [CO2] will not beconsistent or strong relative to ecological and physio-logical responses. This lack of evidence for strongevolutionary effects is surprising given the large effectsof elevated [CO2] on plant phenotypes. Most selectionand quantitative genetics studies to date, however, havebeen conducted in relatively simplistic environmentalconditions, where biotic and abiotic stresses wereavoided. Under more stressful and complex fieldenvironments, it is possible that the genetic changes inphysiological traits observed in numerous studies maychange from conditionally neutral to beneficial, therebyresulting in differential effects on growth and fitness.Given that temperature and potentially drought stresswill increase simultaneously with [CO2], such studiesare needed to identify evolutionary effects and traitsunder selection in future environments. Improvementsin both the number of traits and number of taxa includedin existing trait datasets, as well as identification of likelyadaptive traits through modelling or knowledge of phys-iological responses, may allow for rapid advances inidentifying the physiological, phenological and morpho-logical traits that may lead to increased growth andfitness in future, elevated [CO2] environments. Already,efforts are underway to engineer plants to overcome pre-sent day [CO2] limitations to photosynthesis and carbongain. These include efforts to tackle those inefficienciesof Rubisco that natural selection has failed to overcome,as well as attempts to mimic the evolutionary successesof CO2 concentrating mechanisms and photorespiratoryshunts that allow enhanced carbon gain and greaterresource-use efficiency in some higher plants, algae andbacteria. Looking further into the future, the continuingrise of [CO2] will gradually diminish this limitation tophotosynthesis and optimization of crop productivitywill present a modified set of challenges. Methods totackle this challenge are available and fundamentalunderstanding of plant cellular and physiologicalresponses is improving such that targets for biotechnolo-gical optimization of crop performance under future[CO2] are being proposed and should be tested.

We thank David Beerling for the invitation to participate inthe workshop, CO2 and Plant Evolution, which wasfunded and hosted by the Royal Society at the KavliCentre, and led to the development of this manuscript. Wethank Dana Royer for providing a compilation ofPhanerozoic [CO2] estimates from proxy analysis. We also


thank Dana Royer, Colin Osborne and an anonymousreviewer for helpful comments and suggestions that greatlyimproved this manuscript. This is KBS publication no. 1594.

REFERENCES1 Ainsworth, E. A., Rogers, A. & Leakey, A. D. B. 2008

Targets for crop biotechnology in a future high-CO2and high-O3 world. Plant Physiol. 147, 1319. (doi:10.1104/pp.108.117101)

2 Beerling, D. J. 2009 Coevolution of photosyntheticorganisms and the environment. Geobiology 7, 9799.(doi:10.1111/j.1472-4669.2009.00196.x)

3 Ward, J. K. & Kelly, J. K. 2004 Scaling up evolutionary

responses to elevated CO2: lessons from Arabidopsis.Ecol. Lett. 7, 427440. (doi:10.1111/j.1461-0248.2004.00589.x)

4 Ward, J. K. & Strain, B. R. 1999 Elevated CO2 studies:

past, present and future. Tree Physiol. 19, 211220.(doi:10.1093/treephys/19.4-5.211)

5 Ackerly, D. D. et al. 2000 The evolution of plantecophysiological traits: recent advances and futuredirections. BioScience 50, 979995. (doi:10.1641/0006-3568(2000)050[0979:TEOPET]2.0.CO;2)

6 Ackerly, D. D. & Monson, R. K. 2003 Waking thesleeping giant: the evolutionary foundations ofplant function. Int. J. Plant Sci. 164, S1S6. (doi:10.1086/374729)

7 Reusch, T. B. H. & Wood, T. E. 2007 Molecular ecol-ogy of global change. Mol. Ecol. 16, 39733992.(doi:10.1111/j.1365-294X.2007.03454.x)

8 Beerling, D. J. 2005 Leaf evolution: gases, genes andgeochemistry. Ann. Bot. 96, 345352. (doi:10.1093/aob/mci186)

9 Edwards, E. J., Osborne, C. P., Stromberg, C. A. E.,Smith, S. A. and C4 Grasses Consortium. 2010 Theorigins of C4 grasslands: integrating evolutionary andecosystem science. Science 328, 587591. (doi:10.1126/science.1177216)

10 Feild, T. S. & Arens, N. C. 2005 Form, function andenvironments of the early angiosperms: mergingextant phylogeny and ecophysiology with fossils. NewPhytol. 166, 383408. (doi:10.1111/j.1469-8137.2005.01333.x)

11 Gerhart, L. M. & Ward, J. K. 2010 Plant responses tolow [CO2] of the past. New Phytol. 188, 674695.(doi:10.1111/j.1469-8137.2010.03441.x)

12 Willis, K. J. & McElwain, J. C. 2002 The evolution ofplants. Oxford, UK: Oxford University Press.

13 Berner, R. A. 2006 GEOCARBSULF: a combinedmodel for Phanerozoic atmospheric O2 and CO2.Geochim. Cosmochim. Acta 70, 56535664. (doi:10.1016/j.gca.2005.11.032)

14 Royer, D. L. 2001 Stomatal density and stomatal indexas indicators of paleoatmospheric CO2 concentration.Rev. Palaeobot. Palynol. 114, 128. (doi:10.1016/S0034-6667(00)00074-9)

15 Cerling, T. E. 1991 Carbon dioxide in the atmosphere-evidence from Cenozoic and Mesozoic paleosols.Am. J. Sci. 291, 377400. (doi:10.2475/ajs.291.4.377)

16 Fletcher, B. J., Brentnall, S. J., Anderson, C. W.,

Berner, R. A. & Beerling, D. J. 2008 Atmosphericcarbon dioxide linked with Mesozoic and early Ceno-zoic climate change. Nat. Geosci. 1, 4348. (doi:10.1038/ngeo.2007.29)

17 Freeman, K. H. & Hayes, J. M. 1992 Fractionation of

carbon isolopes by phytoplankton and estimates ofancient carbon dioxide levels. Glob. Biogeochem. Cycles6, 185198. (doi:10.1029/92GB00190)

http://dx.doi.org/10.1104/pp.108.117101http://dx.doi.org/10.1104/pp.108.117101http://dx.doi.org/10.1111/j.1472-4669.2009.00196.xhttp://dx.doi.org/10.1111/j.1461-0248.2004.00589.xhttp://dx.doi.org/10.1111/j.1461-0248.2004.00589.xhttp://dx.doi.org/10.1093/treephys/19.4-5.211http://dx.doi.org/10.1641/0006-3568(2000)050[0979:TEOPET]2.0.CO;2http://dx.doi.org/10.1641/0006-3568(2000)050[0979:TEOPET]2.0.CO;2http://dx.doi.org/10.1086/374729http://dx.doi.org/10.1086/374729http://dx.doi.org/10.1111/j.1365-294X.2007.03454.xhttp://dx.doi.org/10.1093/aob/mci186http://dx.doi.org/10.1093/aob/mci186http://dx.doi.org/10.1126/science.1177216http://dx.doi.org/10.1126/science.1177216http://dx.doi.org/10.1111/j.1469-8137.2005.01333.xhttp://dx.doi.org/10.1111/j.1469-8137.2005.01333.xhttp://dx.doi.org/10.1111/j.1469-8137.2010.03441.xhttp://dx.doi.org/10.1016/j.gca.2005.11.032http://dx.doi.org/10.1016/j.gca.2005.11.032http://dx.doi.org/10.1016/S0034-6667(00)00074-9http://dx.doi.org/10.1016/S0034-6667(00)00074-9http://dx.doi.org/10.2475/ajs.291.4.377http://dx.doi.org/10.1038/ngeo.2007.29http://dx.doi.org/10.1038/ngeo.2007.29http://dx.doi.org/10.1029/92GB00190http://rstb.royalsocietypublishing.org/



18 Pagani, M., Freeman, K. H. & Arthur, M. A. 1999 LateMiocene atmospheric CO2 concentrations and theexpansion of C4 grasses. Science 285, 876879.(doi:10.1126/science.285.5429.876)

19 Pagani, M., Zachos, J. C., Freeman, K. H., Tipple, B. &Bohaty, S. 2005 Marked decline in atmospheric carbondioxide concentrations during the Paleogene. Science309, 600603. (doi:10.1126/science.1110063)

20 Royer, D. L., Berner, R. A. & Beerling, D. J. 2001 Phaner-ozoic atmospheric CO2 change: evaluating geochemicaland paleobiological approaches. Earth Sci. Rev. 54,349392. (doi:10.1016/S0012-8252(00)00042-8)

21 Royer, D. 2006 CO2-forced climate thresholds duringthe Phanerozoic. Geochim. Cosmochim. Acta 70,56655675. (doi:10.1016/j.gca.2005.11.031)

22 Wullschleger, S. D. 1993 Biochemical limitations tocarbon assimilation in C3 plants: a retrospective analy-

sis of the A/Ci curves from 109 species. J. Exp. Bot. 44,907920. (doi:10.1093/jxb/44.5.907)

23 Franks, P. J. & Beerling, D. J. 2009 CO2-forcedevolution of plant gas exchange capacity and water-use efficiency over the Phanerozoic. Geobiology 7,227236. (doi:10.1111/j.1472-4669.2009.00193.x)

24 Knoll, A. H. & Niklas, K. J. 1987 Adaptation, plant evol-ution, and the fossil record. Rev. Palaeobot. Palynol. 50,127149. (doi:10.1016/0034-6667(87)90043-1)

25 Osborne, C. P., Beerling, D. J., Lomax, B. H. &

Chaloner, W. G. 2004 Biophysical constraints on theorigin of leaves inferred from the fossil record. Proc.Natl Acad. Sci. USA 101, 10 36010 362. (doi:10.1073/pnas.0402787101)

26 Brodribb, T. J. & Feild, T. S. 2010 Leaf hydraulicevolution led a surge in leaf photosynthetic capacityduring early angiosperm diversification. Ecol. Lett. 13,175183. (doi:10.1111/j.1461-0248.2009.01410.x)

27 Haworth, M., Elliott-Kingston, C. & McElwain, J. C.

2011 Stomatal control as a driver of plant evolution.J. Exp. Bot. 62, 24192423. (doi:10.1093/jxb/err086)

28 Beerling, D. J., Osborne, C. P. & Chaloner, W. G. 2001Evolution of leaf-form in land plants linked to atmos-pheric CO2 decline in the Late Palaeozoic era. Nature410, 352354. (doi:10.1038/35066546)

29 Franks, P. J. & Beerling, D. J. 2009 Maximum leafconductance driven by CO2 effects on stomatal size anddensity over geologic time. Proc. Natl Acad. Sci. USA106, 10 34310 347. (doi:10.1073/pnas.0904209106)

30 Woodward, W. I. 1998 Do plants really need stomata?J. Exp. Bot. 49, 471480. (doi:10.1093/jexbot/49.suppl_1.471)

31 Boyce, C. K., Brodribb, T. J., Field, T. S. & Zwienieki,

M. A. 2009 Angiosperm leaf vein evolution was physiologi-cally and environmentally transformative. Proc. R. Soc. B276, 17711776. (doi:10.1098/rspb.2008.1919)

32 Beerling, D. J. & Franks, P. J. 2010 The hidden cost oftranspiration. Nature 464, 495496. (doi:10.1038/464495a)

33 McKown, A. D., Cochard, H. & Sack, L. 2010 Decodingleaf hydraulics with a spatially explicit model: principlesof venation architecture and implications for its evol-ution. Am. Nat. 175, 447460. (doi:10.1086/650721)

34 Sack, L. & Holbrook, N. M. 2006 Leaf hydraulics.Annu. Rev. Plant Biol. 57, 361381. (doi:10.1146/annurev.arplant.56.032604.144141)

35 Gould, S. J. & Vrba, E. S. 1982 Exaptation: a missingterm in the science of form. Paleobiology 8, 415.

36 Anderson, L. J., Maherali, H., Johnson, H. B., Polley,H. W. & Jackson, R. B. 2001 Gas exchange and photo-synthetic acclimation over subambient to elevated CO2in a C3C4 grassland. Global Change Biol. 7, 693707.(doi:10.1046/j.1354-1013.2001.00438.x)


37 Tholen, D. & Zhu, X. G. 2011 The mechanistic basis ofinternal conductance: a theoretical analysis of meso-phyll cell photosynthesis and CO2 diffusion. PlantPhysiol. 156, 90105. (doi:10.1104/pp.111.172346)

38 Farquhar, G. D. & Sharkey, T. D. 1982 Stomatal con-ductance and photosynthesis. Annu. Rev. Plant Phys.Plant Mol. Biol. 33, 317345. (doi:10.1146/annurev.pp.33.060182.001533)

39 Niinemets, U., Flexas, J. & Penuelas, J. 2011 Evergreensfavored by higher responsiveness to increased CO2. TrendsEcol. Evol. 26, 136142. (doi:10.1016/j.tree.2010.12.012)

40 Christin, P. A., Besnard, G., Samaritani, E., Duvall,

M. R., Hodkinson, T. R., Savolainen, V. & Salamin, N.2008 Oligocene CO2 decline promoted C4 photosyn-thesis in grasses. Curr. Biol. 18, 3743. (doi:10.1016/j.cub.2007.11.058)

41 Ehleringer, J. R., Cerling, T. E. & Helliker, B. R. 1997

C4 photosynthesis, atmospheric CO2 and climate.Oecologia 112, 285299. (doi:10.1007/s004420050311)

42 Sage, R. F. 2004 The evolution of C4 photosynthesis.New Phytol. 161, 341370. (doi:10.1111/j.1469-8137.2004.00974.x)

43 Urban, M. A., Nelson, D. M., Jimenez-Moreno, G.,Chateauneuf, J. J., Pearson, A. & Hu, F. S. 2010 Isoto-pic evidence of C4 grasses in southwestern Europeduring the Early OligoceneMiddle Miocene. Geology38, 10911094. (doi:10.1130/G31117.1)

44 Kuypers, M. M. M., Pancost, R. D. & Damste, J. S. S.1999 A large and abrupt fall in atmospheric CO2concentration during Cretaceous times. Nature 399,342345. (doi:10.1038/20659)

45 Kuypers, M. M. M., Blokker, P., Erbacher, J., Kinkel,H., Pancost, R. D., Schouten, S. & Damste, J. S. S.2001 Massive expanasion of marine archaea during amid-Cretaceous oceanic anoxic event. Science 239,9294. (doi:10.1126/science.1058424)

46 Silvera, K., Santiago, L. S., Cushman, J. C. &Winter, K. 2009 Crassulacean acid metabolism andepiphytism linked to adaptive radiations in the Orch-idaceae. Plant Physiol. 149, 18381847. (doi:10.1104/pp.108.132555)

47 Arakaki, M., Christin, P.-A., Nyffeler, R., Lendel, A.,Eggli, U., Ogburn, R. M., Spriggs, E., Moore, M. J. &Edwards, E. J. 2011 Contemporaneous and recent radi-ations of the worlds major succulent plant lineages.Proc. Natl Acad. Sci. USA 108, 83798384. (doi:10.1073/pnas.1100628108)

48 Ward, J. K., Harris, J. M., Cerling, T. E., Wiedenhoeft,A., Lott, M. J., Dearing, M. D., Coltrain, J. B. &Ehleringer, J. R. 2005 Carbon starvation in glacial

trees recovered from the La Brea tar pits, southernCalifornia. Proc. Natl Acad. Sci. USA 102, 690694.(doi:10.1073/pnas.0408315102)

49 Geber, M. A. & Dawson, T. E. 1997 Genetic variation instomatal and biochemical limitations to photosynthesis in

the annual plant, Polygonum arenastrum. Oecologia 109,535546. (doi:10.1007/s004420050114)

50 Intergovernmental Panel on Climate Change. 2007 Cli-mate Change 2007: The Physical Science BasisSummaryfor Policymakers, 18 pp. Cambridge, UK: CambridgeUniversity Press.

51 Lammertsma, E. I., de Boer, H. J., Dekker, S. C.,Dilcher, D. L., Lotter, A. F. & Wagner-Cremer, F.2011 Global CO2 rise leads to reduced maximum sto-matal conductance in Florida vegetation. Proc. NatlAcad. Sci. USA 108, 40354040. (doi:10.1073/pnas.1100371108)

52 Lobell, D. B., Schlenker, W. & Costa-Roberts, J. 2011Climate trends and global crop production since 1980.Science 333, 616620. (doi:10.1126/science.1204531)

http://dx.doi.org/10.1126/science.285.5429.876http://dx.doi.org/10.1126/science.1110063http://dx.doi.org/10.1016/S0012-8252(00)00042-8http://dx.doi.org/10.1016/j.gca.2005.11.031http://dx.doi.org/10.1093/jxb/44.5.907http://dx.doi.org/10.1111/j.1472-4669.2009.00193.xhttp://dx.doi.org/10.1016/0034-6667(87)90043-1http://dx.doi.org/10.1073/pnas.0402787101http://dx.doi.org/10.1073/pnas.0402787101http://dx.doi.org/10.1111/j.1461-0248.2009.01410.xhttp://dx.doi.org/10.1093/jxb/err086http://dx.doi.org/10.1038/35066546http://dx.doi.org/10.1073/pnas.0904209106http://dx.doi.org/10.1093/jexbot/49.suppl_1.471http://dx.doi.org/10.1093/jexbot/49.suppl_1.471http://dx.doi.org/10.1098/rspb.2008.1919http://dx.doi.org/10.1038/464495ahttp://dx.doi.org/10.1038/464495ahttp://dx.doi.org/10.1086/650721http://dx.doi.org/10.1146/annurev.arplant.56.032604.144141http://dx.doi.org/10.1146/annurev.arplant.56.032604.144141http://dx.doi.org/10.1046/j.1354-1013.2001.00438.xhttp://dx.doi.org/10.1104/pp.111.172346http://dx.doi.org/10.1146/annurev.pp.33.060182.001533http://dx.doi.org/10.1146/annurev.pp.33.060182.001533http://dx.doi.org/10.1016/j.tree.2010.12.012http://dx.doi.org/10.1016/j.cub.2007.11.058http://dx.doi.org/10.1016/j.cub.2007.11.058http://dx.doi.org/10.1007/s004420050311http://dx.doi.org/10.1111/j.1469-8137.2004.00974.xhttp://dx.doi.org/10.1111/j.1469-8137.2004.00974.xhttp://dx.doi.org/10.1130/G31117.1http://dx.doi.org/10.1038/20659http://dx.doi.org/10.1126/science.1058424http://dx.doi.org/10.1104/pp.108.132555http://dx.doi.org/10.1104/pp.108.132555http://dx.doi.org/10.1073/pnas.1100628108http://dx.doi.org/10.1073/pnas.1100628108http://dx.doi.org/10.1073/pnas.0408315102http://dx.doi.org/10.1007/s004420050114http://dx.doi.org/10.1073/pnas.1100371108http://dx.doi.org/10.1073/pnas.1100371108http://dx.doi.org/10.1126/science.1204531http://rstb.royalsocietypublishing.org/



53 van Mantgem, P. J. et al. 2009 Widespread increase oftree mortality rates in the Western United States. Science323, 521524. (doi:10.1126/science.1165000)

54 Welch, J. R., Vincent, J. R., Auffhammer, M., Moya,P. F., Dobermann, A. & Dawe, D. 2010 Rice yields intropical/subtropical Asia exhibit large but opposing sen-sitivities to minimum and maximum temperatures. Proc.Natl Acad. Sci. USA 107, 14 56214 567. (doi:10.1073/pnas.1001222107)

55 Barnes, J., Bender, J., Lyons, T. & Borland, A. 1999Natural and man-made selection for air pollution resist-ance. J. Exp. Bot. 50, 14231435. (doi:10.1093/jexbot/50.338.1423)

56 Davison, A. W. & Reiling, K. 1995 A rapid change inozone resistance of Plantago major after summers withhigh ozone concentrations. New Phytol. 131, 337344.(doi:10.1111/j.1469-8137.1995.tb03069.x)

57 Franks, S. J., Sim, S. & Weis, A. E. 2007 Rapid evol-ution of flowering time by an annual plant in responseto a climate fluctuation. Proc. Natl Acad. Sci. USA104, 12781282. (doi:10.1073/pnas.0608379104)

58 Pulido, F. & Berthold, P. 2010 Current selection for

lower migratory activity will drive the evolution of resi-dency in a migratory bird population. Proc. Natl Acad.Sci. USA 107, 73417346. (doi:10.1073/pnas.0910361107)

59 Hoffmann, A. A. & Sgro, C. M. 2011 Climate change

and evolutionary adaptation. Nature 470, 479485.(doi:10.1038/nature09670)

60 Ziska, L. H., Morris, C. F. & Goins, E. W. 2004 Quan-titative and qualitative evaluation of selected wheat

varieties released since 1903 to increasing atmosphericcarbon dioxide: can yield sensitivity to carbon dioxidebe a factor in wheat performance? Global ChangeBiol. 10, 18101819. (doi:10.1111/j.1365-2486.2004.00840.x)

61 Manderscheid, R. & Weigel, H. J. 1997 Photosyntheticand growth responses of old and modern spring wheatcultivars to atmospheric CO2 enrichment. Agric.Ecosyst. Environ. 64, 6573. (doi:10.1016/S0167-8809(97)00020-0)

62 Ainsworth, E. A. et al. 2008 Next generation of elevated[CO2] experiments with crops: a critical investmentfor feeding the future world. Plant Cell Environ. 31,13171324. (doi:10.1111/j.1365-3040.2008.01841.x)

63 Bazzaz, F. A., Jasienski, M., Thomas, S. C. & Wayne, P.

1995 Microevolutionary responses in experimentalpopulations of plants to CO2-enriched environments:parallel results from two model systems. Proc. NatlAcad. Sci. USA 92, 81618165. (doi:10.1073/pnas.92.18.8161)

64 Lau, J. A., Shaw, R. G., Reich, P. B., Shaw, F. H. &Tiffin, P. 2007 Strong ecological but weak evolutionaryeffects of elevated CO2 on a recombinant inbredpopulation of Arabidopsis thaliana. New Phytol. 175,351362. (doi:10.1111/j.1469-8137.2007.02108.x)

65 Lau, J. A., Shaw, R. G., Reich, P. B. & Tiffin, P. 2010Species interactions in a changing environment: elev-ated CO2 alters the ecological and potentialevolutionary consequences of competition. Evol. Ecol.Res. 12, 435455.

66 Maxon-Smith, J. W. 1977 Selection for response to CO2-enrichment in glasshouse lettuce. Hort. Res. 17, 1522.

67 Mycroft, E. E., Zhang, J. Y., Adams, G. & Reekie, E.2009 Elevated CO2 will not select for enhanced

growth in white spruce despite genotypic variation inresponse. Basic Appl. Ecol. 10, 349357. (doi:10.1016/j.baae.2008.08.005)

68 Potvin, C. & Tousignant, D. 1996 Evolutionaryconsequences of simulated global change: genetic


adaptation or adaptive phenotypic plasticity. Oecologia108, 683693. (doi:10.1007/BF00329043)

69 Steinger, T., Stephan, A. & Schmid, B. 2007 Predicting

adaptive evolution under elevated atmospheric CO2in the perennial grass Bromus erectus. Global ChangeBiol. 13, 10281039. (doi:10.1111/j.1365-2486.2007.01328.x)

70 Tonsor, S. J. & Scheiner, S. M. 2007 Plastic trait inte-

gration across a CO2 gradient in Arabidopsis thaliana.Am. Nat. 169, E119E140. (doi:10.1086/513493)

71 Ward, J. K., Antonovics, J., Thomas, R. B. & Strain,B. R. 2000 Is atmospheric CO2 a selective agent on

model C3 annuals? Oecologia 123, 330341. (doi:10.1007/s004420051019)

72 Donovan, L. A., Maherali, H., Caruso, C. M., Huber, H.& de Kroon, H. 2011 The evolution of the worldwide leafeconomics spectrum. Trends Ecol. Evol. 26, 8895.(doi:10.1016/j.tree.2010.11.011)

73 Gonzalez, J. A., Bruno, M., Valoy, M. & Prado, F. E.2011 Genotypic variation of gas exchange parametersand leaf stable carbon and nitrogen isotopes in tenQuinoa cultivars grown under drought. J. Agro

review evolutionary context for understanding and manipulating plant responses to past...

Documents