Temporal matching among diurnal ... - Wiley Online Library

Plant, Cell and Environment (2011) 34, 800–810

doi: 10.1111/j.1365-3040.2011.02283.x

Temporal matching among diurnal photosynthetic patterns within the crown of the evergreen sclerophyll Olea europaea L. pce_2283

800..810

C. GRANADO-YELA1, C. GARCÍA-VERDUGO1,2, K. CARRILLO1, R. RUBIO DE CASAS3, L. A. KLECZKOWSKI4 & L. BALAGUER1 1 Departamento de Biología Vegetal I, Universidad Complutense de Madrid, Madrid 28040, Spain, 2Rancho Santa Ana Botanic Garden, Claremont, California 27705, USA, 3National Evolutionary Synthesis Center, Durham, North Carolina 27705, USA and 4Department of Plant Physiology, Umeå Plant Science Centre, Umeå 90187, Sweden

ABSTRACT Trees are modular organisms that adjust their within-crown morphology and physiology in response to within-crown light gradients. However, whether within-plant variation represents a strategy for optimizing light absorption has not been formally tested. We investigated the arrangement of the photosynthetic surface throughout one day and its effects on the photosynthetic process, at the most exposed and most sheltered crown layers of a wild olive tree (Olea europaea L.). Similar measurements were made for cuttings taken from this individual and grown in a greenhouse at contrasted irradiance-levels (100 and 20% full sunlight). Diurnal variations in light interception, carbon fixation and carbohydrate accumulation in sun leaves were negatively correlated with those in shade leaves under field conditions when light intensity was not limiting. Despite genetic identity, these complementary patterns were not found in plants grown in the greenhouse. The temporal disparity among crown positions derived from specialization of the photosynthetic behaviour at different functional and spatial scales: architectural structure (crown level) and carbon budget (leaf level). Our results suggest that the profitability of producing a new module may not only respond to construction costs or light availability, but also rely on its spatio-temporal integration within the productive processes at the whole-crown level. Key-words: CO2 assimilation rate; division of labour; modular organism; non-structural carbohydrates (NSC); phenotypic plasticity; photosynthetically active radiation (PAR); silhouette area; splitting of tasks; sun and shade leaves. Abbreviations: Aarea, instantaneous net CO2 assimilation rate on an area basis; AL, area of the leaf blade; Amass, instantaneous net CO2 assimilation rate on a dry mass basis; Amax, maximum photosynthetic rate; AQ, photosynthesislight response curve; FSh, field shade leaves; FSun, field sun leaves; GSh, greenhouse shade leaves; GSun, greenhouse Correspondence: C. Granado-Yela. E-mail: [email protected] 800

sun leaves; l-AQ, linear phase of the AQ; LSP, light saturation point; NSC, total non-structural carbohydrate content on a dry mass basis; PAR, photosynthetically active radiation at open sky; PARi, photosynthetically active radiation received directly by leaves; Q95-75-50, 95th-75th-50th conditional quantile functions, respectively; SAL, silhouette area of the leaf blade; sat-AQ, light saturated phase of the AQ; tAarea, total diurnal CO2 assimilation on an area basis; tAmass, total diurnal CO2 assimilation on a dry mass basis; tPAR, total diurnal PAR at open sky; tPARi, total diurnal PAR received directly by leaves; %SAL, silhouette area of the leaf blade as percentage of AL; j, apparent photosynthetic efficiency.

INTRODUCTION Plants have been traditionally viewed as colonies of repeated modules that carry out similar, if not identical, physiological roles (White 1979). In this sense, module redundancy has been interpreted as a passive strategy intended to buffer individuals against aging and disturbances at the cost of an increased modular autonomy and disposability (Sprugel, Hinckley & Schaap 1991; Henriksson 2001). Functional benefits for the whole plant, however, appear to increase concomitantly with module population size. A large population of modules enhances resource foraging efficiency, optimizing the use of available resources through active environmental screening and growth in favourable directions (Silvertown & Gordon 1989; Bazzaz 1991; Augspurger & Bartlett 2003). Obviously, increased module population plus the ability to express alternative phenotypes (i.e. phenotypic plasticity) brings about a further increase in the potential relationships among modules. Drawing a parallel with economic systems, colonies of social insects, and modular animals, societies that increase their size by pursuing a common goal are invariably held together by a division-of-labour strategy based on specialization, coordination and profitability (Durkheim 1893; Robinson 1992; Wahl 2002). In vascular plants, division of labour has been characterized only in clonal angiosperms (Stuefer, DeKroon & During 1996; Stuefer 1998). In spatially heterogeneous environments, © 2011 Blackwell Publishing Ltd

Within-crown photosynthetic patterns in O. europaea 801 interconnected modules of clonal plants, i.e. those exchanging water and assimilates, become specialized in the uptake of the most locally abundant resource. Close cooperation by interconnected modules significantly increases the performance of the entire plant in terms of fitness-related traits (Stuefer, DeKroon & During 1996; Stuefer 1998). Synergy between modular demography and modular plasticity appears to be particularly relevant in the case of the tree habit in land plants.Although they do not form a natural group, trees share a number of attributes such as high modularity, large size, long life span, high reproductive output, great phenotypic plasticity and high physiological tolerance (Petit & Hampe 2006). Field evidence has shown, for instance, that exposure to wind may elicit mechanical and hydraulic rearrangements in the outer layers of the tree crown which provide shelter to the inner layers (GarcíaVerdugo et al.2009).Within-crown module plasticity in trees, however, appears to be primarily linked to irradiance (Sack et al. 2006). Sun leaves are generally smaller and thicker and present higher photosynthetic rates per unit leaf area than shade leaves (Jurik, Chabot & Chabot 1979; Valladares & Niinemets 2008). Module plasticity in the expression of leaf inclination angles modulates the quality and quantity of light transmission through the tree crown, enhancing wholeindividual carbon gain (Uemura et al. 2006). High module plasticity at the outer layers enables within-crown habitat modification, smoothing the light environment variation experienced by the inner layers (Rubio de Casas et al. 2007). This interaction among specialized modules, however, enables but does not confirm temporal or spatial splitting of tasks among modules within the tree crown. In the present study we investigated, in an adult wild olive tree (Olea europaea L. ssp. guanchica P. Vargas et al.), the potential spatio-temporal splitting of tasks among modules from two extreme light environments within the tree crown. We extrapolated to trees the experimental approach applied to characterize the division of labour in clonal plants (Stuefer, DeKroon & During 1996): we investigated the performance of the actual interconnected modules (i.e. ‘regular-crown condition’) and the performance of experimentally disconnected modules (i.e. ‘non-crown condition’). Specifically, we analysed the performance of interconnected populations of leaves within a tree crown that experience heterogeneous light environments under field conditions (sun and shade exposures) and that of populations of leaves from cuttings that experience homogeneous light environments at two contrasted levels (sun and shade treatments) in a greenhouse experiment.We first studied how the plant arranged its photosynthetic surface to capture light throughout the day and then determined how variation in the photosynthetic surface arrangement affected CO2 assimilation and carbohydrate content in leaves. We hypothesized that spatial divergence of light environments within the tree crown, mediated by leaf specialization, would lead to temporal segregation of the photosynthetic maxima, and ultimately to the expression of complementary photosynthetic patterns. Conversely, and as a consequence of the non-crown condition, we expect the

diurnal patterns of light interception, carbon assimilation and carbohydrates accumulation in cuttings in the greenhouse to simply track light availability, regardless of the light treatment applied. We specifically addressed the following questions: (1) Does expression of module plasticity optimize photosynthetic performance at the most exposed and most sheltered crown layers?; (2) Do these two crown layers share architectural and photosynthetic temporal patterns?; (3) Do leaves in greenhouse-grown cuttings in two contrasted levels of irradiance share the same patterns?; and (4) Is there any evidence of integration of the local expression of module plasticity between the most exposed and most sheltered crown layers?

MATERIALS AND METHODS Species, site and sampling design The wild olive, O. europaea L. ssp. guanchica P. Vargas et al., is an evergreen, sclerophyllous, medium-sized tree endemic to the Canary Islands. This study was conducted in an adult wild olive and in 3-year-old cuttings taken from this individual and then grown in a common garden. Given the comprehensive measurements required, the analyses in the present study were restricted to one single tree, following the scope of previous studies (e.g. Le Roux et al. 2001; and references therein). The wild olive individual studied was 3.5 m high and located in a natural population on the island of La Palma (28°39′01″ N, 17°45′44″ W). Plant dimensions correspond to the average for the species (García-Verdugo et al. 2009). The population is located upon a plain and neither neighbouring trees nor anthropic structures shade this particular individual. For the greenhouse experiment, 10 mature shoots were cut from the upper part of the crown. Shoots were transported to the greenhouse (La Tahonilla, Tenerife, Canary Islands) under refrigerated and humid conditions, and 10 cm cuttings were taken from them and treated with commercial rooting hormones (0.06% indolebutyric acid and 0.08% a-naphthyl acetamide). Cuttings were planted in 50 L pots filled with a mix of 50% top soil from a nearby site (v/v), 25% peat moss and 25% perlite (for further details, see García-Verdugo et al. 2009). Half of the cuttings were maintained under natural light conditions (sun treatment) and the other half in neutral shade (shade treatment), provided by a white standard raffia cloth, which decreased daily photosynthetic photon flux density by 80%. This shading level was chosen because it corresponds to the annual averaged reduction in light intensity caused by the crown of Mediterranean sclerophyllous species (Balaguer et al. 2001; García-Verdugo 2009). Seven Watermark WEM-II soil moisture sensors were placed in different pots and coupled to a Watermark data logger (Irrometer, Riverside, CA, USA) to control water potential. Water stress was avoided and optimal conditions ensured by means of automatic watering of the pots when water potential reached -10 KPa (García-Verdugo et al. 2009). Plants were supplemented

© 2011 Blackwell Publishing Ltd, Plant, Cell and Environment, 34, 800–810

802 C. Granado-Yela et al.

Figure 1. On the left, the three angles of the leaf lamina (modified from Pearcy & Yang 1996): a, midrib inclination to the horizontal; b, clockwise angle of the direction of the leaf midrib from the north (azimuth); and g, the angle of the lamina through the midrib. On the right, an example of the variation of the silhouette area of the leaf blade (SAL) based on the sun position. With the aim of simplifying, the square plane of length side 1 cm (which represents a model leaf of area AL = 1 cm2) has been set horizontally, and the sun path drawn from east (azimuth 90°) to west (azimuth 270°) reaching the zenith at noon. This horizontal plane of 1 cm2 receives no direct radiation from the sun, either at sunrise or at sunset (situations a1 and a2, respectively) because the plane is horizontal to the sun beam (SAL(a1-2) = 0). On the contrary, the same horizontal plane intercepts the maxima of sun radiation when the sun is at zenith (situation c) because the direct radiation from the sun is perpendicular to the plane of the leaf. Under these circumstances SAL is equal AL, both measuring 1 cm2 in this example. The rest of the day (for instance in situation b), values of SAL might range from 0 cm2 (a1 and a2) to 1 cm2 (c), due to the mismatch of sun position and leaf orientation. Because the sun path runs from east to west and reaches the zenith, the area that would be observed from the sun view (SAL) at any time between sunrise and noon, or after noon to sunset, would be a particular quadrilateral (a trapezium) with an area smaller than 1 cm2 (SALb < 1 cm2).

twice a year with commercial fertilizer (NPK; 7:5:6). Three 3-year-old cuttings measuring approximately 1.1 m high were chosen per treatment and considered as replicates for greenhouse measurements. We characterized the performance of the leaves from the most exposed tree-crown layers (sun leaves) and the most sheltered ones (shade leaves) in the field, and of leaves of the cuttings in sun and shade treatments in the greenhouse, on the 27 and 29 September 2008, respectively. At both sites, we installed a U12 data logger coupled with an S-LIA-M003 PAR sensor (Onset Computer Corp., Bourne, MA, USA) that recorded Photosynthetically active radiation (mmol photons m-2 s-1) at open sky (PAR). Under field conditions, fully exposed leaves from the south-facing sector of the crown top were considered to be sun leaves, and leaves at the basal north facing sector of the crown interior were considered as shade leaves (Sack et al. 2006; Rubio de Casas et al. 2007). In the greenhouse plants, we only sampled leaves from the top part of the crowns in order to avoid self-shading effects, although heterogeneous light conditions within the crowns were unlikely due to the small size of the individuals. Approximately every 6 min, from pre-dawn (0600 h UTC + 1) to post-dusk (2200 h UTC + 1), we haphazardly chose a fully expanded leaf (n = 318), alternating between sun and shade leaves in the field (n = 78 per exposure) and between light treatments and cuttings in the greenhouse (n = 81 per treatment). During these time windows of 6 min between chosen leaves, we traced the shape of each sampled leaf to calculate its area at the laboratory, and we measured its angles (Fig. 1) and its instantaneous CO2 assimilation rate. Leaves were then collected, and two discs of 30 mm2 were taken with a hole-punch from every

leaf. Immediately after collection, leaves were frozen in liquid nitrogen and stored at -80 °C until carbohydrate quantification to preserve each pool of carbohydrates. Discs were weighed at the laboratory with a precision balance (Mettler Toledo, Greifensee, Switzerland), ovendried at 65 °C for 48 h and weighed again to determine the leaf area to dry mass ratio.

Trait measurements

Architectural measurements Light capture of a leaf is frequently assumed to be determined by the area of the leaf blade (AL), i.e. bigger leaves are assumed to intercept more radiation than smaller leaves. Leaf light-capture reaches its maximum values when the direct radiation is perpendicular to the plane of the leaf. Nonetheless, for leaves that do not track sun movement, this situation (leaf blade perpendicular to the sun beam) is anecdotic due to the diurnal and seasonal changes in solar azimuth and altitude. We may conclude that there is a reduction in the potential of light capture most of the time, consistent with the mismatch between sun position and leaf orientation. Several methods have been proposed for describing the actual area of the leaves intercepting light at a precise moment. In the present study, we used a modified version of the silhouette leaf area (Carter & Smith 1985; Smith, Schoettle & Cui 1991) applied to isolated leaves (silhouette area of the leaf blade; SAL), which calculates the one-side area of the leaf blade (AL) that would be observed from the sun view at the sampling time, ignoring possible leaf overlaps. The AL (cm2) was calculated from the traced shapes (n = 318) with Scion Image software (Scion Corp., Frederick,


Within-crown photosynthetic patterns in O. europaea 803 MA,USA).In order to correctly characterize the orientation of the 318 leaves sampled,we measured in situ three angles of each leaf lamina (a, b, g; Fig. 1) using a protractor. The sun path for the 27 and 29 September 2008 was obtained using SP v1.7 (Mike Brackenridge, http://sunposition.info). The planes of the 318 leaves were digitalized in AutoCAD 2007 (Autodesk Inc.) in independent files as squares of 1 ¥ 1 cm2, which were oriented according to the particular three angles of each leaf. Additionally, sun position was also modelled in each file considering the moment of the day when every leaf was sampled (i.e. azimuth bearings and altitude positions previously estimated).The area of the resulting quadrilateral observed from the sun view at sampling time (that range 0–1 cm2) was then estimated graphically (Fig. 1) and multiplied by the AL of each leaf to calculate the silhouette area of the leaf blade (SAL; cm2). We subsequently transformed SAL to a percentage variable (%SAL) with the aim of representing the percentage of the AL of every leaf receiving irradiation at the moment that was sampled.

Photosynthetic measurements We performed 318 measurements (n = 78 per exposure in the field, and n = 81 per treatment in the greenhouse) of instantaneous net CO2 assimilation rate (Aarea; mmol CO2 m-2 s-1 or Amass; mmol CO2 g-1 s-1) following the experimental design previously described. Leaves were measured at nearly ambient light, by maintaining leaf orientation and angles, with a portable infra-red gas analyzer (Li6400, Li-Cor, Lincoln, NE, USA) coupled to a transparent cuvette. The cuvette recorded the amount of photosynthetically active radiation (PARi; mmol photons m-2 s-1) received directly by the oriented plane of the sampled leaves when the instantaneous measurements were made. In addition, we measured 12 photosynthetic-light response curves (n = 3 per exposure/treatment) in order to characterize the average photosynthetic response to light of the leaves from sun and shade exposures/treatments. Photosynthesis-light response curves were obtained with a LED light source (Li-6400-02B, Li-Cor) connected to the Li-Cor 6400 gas exchange system by measuring photosynthesis at increasing light intensities (0, 10, 50, 100, 150, 200, 300, 500, 750, 1000, 1500 and 2000 mmol photons m-2 s-1). During light-curve measurements, air temperature (20 °C) and CO2 concentration (400 mmol CO2 mol air-1) were kept constant within the cuvette. Light curves were fitted (in all cases R2 > 0.90) by nonlinear regression with Sigma Plot 8.0 (SSPS Inc., Chicago, IL, USA) using the Mitscherlich model equation (Potvin, Lechowicz & Tardif 1990). The adjusted parameters of the curves [maximum photosynthetic rate (Amax; mmol CO2 m-2 s-1), apparent photosynthetic efficiency (j; mol/mol) and light saturation point (LSP; mmol photons m-2 s-1)] were averaged for each light environment/ treatment. The LSP represents the threshold that determines whether the photosynthetic activity of a leaf is at the linear (l-AQ) or at the light-saturated portion (sat-AQ) of the light curve.

Carbohydrates quantification Carbohydrate quantification of leaves encompassed both the soluble (glucose, fructose, sucrose and mannitol) and starch fractions, which represents over 95% of total nonstructural carbohydrate content (NSC) in O. europaea L. (Cataldi et al. 2000). Frozen leaves were grounded to a fine powder in a mortar. For soluble sugars extraction, we homogenized 30 mg of leaf powder in 0.45 mL of 80% ethanol–Hepes at 80 °C for 15 min, and then centrifuged for 10 min at 14 000 rpm. The residue was extracted once more in the same conditions, and a third extraction was performed in 0.9 mL of 50% ethanol–Hepes. We then used the pellet from the previous extractions to quantify starch as glucose equivalents. We resuspended the pellet in 0.5 mL of 0.2 m citrate buffer (pH 4.6), heated it at 95 °C for 15 min to gelatinize starch granules, let the samples cool and then added 30 mL of amyloglucosidase from Aspergillus niger (Roche, Basel, Switzerland) that degraded starch to glucose overnight in the heat blocks (50 °C). Carbohydrate content was quantified in three different enzymatic assays coupled to colorimetric measurements: assay for glucose from the enzymatic digestion of starch, assay for mannitol, and assay for glucose, fructose and sucrose. For the colorimetric quantifications of each NSC fraction, we followed the recommendations of the Megazyme Sucrose/D-Fructose/ D-Glucose and D-Mannitol assay kits (Megazyme International Ltd, Bray, Ireland). Measurements of the increase in absorbance at 340 nm resulting from the amounts of NADPH or NADH, depending upon which carbohydrate was being quantified, were performed with a multiplate reader Spectra Max 190 (Molecular Devices, Corp., Sunnyvale, CA, USA). All concentrations are given on a dry mass basis (mg g-1). Content of NSC in each leaf was calculated by adding up carbon equivalents from all carbohydrate fractions measured at each leaf (NSC; mmol Carbon g-1). Total nonstructural carbon of a leaf in a given time is the result of the carbon previously fixed minus the carbon that has been exported.

Data analysis We used B-spline quantile regressions in order to model the diurnal patterns of variation of the measured variables. Quantile regression is a method for estimating functional relations between variables for all portions of a probability distribution (Koenker & Bassett 1978). This statistical tool is widely used in many research areas (for a review, see Yu, Lu & Stander 2003) and has recently been applied to ecological studies (Cade & Noon 2003, and references therein). We performed quantile regression estimates based on linear B-spline expansions (fuller explanation of these smoothing functions is given by Hastie 1991; Wei et al. 2006) using ‘quantreg’ (Koenker 2009) and ‘splines’ (Bates & Venables 2007) libraries in R 2.8.1 (http://www.rproject.org). This method was preferred to conventional


804 C. Granado-Yela et al. statistics because the variation in some response variables over time is too complex to be reasonably modelled with single-slope functions. With the aim of investigating whether module plasticity optimizes the photosynthetic performance of the most exposed and most sheltered leaves within the tree (i.e. regular-crown condition), we modelled diurnal patterns of variation in the 95th, 75th and 50th conditional quantile functions (Q95, Q75 and Q50, respectively) for the variables %SAL, Aarea, and PARi. We followed the same approach to characterize the effects of the non-crown condition (cuttings experiencing two light environments contrasted in light intensity) in the arrangement of the light-capture surface and photosynthetic patterns: we modelled the diurnal patterns of variation in the 95th, 75th and 50th conditional quantile functions for the variables %SAL, Aarea and PARi of leaves from fully exposed and from shaded cuttings. In order to identify evidences of integration of the local expression in module plasticity, we modelled and compared (by means of simple correlations) the variation of the 95th conditional quantile (Q95) of the variables SAL, Amass and NSC in the most exposed and most sheltered crown layers, and in leaves from sun and shade treatments in the greenhouse. For the correlations, we extracted fitted values from the Q95 B-spline regression equations of these variables every 10 min and split each variable into two new variables according to the phase of the photosynthesislight response curves they belonged to (l-AQ or sat-AQ). The variation of SAL and Amass throughout the day

might indicate the investment in AL and photosynthetic/ structural components, respectively, to capture light and fix carbon in each leaf. The diurnal pattern of NSC in leaves results from the interaction of storage and export, and depicts the prevalence of one process over time.

RESULTS In the field, values of %SAL and PARi on sun leaves showed a marked drop at midday when PAR at open sky reached maximum values (Fig. 2). By contrast, %SAL and PARi patterns of shade leaves did not show this drop (Fig. 2). The proximity between the quantile lines, however, pointed to a likely maximum in %SAL around 1400 h in shade leaves (Fig. 2).The values of PARi presented by these leaves followed the PAR pattern at open sky, albeit skewed toward the morning period (Fig. 2). In addition, the pattern of %SAL in sun leaves reached values over 70% from 0800 to 1300 h (5 h interval) and from 1600 to 1830 h (2.5 h interval). This interval was, therefore, twice as long during the morning (when the sun is in the east) as during the afternoon (when the sun is in the west). Carbon assimilation rate per unit area (Aarea) showed no abrupt changes during the day, either in sun or in shade leaves (Fig. 2). The overall PAR received by sun leaves (tPARi) in the field was 3.7 times that of shade leaves (3.4 times on comparing PAR). Total CO2 assimilation on an area basis (tAarea) for the sun leaves was 1.9 times tAarea for the shade leaves (Table 1). Light response curves showed that Amax of sun leaves was 1.2-fold that of the shade leaves, and j

Figure 2. Field diurnal patterns of variation in silhouette area of the leaf blade expressed as a percentage of AL (%SAL), photosynthetically active radiation received by leaves (PARi), and CO2 assimilation rate per unit area (Aarea). Quantile regression lines were modelled by means of B-Splines with 11 d.f. Thin lines correspond to the 95th quantile (Q95, solid line), the 75th quantile (Q75, dashed line) and the 50th quantile (Q50, dotted line). Thick solid lines correspond to PAR at open sky (PAR). The graphs on the left represent patterns for the sun leaves (FSun) and the graphs on the right for the shade leaves (FSh). © 2011 Blackwell Publishing Ltd, Plant, Cell and Environment, 34, 800–810

Within-crown photosynthetic patterns in O. europaea 805 Table 1. Total CO2 assimilation, total PAR and mean values of the estimated parameters of the photosynthesis-light response curve in sun and shade exposures/treatments Site

Expo/Treat

tAarea

tAmass

tPARi

tPAR

Amax

F

LSP

Field

Sun Shade Sun Shade

0.33 0.17 0.33 0.11

2.76 1.29 1.51 1.27

54.93 14.91 31.27 2.24

50.76

4.58 (⫾0.98) 3.69 (⫾1.18) 8.24 (⫾1.78) 3.80 (⫾0.37)

0.010 (⫾0.001) 0.020 (⫾0.004) 0.008 (⫾0.002) 0.011 (⫾0.001)

322.2 (⫾60.8) 91.2 (⫾10.4) 465.8 (⫾112.5) 256.9 (⫾46.1)

Greenhouse

29.89

Values of total diurnal CO2 assimilation on an area basis (tAarea; mol CO2 m-2 day-1), total diurnal CO2 assimilation on a dry mass basis (tAmass; mmol CO2 g-1 day-1), total diurnal PAR received directly by leaves (tPARi; mol photons m-2 day-1), total diurnal PAR received at open sky (tPAR; mol photons m-2 day-1), maximum photosynthetic rate (Amax; mmol CO2 m-2 s-1), apparent photosynthetic efficiency (j; mol/mol) and light saturation point (LSP; mmol photons m-2 s-1). Variables Amax and j represent mean values (n = 3 per site and exposure/treatment). Standard errors are included in brackets (⫾SE).

of the shade leaves was 2.1-fold that of the sun leaves (Table 1). Light-saturating intensity was lower in shade leaves (Table 1). Mean values of carbohydrate content in leaves is shown in Table 2. In the field, NSC variation throughout the day showed opposite pattern in comparison to SAL and Amass (Fig. 3). Thus, an increase in NSC occurred when SAL and Amass experienced a marked decrease. In addition, opposite patterns were also found for sun and shade leaves in all three variables.Accordingly,sun leaves showed an increase in NSC at hours in which Amass and SAL were higher in shade leaves but NSC was low (Fig. 3). Values of SAL and Amass in sun leaves were higher during the morning than during the afternoon (Fig. 3). Because the %SAL was equivalent for the morning and afternoon periods (over 70%; Fig. 1), higher values of SAL during the morning were caused by a higher investment in AL in the leaves facing east. The duration of the midday sat-AQ differed between sun and shade exposures (LSP was achieved later in the morning and earlier in the evening at the shade exposure). We split both sun and shade variables for the correlations according to the shortest period of saturating light (i.e. period from the shade exposure). Sat-AQ occurred from 0950 to 1810 h UTC + 1. Correlations between sun and shade patterns for SAL, Amass and NSC yielded contrasting results depending on the phase of the photosynthesis-light response curves. Thus, sun and shade patterns for the three variables were similar (positively correlated) at the l-AQ whereas at the sat-AQ patterns were complementary (negatively correlated; Fig. 3). All the correlations were significant (P < 0.05).

In the greenhouse, the diurnal %SAL of sun and shade treatments showed a constant and identical pattern throughout the day without any marked drop (Fig. 4). Patterns of PARi and Aarea were also constant throughout the day for the leaves in both sun and shade treatments (Fig. 4). The aspect and location of the greenhouse site resulted in a reduction of approximately 40% of the total diurnal PAR received at open sky when compared to the field site. Overall PAR received by sun cuttings (tPARi) was similar to PAR at open sky, but 13 times higher than the tPARi of the shade treatment (Table 1). Total assimilation on a dry mass basis (tAmass) of sun cuttings was, however, only 1.2 times higher than shade tAmass and 3.1 times on an area basis (tAarea; Table 1). Light saturating intensity was lower in the shade treatment (Table 1). Duration of the midday sat-AQ was shortest for the shade treatment. Once again, we split both sun and shade variables (SAL, Amass and NSC) for the correlations according to the shortest period of saturating light. Sat-AQ occurred from 0920 h to 1610 h UTC+1. In both phases of the light response curve, correlations for the three variables between sun and shade patterns of the B-splines for the 95th quantile regressions models were positive when significant (Table 3).

DISCUSSION It has long been investigated how plants adapt at several levels of architecture (i.e. leaf, shoot and whole crown) to enhance whole-plant performance over a day and a season (for a review, see Smith et al. 2004). The present study is the first, to our knowledge, to substantiate with field data

Table 2. Mean contents on a dry mass basis (mg g-1) of non-structural carbohydrates in sun and shade exposures/treatments Site

Expo/Treat

Man

Glu

Fru

Suc

St

TCarb

TSolSug

Field

Sun Shade Sun Shade

61.6 (⫾1.5) 47.6 (⫾1.4) 46.0 (⫾2.0) 48.4 (⫾2.8)

43.9 (⫾1.5) 53.1 (⫾2.0) 48.5 (⫾2.7) 42.8 (⫾1.8)

7.0 (⫾0.7) 11.1 (⫾0.7) 13.7 (⫾1.4) 9.4 (⫾0.6)

23.4 (⫾2.2) 28.7 (⫾3.0) 24.5 (⫾3.4) 22.2 (⫾1.9)

31.2 (⫾1.9) 52.9 (⫾3.1) 113.0 (⫾4.9) 15.5 (⫾1.5)

171.6 (⫾4.0) 191.2 (⫾5.6) 245.5 (⫾7.6) 143.3 (⫾5.8)

138.6 (⫾3.8) 139.9 (⫾5.0) 138.6 (⫾7.8) 129.9 (⫾6.0)

Greenhouse

Values of mannitol (Man), glucose (Glu), fructose (Fru), sucrose (Suc), starch (St), total carbohydrates (TCarb) and total soluble sugars (TSolSug) are averaged per exposure in the field (n = 78) and per treatment at the greenhouse (n = 81). Standard errors are included in brackets (⫾SE). © 2011 Blackwell Publishing Ltd, Plant, Cell and Environment, 34, 800–810

806 C. Granado-Yela et al.

Figure 3. The graphs on the left represent the B-Splines for the 95th quantile regressions models (Q95) of field diurnal patterns for the silhouette area of the leaf blade (SAL), CO2 assimilation rate on a dry mass basis (Amass) and total non-structural carbohydrate content on a dry mass basis (NSC). Solid lines correspond to sun leaves (FSun) and dashed lines to shade leaves (FSh). From 0950 to 1810 h light intensity was over the light saturation point (LSP) for the shade leaves (91.2 mmol m-2 s-1). Values extracted from the regression equations for this period were considered under the light saturated phase of the photosynthesis-light response curves (sat-AQ). From sunrise to 0950 h [l-AQ(1)] and from 1810 h to sunset [l-AQ(2)] light intensity was below LSP. Values extracted from the regression equations for these periods were considered under the linear phase of the photosynthesis-light response curves (l-AQ). Small graphs represent the scatterplots of the correlations between sun and shade values for each variable at the l-AQ (graphs in the middle, n = 36) and the sat-AQ (graphs on the right, n = 51).

temporal complementarity of responses to the diurnal variation in the light environment between populations of leaves from the two extremes of the light gradient within the crown. Our results show that divergence of temporal patterns within a tree crown is enabled by specialization of the photosynthetic behaviour at different functional and spatial scales: architectural structure at the crown level, and carbon budget at the leaf level. Far from redundancy, functional complementarity between sun and shade modules became apparent at saturating light intensities, as suggested by the negative correlations in SAL, Amass and NSC (Fig. 3). In a fieldgrown adult tree, local expression of plasticity resulted in significant differences in crown architecture between outer and inner crown layers. In sun leaves, %SAL exhibited a marked drop at midday, whereas shade leaves were arranged mainly horizontally and reached a maximum at midday. In sun leaves, such an architectural arrangement enabled sustained carbon assimilation throughout the day regardless of the diurnal variation in solar light intensity (Fig. 2). Most likely, this configuration of the outer crown layers also contributes to minimizing photosynthetic

midday-depression, which was not observed in the present study but has been described as typical in Mediterranean plant species (Tenhunen, Lange & Braun 1981). Conversely, shade leaves experienced almost a fourfold reduction in light interception which only translated into a twofold reduction in total carbon assimilation (Table 1). This higher efficiency of shade leaves was consistent with an increase of the same magnitude in their apparent photosynthetic efficiency, and most likely reflects anatomical specialization of the photosynthetic parenchyma in response to light directionality (Brodersen & Vogelmann 2007; Brodersen et al. 2008). Diffuse light, rather than direct light, may determine daily carbon gain in shaded leaves (Valladares & Pearcy 1998) and modulate the phenotypic expression of morpho-functional traits in Mediterranean species (García-Verdugo et al. 2010). In locations of the tree crown where light is not limiting, other factors such as diffusional limitations to photosynthesis may influence the photosynthetic capacity (DiazEspejo, Nicolas & Fernandez 2007). These constraints are related to stomatal closure aimed at reducing leaf transpiration and limitations in mesophyll conductance


Within-crown photosynthetic patterns in O. europaea 807

Figure 4. Greenhouse diurnal patterns of variation in silhouette area of the leaf blade expressed as a percentage of AL (%SAL), photosynthetically active radiation received by leaves (PARi) and CO2 assimilation rate per unit area (Aarea). Quantile regression lines were modelled by means of B-Splines with 11 d.f. Thin lines correspond to the 95th quantile (Q95, solid line), the 75th quantile (Q75, dashed line) and the 50th quantile (Q50, dotted line). Thick solid lines correspond to PAR at open sky (PAR). The graphs on the left represent patterns for the sun leaves (GSun) and the graphs on the right for the shade leaves (GSh).

(Moriana, Villalobos & Fereres 2002; Centritto, Loreto & Chartzoulakis 2003). Our analyses suggest that the tree crown improved its light-harvesting efficiency by orienting the leaves at the outer layers towards the aspects of most productive irradiance, rather than exposing them horizontally. Leaves nearly reached full sun exposure before and after midday (Fig. 2). Nonetheless, not only leaf angle, but also biomass investment in leaves should be considered with regard to understanding crown strategy intended to maximize

Table 3. Correlations between sun and shade patterns in the greenhouse Variable

l-AQ (n = 51)

sat-AQ (n = 42)

SAL Amass NSC

r = 0.871*** r = 0.974*** r = 0.247

r = 0.475*** r = -0.054 r = 0.311*

Silhouette area of the leaf blade (SAL; cm2), assimilation rate on a dry mass basis (Amass; mmol CO2 g-1 s-1) and total non-structural carbohydrate content on a dry mass basis (NSC; mmol Carbon g-1) at the linear (l-AQ) and the light saturated (sat-AQ) phases of the photosynthesis-light response curves. Significant correlations between sun and shade patterns are indicated by *(P < 0.05) and ***(P < 0.001).

light capture. Optimization theories suggest that plant resources should be allocated in such a way that the photosynthetic capacity is proportional to intercepted light (Field 1983; Farquhar 1989). This can be partly achieved if leaves produced at the outer crown layers are thick, with a large number of parenchyma layers and high investment of nitrogen in photosynthetic enzymes (Murchie & Horton 1997; Niinemets, Tenhunen & Beyschlag 2004). Our results suggest that crown structure at the outer layers is functionally optimized through phenotypic adjustments at three different levels: by increasing the percentage of area of leaf blades with the most productive orientation (east at the study site latitude; Fig. 2), the size of the easterly facing leaves, and the photosynthetic rates on a dry mass basis of these leaves (Fig. 3). In the present study, we assumed that the light reduction within the crown is not gradual but that there were virtually two contrasting light environments, as suggested by the exponential extinction patterns of light described by Uemura et al. (2006). In the former case, we might be overestimating the degree of segregation of the productive processes at whole-crown level, because a gradual variation between the extremes in these processes would be expected in consonance with the linear extinction of light. In the latter case, whole-crown performance would result from the matching of the two extremes we describe.


808 C. Granado-Yela et al. Despite genetic identity (the greenhouse plants were clones of the field individual), none of the patterns observed under field conditions was observed in the plants grown in the greenhouse (Table 3; Fig. 4). A relevant difference that may at least partially account for this striking divergence is that sun and shade modules in the field were integrated within the crown of a single tree (regular-crown condition), while sun and shade saplings in the greenhouse were grown independently (non-crown condition). Indeed, the difference between the total assimilation of the sun and shade leaves was disproportionately larger in the field than in the greenhouse, although shading was more intense in the greenhouse (Table 1). This finding is consistent with previous reports that suggest an integrated-crown assimilation, which is optimized by an enhanced biomass and nutrient investment in sun shoots (Henriksson 2001; Sprugel 2002; Yoshimura 2010). Many other causes, however, may account for the discrepancy between field and greenhouse results, such as availability of resources other than light, spatial and temporal environmental heterogeneity or differences in light quality. In summary, our results strongly support the conclusion, previously drawn (Wayne & Bazzaz 1993; Bouvet, Vigneron & Saya 2005), that the ultimate implications of phenotypic plasticity in tree-crown development and performance can only be observed in mature individuals in the field. The results of the present study provide further evidence that modular specialization, mediated by the local expression of phenotypic plasticity, enables the development of complementary strategies, in contrast to previous approaches that considered modules as redundant units that played similar, if not identical, physiological roles (White 1979). The key step to acknowledge these strategies as evidences of division of labour within tree crowns entails ascertaining to what extent complementarity involves coordination among modules and yields a net benefit at the individual scale. Concerning module coordination, interactive effects among interconnected modules that experience different conditions significantly alter module specialization (de Kroon et al. 2005). These interactive effects do not appear to be mere outcomes of spatial or mechanical interferences. Rather, they appear to be self-organization mechanisms integrated at whole-crown level that enable temporal or spatial task splitting among modules to attain certain control over stress and resource availability. This form of environmentally inducible division of labour is likely to benefit plants that experience a clear patch structure of complementary resources and stable patterns of resource distribution on a small spatial scale (Stuefer 1998). The wild olive tree under field conditions meets these requirements: sun and shade leaf arrangement within the crown gives rise to different light environments that are complementary in the daytime and stable in the medium term, as the wild olive is an evergreen tree whose leaves do not track sun movements.

CONCLUSIONS The profitability of producing a new module within the crown of a wild olive tree depends not only on construction costs or access to light, but may heavily rely on its integration within the spatio-temporal framework of the productive processes at the crown level. We have shown complementarity of productive processes such as light interception, carbon fixation and carbon accumulation/export when integrated at the scale of the tree crown.We also showed that this temporal splitting of tasks among the two crown sectors studied derives from local specialization, which involves coordinated adjustments in crown architecture, biomass investment, and photosynthetic performance. Further research is needed to determine the relevance at the population level of the patterns observed in the present study in a single tree, and at temporal scales longer than a single day. We feel that temporal considerations will form the basis of particularly promising studies in evergreen species, such as those addressing: (1) the trade-off between optimizing leaf orientation for a fraction of the day or for a season of the year; (2) the overall consequences for plant fitness; and ultimately (3) the significance of a putative division of labour among modules in the evolution of the tree habit.

ACKNOWLEDGMENTS We wish to thank M Méndez, MD Jiménez, E Manrique, E Pérez-Corona, A Vázquez, P Vargas, M Meng and S Kunz for their valuable assistance during the experimental design, field work and sample processing. Many thanks to D Martín Flecha for his expertise and help with AutoCAD.We are also grateful to JA Delgado and to La Tahonilla staff for authorizing and assisting the greenhouse study (Excmo. Cabildo de Tenerife). Special thanks to Mr Cormac de Brun for revision of the English. We are grateful for the constructive comments of Dr Sharkey and two anonymous reviewers that significantly improved the focus of the article. This research was funded by the Spanish Ministry of Science and Education (project CGL2009-10392/BOS), by a MECFulbright fellowship to C G-V (FU2009-0068) and by an FPU grant to C. G-Y (AP2006-02283). We are also indebted to the Madrid Regional Govt. (project REMEDINAL-2, S2009/AMB-1783).

REFERENCES Augspurger C.K. & Bartlett E.A. (2003) Differences in leaf phenology between juvenile and adult trees in a temperate deciduous forest. Tree Physiology 23, 517–525. Balaguer L., Martinez-Ferri E., Valladares F., Perez-Corona M.E., Baquedano F.J., Castillo F.J. & Manrique E. (2001) Population divergence in the plasticity of the response of Quercus coccifera to the light environment. Functional Ecology 15, 124– 135. Bates D.M. & Venables W.N. (2007) splines: Regression Spline Functions and Classes. R package version 2.5.1. Bazzaz F.A. (1991) Habitat selection in plants. American Naturalist 137, S116–S130.


Within-crown photosynthetic patterns in O. europaea 809 Bouvet J.M., Vigneron P. & Saya A. (2005) Phenotypic plasticity of growth trajectory and ontogenic allometry in response to density for Eucalyptus hybrid clones and families. Annals of Botany 96, 811–821. Brodersen C.R. & Vogelmann T.C. (2007) Do epidermal lens cells facilitate the absorptance of diffuse light? American Journal of Botany 94, 1061–1066. Brodersen C.R., Vogelmann T.C., Williams W.E. & Gorton H.L. (2008) A new paradigm in leaf-level photosynthesis: direct and diffuse lights are not equal. Plant, Cell & Environment 31, 159– 164. Cade B.S. & Noon B.R. (2003) A gentle introduction to quantile regression for ecologists. Frontiers in Ecology and the Environment 1, 412–420. Carter G.A. & Smith W.K. (1985) Influence of shoot structure on light interception and photosynthesis in conifers. Plant Physiology 79, 1038–1043. Cataldi T.R.I., Margiotta G., Iasi L., Di Chio B., Xiloyannis C. & Bufo S.A. (2000) Determination of sugar compounds in olive plant extracts by anion-exchange chromatography with pulsed amperometric detection. Analytical Chemistry 72, 3902– 3907. Centritto M., Loreto F. & Chartzoulakis K. (2003) The use of low [CO2] to estimate diffusional and non-diffusional limitations of photosynthetic capacity of salt-stressed olive saplings. Plant, Cell & Environment 26, 585–594. Diaz-Espejo A., Nicolas E. & Fernandez J.E. (2007) Seasonal evolution of diffusional limitations and photosynthetic capacity in olive under drought. Plant, Cell & Environment 30, 922–933. Durkheim E. (1893) The Division of Labour in Society. The Free Press, New York, 1933. Farquhar G.D. (1989) Models of integrated photosynthesis of cells and leaves. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 323, 357–367. Field C. (1983) Allocating leaf nitrogen for the maximization of carbon gain-leaf age as a control on the allocation program. Oecologia 56, 341–347. García-Verdugo C. (2009) Procesos y mecanismos microevolutivos en el endemismo canario Olea europaea subsp. guanchica. PhD Thesis, Universidad Complutense de Madrid, Spain. García-Verdugo C., Granado-Yela C., Manrique E., de Casas R.R. & Balaguer L. (2009) Phenotypic plasticity and integration across the canopy of Olea europaea subsp. guanchica (Oleaceae) in populations with different wind exposures. American Journal of Botany 96, 1454–1461. García-Verdugo C., Méndez M., Velázquez-Rosas N. & Balaguer L. (2010) Contrasting patterns of morphological and physiological differentiation across insular environments: phenotypic variation and heritability of light-related traits in Olea europaea. Oecologia 164, 647–655. Hastie T.J. (1991) Generalized additive models. In Statistical Models in S (eds J.M. Chambers & T.J. Hastie), pp. 249–307. Chapman & Hall, London. Henriksson J. (2001) Differential shading of branches or whole trees: survival, growth, and reproduction. Oecologia 126, 482– 486. Jurik T.W., Chabot J.F. & Chabot B.F. (1979) Ontogeny of photosynthetic performance in Fragaria virginiana under changing light regimes. Plant Physiology 63, 542–547. Koenker R. (2009) quantreg: Quantile Regression. R package version 4.30. Koenker R. & Bassett G. (1978) Regression quantiles. Econometrica 46, 33–50. de Kroon H., Huber H., Stuefer J.F. & van Groenendael J.M. (2005) A modular concept of phenotypic plasticity in plants. New Phytologist 166, 73–82.

Le Roux X., Bariac T., Sinoquet H., Genty B., Piel C., Mariotti A., Girardin C. & Richard P. (2001) Spatial distribution of leaf water-use efficiency and carbon isotope discrimination within an isolated tree crown. Plant, Cell & Environment 24, 1021–1032. Moriana A., Villalobos F.J. & Fereres E. (2002) Stomatal and photosynthetic responses of olive (Olea europaea L.) leaves to water deficits. Plant, Cell & Environment 25, 395–405. Murchie E.H. & Horton P. (1997) Acclimation of photosynthesis to irradiance and spectral quality in British plant species: chlorophyll content, photosynthetic capacity and habitat preference. Plant, Cell & Environment 20, 438–448. Niinemets U., Tenhunen J.D. & Beyschlag W. (2004) Spatial and age-dependent modifications of photosynthetic capacity in four Mediterranean oak species. Functional Plant Biology 31, 1179– 1193. Pearcy R.W. & Yang W.M. (1996) A three-dimensional crown architecture model for assessment of light capture and carbon gain by understory plants. Oecologia 108, 1–12. Petit R.J. & Hampe A. (2006) Some evolutionary consequences of being a tree. Annual Review of Ecology Evolution and Systematics 37, 187–214. Potvin C., Lechowicz M.J. & Tardif S. (1990) The statistical analysis of ecological response curves obtained from experiments involving repeated measures. Ecology 71, 1389–1400. Robinson G.E. (1992) Regulation of division of labor in insect societies. Annual Review of Entomology 37, 637–665. Rubio de Casas R., Vargas P., Perez-Corona E., Manrique E., Quintana J.R., García-Verdugo C. & Balaguer L. (2007) Field patterns of leaf plasticity in adults of the long-lived evergreen Quercus coccifera. Annals of Botany 100, 325–334. Sack L., Melcher P.J., Liu W.H., Middleton E. & Pardee T. (2006) How strong is intracanopy leaf plasticity in temperate deciduous trees? American Journal of Botany 93, 829–839. Silvertown J. & Gordon D.M. (1989) A framework for plant behavior. Annual Review of Ecology and Systematics 20, 349– 366. Smith S.D., Naumberg E., Niinemets U. & Germino M. (2004) Leaf to landscape. In Photosynthetic Adaptation Chloroplast to Landscape (eds W.K. Smith, T.C. Vogelmann & C. Critchley), pp. 262–296. Springer, New York. Smith W.K., Schoettle A.W. & Cui M. (1991) Importance of the method of leaf-area measurement to the interpretation of gasexchange of complex shoots. Tree Physiology 8, 121–127. Sprugel D.G. (2002) When branch autonomy fails: Milton’s Law of resource availability and allocation. Tree Physiology 22, 1119– 1124. Sprugel D.G., Hinckley T.M. & Schaap W. (1991) The theory and practice of branch autonomy. Annual Review of Ecology and Systematics 22, 309–334. Stuefer J.F. (1998) Two types of division of labour in clonal plants: benefits, costs and constraints. Perspectives in Plant Ecology Evolution and Systematics 1, 47–60. Stuefer J.F., DeKroon H. & During H.J. (1996) Exploitation of environmental heterogeneity by spatial division of labour in a clonal plant. Functional Ecology 10, 328–334. Tenhunen J.D., Lange O.L. & Braun M. (1981) Midday stomatal closure in Mediterranean type sclerophylls under simulated habitat conditions in an environmental chamber.2. Effect of the complex of leaf temperature and air humidity on gas exchange of Arbutus unedo and Quercus ilex. Oecologia 50, 5– 11. Uemura A., Harayama H., Koike N. & Ishida A. (2006) Coordination of crown structure, leaf plasticity and carbon gain within the crowns of three winter-deciduous mature trees. Tree Physiology 26, 633–641.


810 C. Granado-Yela et al. Valladares F. & Niinemets U. (2008) Shade tolerance, a key plant feature of complex nature and consequences. Annual Review of Ecology Evolution and Systematics 39, 237–257. Valladares F. & Pearcy R.W. (1998) The functional ecology of shoot architecture in sun and shade plants of Heteromeles arbutifolia M. Roem., a Californian chaparral shrub. Oecologia 114, 1–10. Wahl L.M. (2002) Evolving the division of labour: generalists, specialists and task allocation. Journal of Theoretical Biology 219, 371–388. Wayne P.M. & Bazzaz F.A. (1993) Birch seedlings responses to daily time courses of light in experimental forest gaps and shadehouses. Ecology 74, 1500–1515. Wei Y., Pere A., Koenker R. & He X.M. (2006) Quantile regression methods for reference growth charts. Statistics in Medicine 25, 1369–1382.

White J. (1979) The plant as a metapopulation. Annual Review of Ecology and Systematics 10, 109–145. Yoshimura K. (2010) Irradiance heterogeneity within crown affects photosynthetic capacity and nitrogen distribution of leaves in Cedrela sinensis. Plant, Cell & Environment 33, 750–758. Yu K.M., Lu Z.D. & Stander J. (2003) Quantile regression: applications and current research areas. Journal of the Royal Statistical Society – Series D: the Statistician 52, 331–350. Received 9 September 2010; accepted for publication 26 December 2010
