http://orcid.org/0000-0003-4233-3228
Figures
Abstract
Photosynthetic-nitrogen use efficiency (PNUE) is a useful trait to characterize leaf physiology and survival strategy. PNUE can also be considered as part of ‘leaf economics spectrum’ interrelated with leaf nutrient concentrations, photosynthesis and respiration, leaf life-span and dry-mass investment. However, few studies have paid attention to PNUE of N-fixing tree seedlings in subtropical China. In this study, we investigated the differences in PNUE, leaf nitrogen (N) allocation, and mesophyll conductance (gm) in Dalbergia odorifera and Erythrophleum fordii (N-fixing trees), and Betula alnoides and Castanopsis hystrix (non-N-fixing trees). PNUE of D. odorifera and E. fordii were significantly lower than those of B. alnoides and C. hystrix mainly because of their allocation of a lower fraction of leaf N to Rubisco (PR) and bioenergetics (PB). Mesophyll conductance had a significant positive correlation with PNUE in D. odorifera, E. fordii, and B. alnoides, but the effect of gm on PNUE was different between species. The fraction of leaf N to cell wall (PCW) had a significant negative correlation with PR in B. alnoides and C. hystrix seedling leaves, but no correlation in D. odorifera and E. fordii seedling leaves, which may indicate that B. alnoides and C. hystrix seedling leaves did not have enough N to satisfy the demand from both the cell wall and Rubisco. Our results indicate that B. alnoides and C. hystrix may have a higher competitive ability in natural ecosystems with fertile soil, and D. odorifera and E. fordii may grow well in N-poor soil. Mixing these non-N-fixing and N-fixing trees for afforestation is useful for improving soil N utilization efficiency in the tropical forests.
Citation: Tang J, Sun B, Cheng R, Shi Z, Luo D, Liu S, et al. (2019) Seedling leaves allocate lower fractions of nitrogen to photosynthetic apparatus in nitrogen fixing trees than in non-nitrogen fixing trees in subtropical China. PLoS ONE 14(3): e0208971. https://doi.org/10.1371/journal.pone.0208971
Editor: Dafeng Hui, Tennessee State University, UNITED STATES
Received: November 14, 2018; Accepted: February 20, 2019; Published: March 4, 2019
Copyright: © 2019 Tang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This research was sponsored by Fundamental Research Funds of Chinese Academy of Forestry (CAF), http://www.caf.ac.cn/ (funder: Science and technology management office, CAF, no. CAFYBB2018ZA003, received by ZS); the National Key Research and Development Program, http://www.nsfc.gov.cn/ (funder: Ministry of National Science and Technology, China, no. 2016YFC0502104-02, received by ZS) and projects of the National Natural Science Foundation of China, http://www.nsfc.gov.cn/ (funder: Ministry of National Science and Technology, China, no.31290223 and 31570240, received by ZS). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Nitrogen (N) is very important for plants leaves, because main function of leaves- photosynthesis need a lot of N [1,2], and there was a positive correlation between photosynthetic capacity and N content in many species. However, there existed interspecific difference in the photosynthesis–N relationship [3]. Many researchers use photosynthetic-N use efficiency (PNUE, the ratio of light-saturated net CO2 assimilation rate (Amax') to leaf N content per area (Narea) [4]) to show how efficiently N resources are used during photosynthesis, and studies have been conducted on a variety of species [3,5,6]. N-fixing species could convert N from the air through legume bacteria, and always have enough N in leaves [7–9]. Studies have shown that N-fixing trees had lower Amax' and higher Narea, which resulted in a lower PNUE [10, 11]. These contradicting results may imply that some N-fixing species use a different strategy to utilize N compared to non-N-fixing species.
Many factors could affect PNUE, and the most important factor is leaves photosynthetic N allocation [12]. Rubisco is the most abundant enzyme in C3 plants [13], and it is the key factor in carbon assimilation [14]. Many researchers have found a positive correlation between leaf N fraction in Rubisco (PR) and PNUE in various plants [15–16]. Bioenergetics and the light-harvesting components could also influence PNUE in some plants [17]. Apart from photosynthetic, leaf cell walls, which could protect leave cell and influence leaf life-span also need a lot of N to synthesize [18]. Trade-offs may occur between N allocation to cell walls and Rubisco [18–20]. However, some studies have shown that these trade-offs only exist in individuals of the same species [16] or species lacking N in leaves [18, 21].
Carbon dioxide is an important raw material for photosynthesis [22], and CO2 partial pressure is important for Rubisco activity; this is because O2 is a competitive inhibitor of the C assimilatory reaction of Rubisco, promoting the Rubisco oxidation reaction [23]. A significant negative correlation between Ci (intercellular CO2 concentration)-Cc (CO2 concentration at carboxylation site) and PNUE was found in Populus cathayana [24]. Nitrogen is also involved in carbonic anhydrases and aquaporins [25]. These proteins play a role in mesophyll conductance (gm) by changing the nature of the diffusing molecule [26] and facilitating CO2 diffusion through membranes [27]. Therefore, PNUE may be influenced by gm [25]. A significant positive correlation was found between mesophyll conductance (gm) and PNUE in six Populus genotypes [28].
What reason causes the low PNUE in N-fixing plants? One possible explanation is that the percentage of N in the photosynthetic apparatus is lower in the N-fixing trees [10, 11]. However, these studies neglect that gm and the fraction of leaf N to cell wall (PCW) could also influence PNUE [19, 20, 29].We studied the factors that affect PNUE in both N-fixing and non-N-fixing large trees in a previous study and found PR and fraction of leaf N to bioenergetics (PB) to be the main factors; the effects of gm and PCW were relatively small [30], but the effects in N-fixing tree seedlings remained unclear.
Dalbergia odorifera, Erythrophleum fordii, Betula alnoides, and Castanopsis hystrix are suitable for forestation in southern subtropical China and have high economic values [31–34]. D. odorifera and E. fordii are both evergreen N-fixing trees, whereas B. alnoides and C. hystrix are both non-N-fixing, and deciduous and evergreen, respectively. The objectives of our study are as follows: 1) understand how PNUE varies among D. odorifera, E. fordii, B. alnoides, and C. hystrix seedlings; 2) quantify the relationship between PNUE related to leaf N allocation and diffusional conductances to CO2 in seedlings.
Materials and methods
Study area and plant material
This study was carried out in Experimental Center of Tropical Forestry (22°7′19″–22°7′22″N, 106°44′40″–106°44′44″E) of the Chinese Academy of Forestry located in Guangxi Pingxiang, China. The location has a subtropical monsoon climate with distinct dry and wet periods where the mean annual temperature is 21°C. The mean monthly minimum and maximum temperatures are 12.1°C and 26.3°C. The mean annual precipitation is 1400 mm, and it occurs mainly from April to September. Active accumulated temperature above 10°C is 6000–7600°C. The total annual sunshine duration is 1419 hours [35,36].
Seeds of D. odorifera, E. fordii, and C. hystrix were collected from a single tree for each species, and B. alnoides seedlings were somaclone. The seeds of D. odorifera, E. fordii, and C. hystrix were germinated in a seedbed in February 2014 and B. alnoides went through budding at the same time. When the seedlings were approximately 20 cm tall, 30 similarly sized seedlings per species were individually transplanted to pots (5.4 L, filled with washed river sand) and established in an open site at the Experimental Center of Tropical Forestry in March 2014. From April to June, each pot received the same nutrient solution (0.125 g N and 0.11 g P, Hyponex M. Scott & Sons, Marysville, OH, USA) once a week, and was watered every day to keep the soil moist. Natural light (100% of light in the field) was used for illumination.
Determination of gas exchange measurements
Gas exchange parameters were determined with a Li-6400 portable photosynthesis system (LI-COR, Lincoln, NE, USA) on sunny days from 8 am to 10 am in July and August 2014. Seven healthy and newly emerged leaves exposed to the sun in each tree species were chosen (one leaf per individual healthy tree). Photosynthetic response to photosynthetic photon flux density (PPFD) and intercellular CO2 concentration (Ci, μmol mol–1) were determined for each leaf (seven repetitions in each species): Under 380 μmol mol–1 of leaf chamber CO2 concentration (the average air CO2 concentration in the day time), the photosynthetic rates were measured under photon flux densities of 1500, 1200, 1000, 800, 600, 400, 200, 150, 100, 80, 50, 30, 20, 10 and 0 μmol m–2 s–1 [37]. Under a saturated PPFD, the photosynthetic rates were detected using the same leaf-under leaf chamber CO2 concentrations of 380, 200, 150, 100, 80, 50, 380, 600, 800, 1000, 1200, 1500, 1800 and 2000 μmol mol–1 [28]. Relative humidity of the air in the leaf chamber was maintained at 60–70%, and leaf temperature was set at 30°C. The net photosynthetic rate (An, μmol m–2 s–1), stomatal conductance (gs, mol CO2 m–2 s–1), and Ci of each sampled leaf were recorded ten times after 200 s under each PPFD and CO2 concentration. Then light-saturated net CO2 assimilation rate (Amax', μmol m–2 s–1), light-saturated day respiration rate (Rd, μmol m–2 s–1) and light- and CO2-saturated net CO2 assimilation rate (Amax, μmol m–2 s–1) were measured or calculated. For further details see Tang et al. [30].
Determination of chlorophyll fluorescence and mesophyll conductance
Fluorescence yield was measured with a Li–6400 leaf chamber fluorometer (6400–40, LI-COR, Lincoln, Nebraska, USA), using the same leaf with seven repetitions of each species. Chamber temperature was maintained at 28–32°C, and chamber air relative humidity was maintained at 60–70%. Chamber CO2 concentration was set to 380 μmol mol–1. PPFD was set to light saturation point. Constant values of fluorescence yield (⊿F/Fm′) of each leaf sample were recorded 10 times after 200 s [38]. We used Loreto et al. [39] methods to calculate the photosynthetic electron transport rate (Jf, μmol m–2 s–1):
(1)
Leafreflu (leaf absorptance valued) and PARDistPhotosys (the fraction of quanta absorbed by photosystem II) were 0.85 [40] and 0.5 [39], respectively. We used the variable J method described by Harley et al. [41], which has been used in recent years [42–45] to calculate mesophyll conductance (gm, mol CO2 m–2 s–1):
(2)
Where Rd, Ci, and Amax′ were determined from gas exchange measurements. The CO2 photo compensation point (Γ*, μmol mol–1) value was 54.76 at 30°C according to Bernacchi et al [46].
Because the Harley method should calibrate the ETR using Chl fluorescence and gas exchange under low O2, we used the experience value instead (Leafreflu = 0.85) [30]. We also used Ethier and Livingston [47] and the exhaustive dual optimization (EDO) method [48] to calculate gm. We used software based on the Ethier and Livingston method developed by Sharkey et al. [49] to get gm, and uploaded our data through a website (http://www.leafweb.org) to get gm calculated by the EDO method.
Determination of Vcmax and Jmax
The mean value of gm calculated from three methods was used to calculate CO2 concentration in chloroplasts (Cc, μmolmol–1):
(3)
Then Cc was used to fit an An-Cc curve, followed by the maximum carboxylation rate (Vcmax, μmol m–2 s–1) calculated according to Farquhar et al. [14], and the maximum electron transport rate (Jmax, μmol m–2 s–1) calculated according to Loustau et al. [50]. The fitting model used in vivo Rubisco kinetics parameters (Ko, Kc, and their activation energy) measured by Niinemets and Tenhunen [12].
Analysis of quantitative limitations of photosynthetic capacity
The relative controls on photosynthetic capacity imposed by stomatal conductance (ls, %), mesophyll diffusion (lm, %), and biochemical capacity (lb, %) were calculated following the quantitative limitation analysis of Grassi and Magnani [51] as applied in Tomás et al [52], Peguero-Pina et al. [53, 54] and Nha et al. [55]. Different fractional limitations, ls, lm, and lb (ls + lm + lb = 1) were calculated as:
(4)
(5)
(6)
Where gs and gm were used in light-saturated and atmospheric CO2 concentration was 380 μmol mol-1, and gm was the mean value of three methods. The gtot is the total conductance to CO2 from ambient air to chloroplasts (the sum of the inverse CO2 serial conductances gs and gm). The ∂An/∂Cc was calculated as the slope of An–Cc response curves over a Cc range of 50–100 μmol mol−1 [53, 54].
Determination of additional leaf traits
Leaf samples used for gas exchange measurements and leaves which size was similar to leaves used for determine photosynthesis was taken. Leaf areas were measured with a scanner (Perfection v700 Photo, Epson, Nagano-ken, Japan). Leaf dry weights were measured using an analytic balance after being oven-dried at 80°C for 48 h, then leaf mass per area (LMA, g m-2) was calculated.
Dried leaf samples were ground into a dry flour. Organic carbon (C) concentration was determined by the potassium dichromate-sulfuric acid oxidation method (Cmass mg g-1). Nitrogen concentration was determined by a VELP automatic Kjeldahl N determination apparatus (UDK-139, Milano, Italy), and leaf N per mass (Nmass, mg g-1) and per area (Narea g m-2) values were calculated [30]. The PNUE (μmol mol–1 s–1) was then calculated by the formula:
(7)
Where 14 is the atomic mass of nitrogen.
Chlorophylls were extracted by direct immersion: 0.2 g of frozen leaves were cut into small pieces which were 5–10 mg. Leaf pieces were placed into a volumetric flask and 25 mL of 95% (v/v) alcohol was added. The flask was kept in the dark for 24 h. The absorbance of the extracts was measured at 665 nm and 649 nm with a Shimadzu visible-ultraviolet spectrophotometer (UV 2250, Fukuoka, Japan). Cell wall N content was calculated according to Onoda et al. [19]: 1 g of leaves were powdered in liquid N and suspended in sodium phosphate buffer (pH 7.5), the homogenate was centrifuged at 2500 g for 5 min, and the supernatant was discarded. The pellet was washed with 3% (w/v) SDS, amyloglucosidase (35 U ml–1, Rhizopus mold, Sigma, St Louis, MO, USA), and 0.2 M KOH, then heated and centrifuged. The pellet was then washed with distilled water and ethanol, and oven dried (75°C) for 2 days. Nitrogen in the final pellet was determined using an automatic Kjeldahl apparatus (VELP Scientifica, Usmate, Italy). The fraction of leaf N allocated to cell walls (PCW) represents the ratio of cell wall N content to the total N content.
Calculation of N allocation in the photosynthetic apparatus
The fraction of leaf N allocated to Rubisco (PR), bioenergetics (PB), and the light-harvesting components (PL) (g g–1)were calculated from Vcmax, Jmax and chlorophyll contents using the method of Niinemets and Tenhunen [12], which has been widely used in recent years [15, 56–58]:
(8)
(9)
(10)
Where CChl is the chlorophyll concentration (mmol g–1), Vcr is the specific activity of Rubisco (μmol CO2 g–1 Rubisco s–1), Jmc is the potential rate of photosynthetic electron transport (μmol electrons μmol–1Cyt f s–1), and CB is the ratio of leaf chlorophyll to leaf N during light-harvesting (mmol Chl (g N)–1). Where Vcr, Jmc, and CB were calculated according to Niinemets and Tenhunen [12].The fraction of leaf N allocated to the photosynthetic apparatus (PP) was calculated as the sum of PR, PB, and PL.
Statistical analysis
Differences between the seedling leaves were analyzed using one-way analysis of variance (ANOVA), and a post hoc test (Tukey’s test) was conducted if the differences were significant. The significance of the correlation between each pair of variables was tested with a Pearson correlation (two-tailed). All analyses were carried out using Statistical Product and Service Solutions 17.0 (SPSS17.0, Chicago, IL, USA).
Results
PNUE in four seedling leaves
There were significant differences in PNUE between the leaves of the four seedlings (P <0.001, Table 1). The PNUE in B. alnoides and C. hystrix seedling leaves were higher than those in D. odorifera and E. fordii, which was mainly attributed to their lower Narea and Nmass values. The highest PNUE in B. alnoides (120.54 μmol mol–1 s–1) was 2.6 times the lowest, found in E. fordii (45.92 μmol mol–1 s–1). However, Narea and Nmass in B. alnoides were 48.75% and 45.21% lower than in E. fordii, respectively (Table 1). There were no significant differences between B. alnoides, C. hystrix, and D. odorifera seedling leaves in Amax′ and the value in E. fordii (6.60 μmol m–2 s–1) was the smallest (Table 1). The LMA of C. hystrix (100.13 g m-2) was the highest (Table 1). E. fordii and B. alnoides seedling leaves had higher Cmass than D. odorifera and C. hystrix, but C/N was higher in B. alnoides and C. hystrix seedling leaves than D. odorifera and E. fordii (Table 1).
Photosynthetic parameters in four seedling leaves
Analysis of the quantitative limitations of photosynthesis revealed that photosynthetic capacity was mainly limited by diffusional processes (ls and lm), whereas biochemical limitations (lb) were only between 0.33% and 0.45% of the total for all studied species (Table 2).
Photosynthetic parameters were shown in Table 3 and Table 4. The Vcmax and Jmax in E. fordii were higher than the other three species (Table 3) but the statistically significant values (F-ratios) were lower than PNUE (Table 1). Stomatal conductance (gs, 0.100 mol CO2 m–2 s–1) and Ci (292.88 μmol mol–1) in B. alnoides seedling leaves were higher than the other three species (Table 4). Moreover, gm-Harley in B. alnoides (0.136 mol CO2 m–2 s–1) was higher than the other three species but gm-Ethier (0.140 mol CO2 m–2 s–1) and gm-Gu (0.160 mol CO2 m–2 s–1) was highest in D. odorifera (Table 4). The Cc in B. alnoides seedling leaves (all three methods) was higher than the other three species (Table 4).
Leaf N allocation in four species seedling leaves
There were significant differences in leaf N allocation between the four species (P <0.001, Table 5). The PP was higher than PCW in four species seedling leaves (Table 5). The PP was 3.9 times of the PCW in D. odorifera, 5.4 times in E. fordii, 2.0 times in B. alnoides and 1.6 times in C. hystrix. Where PR>PL>PB in D. odorifera, E. fordii, and B. alnoides seedling leaves, and PR>PL = PB in C. hystrix seedling leaves.
The PP in B. alnoides and C. hystrix seedling leaves (both were 0.44 g g–1) were higher than D. odorifera and E. fordii (both were 0.27 g g–1). The PR and PB in B. alnoides and C. hystrix seedling leaves were also higher than in D. odorifera and E. fordii. The PL in B. alnoides was the highest (0.12 g g–1), followed by D. odorifera (0.10 g g–1), C. hystrix (0.07 g g–1), and E. fordii (0.06 g g–1).
Relationship between PNUE and affecting factors
There was a positive relationship between gm and PNUE (P < 0.05), in D. odorifera, E. fordii, and B. alnoides, but not in C. hystrix (Fig 1). Both PP, PR, and PB had a significant positive correlation with PNUE in these species (P < 0.001) (Fig 2).
The determination coefficient (R2) and P-value were also shown. The lines fitted separately for four species were significantly different (P < 0.05) according to the result of a one-way ANCOVA with PNUE as a dependent variable, tree species as fixed factors, and gm as a covariate.
Regression analysis of the fraction of leaf N allocated to (a) Rubisco (PR), (b) bioenergetics (PB), (c) light-harvesting components (PL), and (d) the photosynthetic apparatus (PP) with photosynthetic-N use efficiency (PNUE) in four species seedling leaves. The determination coefficient (R2) and P-value were also shown. Only one line was fitted for four species, because there was no significant difference (P >0.05) according to the result of a one-way ANCOVA with PNUE as a dependent variable, tree species as fixed factors, and PR, PB, PL, or PP as a covariate.
The relationship between PCW and PR in B. alnoides (P = 0.022) and C. hystrix (P = 0.011) seedling leaves were more significant than in D. odorifera (P = 0.409) and E. fordii (P = 0.637). Regression analysis of PCW with PR in B. alnoides seedling leaves was within the shaded zone; C. hystrix was on the shaded zone; D. odorifera and E. fordii were under the shaded zone (Fig 3).
The determination coefficient (R2) and P-value were also shown. The shaded zone was drawn according to this hypothesis: when PCW was 0.300 g g–1, the rest (0.700 g g–1) were soluble and thylakoid protein, Rubisco represented one quarter to one-third of the N in soluble and thylakoid protein, PR valued 0.175–0.233 g g–1(right side of shaded zone). When PCW was valued 0.000 g g–1 in limiting case (does not exist in reality), all the rest (1.000 g g–1) were soluble and thylakoid protein, PR valued 0.250–0.333 g g–1(left side of shaded zone). For more information see Harrison et al. [21]. The lines fitted separately for four species were significantly different (P < 0.05) according to the result of a one-way ANCOVA with PR as a dependent variable, tree species as fixed factors, and PCW as a covariate.
Discussion
The range of PNUE in these tree seedlings was 45.92–120.54 μmol mol–1 s–1(Table 1) which was close to six Fagus sylvatica populations (68.74–122.22 μmol mol–1 s–1) [59] and four Quercus species (approximately 60–150 μmol mol–1 s–1) [20]; lower than P. cathayana (171.64–213.36 μmol mol–1 s–1) [17] and S. alterniflora (171.64–213.36 μmol mol–1 s–1) [15] under different N deposition. Wright et al. summed up PNUE in 710 species and the range was between 10 and 500 μmol mol–1 s–1 [60]; therefore, our results seem reasonable. Shrubs and trees usually have a low PNUE and grasses usually have high value [60]. Fast growing herbaceous species may have a PNUE higher than 200 μmol mol–1 s–1, whereas values for evergreen woody species can be lower than 50 [1]. Our values are within the medium range.
The overall result highlights a substantial difference between N-fixing and non-N-fixing tree seedling leaves in PNUE (Table 1). The variation of PNUE may be attributable to plant evolution and natural selection [61]. Low PNUE species compensate for their low productivity with a long leaf life-span [20]; stress-tolerant species [62] and late successional species [63] usually have low PNUE values. Therefore, low PNUE in D. odorifera and E. fordii may lead to high stress-tolerance traits and increase competitiveness in poor soil [64]. Higher PNUE species such as B. alnoides and C. hystrix could grow faster [20] and have a stronger competitive ability in ecosystems with fertile soil [65]. The PNUE tended to be lower for species at the ‘slow-return’ end of the leaf economics spectrum [60], and according to the ‘leaf economics spectrum’, at the slow-return end are species with long leaf life-span, expensive high-LMA leaf construction, low nutrient concentrations, and low rates of photosynthesis and respiration [4], and therefore, it can be concluded that two N-fixing species were at the ‘slow-return’ end of the leaf economics [4, 60]. Because these species live in the same area, we believe that mix these non-N-fixing and N-fixing trees for afforestation is useful for improving soil N utilization efficiency in this place.
As PNUE is the ratio of Amax′ and Narea, changes of Amax′ and Narea affect PNUE. We found significant F-ratios in Amax′ between the four species’ seedling leaves was 3.441, lower than in Narea which was 36.314. Therefore, a change of Narea was the main reason affecting PNUE in these four species. We suspect that N-fixing species which could gain N from air by legume bacteria [7–9], may have both higher Narea and Amax′, but our results did not support this speculation. Which reason limited Amax′ in two N-fixing species? firstly, relative stomatal (ls), and mesophyll (lm) were main reasons limited photosynthesis ability in these trees (Table 2), secondly, two N-fixing species didn’t show significant higher gs, gm, Cc, Vcmax or Jmax than non-N-fixing species (Tables 3 and 4). Therefore, we believe that a large proportion of N in the leaves of N-fixing plants did not used for photosynthesis.
N-fixing trees D. odorifera and E. fordii had significantly higher Narea than B. alnoides and C. hystrix (Table 1). Because Narea = Nmass× LMA, Narea may also be affected by LMA besides N content Nmass. The difference of LMA between species was far lower than the difference of Nmass (Table 1). Therefore, the significantly higher Nmass, caused the significantly higher Narea in D. odorifera and E. fordii. The low C/N ratio also showed high N in D. odorifera and E. fordii (Table 1). These results agreed with earlier studies [10, 11] and our study in five Fagaceae and five Leguminosae tree species [30]. However, one study reported that N-fixing trees had both higher Narea and Amax′[66]. The relationship of Narea and Amax′varies in different species [60], because different species have their own N allocation patterns. The N allocation in photosynthesis was more important than the total leaf N for photosynthesis [67].
Lower PP, PR, and PB were main reasons that led to lower PNUE in N-fixing tree species (D. odorifera and E. fordii). These results agreed with previous studies [10, 11, 51] and our study on five Fagaceae tree species and five Leguminosae big tree species [30]. Rubisco catalyzes the limiting step for photosynthetic capacity [14]. A positive correlation between Amax′ and Rubisco has been frequently reported [16, 19]. An improved fraction of leaf N allocated to Rubisco could maximize the use of leaf N in photosynthesis. It should be noted that although there was a significant difference in N allocation proportion between N-fixing trees and non-N-fixing trees, there were smaller differences in N allocation quantity in Rubisco, bioenergetics, photosynthetic apparatus, cell wall, and other parts in the four species seedling leaves (mass and area, see S2 Table). The Nmass largely affected the N allocation to the photosynthetic apparatus and PCW.
The gm could also influence the variation in PNUE through N allocation [25]. There was a significant positive relationship between gm and PNUE in D. odorifera, E. fordii, and B. alnoides, but the effect of gm to PNUE was not consistent between species (Fig 1). Broeckx et al. [28] also found this relationship in six poplar (Populus) genotypes, and Nha et al. [55] found gm does not contribute to greater PNUE in temperate forest. We also found gm of ten Fagaceae and Leguminosae species big trees was not significantly related to the PNUE. The effect of gm on PNUE may also age-related.
A significant negative correlation between PCW and PR in B. alnoides and C. hystrix (P < 0.05) suggested a trade-off between N allocation to Rubisco and cell walls, whereas no trade-off was detected in D. odorifera and E. fordii (Fig 3). A similar trade-off was found in Polygonum cuspidatum [19], Quercus species [20], Mikania micrantha and Chromolaena odorata [37]; but this relationship does not exist in some other trees [16]. Some researchers believed that the main influencing factors were whether leaf N could meet the needs of both cell wall N and Rubisco N [15, 21]. We used the method described by Harrison et al. [21] to determine whether leaf N could meet these two needs: the regression analysis of PCW with PR in B. alnoides seedling leaves was within the shaded zone (the shaded zone represents the distribution area of PCW and PR when a trade-off exists), C. hystrix was on the shaded zone which means that B. alnoides and C. hystrix had high PCW and PR and therefore leaf N could not meet both needs, these two factors may affect each other. We believe the high POther (Table 5, possibly composed of free amino acids [68] and inorganic N (NO3–, NH4+) [69]) weakens the correlation between Rubisco and cell wall N. It must be noted that C. hystrix showed a unique relationship between PCW and PR (on the shaded zone), which means higher PCW and PR than the results of Harrison et al. [21]. More trees need to be studied to determine the distribution area of PCW and PR when a trade-off exists.
Excessive storage of N in N-fixing tree species may reduce their PNUE but may be useful for future physiological processes such as reproduction [17]. Storage of N could buffer changes in other N pools such as cell wall N [19, 20, 37] (Fig 3). Evergreen tree leaves with low PNUE have multiple roles in nutrient conservation, nutrient storage, stress tolerance, herbivore deterrence, and photosynthesis [3]. We should consider that some Rubisco can also function as N storage and may not be involve in photosynthesis [70, 71]. This type of Rubisco might lead to greater rates of photosynthesis under suboptimal conditions [3]. Therefore, Rubisco N calculated by the model of Farquhar et al. [14] might be N in activated Rubisco. Using chemical methods to extract and determine Rubisco N content could be useful [20, 72].
We used to do experiment with C. hystrix big trees, and its PNUE was 74.34±8.54 μmol mol–1 s–1 [30], smaller than its seedlings (Table 1). C. hystrix big tree also had higher PCW (0.46 g g-1) than seedlings, but its PP (0.26 g g-1), PR (0.20 g g-1), PB (0.041 g g-1) and PL (0.014 g g-1) [30] were lower than seedlings (Table 1). Seedlings with high PNUE could grow fast, reach the canopy earlier and increased ability of competition for light [5]. Big trees with higher PCW could better resist the environmental stress in canopy, such as typhoon, insect attack and diseases [19]. We suspect that trade-offs for N allocation to photosynthesis versus cell walls may also exist at different stages of a tree's growth, in order to meet the N demand in different growth stages.
Although both the B. alnoides and C. hystrix are non-N-fixing broadleaf plants, and have some similar functional traits, there were significant differences showed in Nmass, LMA, gm, Cc, and PCW (Tables 1, 4 and 5). B. alnoides is a deciduous broad-leaved plant and C. hystrix is an evergreen broad-leaf plant. In order to contribute to a longer leaf life span, evergreen broad-leaf plants should improve leaf tolerance to environmental disturbance [73,74], reflected in higher LMA [60], and PCW [16]. Higher defensive investment could also reduce Nmass [16]. Simultaneously, if higher LMA is a result of mesophyll cell wall thickening, it will reduce gm and Cc [75, 76], and variations in LMA are often inversely correlated with gm and Cc [77, 78], consistent with the results of those two species.
Conclusions
This study indicated that PNUE was significantly lower in two N-fixing trees (D. odorifera and E. fordii) than that in two non-N-fixing trees (B. alnoides and C. hystrix). This finding was mainly attributed to lower PR and PB. B. alnoides and C. hystrix optimized their leaf N allocation to photosynthesis. Although gm had a significant positive correlation with PNUE in D. odorifera, E. fordii, and B. alnoides, the effect of gm on PNUE was different between species. PCW had a significant negative correlation with PR in B. alnoides and C. hystrix seedling leaves, but there was no significant correlation between PCW and PR in D. odorifera and E. fordii seedling leaves, which may indicate that B. alnoides and C. hystrix seedling leaves did not have enough N to satisfy the demand from both the cell wall and Rubisco. B. alnoides and C. hystrix with higher PNUE may have a higher competitive ability in natural ecosystems with fertile soil. Our results indicate that mixing these non-N-fixing and N-fixing trees for afforestation is useful for improving soil N utilization efficiency in the tropical forests.
Supporting information
S1 Table. Chlorophyll contents (chlorophyll a, chlorophyll b, chlorophyll a+b and Chla/b) in four species seedling leaves.
Mean values (± SD) were shown (n = 7). Different letters indicated significant differences between species (Tukey’s test, P<0.05). Statistically significant F-ratios were denoted by *P<0.05, **P<0.01, ***P<0.001.
https://doi.org/10.1371/journal.pone.0208971.s001
(DOCX)
S2 Table. Quantity of leaf N (per area and per mass) allocated to Rubisco (QRarea, QRmass), bioenergetics (QBarea, QBmass), light-harvesting components (QLarea, QLmass), photosynthetic apparatus (QParea, QPmass), cell wall (QCWarea, QCWmass), and other parts (QOther-area, QOther-mass) in four species seedling leaves.
Mean values (± SD) were shown (n = 7). Different letters indicated significant differences between species (Tukey’s test, P<0.05). Statistically significant F-ratios were denoted by *P<0.05, **P<0.01, ***P<0.001.
https://doi.org/10.1371/journal.pone.0208971.s002
(DOCX)
Acknowledgments
The authors would like to thank the Experimental Center of Tropical Forestry, Chinese Academy of Forestry for providing experimental apparatus and help with measurements.
References
- 1.
Field C, Mooney HA. The photosynthesis nitrogen relationship in wild plants. In: Givnish TJ (eds) On the Economy of Plant Form and Function: Cambridge University Press, Cambridge.1986; pp: 25–55.
- 2. Balotf S, Islam S, Kavoosi G, Kholdebarin B, Juhasz A, et al. How exogenous nitric oxide regulates nitrogen assimilation in wheat seedlings under different nitrogen sources and levels. Plos One. 2018; 13: e0190269. pmid:29320529
- 3. Warren CR, Adams MA. Internal conductance does not scale with photosynthetic capacity: implications for carbon isotope discrimination and the economics of water and nitrogen use in photosynthesis. Plant Cell Environ. 2006; 29: 192–201. pmid:17080635
- 4. Wright IJ, Reich PB, Westoby M, Ackerly DD, Baruch Z, et al. The worldwide leaf economics spectrum. Nature. 2004; 428: 821–827. pmid:15103368
- 5. Hikosaka K. Interspecific difference in the photosynthesis-nitrogen relationship: patterns, physiological causes, and ecological importance. J Plant Res. 2004; 117: 481–494. pmid:15583974
- 6. Hikosaka K. Mechanisms underlying interspecific variation in photosynthetic capacity across wild plant species. Plant Biotechnology. 2010; 27: 223–229.
- 7.
Harris W, Baker MJ, Williams WM. Population dynamics and competition. White Clover Wallingford, UK CAB International. 1987; 205–297.
- 8. Chen LY, Zhao J, Zhang RY, Wang SM. Effects of nitrogen and phosphorus fertilization on legumes in Potentilla fruticosa shrub in alpine meadow. Ecol Sci, 2012; 9: 512–517.
- 9. Reed SC. Disentangling the complexities of how legumes and their symbionts regulate plant nitrogen access and storage. New Phytol. 2017; 213: 478–480. pmid:28000933
- 10. Novriyanti E, Watanabe M, Makoto K, Takeda T, Hashidoko Y, et al. Photosynthetic nitrogen and water use efficiency of acacia and eucalypt seedlings as afforestation species. Photosynthetica. 2012; 50: 273–281.
- 11. Zhu JT, Li XY, Zhang XM, Yu Q, Lin LS. Leaf nitrogen allocation and partitioning in three groundwater-dependent herbaceous species in a hyper-arid desert region of north-western China. Aus J Bot. 2012; 60: 61–67.
- 12. Niinemets Ü, Tenhunen JD. A model separating leaf structural and physiological effects on carbon gain along light gradients for the shade-tolerant species Acer saccharum. Plant Cell Environ. 1997; 20: 845–866.
- 13. Evans JR. Photosynthesis and nitrogen relationships in leaves of C3 plants. Oecologia. 1989; 78: 9–19. pmid:28311896
- 14. Farquhar GD, von Caemmerer S, Berry JA. A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta. 1980; 149:78–90. pmid:24306196
- 15. Qing H, Cai Y, Xiao Y, Yao YH, An SQ. Leaf nitrogen partition between photosynthesis and structural defense in invasive and native tall form Spartina alterniflora populations: effects of nitrogen treatments. Biol Invasions. 2012; 14: 2039–2048.
- 16. Hikosaka K, Shigeno A. The role of Rubisco and cell walls in the interspecific variation in photosynthetic capacity. Oecologia. 2009; 160: 443–451. pmid:19288136
- 17. Chen L, Dong T, Duan B. Sex-specific carbon and nitrogen partitioning under n deposition in Populus cathayana. Trees. 2014; 28:793–806.
- 18. Onoda Y, Wright IJ, Evans JR, Hikosaka K, Kitajima K, et al. Physiological and structural tradeoffs underlying the leaf economics spectrum. New Phytol. 2017; 214: 1447–1463. pmid:28295374
- 19. Onoda Y, Hikosaka K, Hirose T. Allocation of nitrogen to cell walls decreases photosynthetic nitrogen-use efficiency. Funct Ecol. 2004; 18: 419–425.
- 20. Takashima T, Hikosaka K, Hirose T. Photosynthesis or persistence: nitrogen allocation in leaves of evergreen and deciduous Quercus species. Plant Cell Environ. 2004; 27: 1047–1054.
- 21. Harrison MT, Edwards EJ, Farquhar GD, Nicotra, AB, et al. Nitrogen in cell walls of sclerophyllous leaves accounts for little of the variation in photosynthetic nitrogen-use efficiency. Plant Cell Environ. 2009; 32: 259–270. pmid:19054350
- 22. Niinemets Ü, Flexas J, Peñuelas J. Evergreens favored by higher responsiveness to increased CO2. Trends Ecol Evol. 2011; 26: 136–142. pmid:21277042
- 23. Li Y, Gao Y, Xu X, Shen Q, Guo S. Light-saturated photosynthetic rate in high-nitrogen rice (Oryza sativa L.) leaves is related to chloroplastic CO2 concentration. J Exp Bot. 2009; 60: 2351–2360. pmid:19395387
- 24. Xu G, Huang TF, Zhang XL, Duan BL. Significance of mesophyll conductance for photosynthetic capacity and water-use efficiency in response to alkaline stress in Populus cathayana seedlings. Photosynthetica. 2013; 51: 438–444.
- 25. Buckley TN, Warren CR. The role of mesophyll conductance in the economics of nitrogen and water use in photosynthesis. Photosynth res. 2014; 119: 77–88. pmid:23609621
- 26. Flexas J, Barbour MM, Brendel O, Cabrera HM, Carriquí M, et al. Mesophyll diffusion conductance to CO2: an unappreciated central player in photosynthesis. Plant Sci. 2012; 193–194: 70–84. pmid:22794920
- 27. Nakhoul NL, Davis BA, Romero MF, Boron WF. Effect of expressing the water channel aquaporin-1 on the CO2 permeability of Xenopus oocytes. Am J Physiol.1998; 274: C543–C548. pmid:9486145
- 28. Broeckx LS, Fichot R, Verlinden MS, Ceulemans R. Seasonal variations in photosynthesis, intrinsic water-use efficiency and stable isotope composition of poplar leaves in a short-rotation plantation. Tree Physiol. 2014; 34:701–715. pmid:25074859
- 29. Warren CR, Adams MA. Evergreen trees do not maximize instantaneous photosynthesis. Trends plant sci. 2004; 9: 270–274. pmid:15165557
- 30. Tang JC., Cheng RM, Shi ZM, Xu GX, Liu SR, et al. Fagaceae tree species allocate higher fraction of nitrogen to photosynthetic apparatus than Leguminosae in Jianfengling tropical montane rain forest, China. Plos One. 2018: 13:e0192040. pmid:29390007
- 31. Luo WY, Luo P, Liu YJ. Choice and development of the fine and valuable hardwood tree species in tropical and south subtropical regions of China. Chinese J Tropical Agriculture. 2010; 30: 15–21 (in Chinese).
- 32. Pang ZH. The study progress of Betula alnoides in China. China J Guangxi Academy Sci. 2011; 27: 243–250 (in Chinese).
- 33. Yang BG, Liu SL, Hao J, Pang SJ, Zhang P. Research advances on the rare tree of Erythrophleum fordii. Guangxi Forestry Sci. 2017; 46: 165–170 (in Chinese).
- 34. You Y, Huang X, Zhu H, Liu S, Liang H, et al. Positive interactions between Pinus massoniana and Castanopsis hystrix species in the uneven-aged mixed plantations can produce more ecosystem carbon in subtropical China. Forest Ecol Manag.
- 35. Wang WX, Shi ZM, Luo D, Liu SR, Lu LH. Characteristics of soil microbial biomass and community composition in three types of plantations in southern subtropical area of China. Chinese J Applied Ecol. 2013; 24: 1784–1792 (in Chinese).
- 36. Tang JC, Shi ZM, Luo D, Liu SR. Photosynthetic nitrogen-use efficiency of Manglietia glauca seedling leaves under different shading levels. Acta Ecol Sinica. 2017; 37:7493–7502 (in Chinese).
- 37. Zhang L, Chen X, Wen D. Interactive effects of rising CO2, and elevated nitrogen and phosphorus on nitrogen allocation in invasive weeds Mikania micrantha and Chromolaena odorata. Biol Invasions. 2016; 18: 1391–1407.
- 38. Feng QH, Cheng RM, Shi ZM, Liu SR, Wang WX, et al. Response of Rumex dentatus foliar nitrogen and its allocation to altitudinal gradients along Balang Mountain, Sichuan, China. Chinese J Plant Ecol. 2013; 37: 591–600 (in Chinese).
- 39. Loreto F, Di Marco G, Tricoli D, Sharkey TD. Measurements of mesophyll conductance, photosynthetic electron transport and alternative electron sinks of field grown wheat leaves. Photosynth Res. 1994; 41: 397–403. pmid:24310154
- 40. Loreto F, Tsonev T, Centritto M. The impact of blue light on leaf mesophyll conductance. J Eep Bot. 2009; 112:1–8.
- 41. Harley PC, Loreto F, Di Marco G, Sharkey TD. Theoretical considerations when estimating the mesophyll conductance to CO2 flux by analysis of the response of photosynthesis to CO2. Plant Physiol. 1992; 98: 1429–1436. pmid:16668811
- 42. Li Y, Ren BB, Ding L, Shen QR, Peng SB, Guo SW. Does chloroplast size influence photosynthetic nitrogen use efficiency?. PloS One. 2013; 8: e62036. pmid:23620801
- 43. Sorrentino G, Haworth M, Wahbi S, Mahmood T, Zuomin S, Centritto M. Abscisic acid induces rapid reductions in mesophyll conductance to carbon dioxide. PloS One. 2016; 11: e0148554. pmid:26862904
- 44. Momayyezi M, Guy RD. Blue light differentially represses mesophyll conductance in high vs low latitude genotypes of Populus trichocarpa Torr. & Gray. J Plant Physiol. 2017; 213: 122–128. pmid:28364640
- 45. Wang X, Du T, Huang J, Peng S, Xiong D. Leaf hydraulic vulnerability triggers the decline in stomatal and mesophyll conductance during drought in rice (Oryza sativa). J Eep Bot. 2018; 69: 4033–4045.
- 46. Bernacchi CJ, Singsaas EL, Pimentel C, Portis JAR, Long SP. Improved temperature response functions for models of rubisco‐limited photosynthesis. Plant Cell Environ. 2001; 24: 253–259.
- 47. Ethier GJ, Livingston NJ. On the need to incorporate sensitivity to CO2 transfer conductance into the farquhar-von caemmerer-berry leaf photosynthesis model. Plant Cell Environ. 2004; 27: 137–153.
- 48. Gu LH, Pallardy SG, Tu K, Law BE, Wullschleger SD. Reliable estimation of biochemical parameters from C3 leaf photosynthesis-intercellular carbon dioxide response curves. Plant Cell Environ. 2010; 33: 1852–1874. pmid:20561254
- 49. Sharkey TD, Bernacchi CJ, Farquhar GD, Singsaas EL. Fitting photosynthetic carbon dioxide response curves for C3 leaves. Plant Cell Environ. 2007: 30: 1035–1040. pmid:17661745
- 50. Loustau D, Brahim MB, Gaudillère JP, Dreyer E. Photosynthetic responses to phosphorus nutrition in two-year-old maritime pine seedlings. Tree Physiol. 1999; 19: 707–715. pmid:12651309
- 51. Grassi G, Magnani F. Stomatal, mesophyll conductance and biochemical limitations to photosynthesis as affected by drought and leaf ontogeny in ash and oak trees. Plant Cell Environ. 2005; 28: 834–849.
- 52. Tomás M, Flexas J, Copolovici L, Galmés J, Hallik L, et al. Importance of leaf anatomy in determining mesophyll diffusion conductance to CO2 across species: quantitative limitations and scaling up by models. J Eep Bot. 2013; 64: 2269–2281.
- 53. Peguero-Pina JJ, Sisó S, Flexas J, Galmés J, Niinemets Ü, et al. Coordinated modifications in mesophyll conductance, photosynthetic potentials and leaf nitrogen contribute to explain the large variation in foliage net assimilation rates across Quercus ilex provenances. Tree Physiol. 2017; 37: 1084–1094. pmid:28541538
- 54. Peguero-Pina JJ., Sisó S, Flexas J, Galmés J, García-Nogales A, et al. Cell-level anatomical characteristics explain high mesophyll conductance and photosynthetic capacity in sclerophyllous Mediterranean oaks. New Phytol. 2017; 214: 1–12.
- 55. Nha B, Hayes L, Scafaro AP, Atkin OK, Evans JR. Mesophyll conductance does not contribute to greater photosynthetic rate per unit nitrogen in temperate compared with tropical evergreen wet-forest tree leaves. New Phytol. 2018; 218: 1–13.
- 56. Feng YL, Wang JF, Sang WG. Biomass allocation, morphology and photosynthesis of invasive and noninvasive exotic species grown at four irradiance levels. Acta Oecologica. 2007; 31: 40–47.
- 57. Bahar NH, Ishida FY, Weerasinghe LK, Guerrieri R, O’Sullivan OS, et al. Leaf-level photosynthetic capacity in lowland Amazonian and high-elevation Andean tropical moist forests of Peru. New Phytol. 2016; pmid:27389684
- 58. Yao HS, Zhang YL, Yi XP, Zhang XJ, Fan DY, et al. Diaheliotropic leaf movement enhances leaf photosynthetic capacity and photosynthetic use efficiency of light and photosynthetic nitrogen via optimizing nitrogen partitioning among photosynthetic components in cotton (Gossypium hirsutum l.). Plant Biol. 2018; 20: 213–222. pmid:29222927
- 59. Sánchez-Gómez D, Robson TM, Gascó A, Gil-Pelegrín E, Aranda I. Differences in the leaf functional traits of six beech (Fagus sylvatica, L.) populations are reflected in their response to water limitation. Environ Exp Bot. 2013; 87: 110–119.
- 60. Wright IJ, Reich PB, Cornelissen JH, Falster DS, Garnier E, et al. Assessing the generality of global leaf trait relationships. New phytol. 2005; 166: 485–496. pmid:15819912
- 61. Reich PB, Oleksyn J, Wright IJ. Leaf phosphorus influences the photosynthesis-nitrogen relation: a cross-biome analysis of 314 species. Oecologia. 2009; 160: 207–212. pmid:19212782
- 62. Hikosaka K, Hirose T. Photosynthetic nitrogen-use efficiency in evergreen broad-leaved woody species coexisting in a warm-temperate forest. Tree Physiol. 2000; 20: 1249–1254. pmid:12651488
- 63. Ellsworth DS, Reich PB. Photosynthesis and leaf nitrogen in five Amazonian tree species during early secondary succession. Ecology. 1996; 77: 581–594.
- 64. Kikuzawa K. A cost-benefit analysis of leaf habit and leaf longevity of trees and their geographical pattern. Am Nat. 1991; 138: 1250–1263.
- 65. Robinson DE, Wagner RG, Bell FW, Swanton CJ. Photosynthesis, nitrogen-use efficiency, and water-use efficiency of jack pine seedlings in competition with four boreal forest plant species. Can J Forest Res. 2001; 31: 2014–2025.
- 66. Moon M, Kang KS, Park IK, Kim T, Kim HS. Effects of leaf nitrogen allocation on the photosynthetic nitrogen-use efficiency of seedlings of three tropical species in Indonesia. J Korean Soc Appl Bi. 2015; 58: 511–519.
- 67. Qin RM, Zheng YL, Valiente-Banuet A, Callaway RM, Barclay GF, et al. The evolution of increased competitive ability, innate competitive advantages, and novel biochemical weapons act in concert for a tropical invader. New Phytol. 2013; 197: 979–988. pmid:23252450
- 68. Ruan J, Haerdter R, Gerendás J. Impact of nitrogen supply on carbon/nitrogen allocation: a case study on amino acids and catechins in green tea [Camellia sinensis (L.) O. Kuntze] plants. Plant Biology.2010; 12: 724–734. pmid:20701695
- 69. Funk JL, Glenwinkel LA, Sack L. Differential allocation to photosynthetic and non-photosynthetic nitrogen fractions among native and invasive species. PloS one.2013; 8: e64502. pmid:23700483
- 70. Stitt M., Schulze ED. Does Rubisco control the rate of photosynthesis and plant growth? An exercise in molecular ecophysiology, Plant Cell Environ.1994; 17: 465–487.
- 71. Warren CR, Adams MA. Distribution of N, Rubisco and photosynthesis in Pinus pinaster and acclimation to light. Plant Cell Environ. 2001; 24: 597–609.
- 72. Carmo‐Silva E, Scales JC, Madgwick PJ, Parry MAJ. Optimizing Rubisco and its regulation for greater resource use efficiency. Plant Cell Environ. 2015; 38: 1817–1832. pmid:25123951
- 73. Wright IJ, Cannon K. Relationships between leaf lifespan and structural defences in a low-nutrient, sclerophyll flora. Funct Ecol.2001; 15: 351–359.
- 74. Onoda Y, Schieving F, Anten NP. Effects of light and nutrient availability on leaf mechanical properties of Plantago major: a conceptual approach. Ann Bot. 2008; 101: 727–736. pmid:18272529
- 75. Warren CR. Stand aside stomata, another actor deserves centre stage: the forgotten role of the internal conductance to CO2 transfer. J Exp Bot. 2008; 59:1475–1487. pmid:17975206
- 76. Scafaro AP, Von CS, Evans JR, Atwell BJ. Temperature response of mesophyll conductance in cultivated and wild Oryza species with contrasting mesophyll cell wall thickness. Plant Cell Environ. 2011;34: 1999–2008. pmid:21752031
- 77. Piel C, Frak E, Le Roux X, Genty B. Effect of local irradiance on CO2 transfer conductance of mesophyll in walnut. J Exp Bot. 2002; 53: 2423–2430. pmid:12432034
- 78. Tosens T, Niinemets U, Vislap V, Eichelmann H, Díez PC. Developmental changes in mesophyll diffusion conductance and photosynthetic capacity under different light and water availabilities in Populus tremula: how structure constrains function. Plant Cell Environ. 2012; 35: 839–856. pmid:22070625