Effects of age and ontogeny on photosynthetic responses of a determinate annual plant to elevated CO2 concentrations

Authors


Correspondence: James D. Lewis. Fax:914 2736346; e-mail: jdlewis@fordham.edu

Abstract

Plant responses to elevated CO2 concentrations ([CO2]) may be regulated by both accelerated ontogeny and allocational changes as plants grow. However, isolating ontogeny-related effects from age-related effects are difficult because these factors are often confounded. In this study, the roles of age and ontogeny in photosynthetic responses to elevated [CO2] were examined on Xanthium strumarium L. grown at ambient (365 µmol mol−1) and elevated (730 µmol mol−1) [CO2]. To examine age-related effects, six cohorts were planted at 5-day intervals. To examine ontogeny-related effects, all plants were induced to flower at the same time; ontogeny in Xanthium is relatively unaffected by growth in elevated [CO2]. Growth in elevated [CO2] increased net photosynthetic rates by approximately 30% throughout vegetative growth (i.e. active carbohydrate sinks), approximately 10% during flowering (i.e. minimal sink activity), and approximately 20% during fruit production (i.e. active sinks). At the harvest, the ratio of source to sink tissue significantly decreased with increasing plant age and was correlated with leaf soluble sugar concentration. Leaf soluble sugar concentration was negatively correlated with the relative photosynthetic response to elevated [CO2]. These results suggest that age and ontogeny independently affect photosynthetic responses to elevated [CO2] and the effects are mediated by reversible changes in source : sink balance.

Introduction

By the end of the twenty-first century, atmospheric CO2 concentrations ([CO2]) are expected to double from the current concentration of approximately 365 µmol mol−1 (Houghton et al. 1995). Although net photosynthetic rates in C3 plants typically increase as the growth [CO2] increases, the relative response varies greatly both within and across species (Lloyd & Farquhar 1996; Poorter, Roumet & Campbell 1996; Curtis & Wang 1998; Saxe, Ellsworth & Heath 1998; Norby et al. 1999). Mechanisms to explain within-species variation often focus on changes in the balance between carbohydrate production (source) and the capacity to use and store carbohydrates (sink) as plants age (Griffin & Seemann 1996; Moore et al. 1999; Morison & Lawlor 1999). For example, if increased carbohydrate production associated with elevated [CO2] exceeds the capacity to produce new sinks as plants mature, net photosynthetic rates may decline to balance source activity and sink capacity (e.g. Thomas & Strain 1991).

Changes in photosynthetic responses to elevated [CO2] due to changes in source : sink balance may result from accelerated ontogeny (the timing and duration of phenologic events) (Loehle 1995) or morphological changes associated with increasing plant age (Centritto, Lee & Jarvis 1999; Bruhn, Leverenz & Saxe 2000). However, because elevated [CO2] may alter morphology by accelerating ontogeny, these factors are often confounded (Coleman, McConnaughay & Ackerly 1994), and identifying the direct effects of these two factors is difficult (Reekie 1996). Few studies have manipulated ontogeny and plant age independently to isolate the direct effects of these processes on plant responses to elevated [CO2].

Age-related effects on plant responses to elevated [CO2] are not a function of accelerated ontogeny in many determinate plants because their developmental rates are relatively inflexible due to genetic constraints (Harper 1977). For example, developmental rates in Xanthium strumarium L. (common cocklebur) and other determinate short-day plants are generally unaffected, or even slowed, by growth in elevated [CO2] (Reekie, Hicklenton & Reekie 1994). However, ontogeny in these plants can be manipulated through changes in photoperiod (Raven, Evert & Eichhorn 1999). By inducing plants to flower, and to subsequently fruit, at different ages the direct effects of accelerated ontogeny on plant responses to elevated [CO2] (comparing plants of the same age at different phenologic stages) can be separated from direct effects of plant age (comparing plants of different ages at the same phenologic stage).

Examining the effects of flowering, and time to flowering, on plant responses to elevated [CO2] also presents a way of examining how changes in source : sink balance affect plant responses to elevated [CO2]. In determinate plants, where photosynthetic responses to elevated [CO2] are sink-limited (Morison & Lawlor 1999), the relative photosynthetic response to elevated [CO2] should be directly proportional to the ratio of sink capacity to source activity. Determinate plants are characterized by a shift from vegetative to reproductive growth once flowering begins because flowering occurs from the terminal meristems (Harper 1977). This shift can be very abrupt; vegetative growth in X. strumarium var. canadense was minimal after flowering began in plants with a 68 d vegetative period (Shitaka & Hirose 1999). At the onset of flowering, sink activity drops substantially as vegetative growth slows and the primary sink tissues produced are flowers, which have relatively low carbon demand. Once fruit production begins, sink capacity begins to increase. Thus, the transition from vegetative to reproductive growth presents a natural, reversible manipulation of source : sink balance, where sink capacity goes from relatively high (vegetative growth) to relatively low (flowering) back to relatively high (fruit production).

In the present study, we examined the direct effects of accelerated ontogeny (e.g. the effect of inducing flowering simultaneously in young and older plants) and age-related changes in morphology on photosynthetic responses to elevated [CO2]. The effects of these factors and the reversibility of source : sink effects on photosynthetic responses to elevated [CO2] were examined by monitoring net photosynthetic rates from early vegetative growth through the onset of fruit production. To examine age-related effects, six cohorts of Xanthium strumarium were planted at intervals of 5 days. To examine ontogeny-related effects, all cohorts were induced to flower at 45 d after emergence of the first cohort. Specifically, we hypothesized that: (1) growth in elevated [CO2] would increase net photosynthetic rates during vegetative growth (i.e. high sink activity) across cohorts; (2) elevated [CO2] would not affect net photosynthetic rates during flowering (i.e. minimal sink activity) across cohorts; (3) elevated [CO2] would increase net photosynthetic rates during fruit production (i.e. high sink activity) across cohorts; and (4) during periods of high sink activity, the relative photosynthetic response to elevated [CO2] would vary among cohorts based on differences in source : sink balance among cohorts.

Materials and methods

Growth conditions

Seeds of Xanthium strumarium L. (common cocklebur) were obtained from a single seed source in Lubbock, Texas, USA. Plants were germinated and grown in 8·4 L pots filled with sand in four 1·4 m2 environmental growth chambers (Conviron Inc., Winnipeg, Manitoba, Canada) at Lamont-Doherty Earth Observatory, New York, USA (Wang et al. 2001). To examine the interactive effects of ontogeny and age on plant responses to elevated [CO2], six cohorts were planted at 5 d intervals (Fig. 1). Seedling emergence typically occurred 5 d after planting, so that one cohort emerged on the same day that the next cohort was planted. Following germination, seedlings were thinned to one per pot. The [CO2] in each of two chambers per CO2 treatment were automatically monitored and controlled to either 365 (ambient CO2 treatment) or 730 µmol mol−1 (elevated CO2 treatment). Air temperature in the chambers was maintained at 28/22 °C (day/night) with a photosynthetic photon flux density (PPFD) at the leaf surface of approximately 400 µmol m−2 s−1 during the 18 h photoperiod and thermoperiod. Relative humidity was maintained at approximately 50%. Flowering was simultaneously induced in all cohorts by changing the photoperiod to 12 h beginning 45 d after the first cohort emerged. Two days later, the photoperiod was changed back to 18 h and was maintained at 18 h until the end of the experiment. All plants simultaneously began flowering 12 d after induction (Wang et al. 2001). All pots were watered to saturation every morning with de-ionized water. Soil nutrients were supplemented by adding Osmocote Plus (15-11-13, 90269, Scotts-Sierra Horticultural Products Company, Marysville, Ohio, USA).

Figure 1.

Timing of developmental manipulations and physiological measurements.

Gas exchange measurements

Net photosynthetic rates and conductance were measured using infrared gas analysers built into a leaf cuvette in an open-flow gas exchange system (LI-6400; Li-Cor Inc., Lincoln, NE, USA). All measurements were made on the youngest mature, unshaded leaf. All measurements were made using a red : blue light source (LI-6400–02B; Li-Cor Inc.). The PPFD values at the upper leaf surface were typically 1500 µmol photons m−2 s−1. The airstream entering the cuvette was maintained at the desired [CO2] (either 365 or 730 µmol mol−1) using a computer-controlled CO2 mixing system supplied with the LI-6400. Leaf, cuvette and air temperatures were measured with thermocouples linked to the LI-6400 computer. Cuvette block temperature was maintained at 28 °C using a computer-controlled Peltier module mounted on the cuvette. Relative humidity inside the cuvette was maintained at approximately 50%.

Gas exchange measurements began 15 d after emergence of the first cohort, and were made approximately every 5 d until the induction of flowering. After the induction of flowering, measurements were made approximately every 10 d. Only the first cohort was measured at the first measurement day because the other cohorts had not emerged or had no fully developed true leaves. Measurements on each new cohort were begun 15 d after emergence, so that all cohorts were measured at each measurement day beginning 40 d after emergence of the first cohort. Prior to each measurement, leaf gas exchange was equilibrated in the cuvette at the measurement PPFD, the growth [CO2], and the measurement temperature. Leaf gas exchange was considered equilibrated if parameters were stable for 1 min. Measurements were initiated 1 h after the lights were turned on in the growth chambers, and typically were completed within 3 h.

To examine possible acclimation to elevated [CO2], photosynthesis was measured at the growth [CO2] and then at the opposite [CO2] during all but one measurement day (measurements at 37 d after emergence of the first cohort were made only at the growth [CO2]). Additionally, net photosynthetic rate versus intercellular [CO2] (A/Ci) response curves were measured at 43, 47, 59 and 71 d after emergence of the first cohort. The first two measurement days occurred during vegetative growth, measurements at day 59 were made during flowering, whereas measurements at day 71 were made during fruit development. To measure A/Ci response curves, external [CO2] was supplied in 12 steps ranging from 50 to 1500 µmol mol−1. Gas exchange parameters were allowed to equilibrate at each new [CO2] (typically requiring 2–3 min) prior to measurement, and were recorded automatically once the total coefficient of variation was less than 1%. Other than changing the [CO2], all A/Ci response curves were measured following the same protocols as the point measurements.

Photosynthetic response curves were analysed by calculating two parameters potentially limiting to photosyn­thesis: Vcmax[maximum carboxylation rate of ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco)] and Jmax (RuBP regeneration capacity mediated by maximum electron transport rate). This was achieved using the biochemical model describing A by Farquhar, von Caemmerer & Berry (1980) and von Caemmerer & Farquhar (1981):

inline image 1

where Ac and Aq are the net photosynthetic rates limited by Rubisco activity and electron transport rate, respectively. Rd is the daytime respiration rate resulting from processes other than photorespiration. The model was parameterized and fitted to the experimental data as in Lewis et al. (1994) using Photosyn Assistant (Dundee Scientific, Dundee, Scotland, UK). In brief, the values for Kc, Ko and VO,max/VC,max were calculated using the approach of Harley et al. (1992) and Wullschleger (1993). Initial estimates for Vcmax and Rd were then obtained using data collected at the six lowest [CO2]. After incorporating the initial estimates into the model, Vcmax and Jmax were simultaneously calculated using an iterative approach that minimizes the sum of squares.

At the final measurement day, the relative photosynthetic response to elevated [CO2] was calculated as the ratio of the mean net photosynthetic rate in the elevated CO2 treatment to the mean net photosynthetic rate in the ambient CO2 treatment measured at the growth [CO2].

Leaf biochemical measurements

At the end of the last gas exchange measurements, all plants were harvested and leaves were separated from stems and roots. Leaf area was measured using a leaf area meter (LI-3000 A; Li-Cor Inc.). After leaf area measurements, leaves were dried and ground. Leaf area ratio was calculated as the leaf area per unit total plant dry mass. Leaf N and C concentration were determined using a NCS autoanalyser (Carlo Erba NCS 2500, Milan, Italy). Soluble sugar content and starch content of leaves were determined colorimetrically using a phenol–sulphuric acid technique (Tissue & Wright 1995). Total non-structural carbohydrate (TNC) content, expressed as a percentage of leaf dry mass, was calculated as the sum of soluble sugar and starch.

Statistical analysis

The experimental design was a three-factor design (CO2 supply, cohort, chamber) with one nested factor (chamber nested within CO2 treatment). Chamber was nested within the CO2 treatment because there were two chambers per CO2 treatment, but each chamber received only one CO2 treatment. All cohorts were replicated in each chamber, and thus cohort was orthogonal to both CO2 treatment and chamber. There were three replicates per cohort by chamber combination for a total of 72 plants (six cohorts × two CO2 treatments × two chambers per CO2 treatment × three replicates per chamber). Measurements were made on two to three replicates per cohort by chamber combination except at 47 d after emergence of the first cohort, when only one plant per cohort by chamber combination was measured. To examine treatment effects at a given measurement day, data were analysed following Winer (1971) and Lindman 1991) using MGLH in systat (SPSS Inc., Chicago, IL, USA) with CO2 supply, cohort and chamber as fixed factors. To examine temporal changes in net photosynthetic rates during the transition from vegetative growth to fruit production, data were analysed using repeated measures analysis of variance (anova) with CO2 supply and cohort as the between-subjects factors and measurement day as the within-subjects factor. Data were examined for normality by analysing histograms, normal probability plots and by plotting Studentized residuals versus predicted values. Least square means were compared using Tukey’s test for pairwise comparisons. Treatment effects were considered to be significant if P ≤ 0·05. Although there were significant interactions between CO2 supply and chamber for several measurement days, data presented here are combined across chambers because CO2 treatment effects generally were consistent across chambers.

Results

Changes in developmental stage were associated with changes in the relative response to elevated CO2 (Fig. 2a). Growth in elevated CO2 increased net photosynthetic rates by approximately 30% throughout vegetative growth (i.e. active carbohydrate sinks), by approximately 10% during flowering (i.e. minimal sink activity), and by approximately 20% during fruit production (i.e. active sinks). Although the relative response varied between developmental stages, plants grown in elevated CO2 consistently exhibited significantly higher net photosynthetic rates than plants grown in ambient CO2 when measured at their respective growth [CO2] (Table 1). Across the 56 d vegetative growth period, mean net photosynthetic rates in the elevated CO2 treatment varied by less than 7%, ranging between 41·7 and 44·5 µmol m−2 s−1, whereas the mean net photosynthetic rates in the ambient CO2 treatment ranged from 31·3 to 38·7 µmol m−2 s−1. Net photosynthetic rates in both CO2 treatments significantly declined after the onset of flowering compared to rates at the end of vegetative growth, based on repeated measures anova (P ≤ 0·001). After the onset of fruit production, net photosynthetic rates across CO2 treatments significantly increased compared with rates during flowering (P = 0·003). Net photosynthetic rates did not significantly vary among cohorts on nine of 10 measurement days, and there were no significant interactions between CO2 supply and planting date (P ≥ 0·139 in all cases). For the repeated measures anova, there also were no significant interactions between CO2 supply, cohort and measurement day (P ≥ 0·083 in both cases).

Figure 2.

Figure 2.

Effect of CO2 treatment on mean (± SE) net photosynthetic rates measured at the growth [CO2] (a), at 365 µmol mol−1 CO2 (b), and at 730 µmol mol−1 CO2 (c), along with Ci : Ca ratios at the growth CO2 (d) over the course of the experiment. Periods corresponding to vegetative growth, flowering and fruit production for all cohorts are also shown. Measurements made between 15 and 52 d after emergence of the first cohort were made during vegetative growth, measurements at 59 d were made during flowering, whereas measurements at 71 d were made during fruit development. Asterisks (*) indicate treatment means that are significantly different at P ≤ 0·05. ‘+’ indicates a significant interaction between CO2 supply and planting date. Otherwise, there were no significant interactions between CO2 supply and planting date (P ≥ 0·062 in all cases). n= 6–30 for each CO2 treatment at each measurement day.

Figure 2.

Figure 2.

Effect of CO2 treatment on mean (± SE) net photosynthetic rates measured at the growth [CO2] (a), at 365 µmol mol−1 CO2 (b), and at 730 µmol mol−1 CO2 (c), along with Ci : Ca ratios at the growth CO2 (d) over the course of the experiment. Periods corresponding to vegetative growth, flowering and fruit production for all cohorts are also shown. Measurements made between 15 and 52 d after emergence of the first cohort were made during vegetative growth, measurements at 59 d were made during flowering, whereas measurements at 71 d were made during fruit development. Asterisks (*) indicate treatment means that are significantly different at P ≤ 0·05. ‘+’ indicates a significant interaction between CO2 supply and planting date. Otherwise, there were no significant interactions between CO2 supply and planting date (P ≥ 0·062 in all cases). n= 6–30 for each CO2 treatment at each measurement day.

Table 1.  Probability values from anova of gas exchange measurements at each measurement day. (A) Net photosynthetic rates were compared at the growth [CO2], 365 µmol mol−1, and 730 µmol mol−1 CO2. (B) Vcmax, Jmax and Rd were calculated from A/Ci measurements made 43, 47, 59 and 71 d after emergence of the first cohort. Measurements made between 15 and 52 d after emergence of the first cohort were made during vegetative growth, measurements at 59 d were made during flowering, whereas measurements at 71 d were made during fruit development.
(A)
Days after emergence
of first cohort
Photosynthesis
measured at growth
CO2
Photosynthesis
measured at
365 µmol mol−1
Photosynthesis
measured at
730 µmol mol−1
Ci/Ca measured at
growth CO2
CO2CohortCO2CohortCO2CohortCO2Cohort
15  0·007   0·978   0·424   0·077 
20<0·0010·175  0·7300·426  0·1140·099<0·001  0·140
25<0·0010·063  0·1500·023  0·3270·075<0·001  0·028
30<0·0010·012  0·0960·002  0·1570·044  0·084<0·001
37  0·0360·833      0·478  0·974
40<0·0010·326  0·0090·069  0·0060·190  0·002  0·059
43<0·0010·501++  0·0260·506<0·001  0·005
47  0·0390·794  0·1830·947  0·2590·642  0·167  0·593
52<0·0010·057<0·0010·876  0·1210·005  0·723  0·201
59 (flowering)  0·0260·715<0·0010·992<0·0010·632  0·538  0·303
71 (fruit development)<0·0010·528<0·0010·366<0·0010·937  0·002  0·421
(B)
Days after emergence
of first cohort
Vcmax (µmol m−2 s−1)Jmax (µmol m−2 s−1)Rd (µmol m−2 s−1)
CO2DateCO2DateCO2Date
  1. Significant effects of CO2 supply or planting date are boldface. ‘+’ indicates a significant interaction between CO2 supply and planting date (please see text). Otherwise, there were no significant interactions between CO2 supply and planting date (P ≥0·062 in all cases). Only one cohort was measured 15 d after the emergence of the first cohort. At 37 d after emergence of the first cohort, photosynthesis was measured only at the growth [CO2].

20
30
40
430·4020·992++  0·3170·212
470·1460·784  0·3210·542  0·3910·363
52
59 (flowering)0·0020·479<0·0010·428<0·0010·648
71 (fruit development)0·2930·464  0·0190·800<0·0010·988

Net photosynthetic rates measured at 365 µmol mol−1 CO2 did not significantly vary between CO2 treatments during the first four measurement days (Fig. 2b; Table 1). However, at 40 and 52 d after emergence of the first cohort, during flowering and during fruit production, net photosynthetic rates measured at 365 µmol mol−1 CO2 were significantly lower in the elevated CO2 treatment relative to the ambient CO2 treatment. Also, at 43 d after emergence of the first cohort, there was a significant interaction between CO2 supply and planting date on net photosynthetic rates measured at 365 µmol mol−1 (P = 0·044). At this measurement day, net photosynthetic rates measured at 365 µmol mol−1 CO2 were significantly higher in cohort 1 in the ambient CO2 treatment compared to cohorts 2 and 4 in the elevated CO2 treatment, and were significantly higher in cohort 4 in the ambient CO2 treatment compared to cohort 2 in the elevated CO2 treatment. There were no other significant differences at this day. In addition, there were no other significant interactions between CO2 supply and planting date on net photosynthetic rates measured at 365 µmol mol−1 CO2 (P ≥ 0.218 in all cases).

Net photosynthetic rates measured at 730 µmol mol−1 CO2 did not significantly vary between CO2 treatments at six of eight measurement days during vegetative growth (Fig. 2c; Table 1). However, during flowering and fruit production, net photosynthetic rates measured at 730 µmol mol−1 CO2 were significantly lower in the elevated CO2 treatment relative to the ambient CO2 treatment. Also, net photosynthetic rates measured at 730 µmol mol−1 CO2 were significantly lower in the elevated CO2 treatment relative to the ambient CO2 treatment at 40 and 43 d after emergence of the first cohort. There were no significant interactions between CO2 supply and planting date on net photosynthetic rates measured at 730 µmol mol−1 CO2 (P ≥ 0·062 in all cases). In addition, net photosynthetic rates measured at 365 or 730 µmol mol−1 CO2 did not significantly vary among cohorts at seven of nine measurement days.

The ratio of intercellular CO2 to ambient CO2 (Ci/Ca) was significantly higher in the elevated CO2 treatment at five of 10 measurement days, but did not significantly vary with CO2 supply on the other days and showed no clear pattern within or between the vegetative and reproductive growth periods (Fig. 2d, Table 1). Ci/Ca did not significantly vary among cohorts at seven of 10 measurement days, and there were no significant interactions between CO2 supply and planting date (P ≥ 0·086 in all cases).

Leaf physiological parameters calculated from A/Ci curves generally did not significantly vary between CO2 treatments during vegetative growth (Table 1). However, Vcmax and Jmax were significantly lower in the elevated CO2 treatment relative to the ambient CO2 treatment during flowering (59 d after emergence of the first cohort), whereas Rd was significantly higher in the elevated CO2 treatment (Table 2). Vcmax did not significantly vary between CO2 treatments during fruit production (71 d after emergence of the first cohort), whereas Jmax was significantly lower and Rd significantly higher in the elevated CO2 treatment relative to the ambient CO2 treatment. In addition, there was a significant interaction between CO2 supply and planting date on Jmax (P = 0·022) at 43 d after emergence of the first cohort. On this measurement day, Jmax was significantly higher in cohort 1 in the ambient CO2 treatment compared to cohort 1 in the elevated CO2 treatment. Jmax did not significantly vary between other treatment combinations at this day. In addition, there were no other significant interactions between CO2 supply and planting date (P ≥ 0·061 in all cases) or significant differences among cohorts (P ≥ 0·212 in all cases).

Table 2.  Comparison of means (± SE) in the ambient and elevated CO2 treatments for Vcmax, Jmax and Rd calculated from A/Ci measurements made 43, 47, 59 and 71 d after emergence of the first cohort. Measurements on the first two days were made during vegetative growth, measurements at 59 d were made during flowering, whereas measurements at 71 d were made during fruit development
 Vcmax
(µmol m−2 s−1)
Jmax
(µmol m−2 s−1)
Rd
(µmol m−2 s−1)
  1. Asterisks (*) and different letters indicate means that are significantly different at P ≤0·05. ‘+’ indicates a significant interaction between CO2 supply and planting date. n= 12–24 for each [CO2].

43 d + 
Ambient CO2134·5 ± 3·1342·7 ± 11·71·8 ± 0·3
Elevated CO2131·3 ± 3·1310·5 ± 11·72·1 ± 0·3
47 d
Ambient CO2154·2 ± 4·5364·2 ± 13·52·5 ± 0·3
Elevated CO2147·1 ± 4·5342·5 ± 13·53·3 ± 0·3
59 d***
Ambient CO2148·8 ± 3·8a343·8 ± 13·1a1·7 ± 0·3b
Elevated CO2129·6 ± 3·8b252·6 ± 13·1b3·6 ± 0·3a
71 d **
Ambient CO2136·9 ± 3·4314·4 ± 11·5a0·7 ± 0·3b
Elevated CO2131·9 ± 3·4276·9 ± 11·5b2·1 ± 0·3a

Leaf area ratio (LAR; leaf area per unit plant dry mass), which can be used as an estimate of the ratio of source tissue to sink tissue (Lewis & Strain 1996), significantly increased from the oldest to the youngest cohort at the final harvest (P = 0·003; Fig. 3). In parallel with the changes in LAR, leaf soluble sugar (SS) and total non-structural carbohydrate (TNC) concentrations significantly increased from the oldest to the youngest cohort (P ≤ 0·022 in both cases). Increasing LAR was associated with a significant increase in SS concentrations (P = 0·004; corrected r2 = 0·527), and increasing SS concentrations in elevated CO2 plants were associated with a significant decrease in the relative photosynthetic response to elevated CO2 (Fig. 4). A logarithmic equation provided the best fit for the relationship between SS concentration and the relative photosynthetic response to elevated CO2 (corrected r2 = 0·726). Leaf SS and nitrogen concentrations significantly decreased with increasing [CO2] (P ≤ 0·038 in both cases). Leaf TNC concentration and LAR did not significantly vary with CO2 supply, leaf nitrogen concentration did not significantly vary among cohorts, and CO2 supply and cohort did not significantly affect leaf starch and total carbon concentrations or the carbon : nitrogen ratio (P ≥ 0·111 in all cases). There were no significant interactions between cohort and CO2 supply on LAR or leaf biochemical parameters (P ≥ 0·387 in all cases).

Figure 3.

Effects of planting date on mean (± SE) leaf soluble sugar (SS) concentrations, leaf total non-structural carbohydrate (TNC) concentrations and leaf area ratio (LAR; leaf area per unit plant dry mass). Note difference in y-axis scale for LAR. n= 8 for each treatment.

Figure 4.

The relationship between mean leaf area ratio and mean leaf soluble sugar concentrations at the final measurement day for all CO2 treatment by cohort combinations (a), and between mean leaf soluble sugar concentrations for cohorts in the elevated CO2 treatment and the relative photosynthetic response to elevated CO2 at the final measurement day (b). Leaf area ratio reflects the relative balance between carbohydrate source and sink tissue.

Discussion

These results clearly demonstrate that both accelerated ontogeny and changes in allocation as plants age affect photosynthetic responses to elevated [CO2], and that these responses are at least partially driven by changes in carbohydrate source : sink balance. As we hypothesized, the photosynthetic response to elevated [CO2] was reduced by the onset of flowering, associated with minimal sink activity, and the reduction was reversed by fruit production, associated with increased sink activity (Fig. 2). The reduction in relative response to elevated [CO2] during flowering and the reversal during fruiting occurred regardless of plant age, suggesting accelerated ontogeny directly influenced plant responses to elevated [CO2]. In other words, by accelerating the onset of flowering and fruiting, the effects of these processes on photosynthetic responses to elevated [CO2] occurred earlier in younger cohorts relative to older cohorts. These results are consistent with other studies that have demonstrated changes in plant responses to elevated [CO2] due to accelerated ontogeny (Ackerly et al. 1992; Coleman, McConnaughay & Bazzaz 1993; Bazzaz, Miao & Wayne 1993; Seneweera et al. 1995; Gebauer, Reynolds & Strain 1996; Miller et al. 1997).

Two unique aspects of this study are that the changes in ontogeny were not due to elevated [CO2] and that the effect of changes in ontogeny was reversible. As a result, changes in the photosynthetic response to elevated [CO2] could be linked to specific phenologic events, and the effects of these events could be linked to reversible changes in source : sink balance. This study is also unusual in that few studies have manipulated source : sink balance within individual plants to produce reversible changes in relative photosynthetic response to elevated [CO2]. Most manipulative studies that have examined effects of changes in source : sink balance on photosynthetic responses to elevated [CO2] could not produce reversible changes because the manipulation involved tissue removal. Tissue removal also creates wound responses that may be confounded with changes in source : sink balance (Reekie et al. 1998). Our technique presents a non-invasive method to reversibly examine the role of changing source : sink balance associated with accelerated ontogeny in the response of plants to elevated [CO2].

As with accelerated ontogeny, our results suggest that the effects of morphological changes associated with plant age on the relative response to elevated [CO2] during fruiting were mediated through effects on the balance between source activity and sink capacity. Leaf area ratio (leaf area per unit plant dry mass) was correlated with increased leaf soluble sugar concentration and decreased with age at the final harvest, suggesting that relative source activity decreased with increasing age (Fig. 4). Leaf soluble sugar concentration was negatively correlated with the relative photosynthetic response to elevated [CO2]. In agreement with our hypothesis, these correlations suggest that the relative photosynthetic response to elevated [CO2] decreased due to changes in source : sink balance as the ratio of source tissue to sink tissue increased. These results are consistent with other studies which have shown that plant responses to growth in elevated [CO2], both within and across related species, decline in conjunction with reductions in the capacity to generate additional sinks (e.g. new vegetative or reproductive structures) to utilize the increased supply of carbohydrates (Arp 1991; Hunt et al. 1991; Thomas & Strain 1991; Poorter 1993; Poorter 1998; Reekie et al. 1998; Atkin et al. 1999). Further, because differences in source : sink balance associated with plant age occurred at the same developmental stage (fruiting), age-related effects could be isolated from effects related to accelerated ontogeny. These results also suggest that changes in morphology may have indirectly affected plant responses to accelerated ontogeny through effects on changes in source : sink balance.

Reductions in the relative response to elevated [CO2] associated with reduced sink activity and increased leaf soluble sugar concentrations are consistent with the sugar-signalling feedback mechanism that has been proposed to account for regulation of photosynthetic responses to elevated [CO2] (Sheen 1990; Griffin & Seemann 1996; Drake, Gonzàlez-Meler & Long 1997; Midgley, Wand & Pammenter 1999; Moore et al. 1999). Nitrogen metabolism may also play an important role in co-ordinating source activity and sink capacity (Stitt & Krapp 1999). Leaf nitrogen concentration at the final harvest significantly decreased with increasing CO2, but did not significantly vary among cohorts. Although the leaf nitrogen data do not specifically address whether carbohydrates or nitrogen drove changes in the relative photosynthetic responses to elevated [CO2], they suggest that nitrogen metabolism was not the primary factor regulating photosynthetic responses in this study. The net result of these signalling pathways is the maintenance of a dynamic balance between carbohydrate production and utilization that may substantially account for the differential acclimation responses observed within and between species (Midgley et al. 1999; Moore et al. 1999). Reductions in sink capacity lead to proportionate reductions in the photosynthetic response to elevated [CO2], as observed during flowering in this study. Conversely, increasing sink capacity in plants that are sink limited will increase the response to elevated [CO2], as occurred during fruit production.

Changes in photosynthetic responses to elevated [CO2] associated with changes in sink activity should be accompanied by changes in Rubisco activity (Vcmax) because changes in sink activity are predicted to alter photosynthesis through effects on Rubisco (Moore et al. 1999). In addition, changes in photosynthetic responses to elevated [CO2] may be paralleled by changes in RuBP regeneration mediated by electron transport capacity (Jmax) to balance RuBP carboxylation activity with RuBP regeneration capacity (Sage 1994). Consistent with these predictions, photosynthetic adjustment was observed in the elevated CO2 treatment during flowering, based on significant reductions in both Vcmax and Jmax relative to the ambient CO2 treatment, as well as in net photosynthetic rates measured at a common [CO2]. The photosynthetic adjustment observed during flowering was reversed during fruit production, based on the lack of differences between CO2 treatments in Vcmax. However, the reversal was not complete, as Jmax and net photosynthetic rates measured at a common [CO2] were lower in the elevated CO2 treatment relative to the ambient CO2 treatment. It is unclear whether the reversal was incomplete because fruit production is a small sink relative to vegetative growth or because measurements were made shortly after fruiting began and hence there was not enough time for complete recovery of the processes regulating Jmax, which are inherently less dynamic than changes in Rubisco activity.

Although this study focused on the effects of the timing of reproduction in an annual plant, plant-level changes in source : sink balance may also play an important role in photosynthetic responses to elevated [CO2] in longer-lived plants. For example, changes in sink activity due to seasonal growth patterns (Tissue, Thomas & Strain 1997; Lewis, Olszyk & Tingey 1999; Myers, Thomas & DeLucia 1999) and mycorrhizal colonization (Lewis & Strain 1996; Loewe et al. 2000) have been associated with changes in the photosynthetic response of trees to elevated [CO2]. Additionally, reductions in photosynthetic responses to elevated [CO2] observed in some species during long-term exposure (3 or more years) have been attributed to changes in source : sink balance as trees age (Griffin et al. 2000).

Effects of plant-level changes in source : sink balance on photosynthetic responses to elevated [CO2] are likely to interact with effects of leaf development (Farrar 1996). Several studies have demonstrated that the photosynthetic response to elevated [CO2] declines as leaves mature (Porter & Grodzinski 1984; Besford, Ludwig & Withers 1990; Kelly, Hicklenton & Reekie 1991; Miller et al. 1997; Turnbull et al. 1998; Tissue et al. 2001). These changes often reflect photosynthetic adjustment to balance leaf-level source activity and sink capacity (Van Oosten & Besford 1995; Pearson & Brooks 1995; Wait et al. 1999), because young leaves are typically strong sinks for carbohydrates, whereas mature leaves are generally strong sources (Vogelmann, Larson & Dickson 1982). In the current study, effects of leaf development on responses to elevated [CO2] were minimized by performing all measurements on the most-recently fully expanded leaf on each plant.

In summary, we used a novel experimental technique to isolate effects related to plant age from effects related to accelerated ontogeny on photosynthetic responses to elevated [CO2]. Our results suggest that both age and ontogeny directly affect plant responses to elevated [CO2] through changes in source : sink balance, and that the effects of changes in source : sink balance are reversible. Additionally, these results are consistent with other studies that have shown that changes in photosynthetic responses to elevated [CO2] are due to both genetically and environmentally driven changes in source : sink balance. As a result, source : sink balance may be useful for integrating the effects of genetic and environmental factors to predict plant responses to elevated [CO2] at a given point in time.

Acknowledgments

We thank Ardis Thompson for her technical support in the design and execution of this study. Dr Jacqui Johnson, Dr Steve Long and two anonymous reviewers improved earlier drafts of this manuscript. This project was supported in part by funding from Fordham University to J.D.L and by a National Science Foundation grant (IBN-9603940) to K.L.G. This is contribution number 206 from the Louis Calder Center and Biological Station, Fordham University.

Received 5 April 2001;received in revised form 2 October 2001; accepted for publication 2 October 2001

Ancillary

Advertisement