JXB Advance Access originally published online on August 1, 2006
Journal of Experimental Botany 2006 57(12):3057-3067; doi:10.1093/jxb/erl067
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Temperature response of photosynthesis and internal conductance to CO2: results from two independent approaches
1School of Forest and Ecosystem Science, University of Melbourne, Water Street, Creswick, VIC 3363, Australia
2INRA, UMR INRA-UHP, Ecologie et Ecophysiologie Forestières, F-54280 Champenoux, France
*To whom correspondence should be addressed. E-mail: charles.warren{at}bio.usyd.edu.au
Received 27 March 2006; Accepted 30 May 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The internal conductance to CO2 transfer from intercellular spaces to chloroplasts poses a major limitation to photosynthesis, but few studies have investigated its temperature response. The aim of this study was to determine the temperature response of photosynthesis and internal conductance between 10 °C and 35 °C in seedlings of a deciduous forest tree species, Quercus canariensis. Internal conductance was estimated via simultaneous measurements of gas exchange and chlorophyll fluorescence (‘variable J method’). Two of the required parameters, the intercellular photocompensation point (Ci*) and rate of mitochondrial respiration in the light (Rd), were estimated by the Laisk method. These were used to calculate the chloroplastic photocompensation point (
*) in a simultaneous equation with gi. An independent estimate of internal conductance was obtained by a novel curve-fitting method based on the curvature of the initial Rubisco-limited portion of an A/Ci curve. The temperature responses of the rate of Rubisco carboxylation (Vcmax) and the RuBP limited rate of electron transport (Jmax) were determined from chloroplastic CO2 concentrations. The rate of net photosynthesis peaked at 24 °C. Ci* was similar to reports for other species with a Ci* of 39 µmol mol–1 at 25 °C and an activation energy of 34 kJ mol–1. * was very similar to the published temperature response for Spinacia oleracea from 20 °C to 35 °C, but was slightly greater at 10 °C and 15 °C. Jmax peaked at 30 °C, whereas Vcmax did not reach a maximum between 10 °C and 35 °C. Activation energies were 49 kJ mol–1 for Vcmax and 100 kJ mol–1 for Jmax. Both methods showed that internal conductance doubled from 10 °C to 20 °C, and then was nearly temperature-independent from 20 °C to 35 °C. Hence, the temperature response of internal conductance could not be fitted to an Arrhenius function. The best fit to estimated gi was obtained with a three-parameter log normal function (R2=0.98), with a maximum gi of 0.19 mol m–2 s–1 at 29 °C. Key words: Carbon dioxide, diffusion, internal resistance, mesophyll resistance, photosynthesis, temperature responses, transfer conductance, transfer resistance
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
For photosynthesis to occur, CO2 must diffuse from the atmosphere to the sites of carboxylation. The equivalent concentration of CO2 at the sites of carboxylation (Cc) is less than atmospheric (Ca) owing to a series of gas-phase (air) and liquid-phase (mesophyll cells) resistances. In the gaseous phase, CO2 must diffuse across a boundary layer in the air above the foliage surface, through stomatal openings, and across intercellular air spaces surrounding mesophyll cells. In the liquid phase, there are resistances as CO2 enters the liquid phase at the surface of mesophyll cells, as CO2 diffuses within the cell to the chloroplast membrane, and from there to the sites of carboxylation (Gaastra, 1959; Aalto and Juurola, 2002).
Carbon dioxide concentrations at Cc are significantly less than Ci (Evans et al., 1986; Parkhurst and Mott, 1990; Lloyd et al., 1992; Epron et al., 1995). Internal conductance (gi) describes this drawdown in CO2 concentration between Ci and Cc as a function of net photosynthesis [gi=A/(Ci–Cc)]. Internal conductance may limit photosynthesis by 20% or more and has a large effect on in vivo estimates of the maximum rate of carboxylation (Vcmax), and of the Michaelis–Menten constants for carboxylation (Kc) and oxygenation (Ko) (von Caemmerer et al., 1994; Epron et al., 1995; Bernacchi et al., 2002; Warren et al., 2003b; Ethier and Livingston, 2004).
There are several sources of variation in gi. Among species, gi is typically two to three times greater in annuals than in woody species (von Caemmerer and Evans, 1991; Loreto et al., 1992; Epron et al., 1995; Warren et al., 2003b). gi is also correlated with morphological and anatomical traits (Vitousek et al., 1990; Evans et al., 1994; Kogami et al., 2001), and physiological parameters such as maximum stomatal conductance (gs) (Loreto et al., 1992), rates of light-saturated photosynthesis (Amax) (Warren et al., 2003b), carbonic anhydrase activity (Makino et al., 1992; Price et al., 1994), and aquaporin content (Uehlein et al., 2003; Hanba et al., 2004).
Temperature is one source of variation in gi that has received little attention, despite these data being critical for photosynthesis models, the correct determination of Vcmax, and understanding the major limitations of photosynthesis (Bernacchi et al., 2002). Such data may also provide clues as to the processes determining gi. For example, the temperature response of gi in Nicotiana tabacum has a temperature coefficient (Q10) of 
2.2 and decreases at temperature above 35 °C—this was argued to indicate control by a protein-facilitated process (Bernacchi et al., 2002). By contrast, internal conductance of Eperua grandiflora was found to be temperature independent from 28 °C to 38 °C (Pons and Welschen, 2003). The primary aim of this study was to determine the temperature response of photosynthesis and gi between 10 °C and 35 °C in saplings of a deciduous oak species with a Mediterranean distribution, Quercus canariensis. gi was determined with the variable J method (Di Marco et al., 1990; Loreto et al., 1992) and a novel curve-fitting technique based on the curvature of the Rubisco-limited portion of an A/Ci curve (Ethier and Livingston, 2004). The Laisk method estimated the temperature response of the intercellular photocompensation point (Ci*) and mitochondrial respiration in the light (Rd). gi is highly sensitive to the value of the chloroplastic photocompensation point (*) used; therefore, gi was calculated using three different * estimates: (i) the measured Ci* for Q. canariensis; (ii) * calculated as a simultaneous equation with gi(Ci*+Rd/gi=*); and (iii) the estimate of * by Bernacchi et al. (2002). Finally, Cc was used, to estimate the temperature responses of the rates of Rubisco carboxylation (Vcmax) and RuBP-limited electron transport (Jmax).
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material
Seedlings of Quercus canariensis Willd. (Zeen oak, Fagaceae, from the tribe ‘robur’, i.e. European deciduous white oaks) were collected during autumn 2002 from a natural adult stand close to Ain Drahem, Northern Tunisia (800 m a.s.l.). They were treated with a fungicide and kept at –1 °C over winter. In early 2003, seedlings were transplanted into pots (10 l) containing a 2:1 mixture of sand with blond peat. Seedlings were placed in a greenhouse at Champenoux (Nancy, north-east France) under natural illumination. The absolute minimum and maximum temperatures during the growth season were 13 °C and 32 °C, respectively. Each year seedlings received a slow-release fertilizer (Nutricote 100, 13/13/13 with oligoelements, 40 g per pot). Measurements were conducted during September 2004 on fully mature current-year leaves.
Experimental protocol
Measurements of gas exchange and chlorophyll a fluorescence were made between 8 September and 13 October on one attached leaf from each of six Q. canariensis seedlings at leaf temperatures of 10, 15, 20, 25, 30, and 35 °C. For each plant, the same leaf was used at each temperature to minimize variability in photosynthetic capacity. Measurements were made in two walk-in climate chambers with air temperature controlled within 2 °C of the target leaf temperature. Inside the climate chamber, seedlings experienced a 16 h/8 h photoperiod with a PPFD of 300 µmol m–2 s–1 at leaf height. Seedlings were transferred from the greenhouse to the climate chamber 12 h before measurements commenced. To minimize the possibility of seedlings acclimating to temperatures other than those of the greenhouse, they were immediately returned to the greenhouse once measurements (at that temperature) were complete. Measurements were made in the order 15, 10, 20, 25, 30, 35 °C, and seedlings spent 13 d of the 35 d measurement period in the climate chambers.
Gas exchange and chlorophyll fluorescence systems
Three chief measurements were made: (i) a calibration of the relationship between linear electron transport estimated from chlorophyll fluorescence and linear electron transport estimated from gas exchange under non-photorespiratory conditions; (ii) the CO2 response of gas exchange and fluorescence; and (iii) simultaneous estimation of Ci* and Rd via the Laisk method (Laisk, 1977) and of gi with the curve-fitting method developed by Ethier and Livingston (2004). Simultaneous gas exchange and fluorescence measurements (i.e. i and ii) were made with an LI-6400 gas exchange system with integrated fluorescence chamber (LI-6400-40; Li-Cor, Lincoln, NE, USA). Ci*, Rd, and curve-fit estimates of gi required a different LI-6400 that had a modified 2x3 cm broadleaf chamber and an integrated light source (LI-6400-02B; Li-Cor). Using the 2x3 cm broadleaf chamber, as opposed to the fluorescence chamber, reduced the influence of diffusional leaks, and increased 3-fold the enclosed leaf area and the signal-to-noise ratio—critical factors when estimating Ci* and Rd, and also essential for capturing the curvature of an A/Ci curve.
The 2x3 cm broadleaf chamber was modified to reduce diffusional leaks by enclosing it in a secondary chamber flushed with the reference gas of the gas exchange system. The secondary chamber comprised upper and lower U-shaped pieces of 2 cm width that were fitted with neoprene gaskets (1 cm wide) on the outer surface. The U-shaped chambers were glued around the upper and lower chamber ‘lips’ of the Li-6400, forming a second chamber with a small air-filled cavity that was flushed with reference gas at a low flow rate. This procedure reduced significantly the diffusion gradient between the inside and the outside of the chamber.
Before making any measurements both LI-6400s were calibrated (zero CO2 and H2O, and 501 µmol mol–1 CO2 span), and each day both were re-zeroed using freshly regenerated drierite and new soda lime. This procedure was necessary to take into account the (minor) temperature-induced shifts of zeros. With both instruments, relative humidity was controlled between 50% and 70% using a custom-built water trap that controlled the dew point temperature of incoming air at ±0.1 °C via Peltier cooling. Measurements were made with leaf temperature controlled ±1 °C of the target temperature.
Calibration of the relationship between chlorophyll fluorescence and rates of electron transport
The photochemical efficiency of photosystem II (
PSII) was computed from steady-state fluorescence (F') and maximal fluorescence (F'm) during a light-saturating pulse (Genty et al., 1989):
 | (1) |
PSII:
 | (2) |
is the total leaf absorptance and the factor 0.5 describes the expected distribution of light between the two photosystems. Fluorescence estimates of J(Jf) are not strictly related to linear electron transport because fluorescence primarily measures upper cell layers and is thus not representative of the whole leaf. Furthermore, the distribution of light between photosystems is set to be equal, but this might not be the case. Owing to these uncertainties, no a priori assumptions were made regarding relationships between Jf and J. Instead, an empirical relationship was determined. Such a ‘calibration’ procedure obviates the need to measure leaf absorptance and to make assumptions regarding the distribution of light between photosystems and representativeness of fluorescence to whole-leaf processes. The relationship between Jf and J was determined under non-photorespiratory conditions (1% O2, 1000 µmol mol–1 CO2) on one leaf from each of the six seedlings at 20, 30, and 35 °C. This method assumes that under non-photorespiratory conditions the rate of linear electron transport is wholly associated with gross photosynthesis, i.e. J=4(A+Rd). Rd was determined via the Laisk method (Laisk, 1977; see below). Mass flow controllers mixed atmospheric air with N2 to produce 1% O2. This mixture was scrubbed of CO2 using the soda lime column of the LI-6400, and CO2 was added to a concentration of 1000 µmol mol–1 using the LI-6400's CO2 mixer. Measurements were made inside a climate chamber maintained at the target leaf temperature, and with leaf temperature controlled more precisely (via the LI-6400) at 20, 30, and 35 °C. One leaf (per seedling) was placed in the fluorescence chamber (flow rate=300 µmol s–1) and acclimated to darkness for at least 30 min before gas exchange was recorded and a saturation pulse given to determine fluorescence. This procedure was repeated at 500 µmol m–2 s–1 and then 1500 µmol m–2 s–1, with a 30 min wait at each PPFD before measuring gas exchange and fluorescence.
The CO2-response of gas exchange and fluorescence
One leaf (per seedling) was carefully placed inside the fluorescence chamber and exposed to a saturating PPFD, chamber flow rate of 300 µmol s–1, and 400 µmol mol–1 CO2 (ambient O2) for at least 15 min or until rates of gas exchange and fluorescence were steady. The PPFD used was light saturating at all temperatures: 1500 µmol m–2 s–1 for 20–35 °C, 1000 µmol m–2 s–1 at 15 °C, and 750 µmol m–2 s–1 at 10 °C. Lower PPFD was used at the two lowest temperatures due to lower light-saturation points and concerns about photoinhibition. Once gas exchange and fluorescence were steady, a CO2-response curve was generated by increasing Ca, in eight steps, from 400 to 2000 µmol mol–1. At each ‘step’ gas exchange and fluorescence were allowed to stabilize for 5–7 min and then a saturating pulse was given and fluorescence and gas exchange date were recorded. Ca was subsequently decreased to 400 µmol mol–1 and the response curve was continued by decreasing Ca, in seven steps, to 50 µmol mol–1.
Measurement of Ci*, Rd, and curve-fit estimates of gi
Ci* and Rd were estimated using the Laisk method (Laisk, 1977) on one leaf from each of six seedlings at each of the six measurement temperatures. Measurements were made with the modified 2x3 cm broadleaf chamber at a flow rate of 250 µmol s–1. One leaf (per seedling) was carefully placed inside the chamber and exposed to a saturating PPFD (500 µmol m–2 s–1 from 15–35 °C, 200 µmol m–2 s–1 at 10 °C), chamber flow rate of 250 µmol s–1 and 400 µmol mol–1 CO2 (ambient O2) for at least 15 min or until rates of gas exchange and fluorescence were steady. Ci* and Rd were estimated from three partial CO2-response curves (40–125 µmol mol–1 CO2) measured at three PPFD. A PPFD of 75, 200, and 500 µmol m–2 s–1 was used from 15–35 °C, whereas, 50, 100, and 200 µmol m–2 s–1 was used at 10 °C. These light intensities were chosen following preliminary trials to ensure a large difference in slope of the three A–Ci curves. Each partial CO2-response curve comprised at least six points, although sometimes only four or five points were used in regressions because at the lowest PPFD there was sometimes curvature in the A–Ci relationship above 100 µmol mol–1 CO2. The intersection of the three lines identified Ci* (x-axis) and Rd (y-axis).
The same gas exchange system and six leaves per plant and temperature were used to determine gi via the curve-fitting method developed by Ethier and Livingston (2004). The Rubisco-limited portion of CO2-response curves (50–400 µmol mol–1 CO2) was determined at the highest PPFD (500 µmol m–2 s–1 from 15 to 35 °C, 200 µmol m–2 s–1 at 10 °C). Each curve comprised seven or eight points.
Calculation of gas exchange parameters
Data for both gas exchange systems were corrected for diffusion of CO2 into and out of the leaf chamber according to the manufacturer's advice (Anon., 2001). Diffusion leaks are proportional to the gradient in CO2 concentrations between the inside and the outside of the chamber and the flow rate of air through the chamber. This was accounted for by a diffusion coefficient which was determined by measuring the diffusion of CO2 into empty chambers (Ca=0 µmol mol–1) as a function of the flow rate of air through the chamber and the inside–outside gradient of CO2. The CO2 concentration of the sample cell was measured by the reference infrared gas analyser using the match valve, and the diffusion coefficient was used to recalculate the gas exchange data (Anon., 2001). As expected, the diffusion co-efficient for the modified 2x3 cm chamber was around one-third that of the smaller fluorescence chamber.
For measurements made under non-photorespiratory conditions (1% O2) it was necessary to recalculate all gas exchange data. The CO2 and H2O sensitivity of the LI-6400 is affected by O2 concentration of the analysis gas. The effect of O2 on CO2 sensitivity is small enough to be ignored, whereas the change in H2O sensitivity of 3% (between 1% and 21% O2) has a measurable effect on the estimation of stomatal conductance and Ci. Therefore, the measured H2O concentration was corrected using an empirical correction reported by Ghannoum et al. (1998) and all gas exchange parameters recalculated.
Calculation of gi with the variable J method
It was not possible to use the ‘constant J method’ because J (and A) decreased at elevated [CO2], especially at lower temperatures. Hence, gi was estimated with the ‘variable J method’ (Di Marco et al., 1990; Loreto et al., 1992) using A, Ci, and Jf measured at 400 µmol mol–1 CO2 and Ci* and Rd determined by the Laisk method (Laisk, 1977). Fluorescence estimates of electron transport (Jf) were converted into actual rates of electron transport (J) using the empirical relationship determined under non-photorespiratory conditions (Fig. 1).
 | (3) |
*) used. Therefore, gi was calculated with three different estimates of *: (i) the measured Ci* for Q. canariensis, which was in between the two published * responses (Figs 2, 3); (ii) * calculated as a simultaneous equation with gi (Ci*+Rd/gi=*); and (iii) the estimate of * by Bernacchi et al. (2002).
|
|
|
Using estimated gi and measured A and Ci, Cc was calculated as:
 | (4) |
Calculation of gi with the curve-fitting procedure
Ethier and Livingston (2004) introduced a new method for estimating gi that is based on the curvature of the Rubisco-limited portion of an A/Ci curve. The fundamental premise of the ‘curvature method’ is that gi reduces the curvature of the Rubisco-limited portion of an A/Ci response curve. Under Rubisco-limited conditions, the response of photosynthesis to CO2 is a function of the kinetic properties of Rubisco (Kc, Ko, *) and Cc:
 | (5) |
 |
 | (6) |
250 µmol mol–1). Values of Kc and Ko and their temperature responses were the Cc-based in vivo values of Bernacchi (2002) for tobacco, while fitted values of Ci* were used as a surrogate for *. Rd was not fixed but was estimated by the curve-fitting procedure. Equation 6 yielded estimates of Vcmax and gi, but Vcmax data are not reported since the PPFD used was lower than for the complete A/Ci curves.
Calculation of Vcmax and Jmax
Vcmax and Jmax were determined on a Cc basis from CO2-response data and gi determined with the combined gas exchange-fluorescence system. Data were fitted to the photosynthesis model of Farquhar et al. (1980), essentially as described previously (Warren et al., 2003a). Values of Kc and Ko and their temperature responses were the Cc-based in vivo values of Bernacchi (2002) for tobacco, while fitted values of Ci* were used (Fig. 2) as a surrogate for *. Maximum quantum yield was assumed to be 0.24 mol (electrons) mol–1 (photons) (Harley and Tenhunen, 1991).
Temperature response models of Ci*, gi, Vcmax, and Jmax
Temperature responses were fitted to gi, Vcmax, and Jmax determined with the combined gas exchange-fluorescence system. Temperature responses were also fitted to Ci* determined by the Laisk method (Laisk, 1977), but responses were not fitted to Rd because it was temperature insensitive (Fig. 2).
The temperature dependencies of Rd, Ci*, and Vcmax did not peak between 10 °C and 35 °C. They were described following the general equation (Sharpe and DeMichele, 1977; Leuning, 1997):
 | (7) |
 | (8) |
S (J K–1 mol–1) is an entropy term, Ed (J mol–1) is the deactivation energy, and P(Tref) is the potential value that the parameter would have at the temperature Tref in the absence of deactivation (Leuning, 1997; Wohlfahrt et al., 1999)
A simple derivation shows that the optimal temperature can be computed as:
 | (9) |
 | (10) |
Statistical treatment of temperature responses
To fit temperature responses, it was assumed that all replicates have similar temperature responses but different values at the reference temperature (25 °C) (Dreyer et al., 2001). Five dummy variables F1 to F5 were used for referencing each replicate and the following model was fitted to the temperature response of the parameter P:
 | (11) |
i is the difference between the value of the parameter at Tref for replicate i and P6(Tref); and u(T) corresponds to one of the three models (7, 8, or 10) described above.
In addition, S was replaced in u(T) by the expression derived from equation 9:
 | (12) |
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Relationship of fluorescence with electron transport
There was a strong positive relationship between fluorescence (Jf) and gas exchange (J) estimates of linear electron transport (Fig. 1; J=0.939Jf+4.06; r2=0.80). Temperature did not affect relationships (ANCOVA, P >0.05), as has been found previously (Genty et al., 1989). The single empirical relationship of J with Jf was therefore used in subsequent calculations of gi. The relationship was assumed to hold at lower temperatures (down to 10 °C).
Ci* and Rd
The temperature responses of the intercellular photocompensation point (Ci*) in Quercus canariensis was well described by the simple Arrhenius model (R2=0.8), with a (modelled) Ci* of 39 µmol mol–1 at 25 °C and an activation energy of 34 kJ mol–1 (Fig. 2; Table 1). The temperature response of Ci*, as quantified by Ea, was generally similar to those reported for other species, while Ci* at 25 °C was within 6 µmol mol–1 of other published values for Ci* (Fig. 2). Using measured Ci*, * was estimated as a simultaneous equation with gi (Fig. 3). The mean * at 25 °C was 45 µmol mol–1, which was 6 µmol mol–1 greater than Ci*. The estimates of * in the present study were very close to those for Spinacia oleracea from 20 to 35 °C, but at 10 and 15 °C they were somewhat higher (Fig. 3).
|
There was large variability among replicates in the rate of mitochondrial respiration in the light (Rd), and this to some extent obscured the temperature response (Fig. 4). Rd was
0.5 µmol m–2 s–1 between 10 and 20 °C, and then generally increased with increasing temperature.
|
gs, gi, Ci, and Cc
Stomatal conductance (gs) was unrelated to temperature and varied between 0.12 and 0.18 mol m–2 s–1 (data not shown). By contrast, internal conductance (gi) estimated by the variable J method more than doubled from 10 °C to 20 °C, with a plateau from 20 °C to 35 °C (Fig. 5). There was little difference between gi estimated with measured Ci* or calculated
*, with the latter producing, in general, slightly larger values at all temperatures. The same was generally true of gi estimated with the * of Bernacchi et al. (2002), except at 35 °C where there was a pronounced decrease in gi that was not evident with measured Ci* or calculated *.
|
These results with the variable J method were supported by estimates of gi with the curve-fitting method (Fig. 6). The two methods yielded similar temperature responses and similar means at the highest temperature (around 0.18 mol m–2 s–1). Estimates with the variable J method were less variable than with the curve-fitting method, and thus all subsequent analyses were based on gi estimated by the variable J method using measured Ci*. The temperature response of gi (variable J) was described well by a three-parameter log normal function (R2=0.92), with a maximum at 29 °C and mean of 0.19 mol m–2 s–1. The temperature response of gi in Q. canariensis was very different from that reported for Nicotiana tabacum (Bernacchi et al., 2002) (dashed curve in Fig. 6), but a similar plateau was observed from 28 °C to 38 °C in Eperua grandiflora (Pons and Welschen, 2003) (three crosses in Fig. 6).
|
The concentration of CO2 in the substomatal cavities decreased from 304 µmol mol–1 at 10 °C to 215–230 µmol mol–1 from 25 °C to 35 °C (Fig. 7). By contrast, the concentration of CO2 in the chloroplast (Cc) was generally unaffected by temperature and varied between 117 and 166 µmol mol–1. The draw-down from Ci to Cc was greatest at 10 °C (169 µmol mol–1) and decreased with increasing temperature to 64 µmol mol–1 at 35 °C.
|
Amax, Vcmax, and Jmax
The maximum rate of light-saturated net photosynthesis (Amax) peaked at 25 °C (16 µmol m–2 s–1) and declined at lower and higher temperatures (Fig. 8). Vcmax (on a Cc basis) increased from 10 µmol m–2 s–1 at 10 °C to 132 µmol m–2 s–1 at 35 °C and was described well by an Arrhenius equation (Ea=49.4 kJ mol–1; R2=0.86) (Fig. 8; Table 1). Jmax increased from 28 µmol m–2 s–1 at 10 °C to 314 µmol m–2 s–1 at 30 °C, before decreasing (slightly) to 305 µmol m–2 s–1 at 35 °C (Fig. 8). The slight decrease in Jmax at 35 °C was accounted for by including a de-activation term (Ed) in the model (Ea=100 kJ mol–1, Ed=176 kJ mol–1; R2=0.96).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Temperature response of gi and photosynthesis
The temperature response of Amax (Fig. 8) was rather similar to results for other species (Berry and Björkman, 1980; Cunningham and Read, 2002), and is at least partly a function of internal conductance to CO2 transfer (Fig. 6). Two independent estimates showed that the internal conductance of mature leaves of a deciduous oak, Q. canariensis, increases from 10 °C to 20 °C and is nearly temperature-independent from 20 °C to 35 °C. A similar temperature-independence was reported by Pons and Welschen (2003), who found that internal conductance of Eperua grandiflora did not vary systematically between 28 °C and 38 °C. However, the temperature response of gi in N. tabacum (Bernacchi et al., 2002) was very different. In N. tabacum, gi increased from 10 °C to a maximum between 35 °C and 37.5 °C, before declining at higher temperatures. That a decline in gi at high temperatures was not observed might be because the highest temperature (35 °C) was too low to cause reductions in gi. There is, however, no ready explanation for why gi was temperature-independent between 20 °C and 35 °C in Q. canariensis (this study) and from 28 °C and 38 °C in Eperua grandiflora (Pons and Welschen, 2003), but increased steadily from 10 °C to 35 °C in N. tabacum (Bernacchi et al., 2002). Additional data from other species are required before a generalization can be made about the temperature response of gi.
Estimates of internal conductance by the variable J method are highly sensitive to the chloroplastic photocompensation point * (Harley et al., 1992) but, in the case of Q. canariensis, the temperature response of gi was affected little by three different estimates of * (Fig. 5). gi was around 0.025 mol m–2 s–1 greater when estimated with calculated * than with measured Ci*, but the shape of the temperature response did not differ—a not altogether unexpected result given the similarity of the Ci* and * temperature responses. The * of Bernacchi et al. (2002) was less temperature sensitive than either measured Ci* or calculated * (Fig. 3), and thus gi estimated with Bernacchi et al.'s * decreased at 35 °C, whereas it did not with the other two estimates. It is worthwhile noting that using the * estimates of (Bernacchi et al., 2002) did not result in a temperature response of gi like that observed for N. tabacum. Hence, it is concluded that the temperature response of gi in Q. canariensis differs from that of N. tabacum, and this is not due to different estimates of *.
Adaptation and acclimation affect the temperature response of photosynthetic parameters (Berry and Björkman, 1980; Bernacchi et al., 2003), but it is unlikely that the different temperature responses of gi can be attributed to adaptation or acclimation. The differences in shape of gi temperature responses are far larger than one would expect based on temperature responses for other parameters, which generally have very similar shapes. Temperature responses of Amax, for example, generally have a similar shape despite large differences in the optimal temperature (Mooney, 1986).
The temperature response of gi may provide clues as to its nature. The 3-fold increase in gi from 10 °C to 20 °C is inconsistent with simple diffusion in water, which increases by around 25% over the same range in temperatures (Tamimi et al., 1994). This was also noted by Bernacchi et al. (2002) who argued that gi must be a protein-facilitated process. This argument is defensible in the case of N. tabacum, which had the expected temperature response for a protein-facilitated process. Other support for a dominant role of proteins comes from earlier studies showing that gi is related, at least in part, to carbonic anhydrase activity (Makino et al., 1992; Price et al., 1994) and aquaporin content (Uehlein et al., 2003; Hanba et al., 2004). However, the relative constancy of gi from 20 °C to 35 °C in Q. canariensis and from 28 °C and 38 °C in Eperua grandiflora (Pons and Welschen, 2003) argues against gi being determined by a simple protein-facilitated process. Instead it seems more likely that gi is determined by multiple processes with different temperature sensitivities which results in a complex temperature response. This indirect evidence as to the nature of gi needs to be supported by direct experimental evidence, and this should be a priority for future research.
Internal conductance affects not only absolute values of Vcmax and Jmax (Warren et al., 2003b), but also their temperature response. A recent paper modelled the temperature response of gi and examined its effect on the temperature response of Vcmax and Jmax (Juurola et al., 2005); however, the present study is the first to determine experimentally the temperature responses of Vcmax and Jmax on a Cc basis. Cc-based Vcmax and Jmax increased 10-fold from 10 °C to 30 °C, which is significantly greater than the 6-fold increase in Ci-based Vcmax and Jmax observed in Q. canariensis (data not shown) and many other species (Dreyer et al., 2001; Medlyn et al., 2002).
The temperature response of internal conductance further complicates conversion from Ci-based to Cc-based Vcmax and Jmax (or Ea). Choice (and interpretation) of kinetic constants (Kc, Ko, *) for Cc-based Vcmax and Jmax is comparatively straightforward—one may use either in vitro values (Jordan and Ogren, 1984), or in vivo values determined on a Cc basis (Bernacchi et al., 2002). For Ci-based Vcmax and Jmax the appropriate kinetic constants are a function of the relationship between internal conductance and photosynthetic capacity (Vcmax and Jmax). This problem has been discussed at length (Bernacchi et al., 2002; Ethier and Livingston, 2004), but is perhaps best understood by using a simple example. The widely used Ci-based kinetic constants (apparent Kc, apparent Ko, Ci*) of Bernacchi et al. (2001) are only appropriate for plants that have the same relationship between internal conductance and photosynthetic capacity (Vcmax, Jmax), and the same temperature response of internal conductance. Internal conductance relative to Vcmax varies more than 2-fold among species (Ethier and Livingston, 2004; Warren and Adams, 2006), and thus the kinetic constants of Bernacchi et al. (2001) are not universally applicable. An additional caveat in using Ci-based kinetic constants (at temperatures other than 25 °C) is that the temperature response of internal conductance must be the same otherwise there will be systematic temperature-related errors in apparent Kc, apparent Ko, and Ci*. Evidence from three species (Q. canariensis, this study; Eperua grandiflora, Pons and Welschen 2003; N. tabacum, Bernacchi et al., 2002) suggests that the temperature response of internal conductance does vary among species, which makes Ci-based kinetic constants even more highly species-specific.
Methodological considerations
Estimating internal conductance is not easy and this is perhaps borne out by the poor precision of the present estimates with relative standard deviations (standard deviation/mean) between 13% and 30%. Nonetheless, the variability in the present study is broadly similar to that reported in other studies of gi using chlorophyll fluorescence (e.g. 19–22% RSD, Harley et al., 1992; 34–38% RSD, Epron et al., 1995) and isotopic methods (6–13% RSD, Lloyd et al., 1992; 11–50% RSD, Loreto et al., 1992; 14% RSD, Warren et al., 2003b).
Precision of estimates may well have been poor, but two independent methods yielded similar results, lending considerable weight to the present data (Fig. 6). The variable J method and the curve-fitting method were used, but these are not totally independent because they share a few common assumptions, namely the accuracy of A and Ci. It was not possible to use the constant J method because J was not constant but decreased at elevated CO2. In any case, the two fluorescence methods share many common assumptions and using both methods adds little credibility.
Neglecting cuticular conductance may have led to an overestimate of Ci (Boyer et al., 1997) and an underestimate of gi (Warren et al., 2004), but the effect is rather small. Unfortunately, there are no estimates of cuticular conductance for Q. canariensis, but it is likely to be <0.01 mol m–2 s–1 (Tausz et al., 2005). Inclusion of such a cuticular conductance decreases estimates of Ci by <4 µmol mol–1 and increases internal conductance by <0.002 mol m–2 s–1 (data not shown). Hence the effect of cuticular conductance on estimates of internal conductance is negligible, which was not entirely unexpected given that stomatal conductance varied between 0.13 and 0.17 mol m–2 s–1 and thus was more than an order of magnitude greater than cuticular conductance.
|
Acknowledgements |
|---|
Charles Warren acknowledges the financial support of the Australian Research Council in the form of a Discovery Grant and APD fellowship, and of INRA in the form of a short visit grant. Both authors gratefully acknowledge additional financial support in the form of a linkage-International award from the Australian Research Council. We appreciate the constructive comments of Gilbert Ethier on a previous draft of this manuscript, and the help of Pierre Montpied with statistical analyses. The trees were grown by Jean Marie Gioria at INRA Nancy.
|
Footnotes |
|---|
Present address: School of Biological Sciences, University of Sydney, NSW 2006, Australia 
|
Abbreviations |
|---|
A, rate of net photosynthesis; Ca, ambient CO2 concentration; Ci, intercellular CO2 concentration; Ci*, intercellular CO2 concentration at which the chloroplastic CO2 concentration=
*; Cc, chloroplastic CO2 concentration; *, CO2 compensation concentration in the absence of mitochondrial respiration (photocompensation point); Jmax, RuBP limited rate of electron transport; Kc and Ko, Michaelis–Menten constants for RuBP carboxylation and oxygenation, respectively; PPFD, photosynthetic photon flux density; PSII, photosystem II; Rd, mitochondrial respiration in the light (day respiration); gi, internal conductance; gs, stomatal conductance; O, O2 concentration in chloroplasts; Vcmax, maximum rate of carboxylation. |
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Aalto T and Juurola E. (2002) A three-dimensional model of CO2 transport in airspaces and mesophyll cells of a silver birch leaf. Plant, Cell and Environment 25:1399–1409.[CrossRef]
Anon E. (2001) Using the LI-6400 portable photosynthesis system (Li-Cor Biosciences Inc, Lincoln, NE, USA).
Atkin OK, Evans JR, Ball MC, Lambers H, Pons TL. (2000) Leaf respiration of snow gum in the light and dark: interactions between temperature and irradiance. Plant Physiology 122:915–923.
Bernacchi CJ, Pimentel C, Long SP. (2003) In vivo temperature response functions of parameters required to model RuBP-limited photosynthesis. Plant, Cell and Environment 26:1419–1430.[CrossRef]
Bernacchi CJ, Portis AR, Nakano H, von Caemmerer S, Long SP. (2002) Temperature response of mesophyll conductance: implications for the determination of Rubisco enzyme kinetics and for limitations to photosynthesis in vivo. Plant Physiology 130:1992–1998.
Bernacchi CJ, Singsaas EL, Pimentel C, Portis AR, Long SP. (2001) Improved temperature response functions for models of Rubisco-limited photosynthesis. Plant, Cell and Environment 24:253–259.[CrossRef]
Berry J and Björkman O. (1980) Photosynthetic response and adaptation to temperature in higher plants. Annual Review of Plant Physiology 31:491–543.[ISI]
Boyer JS, Wong SC, Farquhar GD. (1997) CO2 and water vapor exchange across leaf cuticle (epidermis) at various water potentials. Plant Physiology 114:185–191.[Abstract]
Brooks A and Farquhar GD. (1985) Effect of temperature on the CO2/O2 specificity of ribulose-1,5-bisphosphate carboxylase oxygenase and the rate of respiration in the light: estimates from gas-exchange measurements on spinach. Planta 165:397–406.[CrossRef][ISI]
Cunningham SC and Read J. (2002) Comparison of temperate and tropical rainforest tree species: photosynthetic responses to growth temperature. Oecologia 133:112–119.[CrossRef]
Di Marco G, Manes F, Tricoli D, Vitale E. (1990) Fluorescence parameters measured concurrently with net photosynthesis to investigate chloroplastic CO2 concentration in leaves of Quercus ilex L. Journal of Plant Physiology 136:538–543.
Dreyer E, Le Roux X, Montpied P, Daudet FA, Masson F. (2001) Temperature response of leaf photosynthetic capacity in seedlings from seven temperate tree species. Tree Physiology 21:223–232.[ISI][Medline]
Epron D, Godard D, Cornic G, Genty B. (1995) Limitation of net CO2 assimilation rate by internal resistances to CO2 transfer in the leaves of 2 tree species (Fagus sylvatica L and Castanea sativa Mill). Plant, Cell and Environment 18:43–51.[Medline]
Ethier GJ and Livingston NJ. (2004) On the need to incorporate sensitivity to CO2 transfer conductance into the Farquhar-von Caemmerer-Berry leaf photosynthesis model. Plant, Cell and Environment 27:137–153.[CrossRef]
Evans JR, Sharkey TD, Berry JA, Farquhar GD. (1986) Carbon isotope discrimination measured concurrently with gas-exchange to investigate CO2 diffusion in leaves of higher-plants. Australian Journal of Plant Physiology 13:281–292.
Evans JR, von Caemmerer S, Setchell BA, Hudson GS. (1994) The relationship between CO2 transfer conductance and leaf anatomy in transgenic tobacco with a reduced content of Rubisco. Australian Journal of Plant Physiology 21:475–495.
Farquhar GD and von Caemmerer S, et al. (1980) A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta 149:78–90.[CrossRef][ISI]
Gaastra P. (1959) Photosynthesis of crop plants as influenced by light, carbon dioxide, temperature and stomatal diffusion resistance. Mededelingen van de Landbouwhogeschool Wageningen-Nederland 59:1–68.
Genty B, Briantais JM, Baker NR. (1989) The relationship between the quantum yield of photosynthetic electron-transport and quenching of chlorophyll fluorescence. Biochimica et Biophysica Acta 990:87–92.
Ghannoum O, Siebke K, von Caemmerer S, Conroy JP. (1998) The photosynthesis of young Panicum C-4 leaves is not C-3-like. Plant, Cell and Environment 21:1123–1131.[CrossRef]
Hanba YT, Shibasaka M, Hayashi Y, Hayakawa T, Kasamo K, Terashima I, Katsuhara M. (2004) Overexpression of the barley aquaporin HvPIP2;1 increases internal CO2 conductance and CO2 assimilation in the leaves of transgenic rice plants. Plant and Cell Physiology 45:521–529.
Harley PC, Loreto F, Dimarco G, Sharkey TD. (1992) Theoretical considerations when estimating the mesophyll conductance to CO2 flux by analysis of the response of photosynthesis to CO2. Plant Physiology 98:1429–1436.
Harley PC and Tenhunen JD. (1991) Modeling the photosynthetic response of C3 leaves to environmental factors. Modeling crop photosynthesis – from biochemistry to canopy (American Society of Agronomy and Crop Science Society of America, Madison, WI) pp. 17–39.
Johnson F, Eyring H, Williams R. (1942) The nature of enzyme inhibitions in bacterial luminescence: sulfanilamide, urethane, temperature and pressure. Journal of Cell Comparative Physiology 20:247–268.[CrossRef]
Jordan DB and Ogren WL. (1984) The CO2/O2 specificity of ribulose 1,5-bisphosphate carboxylase oxygenase: dependence on ribulosecbisphosphate concentration, pH and temperature. Planta 161:308–313.[CrossRef][ISI]
Juurola E, Aalto T, Thum T, Vesala T, Hari P. (2005) Temperature dependence of leaf-level CO2 fixation: revising biochemical coefficients through analysis of leaf three-dimensional structure. New Phytologist 166:205–215.[CrossRef][ISI][Medline]
Kogami H, Hanba YT, Kibe T, Terashima I, Masuzawa T. (2001) CO2 transfer conductance, leaf structure and carbon isotope composition of Polygonum cuspidatum leaves from low and high altitudes. Plant, Cell and Environment 24:529–538.[CrossRef]
Laisk AK. (1977) Kinetics of photosynthesis and photorespiration in C3-plants (Nauka Publishing, Moscow).
Leuning R. (1997) Scaling to a common temperature improves the correlation between the photosynthesis parameters J(max) and V-cmax. Journal of Experimental Botany 48:345–347.[ISI]
Lloyd J, Syvertsen JP, Kriedemann PE, Farquhar GD. (1992) Low conductances for CO2 diffusion from stomata to the sites of carboxylation in leaves of woody species. Plant, Cell and Environment 15:873–899.[CrossRef]
Loreto F, Harley PC, Dimarco G, Sharkey TD. (1992) Estimation of mesophyll conductance to CO2 flux by 3 different methods. Plant Physiology 98:1437–1443.
Makino A, Sakashita H, Hidema J, Mae T, Ojima K, Osmond B. (1992) Distinctive responses of ribulose-1,5-bisphosphate carboxylase and carbonic-anhydrase in wheat leaves to nitrogen nutrition and their possible relationships to CO2-transfer resistance. Plant Physiology 100:1737–1743.
Medlyn BE, Dreyer E, Ellsworth D, et al. (2002) Temperature response of parameters of a biochemically based model of photosynthesis. II. A review of experimental data. Plant, Cell and Environment 25:1167–1179.[CrossRef]
Mooney H. (1986) Photosynthesis. In Crawley MJ (Ed.). Plant ecology (Blackwell, Oxford) pp. 345–373.