Errata: Corrigendum Volume 172, Issue 2, 378, Article first published online: 22 September 2006
Author for correspondence: Jinxing Lin Fax: +086 10 62590833 Email: firstname.lastname@example.org
• Leaves of Arabidopsis thaliana grown under elevated or ambient CO2 (700 or 370 µmol mol−1, respectively) were examined for physiological, biochemical and structural changes.
• Stomatal characters, carbohydrate and mineral nutrient concentrations, leaf ultrastructure and plant hormone content were investigated using atomic absorption spectrophotometry, transmission electron microscopy and enzyme-linked immunosorbent assay (ELISA).
• Elevated CO2 reduced the stomatal density and stomatal index of leaves, and also reduced stomatal conductance and transpiration rate. Elevated CO2 increased chloroplast number, width and profile area, and starch grain size and number, but reduced the number of grana thylakoid membranes. Under elevated CO2, the concentrations of carbohydrates and plant hormones, with the exception of abscisic acid, increased whereas mineral nutrient concentrations declined.
• These results suggest that the changes in chloroplast ultrastructure may primarily be a consequence of increased starch accumulation. Accelerated A. thaliana growth and development in elevated CO2 could in part be attributed to increased foliar concentrations of plant hormones. The reductions in mineral nutrient concentrations may be a result of dilution by increased concentrations of carbohydrates and also of decreases in stomatal conductance and transpiration rate.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
The effects of rising atmospheric CO2 concentrations on the global climate have attracted considerable attention. Plants respond both physiologically and anatomically to elevated CO2. Many studies have investigated plant responses to elevated CO2 on ecosystem, community, population, plant, leaf, physiological, biochemical and molecular scales (Cheng et al., 1998; Norby et al., 1999; Gibeaut et al., 2001; Lin et al., 2001); however, relatively little information exists on the effects of elevated CO2 on the ultrastructure of the photosynthetic apparatus. Whereas several studies have investigated the responses of chloroplast ultrastructure to elevated CO2 (Robertson & Leech, 1995; Bockers et al., 1997; Watling et al., 2000; Zuo et al., 2002), nearly all of these have focused on pure morphological descriptions of changes in leaf ultrastructure, and seldom have the changes been investigated from a statistical perspective.
There is accumulating evidence that elevated CO2 can accelerate plant growth and development by affecting cell division, elongation and differentiation within apical meristems (Masle, 2000; Yong et al., 2000; Ferris et al., 2001; Li et al., 2002; Vu et al., 2002; Taylor et al., 2003; Luomala et al., 2005). These cellular processes are usually regulated by plant hormones, including auxins, gibberellins (GAs), cytokinins (CKs) and abscisic acid (ABA) (Yong et al., 2000; Li et al., 2002). Thus, changes in the levels of these hormones probably play an important role in regulating the ontogeny of plants grown under elevated CO2. However, hormone concentrations in plants grown in elevated CO2 have not been studied in depth. Examination of the effects of elevated CO2 on plant hormone concentrations will lead to an improved understanding of the mechanism by which plant ontogeny is accelerated by elevated CO2.
The effects of elevated CO2 on the growth and development of Arabidopsis thaliana, a model plant widely used in molecular, genetic and developmental biology, have been examined previously (Ward & Strain, 1997; Cheng et al., 1998; Van Der Kooij et al., 1999; Andalo et al., 2001; Ward & Kelly, 2004). However, most of these studies have focused on the responses of growth rate, carbohydrate content, cell wall composition, biomass production, reproduction, evolutionary selection, Rubisco content and mRNA accumulation to elevated CO2. The effects of CO2 enrichment on leaf ultrastructure, mineral nutrition and plant hormone concentrations have not been extensively studied in this model plant; nevertheless, these aspects, in particular leaf ultrastructure and plant hormone concentrations, are very important for an integrative understanding of plant responses to increased atmospheric CO2. Therefore, to better understand the effects of CO2 enrichment on physiological, biochemical and ultrastructural characteristics, we investigated the effects of elevated CO2 on carbohydrate and mineral nutrient concentrations and stomatal characters of A. thaliana leaves, with particular attention to cellular fine structure and plant hormone concentrations.
Materials and Methods
Experimental design and growth conditions
Wild-type Columbia (Col-0) seeds of Arabidopsis thaliana (L.) Heynh. (Nottingham Arabidopsis Stock Centre, Nottingham University, Nottingham, UK) were exposed to a dark, cold and wet treatment at 4°C for 2 d before planting to promote uniform germination. Thereafter they were sown in 200-cm3 plastic pots filled with medium consisting of a 1 : 1 [volume/volume (v/v)] mixture of vermiculite and peat (Kaiyin Company, Beijing, China). Plants were grown in CO2-controlled growth chambers (HPG-280H; Harbin Donglian Electronic & Technology Development Co. Ltd, Harbin, China) that can automatically and accurately control temperature, light, relative humidity, photosynthetically active radiation (PAR) and CO2 concentration according to preset data. The chambers were maintained at CO2 concentrations of 370 ± 30 or 700 ± 50 µmol mol−1 and the purity of CO2 applied in this study was 99.999% (Haiwen Special Gas Co. Ltd, Beijing, China). The CO2 concentration was monitored using an infrared CO2 controller (eSENSE-D; Sense Air, Delsbo, Sweden), which can continuously and automatically control the electromagnetism valve attached to the CO2 cylinder. The plants were grown under a 16-h photoperiod with 400 µmol m−2 s−1 PAR supplied by fluorescent tubes (Philips Electronics Trading & Services Co. Ltd, Shanghai, China), 75% relative humidity, and day:night temperatures of 23 : 19°C. The plants were alternately watered to saturation with 1/2 Murashige–Skoog (MS) solution (Murashige & Skoog, 1962) or deionized water weekly. Following emergence, seedlings were thinned to one individual closest to the centre of each pot. When bolting had just commenced, i.e. at stage 5.10 (Boyes et al., 2001), the leaves were sampled for the following investigations.
According to the method described by Leishman et al. (1999) and Gibeaut et al. (2001), we used two chambers in the experiment (treatments and plants were swapped between chambers during the course of the experiment). In order to keep the potential for interactive effects between the chambers and the developmental stage of the plants to a minimum, the CO2 concentrations of the two chambers were swapped and the pots were moved between chambers and randomly rearranged weekly. Except for the concentration of CO2, other environmental factors such as temperature, light and relative humidity were common to both growth chambers to average out any possible effects resulting from the chambers and pot positions within the chambers.
Determination of stomatal characteristics and transpiration rate
Fully expanded rosette leaves were sampled for the counting of stomata and epidermal cells. Impressions of abaxial (lower) and adaxial (upper) epidermis were taken using colourless nail polish and adhesive cellophane tape. The imprints were observed and photographed using a microscope (Zeiss Axioskop 40; Carl Zeiss, Beijing, China) equipped with a digital camera (Axiocam MRC; Carl Zeiss). Stomata and epidermal cells were counted on the images. Stomatal density, stomatal index and epidermal cell density were calculated according to the methods outlined by Ceulemans et al. (1995). Data evaluation was based on the imprints from 10 individual plants in each treatment. Three leaves were sampled from each of 10 plants and 20 separate fields of 0.16 mm2 were analysed in each leaf. Additionally, three fully expanded leaves from each of five plants were sampled for the measurement of stomatal conductance and leaf transpiration rate using an LI-6400 Portable Photosynthesis System (Li-Cor Inc., Lincoln, NB, USA). The measurements for both sets of plants were, respectively, carried out at 1500 µmol m−2 s−1 PAR, 2.0–2.5 KPa vapour pressure deficit (VPD), 23°C and 370 or 700 µmol mol−1 CO2.
Preparations for microscopy
Fully expanded leaves on a rosette were dissected and immediately fixed in 2.5% (v/v) glutaraldehyde (in 0.1 m phosphate buffer, pH 7.0) for 2 h at 4°C. Then the samples were washed five times with the same buffer and postfixed in 1% osmium tetroxide for 3 h. After being washed with the same buffer, leaf tissues were passed through an ethanol dehydration series, and then infiltrated and embedded in Spurr's resin. Sections were cut using an LKB-V ultramicrotome (LKB, Bromma, Sweden). Thin sections were stained with uranyl acetate and lead citrate, then observed and photographed under a transmission electron microscope (JEOL Ltd, Tokyo, Japan).
Leaves at the end of light period were sampled, oven-dried at 60°C, and ground using a mortar and pestle. Approximately 50 mg of the oven-dried leaf power of each sample was extracted with 80% ethanol (v/v) at 85°C for 1 h. The solutions were then centrifuged at 12 000 g for 10 min. The ethanol extraction step was repeated three times. The three resulting supernatants were combined, treated with activated charcoal, and evaporated to dryness in a vacuum evaporator. The residues were redissolved in distilled water, and subjected to soluble sugar analysis using the anthrone-sulfuric acid method (Ebell, 1969). Following removal of soluble sugars, the remaining pellets were oven-dried overnight at 60°C and retained for starch analysis according to the procedures described in previous publications (Lindroth et al., 2002; Vu et al., 2002). Total nonstructural carbohydrate (TNC) was calculated as the sum of soluble sugar and starch. Cellulose content was determined by the method of Updegraff (1969). All the analyses were repeated five times.
FTIR analysis was carried out according to the method of Chen et al. (1998), with some modifications. The youngest mature leaves were excised from plants before the end of the light period. The leaves were immersed in 80% ethanol (v/v) at 85°C for 20 min to extract chlorophyll, soluble sugars and other small molecules. The extraction step was repeated three times until the leaves became white in colour. The samples were then washed three times with double-distilled water and dried at room temperature on a barium fluoride window (13 mm diameter × 2 mm). An area of leaf (100 × 100 µm2) away from the mid-vein was selected for spectral collection. Infrared spectra of the leaves were obtained using a Magna 750 FTIR spectrometer (Nicolet Corp., Tokyo, Japan) equipped with a mercury-cadmium-telluride (MCT) detector. The spectra were recorded at a resolution of 8 cm−1 with 128 coadded interferograms and were normalized to obtain the relative absorbance. Three spectra were collected from different areas of each A. thaliana leaf and then averaged and baseline-corrected.
Determination of plant hormones
All leaves in a rosette were sampled just at the end of the light period and immediately frozen in liquid nitrogen after their fresh weight had been determined, and then stored at −80°C until analysis. The samples were ground in liquid nitrogen using a mortar and pestle, extracted with ice-cold 80% methanol (v/v) containing 1 mmol l−1 butylated hydroxytoluene to prevent oxidation, and then stored overnight at 4°C. The extracts were then centrifuged at 12 000 g for 15 min at 4°C. The residues were suspended in the same ice-cold extraction solution and stored at 4°C for 1 h, and then centrifuged again at 12 000 g for 15 min at 4°C. The two resulting supernatants were combined and passed through a C18 Sep-Pak cartridge (Waters, Milford, MA, USA). The efflux was collected and dried in N2. The residues were then dissolved in 0.01 mol l−1 phosphate buffer solution (pH 7.4) and concentrations of indole-3-acetic acid (IAA), gibberellic acid (GA3), zeatin riboside (ZR), dihydrozeatin riboside (DHZR), isopentenyladenosine (iPA) and ABA were determined in an enzyme-linked immunosorbent assay (ELISA) following methods described in previous publications (Chen et al., 1997; Yang et al., 2001; Zhu et al., 2005), with some modifications. In brief, microtitration plates (Nunc, Roskilde, Denmark) were coated with synthetic IAA, GA3, ZR, DHZR, iPA or ABA–ovalbumin conjugates in 50 mmol l−1 NaHCO3 buffer solution (pH 9.6) and stored overnight at 37°C. Ovalbumin solution (10 mg ml−1) was added to each well of the plates for the purpose of blocking nonspecific binding. After incubation for 30 min at 37°C, standard IAA, GA3, ZR, DHZR, iPA, ABA, samples and antibodies were added and incubated for an additional 45 min at 37°C. The antibodies against IAA, GA3, ZR, DHZR, iPA or ABA were produced according to the method described by Weiler et al. (1981). Horseradish peroxidase-labelled goat anti-rabbit immunoglobulin was then added to each well and incubated for 1 h at 37°C. After the incubation, orthophenylenediamino (OPD) substrate solution was added to each well of the plates. The plates were then kept in the dark at 37°C for the substrate reaction. Fifteen minutes later, 3 mol l−1 H2SO4 was added to each well to terminate the substrate reaction. The absorbance of each well was measured at 490 nm using the ELISA Recorder (Model DG-3022 A; Huadong Electron Tube Factory, Shanghai, China). Calculations for the ELISA data were performed according to the method described by Weiler et al. (1981). In the current study, the percentage recovery of each hormone was calculated by adding known quantities of standard hormone to a split extract. All the percentage recoveries were > 90%, and all the sample extract dilution curves paralleled the standard curves, showing the absence of nonspecific inhibitors in the extracts.
Analysis of macroelement contents
Total carbon (C) and nitrogen (N) contents in leaves were determined using the H2SO4-K2CrO7 method and the microKjeldahl technique, respectively, and the phosphorus (P) content was measured spectrophotometrically by the molybdenum blue method (Xu & Yang, 2000). Potassium (K), calcium (Ca) and magnesium (Mg) contents were determined using an atomic absorption spectrophotometer (AA-6300; Shimadzu, Kyoto, Japan).
The data are shown as mean ± standard deviation. Data were subjected to one-way analysis of variance and t-test using software SPSS 10.0 (SPSS Inc., Chicago, IL, USA).
Under both ambient and elevated CO2 conditions, most of the seeds germinated within 3 d after the dark, cold and wet treatment. Elevated CO2 had no significant effects on plant growth before stage 1.0. However, elevated CO2 significantly enhanced A. thaliana growth from stage 1.0 to stage 5.10 (Boyes et al., 2001). The average plant growth rate from stage 1.0 to stage 5.10 in elevated CO2 was approx. 29% higher than that of plants grown in ambient CO2.
Stomatal characters and leaf transpiration rate
Both the stomatal density (SD) and the stomatal index (SI) of fully developed rosette leaves were reduced on adaxial and abaxial leaf surfaces under elevated CO2 (P < 0.01), whereas epidermal cell density was not significantly affected. SD was reduced by c. 19 and 14% on the adaxial and abaxial surfaces, respectively (Fig. 1a). Similarly, SI was also significantly reduced by c. 12 and 9% on the adaxial and abaxial surfaces, respectively (Fig. 1b). Only a slight decrease in epidermal cell density was observed under elevated compared with ambient CO2 (Fig. 1c). Elevated CO2 also reduced stomatal conductance (Gs) and the transpiration rate (Tr) of A. thaliana leaves. From Fig. 1(d), it was evident that the elevated CO2 treatment significantly reduced both the Gs and Tr of A. thaliana leaves when the measurements were taken at the same CO2 concentration, i.e. 370 or 700 µmol mol−1 CO2. In addition, the Gs and Tr of plants grown in either ambient or elevated CO2 measured at 700 µmol mol−1 CO2 were significantly lower than those of plants grown in either ambient or elevated CO2 measured at 370 µmol mol−1 CO2, demonstrating that elevated CO2 applied during measurements also significantly reduced Gs and Tr, probably mainly through reductions in stomatal apertures.
Chloroplast ultrastructure and leaf structure
Elevated CO2 significantly increased the number of chloroplasts per mesophyll cell, chloroplast width and profile area, and leaf and cell wall thickness (Fig. 2g,h), but had little effect on chloroplast length and cell size (Table 1). Relative to ambient CO2, elevated CO2 increased the number of chloroplasts per cell by 17.9%, chloroplast width by 19.8%, cell wall thickness by 23.8%, and leaf thickness by 5.3%. The concomitant rise (23.9%) in chloroplast profile area was a result of the increased chloroplast width, given the lack of change in chloroplast length.
Table 1. Chloroplast ultrastructure and leaf structure of Arabidopsis thaliana grown at elevated or ambient CO2 concentration
Ambient CO2 (370 µmol mol−1)
Elevated CO2 (700 µmol mol−1)
Values given are mean ± standard deviation for five plants (270 cells were recorded from three leaves of each plant). The fully expanded rosette leaves were sampled at stage 5.0. Mean values were compared by t-test.
The chloroplast length is the longest dimension of the chloroplast, and the chloroplast width is its widest dimension.
Ratio of total starch grains per chloroplast relative to chloroplast area (%)
Number of grana thylakoid membranes
9.01 ± 3.15
6.52 ± 2.04
Number of plastoglobuli per chloroplast
10.14 ± 4.43
10.98 ± 3.51
Cell wall thickness (µm)
0.3156 ± 0.0283
0.2549 ± 0.0236
Cell size (µm2)
727 ± 266
753 ± 275
Leaf thickness (µm)
139.9 ± 13.5
147.3 ± 14.9
The size and number of starch grains were also significantly increased under elevated CO2 (Table 1). The average size per starch grain increased from 0.83 µm2 in ambient CO2 to 1.26 µm2 in elevated CO2, an increase of 51.8%. The average number of starch grains per chloroplast increased by 40.7%. Approximately 30% of chloroplasts did not contain starch grains in ambient CO2, whereas only c. 10% of chloroplasts lacked starch grains in elevated CO2. Starch grains accounted for an average of 34.4% of the chloroplast profile in elevated CO2 (Fig. 2b), significantly more than the 21.8% in ambient CO2 (Fig. 2a). In elevated CO2, in some chloroplasts starch grains accounted for as much as 60% of the chloroplast profile (Fig. 2c). In most cases, chloroplasts containing large grains expanded and became rounder in shape relative to those in ambient CO2, and also showed distortion of the thylakoid membranes (Fig. 2d). In addition, elevated CO2 seemed to increase the ratio of stroma thylakoid to grana thylakoid and decrease the number of grana thylakoid membranes by 27.6% (Fig. 2a–f, Table 1). Distorted thylakoid membranes such as those shown in Fig. 2(d) at elevated CO2 were probably caused by pressure from large starch grains deposited in the chloroplast. In the early stages, the size of starch grains deposited between thylakoid membranes was very small. As starch accumulation increased gradually and the number and size of starch grains reached a certain level, the large starch grains mechanically pressed against, separated and finally distorted thylakoid membranes, leading to a more elliptical or football-shaped chloroplast.
The predominant nonstructural carbohydrates found in leaves are total soluble sugars and starch. Exposure to elevated CO2 not only caused a significant increase in the soluble sugar and starch contents, but also caused an increase in the cellulose content (Table 2). The total soluble sugar and starch contents per unit leaf dry weight in A. thaliana grown in elevated CO2 increased by more than 70% (P < 0.001) and approx. 80% (P < 0.001), respectively. Thus, considering foliar soluble sugars and starch together, the TNC content increased by c. 76% (P < 0.001) in leaves. In addition, elevated CO2 dramatically increased the content of some structural carbohydrates, such as cellulose, in A. thaliana leaves (P = 0.002). However, the increase in cellulose content (c. 22%) was approximately one-third that of TNC (c. 76%).
Table 2. Content of carbohydrates in rosette leaves of Arabidopsis thaliana grown at ambient or elevated CO2 concentration
Content (µg mg−1 DW)
Ambient CO2 (370 µmol mol−1)
Elevated CO2 (700 µmol mol−1)
Values given are mean ± standard deviation. The rosette leaves at stage 5.0 were sampled and mixed for the determination of carbohydrates. Mean values (n = 5 samples, with five plants per sample) were compared by Student's t-test. Total nonstructural carbohydrates = soluble sugars + starch.
DW, dry weight.
21.39 ± 0.81
36.76 ± 1.24
37.64 ± 1.47
67.26 ± 2.62
Total nonstructural carbohydrates
59.03 ± 0.66
104.02 ± 3.86
98.46 ± 3.27
120.42 ± 4.26
Given that general carbohydrate vibrations are mainly distributed between 900 and 1200 cm−1 (Michael & Maureen, 2000), for our investigations of the carbohydrate contents of foliar cell walls, we paid particular attention to this range of wave numbers. The FTIR spectra of the cell walls of leaves grown under elevated and ambient CO2 had four peaks (1020, 1043, 1082 and 1157 cm−1; Fig. 3a). Figure 3(b) is the difference spectrum generated by the digital subtraction of the ambient from the elevated CO2 spectrum in Fig. 3(a), demonstrating that elevated CO2 increased the cellulose (1043 and 1157 cm−1) (Mouille et al., 2003) and pectin (1020 and 1082 cm−1) (Oomen et al., 2004) contents in cell walls.
Foliar concentrations of plant hormones
There was a significant increase in IAA, GA3, ZR, DHZR and iPA concentrations in leaves grown in elevated CO2 compared with those grown in ambient CO2. However, elevated CO2 led to a significant decrease in the ABA content of leaves (Table 3). Specifically, relative to ambient CO2, elevated CO2 increased IAA, GA3, ZR, DHZR and iPA concentrations in A. thaliana leaves by 13.7, 55.4, 15.6, 55.9 and 74.6%, respectively, whereas the ABA content was reduced by 15.2%.
Table 3. Concentrations of plant hormones in rosette leaves of Arabidopsis thaliana grown at ambient or elevated CO2 concentration
Concentration (ng g−1 FW)
Ambient CO2 (370 µmol mol−1)
Elevated CO2 (700 µmol mol−1)
Values given are mean ± standard deviation. The rosette leaves at stage 5.0 were sampled and mixed for the determination of the four plant hormones. Mean values (n = 5 samples, with five plants per sample) were compared by Student's t-test.
Elevated CO2 significantly decreased the concentrations of most foliar nutrients on a dry weight (DW) basis. The average concentrations of N, P, K, Ca and Mg in the leaves in elevated CO2 were significantly reduced, by 11.2, 17.9, 20.7, 19.2 and 14.8%, respectively (Table 4). By contrast, the foliar C concentration increased to 455.6 mg g−1 DW in elevated CO2, a significant increase of 10.7% (Table 4). Because of the increased C and decreased N concentrations, the C:N ratio was significantly increased from 13.54 to 16.90, an increase of 24.8%.
Table 4. Nutrient concentrations in rosette leaves of Arabidopsis thaliana grown at ambient or elevated CO2 concentration
Concentration (mg g−1 DW)
Ambient CO2 (370 µmol mol−1)
Elevated CO2 (700 µmol mol−1)
Values given are mean ± standard deviation. The rosette leaves at stage 5.0 were sampled and mixed for the determination of macroelements. Mean values (n = 5 samples, with five plants per sample) were compared by Student's t-test.
DW, dry weight.
411.7 ± 11.2
455.6 ± 13.0
30.4 ± 1.3
27.0 ± 0.8
3.9 ± 0.3
3.2 ± 0.3
22.2 ± 1.1
17.6 ± 0. 9
15.1 ± 0.9
12.2 ± 0.8
2.7 ± 0.2
2.3 ± 0.2
13.54 ± 0.25
16.90 ± 0.44
Stomatal characters and transpiration rate
The stomatal responses to CO2 enrichment may be species-dependent and nonuniform across leaves (Ceulemans et al., 1995; Lin et al., 2001). A wide range of species grown under CO2 enrichment have shown reduced SD (Woodward, 1987; Woodward & Kelly, 1995; Beerling et al., 1998; Lin et al., 2001); however, no changes in SD were observed in some species (Ceulemans et al., 1995; Poole et al., 2000) and some investigations have even reported increases in SD under CO2 enrichment (Atkinson et al., 1997; Lawson et al., 2002). We found that the SD of A. thaliana leaves decreased by 19 and 14% on the adaxial and abaxial surfaces, respectively, in elevated CO2; this is consistent with the previously reported average reduction in SD of 14.3% observed for 100 species grown under CO2 enrichment (Woodward & Kelly, 1995). The SI of A. thaliana leaves was also reduced by elevated CO2. Similar results were found in other species, such as Rumex crispus, Vaccinium myrtillus and Lycopersicon esculentum (Woodward, 1987). By contrast, Lawson et al. (2002) recently reported that SD increased under elevated CO2. Our data suggest that the reduction in the SI of A. thaliana leaves was mainly caused by the reduction in SD, given that the epidermal cell density changed only slightly with CO2 enrichment. Interestingly, the adaxial stomata were more sensitive to elevated CO2 than the abaxial stomata, a pattern also observed in R. crispus and V. myrtillus (Woodward, 1987). A. thaliana stomata seem to be more sensitive than epidermal cells to high CO2 concentrations. In other words, elevated CO2 may have directly affected the initiation of the number of stomata during ontogenesis, but had little effect on the differentiation and expansion of epidermal cells. In addition, stomata on the adaxial and abaxial leaf surfaces respond differently to CO2 enrichment. In an investigation of the mechanism of the effect of elevated CO2 on the stomatal index and density of the model plant A. thaliana, Lake et al. (2001) found that it was mature leaves that detected CO2 concentration first and consequently transmitted a long-distance signal to control stomatal development in developing leaves. Young leaves, however, seemed to lack the capacity to respond to CO2 concentration. More recently, the same group of researchers further confirmed that the response of mature leaves to changing CO2 influenced stomatal development in expanding leaves and suggested that the differences in sugar content and plant hormone concentrations between mature and young leaves may be the sources of the signal transmitted from mature to young leaves (Coupe et al., 2006). In accordance with early reports (Woodward, 2002; Wullschleger et al., 2002; Hetherington & Woodward, 2003), the stomatal conductance and transpiration rate of A. thaliana leaves were significantly reduced by CO2 enrichment. We here show that the decreases in leaf stomatal conductance and transpiration rate under elevated CO2 were mainly the result of reduced SD.
The number of chloroplasts per cell was significantly higher in A. thaliana leaves grown in elevated CO2. This result agrees with that of Wang et al. (2004), who found that elevated CO2 increased the number of chloroplasts per unit cell area by 71% in Nicotiana sylvestris leaves. Similarly, Bockers et al. (1997) observed that high CO2 concentrations increased the number of chloroplasts in Marchantia polymorpha. However, the number of chloroplasts per mesophyll cell was not affected by elevated CO2 in leaves of Triticum aestivum (Roberton & Leech, 1995). These differences can probably be attributed to the duration of exposure to elevated CO2 and the developmental stages of the leaves sampled. For example, the youngest mature leaves were sampled from N. sylvestris and A. thaliana after c. 5 wk of exposure to elevated CO2, whereas 7-d-old leaves of T. aestivum were sampled. Whereas most studies have demonstrated that elevated CO2 increases the number of chloroplasts, the mechanism by which elevated CO2 regulates chloroplast number in mesophyll cells remains unknown (Bockers et al., 1997). Perhaps elevated CO2 increases the number of chloroplasts by stimulating chloroplast biogenesis (Wang et al., 2004).
The results presented here show that the chloroplast profile area and chloroplast width increased under CO2 enrichment. The increase in chloroplast width may have primarily resulted from greater starch accumulation in elevated CO2. There was a corresponding increase in chloroplast area, largely because of the increase in chloroplast width, given that chloroplast length did not vary much (Table 2). Although similar results have been reported previously (Pritchard et al., 1997; Ramonell et al., 2001; Wang et al., 2004), our results were not consistent with those of Robertson & Leech (1995), who found that higher CO2 concentrations had no effect on chloroplast shape in mesophyll cells of 7-d-old T. aestivum. It is possible that the T. aestivum leaves were too young to have accumulated much starch, because chloroplast shape is often altered by significant starch accumulation.
We found that the average size and number of starch grains in leaf chloroplasts were significantly increased in elevated CO2. In addition, starch grains in leaves grown in elevated CO2 took up as much as 34.4% of the chloroplast profile on average, significantly more than those grown in ambient CO2. Biochemical analyses also revealed that the starch content of A. thaliana leaves grown in elevated CO2 was significantly higher than that of leaves grown in ambient CO2 (Table 1). The increased starch accumulation in chloroplasts in elevated CO2 may act as a mechanism for storing C, and thereby expand sink capacities (Wolfe et al., 1998). This result is in accordance with some previous reports that CO2 enrichment significantly increases the size and number of starch grains in chloroplasts of mesophyll cells (Oksanen et al., 2001; Zuo et al., 2002; Wang et al., 2004). For instance, elevated CO2 treatment led to a significant increase in starch grain size in the chloroplasts of three aspen clones (Oksanen et al., 2001). In 36-d-old N. sylvestris leaves grown in elevated CO2, the chloroplasts were larger and contained more starch grains than leaves grown in ambient CO2 (Wang et al., 2004). By contrast, Robertson & Leech (1995) found more starch accumulation in chloroplasts of 7-d-old leaves grown in ambient CO2, with starch occupying as much as 60% of the chloroplast profile; small starch grains dispersed throughout the stroma of chloroplasts were observed in leaves grown in elevated CO2. A possible reason for this is that seedlings grown in elevated CO2 had a higher demand for energy and carbon skeletons than those grown in ambient CO2 during the early developmental stages; thus, more starch was consumed by the rapid seedling growth, leaving fewer starch grains to be stored in leaves in elevated CO2 (Robertson & Leech, 1995). In fact, biochemical analyses showed that the starch content of 10-d-old leaves of A. thaliana grown in ambient CO2 was significantly higher than that of young leaves of plants grown in elevated CO2 (data not shown).
Elevated CO2 reportedly increases the proportion of stroma (nonappressed) to grana (appressed) thylakoids (Kutik et al., 1995; Griffin et al., 2001). In sugar beets (Beta vulgaris), elevated CO2 increased the ratio of stroma thylakoids to chloroplast profile area when adequate nitrogen was supplied (Kutik et al., 1995). Similarly, Griffin et al. (2001) found a significant increase in the proportion of stroma thylakoid membranes to grana thylakoid membranes in leaves of four tree species (Acer rubrum, Cercis canadensis, Liquidambar styraciflua and Piper auritum) grown in elevated CO2 compared with leaves grown in ambient CO2. We found that elevated CO2 led to an increase in the ratio of stroma to grana thylakoids, similar to the findings of Kutik et al. (1995) and Griffin et al. (2001), but in contrast to those of Zuo et al. (2002), who found that elevated CO2 increased the degree of stacking of grana thylakoids. In addition, we found that the average number of grana thylakoid membranes decreased from 9.01 in ambient CO2 to 6.52 in elevated CO2 (Table 1), a pattern that has never previously been reported. Griffin et al. (2001) recently hypothesized that the changes in chloroplast ultrastructure in plants grown in elevated CO2 may be attributed to the higher energy demand required for faster growth, because more ATP is produced in chloroplasts with higher proportions of stroma thylakoid membranes to grana thylakoid membranes. In addition to the explanation suggested by Griffin et al. (2001), the increase in the number and size of starch grains in chloroplasts in elevated CO2 may also contribute to a reduction in the number of grana thylakoid membranes, because the more numerous and larger starch grains may press against and separate grana thylakoids, leading to a reduced number of grana thylakoid membranes or to some degree of damage to these membranes (Fig. 2a–f).
Many studies have shown that plants grown in elevated CO2 usually have increased soluble sugar and starch contents in leaves because of assimilation rates in excess of carbohydrate consumption rates (Delucia et al., 1985; Long & Drake, 1992; Moore et al., 1997). On average, soluble sugar and starch contents reportedly increase by c. 50 and 160%, respectively (Long & Drake, 1992). In orange (Citrus reticulata × C. paradisi) trees grown in elevated CO2 (720 µmol mol−1), total soluble sugars in leaves increased up to 50%, and starch and TNC increased by 424 and 166%, respectively (Vu et al., 2002). Obrist et al. (2001) found that elevated CO2 significantly increased soluble sugar, starch and TNC contents by 25, 53 and 40%, respectively. In tobacco (Nicotiana sylvestris), snapdragon (Antirrhinum majus) and parsley (Petroselinum hortense) grown in elevated CO2, the TNC content in leaves significantly increased by 108, 69 and 62%, respectively (Moore et al., 1997). We found that elevated CO2 also increased TNC content in A. thaliana leaves by c. 76%; whereas this result agrees with previous reports, it does not correspond with the results of Cheng et al. (1998), who found a two-fold increase in TNC in elevated CO2. This apparent discrepancy may be attributable to the different elevated CO2 concentrations used in the experiments, i.e. 1000 µmol mol−1 (Cheng et al., 1998) vs 700 µmol mol−1. In fact, CO2 concentrations over 1000 µmol mol−1 could be expected to further enhance the accumulation of carbohydrates in A. thaliana leaves. In comparison with the multitude of studies on the effects of CO2 enrichment on the TNC content, there have been few investigations of the effects of elevated CO2 on the foliar cellulose content. Staudt et al. (2001) reported that elevated CO2 resulted in a significant increase in foliar cellulose concentration in Quercus ilex. We also found that the foliar cellulose content in A. thaliana increased in elevated CO2, probably because of the increase in cell wall thickness (Table 1 and Fig. 2g,h), and there was no significant change in cell size (Table 1). This result is in agreement with the observations of Gibeaut et al. (2001), who speculated that the increased cellulose content may be caused by a higher allocation of carbohydrates to cell walls or an increase in the activity of cellulose synthase in plants grown in elevated CO2.
FTIR microspectroscopy is a fast, nondestructive and reliable method for examining cell wall components (Chen et al., 1998; Szyjanowicz et al., 2004; Hao et al., 2005). FTIR analysis showed that the cellulose content (1043 and 1157 cm−1) of cell walls was increased in elevated CO2, which is consistent with the results of biochemical analyses. In addition, FTIR spectra indicated that elevated CO2 increased the pectin content of cell walls. FTIR microspectroscopy provided an alternative method for measuring leaf cell wall components and reconfirmed that the cellulose content in foliar cell walls was indeed higher in elevated CO2.
Foliar concentrations of plant hormones
Elevated CO2 significantly increased the IAA, GA3, ZR, DHZR and iPA contents of leaves, but significantly reduced the ABA content (Table 3). Elevated CO2 also stimulated the growth of A. thaliana. This is interesting because plant hormones, such as CK, auxins and GA, can enhance plant growth and development by stimulating cell division, cell elongation and protein synthesis (Yong et al., 2000), whereas ABA is considered an inhibitor of leaf growth (Zhang & Davies, 1990). It is well known that carbohydrates are the most readily available form of energy in plant body and plant hormone metabolism is dependent on the supply of carbohydrates (Taiz & Zeiger, 1998). Therefore, we propose that higher carbohydrate production may result in higher hormone concentrations, which in turn may enhance plant growth. Indeed, higher plant growth rate accompanied by higher concentrations of carbohydrate and some hormones (not ABA) under elevated CO2 was reported previously (Jitla et al., 1997; Yong et al., 2000; Li et al., 2002).
In conclusion, using various methods and techniques, including TEM, FTIR, ELISA, atomic absorption spectrophotometry, photosynthesis systems and biochemical analysis, we investigated physiological, biochemical and structural changes in A. thaliana leaves grown in elevated CO2. Three findings are worth noting. First, the increased starch accumulation in chloroplasts in leaves grown in elevated CO2 may be responsible for the changes in chloroplast ultrastructure. Secondly, the significant increase in plant hormone concentrations in leaves grown in elevated CO2 could be an important factor in the accelerated growth and development of A. thaliana. Thirdly, the reductions in mineral nutrient concentrations in leaves grown in elevated CO2 were mainly attributable to dilution by carbohydrates and decreases in leaf stomatal conductance and transpiration rate.
This work was supported by the National Science Fund of China (90211005) and the National Science Fund of China for Distinguished Young Scholars (30225005). We thank Professor Yuxi Hu and Dr Jinsheng He for valuable discussion at the early stages of these experiments.