Spinach (Spinacia oleracea) plants were grown under the day/night temperature regime of 15/10 °C (LT) or 30/25 °C (HT). The plants were also transferred from HT to LT when the sample leaves were at particular developmental stages (HL-transfer). With fully mature leaves, the light-saturated photosynthetic rate (A) at the ambient CO2 concentration (Ca) of 1500 µL L−1 (A1500) and the initial slope of A versus intercellular CO2 concentration (Ci) at low Ci region (IS) were obtained to assess capacities of RuBP regeneration and carboxylation. Photosynthetic components including Rubisco and cytochrome f (Cyt f) were also determined. The optimum temperatures for A at Ca of 360 µL L−1 (A360), A1500 and IS in HT leaves were 27, 36 and 24 °C, whereas those in LT leaves were 18, 30 and 18 °C. The optimum temperatures in HL-transfer leaves approached those of LT leaves with the increase in the duration at LT. The shift in the optimum temperature was greater and quicker for IS than A1500. By the HL-transfer, the maximum values of A1500 and IS also increased. The maximum A1500 and Cyt f content increased more promptly than IS and Rubisco content. Changes in the Cyt f/Rubisco ratio were reflected to those in the A1500/IS ratio. Taken together, photosynthetic acclimation to low temperature in spinach leaves was due not only to the change in the balance of the absolute rates of RuBP regeneration and carboxylation but also to the large change in the optimum temperature of RuBP carboxylation.
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.
Temperature influences leaf photosynthetic rates in both short and long-terms. In the long term, a shift in growth temperature causes a shift in the optimum temperature of leaf photosynthesis, which allows the plants to perform more efficient photosynthesis at their new growth temperature (Berry & Björkman 1980).
The shift of the optimum temperature for photosynthesis has been analysed based on the model of C3 photosynthesis (Farquhar, von Caemmerer & Berry 1980). This model assumes that a photosynthetic rate is equal to the minimum rate of two partial reactions, RuBP regeneration and RuBP carboxylation. If temperature dependencies of these partial reactions are different, the temperature dependence of photosynthetic rate will be delimited by the lower one of these temperature dependencies and the optimum temperature occurs at the node of these two curves. Farquhar & von Caemmerer (1982) explained that the shift of the optimum temperature for photosynthesis is a result of changes in balance between RuBP regeneration and carboxylation. If the maximum rate of the one partial reaction with the lower optimum temperature increases relative to that of the other partial reaction in response to a shift of growth temperature, the optimum temperature of photosynthesis (the node of the two curves) will shift towards the higher temperature. The optimum temperature can shift in this way even when the optimum temperatures for these two partial reactions do not change.
Shifts of the optimum temperatures of these photosynthetic partial reactions with the changes in growth temperature have been studied in vivo in Quercus myrsinaefolia (Hikosaka et al. 1999), eight herbaceous C3 species (Bunce 2000), Nerium oleander (Hikosaka & Hirose 2001), seven tree species (Dreyer et al. 2001) and Pinus pinaster (Medlyn, Loustau & Delzon 2002). These studies were all based on the gas exchange data and the above-mentioned model of C3 photosynthesis, and largely descriptive. Since there are few studies in which biochemical analyses were made concomitant with gas exchange analyses, molecular mechanisms underlying the shift of the temperature dependence of photosynthesis are still unclear.
In the present study, we focused on acclimation of photosynthesis to low growth temperatures. Capacities of both RuBP regeneration and carboxylation, measured at optimum temperatures, generally increase when plants are grown at low temperatures. Plants grown at low temperatures have higher amounts of enzymes of photosynthesis, such as enzymes of the photosynthetic carbon reduction cycle, including Rubisco, stromal fructose-1,6-bisphosphatese and sedoheptulose-1,7-bisphosphatase (Holaday et al. 1992; Hurry et al. 1994, 1995; Strand et al. 1997, 1999) and those of sucrose synthesis, including cytosolic fructose-1,6-bisphosphatase and sucrose phosphate synthase (Guy, Huber & Huber 1992; Holaday et al. 1992; Hurry et al. 1994, 1995; Strand et al. 1997, 1999). The large amounts of these enzymes would be needed to compensate for decreased activities of these enzymes at low temperatures. Also, when measured at their optimum temperatures, the maximum electron transport rate in the thylakoids isolated from low temperature-grown plants is generally higher, on the chlorophyll (Chl) basis, than that in thylakoids from high temperature-grown ones (Badger et al. 1982; Mitchel & Barber 1986). Therefore, we need to investigate biochemical changes, in both quantitative and qualitative senses, and examine whether such changes agree with the changes of RuBP regeneration and carboxylation capacities estimated by the gas exchange techniques.
The aim of our studies was to conduct a series of experiments encompassing from ecophysiology to biochemistry to clarify molecular mechanisms of the shift of the temperature dependence for photosynthesis. We selected spinach plants, because their leaves are suitable for biochemical analyses. In this study, we grew plants at 15/10 or 30/25 °C (day/night temperature). We also transferred some plants that had been grown at 30/25 to 15/10 °C to follow sequential changes in photosynthetic properties of the leaves. We not only conducted gas exchange measurements, but also determined various photosynthetic components. Based on these results, we addressed several key questions: (1) do the temperature dependencies of RuBP regeneration and carboxylation capacities shift towards lower temperatures in response to the downward shift of growth temperature? In particular, we paid attention to the temperature shift of the capacity for RuBP carboxylation, because the temperature dependence of RuBP carboxylation is thought to be stable. (2) Does the balance between the RuBP regeneration capacity and the RuBP carboxylation capacity change with the shift of growth temperature? If the change is the case, biochemical changes that are responsible for the change of balance should be sought. (3) To what extents do such changes in the temperature dependencies of the capacities for partial reactions and those in the balance contribute to the shift of the optimum temperature for photosynthesis?
MATERIALS AND METHODS
Plant materials and growth conditions
Spinach (Spinacia oleracea L. cv. Torai) plants were grown in vermiculite in 1.3 L plastic pots (one plant per pot) in growth chambers (KG-50HLA-S; Koito, Osaka, Japan). Day and night lengths were 8 and 16 h, respectively. Photosynthetically active photon flux density (PPFD) in the day time was 230 µmol m−2 s−1. The day/night air temperatures were either 30/25 °C or 15/10 °C. These are referred to as high temperature (HT) and low temperature (LT) conditions, respectively. The plants/leaves grown at HT (LT) are called HT (LT)-plants/leaves. The seventh true leaves were used for all the measurements. It took about 1.5 and 4 months for the seventh leaves to fully expand in HT and LT plants, respectively. Some plants grown at HT were transferred to the LT condition when the seventh true leaves were at either of the particular developmental stages. The plants/leaves experiencing the transfer were called HL-plants/leaves. HL-mature plants were transferred to LT when their seventh leaves were just fully expanded; they were kept at LT for another 2 weeks. HL-young plants were transferred to LT when their seventh leaves were about one-fifth the area of the fully expanded leaves; they were kept for about 1 month at LT until full maturation of their seventh leaves. HL-new plants were transferred to LT when their seventh leaves were about to emerge; they were kept at LT for about 2 months. The plants were watered once a week and fertilized with 200 mL of a nutrient solution containing 2 mm KNO3, 2 mm Ca(NO3)2, 0.75 mm MgSO4, 0.665 mm NaH2PO4, 25 µm Fe-EDTA, 5 µm MnSO4, 0.5 µm ZnSO4, 0.5 µm CuSO4, 25 µm H3BO4, 0.25 µm Na2MoO4, 50 µm NaCl, and 0.1 µm CoSO4 once a week.
Gas exchange measurements
CO2 gas exchange of leaves was measured with a portable gas exchange system (LI-6400; Li-Cor Inc., Lincoln, NE, USA). The whole portable gas exchange system was enclosed in a 220 L temperature-controlled chamber (System Biotron; Nippon Medical & Chemical Instruments, Tokyo, Japan). When CO2 concentration of air entering the cuvette of the gas exchange system was markedly different from that of the ambient air, leakage of CO2 through the slits between the sealing pads and the leaf was not negligible. To minimize such a CO2 leak, we routinely used a laboratory-made skirt (Miyazawa & Terashima 2001) with which the air once exhausted from the cuvette was again blown to the slits from the outside.
With one leaf, we measured rates of dark respiration and photosynthesis at every 3 °C from 9 to 39 °C. The leaf respiration rate was determined at an ambient CO2 concentration of 360 µL L−1. We measured the rate of dark respiration after a sufficient dark period (approximately 9 h), because the rate of dark respiration of spinach leaves was enhanced by accumulation of photosynthates (Noguchi, Sonoike & Terashima 1996). Subsequently, rates of photosynthesis were measured at saturating light of 1500 µmol m−2 s−1. The CO2 dependence of photosynthesis was examined based on the measurements at CO2 concentrations in the ambient air (Ca) of 50, 100, 150, 360 and 1500 µL L−1. At each CO2 concentration, net photosynthetic rate (A) and intercellular CO2 concentration (Ci) were calculated. RuBP regeneration rate was estimated from A at the highest Ca of 1500 µL L−1 CO2 (A1500). RuBP carboxylation rate was estimated from the initial slope of the A versus Ci curve (IS) obtained with the data measured at Ca of 50, 100 and 150 µL L−1. Maximum rates of the photosynthetic electron transport (Jmax) and RuBP carboxylation (Vcmax) were calculated from A1500 and IS, respectively (see Appendix for mathematical equations).
Variation in the vapour pressure deficit (VPD) was minimized by using a dew point generator (Li-610; Li-Cor, Lincoln, NE, USA). Water condensation within the gas exchange system was avoided by regulating the temperature in the 220 L chamber enclosing the gas-exchange system. The VPD was maintained between 0.5 and 2.5 kPa to prevent stomatal closure. Although the VPD was sometimes above 3.0 kPa at the highest leaf temperature of 39 °C, the stomatal conductance was no less than 0.35 mol H2O m−2 s−1.
Determinations of Chl, Rubisco, cytochrome f, carbon and nitrogen
Immediately after the measurements of gas exchange, discs of 0.8 cm diameter (approximately. 0.5 cm2) were taken from each of the seventh leaves with major veins being avoided. The leaf discs were immersed in liquid nitrogen and stored at −80 °C until determinations of Chl, Rubisco and cytochrome f (Cyt f). Some leaf discs were dried at 70 °C for more than 3 d for measurements of dry weight, carbon and nitrogen contents.
The frozen leaf sample was ground in liquid nitrogen and homogenized in an extraction buffer containing 100 mm sodium-phosphate buffer (pH 7.0), 1.0% (w/v) polyvinylpyrrolidone, 0.1% (v/v) Triton X-100, 1 mm phenylmethyl sulfonyl fluoride, and 1.0%β-mercaptoethanol. Chl was extracted with 80% (v/v) acetone and determined by the procedure of Porra, Thompson & Kriedemann (1989). The rest of the homogenate was used for quantification of Rubisco. The Rubisco content was measured according to Makino, Mae & Ohira (1986) with slight modifications. The extract was centrifuged (10 000 × g, 15 min) and the supernatant was used for the determination of Rubisco. A portion of the supernatant was mixed with 2 × Laemmli buffer (Laemmli 1970) containing 0.5 m Tris-HCl (pH 6.8), 10% sodium dodecyl sulfate, 10%β-mercaptoethanol, 10% glycerol and 0.1% bromophenol blue, and kept at room temperature for 1 h. Rubisco was separated by polyacrylamide gel electrophoresis with 10% resolving gel and 4.75% stacking gel (Laemmli 1970). The gel was stained with Coomassie Brilliant Blue R-250. The amount of Rubisco large subunit was determined by scanning the gel at 560 nm using a gel-densitometer (FD-A-V; Fujiox, Tokyo, Japan).
Cyt f content was estimated from the hydroquinone-reduced, ferricyanide-oxidized difference spectrum of the thylakoid membranes according to Bendall, Davenport & Hill (1971). The difference spectrum was recorded with a spectrophotometer (U-3310; Hitachi, Tokyo, Japan). The absorbance at 554 nm above the line that was drawn between the absorbances of 540 and 570 nm was measured. The millimolar extinction coefficient of 20 mm−1 cm−1 was used, according to Evans & Terashima (1987).
Leaf carbon and nitrogen contents were measured with an NC analyser (CHNOS Elemental analyser, Vario EL III; Elementar, Hanau, Germany).
Data are presented as means ± SE. Scheffe's multiple comparison test was performed using a statistics software (Statview ver. 4.58; SAS Institute Inc., Cary, North Carolina, USA). Significant differences between regression lines were detected with ancova according to Sokal & Rohlf (1995).
Table 1 shows effects of growth temperature and the transfers on leaf characteristics and the amounts of photosynthetic components. With the increase in the growth period at LT, leaf area and leaf mass per area (LMA) increased, whereas water content gradually decreased. Although nitrogen per leaf area increased with the increase in the LT period, the greater increase in carbon content resulted in the increase in C/N ratio.
Table 1. Effect of growth temperature on leaf properties
LMA, leaf mass per area; C/N, the carbon to nitrogen ratio; Rubisco/N, Rubisco content per nitrogen content; Cyt f/N, cytochrome f per nitrogen content; Chl/N, chlorophyll content per nitrogen content; Rubisco/Chl, Rubisco content per chlorophyll content; Cyt f/Chl, cytochrome f per chlorophyll content; Chl a/b, chlorophyll a/b ratio and Cyt f/Rubisco, cytochrome f per Rubisco content. Water content was determined from the ratio of fresh weight minus dry weight relative to fresh weight. Data represent means ± SE, n = 5. Different alphabets indicate statistically significant differences (Scheffe's multiple comparison test, P < 0.05).
Leaf area (cm2)
22.1 ± 0.9a
22.9 ± 0.7a
25.5 ± 0.9a
37.1 ± 0.8b
40.7 ± 1.8b
LMA (g m−2)
21.6 ± 0.8a
44.8 ± 0.4b
58.9 ± 1.1c
72.2 ± 0.7d
79.8 ± 0.9e
Water content (%)
90.0 ± 0.2a
86.2 ± 0.2b
82.7 ± 0.3c
80.3 ± 0.2d
80.2 ± 0.6d
Carbon (mol m−2)
0.65 ± 0.03a
1.40 ± 0.03b
1.85 ± 0.08c
2.47 ± 0.05d
2.73 ± 0.02e
Nitrogen (mmol m−2)
81.1 ± 4.2a
103.8 ± 3.0b
129.0 ± 2.0c
144.8 ± 2.3d
126.9 ± 1.0c
C/N (mol mol−1)
8.0 ± 0.2a
13.4 ± 0.7b
14.1 ± 1.0b
17.1 ± 1.3c
21.5 ± 0.5d
Rubisco (µmol m−2)
1.88 ± 0.13a
2.99 ± 0.13b
3.48 ± 0.15bc
3.78 ± 0.10c
3.64 ± 0.06c
Cyt f (µmol m−2)
0.47 ± 0.02a
0.89 ± 0.03b
0.98 ± 0.02bc
1.10 ± 0.03c
1.09 ± 0.05c
Chl (mmol m−2)
0.38 ± 0.01a
0.45 ± 0.01b
0.49 ± 0.02b
0.56 ± 0.01c
0.51 ± 0.01b
Rubisco/N (mmol mol−1)
23.3 ± 1.0a
29.1 ± 1.2b
27.3 ± 1.0a
26.5 ± 0.7a
28.9 ± 0.8b
Cyt f/N (mmol mol−1)
5.89 ± 0.15a
8.58 ± 0.25b
7.63 ± 0.14b
7.58 ± 0.20b
8.76 ± 0.39b
Chl/N (mmol mol−1)
4.87 ± 0.11a
4.54 ± 0.16ac
3.73 ± 0.14b
3.86 ± 0.12b
3.98 ± 0.12bc
Rubisco/Chl (mmol mol−1)
4.97 ± 0.19a
5.81 ± 0.20a
7.47 ± 0.28b
6.83 ± 0.17b
7.18 ± 0.19b
Cyt f/Chl (mmol mol−1)
1.25 ± 0.03a
1.98 ± 0.06b
2.01 ± 0.04b
1.96 ± 0.05b
2.13 ± 0.10b
Chl a/b (mol mol−1)
3.23 ± 0.05a
3.52 ± 0.05b
3.55 ± 0.02b
3.76 ± 0.05c
3.85 ± 0.01c
Cyt f/Rubisco (mol mol−1)
0.25 ± 0.01a
0.30 ± 0.01b
0.28 ± 0.01a
0.29 ± 0.01b
0.30 ± 0.01b
Amounts of photosynthetic components showed a trend similar to that of nitrogen content. LT and HL (HL-new, HL-young and HL-mature) leaves contained approximately 1.6- to 2.0-fold more Rubisco and 1.9- to 2.3-fold Cyt f than HT leaves. Chl content increased less markedly. LT and HL leaves had approximately 1.2- to 1.5-fold more Chl than HT leaves. In HL leaves, increase in Cyt f occurred faster than that in Rubisco.
Rubisco/N and Cyt f/N in HT leaves were lower than those in other leaves. On the other hand, Chl/N was slightly greater in HT and HL-mature leaves than in other leaves. Rubisco/Chl and Cyt f/Chl in HT leaves were lower than those in other leaves. Chl a/b increased with the increase in the period at LT during leaf development. All of these results indicate that LT and HL leaves showed more sun type characteristics than HT leaves. Figure 1 shows the sample plants. It is obvious that the leaves became stiffer and larger, and the petioles shorter with the increase in the period at LT during leaf development. Sun-type characteristics of leaves in plants grown at low temperature have been reported (for a review, see, Huner, Öquist & Sarhan 1998). In this study, we confirmed this in spinach leaves.
Temperature dependencies of the dark respiration rates per unit leaf area are shown in Fig. 2. The dark respiration rate increased with the increase in the measurement temperature. The dark respiration rates of LT and HL leaves were similar, and greater than that of HT at any temperature.
Light-saturated photosynthesis at Ca of 360 µL L−1
Figure 3a shows temperature dependencies of the light-saturated rates of net photosynthesis at Ca of 360 µL L−1 (A360). The rates of net photosynthesis relative to those measured at 21 °C are shown in Fig. 3b. The optimum temperatures of photosynthesis were 27, 24, 21, 21 and 18 °C for HT, HL-mature, HL-young, HL-new and LT leaves, respectively. The temperature dependencies of photosynthesis differed depending on the growth temperature and on duration at LT during leaf development.
RuBP regeneration and RuBP carboxylation
Temperature dependencies of A at the highest ambient CO2 concentration of 1500 µL L−1 (A1500) and those of the initial slopes of the A–Ci curves (IS) are shown in Figs 4 and 5. Figures 4a and 5a show temperature dependencies of A1500 and IS, respectively. The values relative to the rates at 21 °C are shown in Figs 4b and 5b.
For A1500, LT leaves showed the optimum temperature at 30 °C, while that of HT leaves was at 36 °C (Fig. 4). In HL-mature leaves, the optimum temperature did not shift lower and was the same as that of HT leaves, while the absolute value of A1500 markedly increased. In HL-new and HL-young leaves, the optimum temperatures shifted to 33 °C which were still higher than that of LT leaves. The capacities of RuBP regeneration of LT and HL leaves, judging from the absolute values of A1500, were greater than that of HT leaves at any temperature.
For IS, LT leaves showed the optimum temperature at 18 °C, whereas HT leaves did so at 24 °C (Fig. 5). In all the HL leaves, the optimum temperatures shifted downwards to 18 °C (Fig. 5). These HL leaves showed apparent shoulders in the curves at around 24–27 °C, near the optimum temperature for HT leaves.
Relationships between values of A1500 (Jmax) and IS (Vcmax)
The temperature dependencies of the ratio of A1500 to IS (A1500/IS) were shown in Fig. 6. In all the leaves, A1500/IS values showed little change at the measuring temperatures below 18 °C. Above 18 °C, the ratio increased with the increase in the measurement temperature. The temperature dependencies of A1500/IS were strongly affected by the growth temperature. LT and HT leaves showed the highest and lowest A1500/IS values, respectively, at most of the measurement temperatures. All the HL-leaves showed similar values. Below 18 °C, their values were similar to those of HT leaves, whereas, above 18 °C, their values lay between those of HT and LT leaves. Interestingly, the overshoots of A1500/IS ratios in HL-mature leaves were observed at 39 °C, the highest temperature.
Using the equations in Appendix, we calculated Jmax and Vcmax(Fig. 7). We used the same kinetic parameters of Rubisco, irrespective of the sample leaves. Day respiration rate (Rd) was assumed to be half the dark respiration rate. There were linear relationships between Jmax and Vcmax. Also noted were the positive intercepts on the Jmax axis. The positive intercept causes the marked decrease in Jmax/Vcmax with the increase in Vcmax. For HT leaves, Jmax/Vcmax ratios were 3.62, 1.65 and 1.75 at 9, 24 and 39 °C, respectively. For LT leaves, Jmax/Vcmax ratios were 4.74, 2.13 and 2.32 at 9, 24 and 39 °C, respectively. Relationships between Jmax and Vcmax were different between HT and LT (the slope of the regression line of LT statistically differed from that of HT, ancova, P = 0.00056). All the HL leaves showed similar slopes between those of HT and LT leaves, except for HL-mature leaves at the highest temperatures, which showed Jmax/Vcmax values comparable to those in LT leaves.
Acclimation capacity between respiration and photosynthesis
The temperature dependencies of the light-saturated rates of net photosynthesis at Ca of 360 µL L−1 (A360) were different depending on the growth temperature and on the duration at LT (Fig. 3). The temperature dependencies of the respiration rates were also different depending on the growth temperature (Fig. 2). However, the degree of acclimation of respiration was similar regardless of the timing of transfer. All HL leaves showed respiration rates similar to those of LT leaves. Clearly, respiration acclimated to LT much faster than photosynthesis. Such a difference in acclimation capacity between respiration and photosynthesis was also observed by Sims & Pearcy (1991), although they transferred Alocasia macrorrhiza, a shade-tolerant species, between high and low light conditions. The leaf respiration of A. macrorrhiza acclimated to the new growth condition within 3 d after the transfer, whereas the photosynthetic capacities did not show clear acclimation. Noguchi, Nakajima & Terashima (2001) followed changes in respiration in leaves of Alocasia odora alter the transfer in more detail and found a significant respiratory acclimation even within 24 h. Atkin, Holly & Ball (2000) showed that substantial acclimation in respiration occurred within 1 week after the change in growth temperature. However, they did not investigate differences in acclimation capacity between respiration and photosynthesis. In the present study, it was shown that the acclimation capacity of the respiration is greater than that of photosynthesis in temperature acclimation.
Temperature dependence of RuBP regeneration
At high CO2 concentration, in particular in combination with high irradiance and/or low temperature, the rate of CO2 assimilation can sometimes be limited by triose phosphate utilization (Sharkey 1985; Leegood & Furbank 1986; Labate & Leegood 1988; Sage, Sharkey & Pearcy 1990). In this study, as well, it is possible that A1500 was limited by triose phosphate utilization. However, this was not mainly concerned in this study, because the maximum A1500 was strongly correlated with the Cyt f content (Pearson's correlation coefficient: r = 0.92, P = 0.0236, data not shown).
We observed clear differences in A1500 among the leaves (Fig. 4). In LT and HL leaves including HL-mature leaves, A1500 were greater than that of HT leaves at any temperature. The optimum temperature for the HL-mature leaves was the same as that of HT leaves, but, those for the HL-young and HL-new leaves shifted downwards slightly. These results indicate that the changes in the absolute rate of the electron transport occurred more rapidly than that in the optimum temperature, although the reason for this time difference is unknown. The temperature acclimation of electron transport rate was often associated with changes in plastoquinone pool (Griffith, Elfman & Camm 1984; Huner et al. 1993), lipid and protein composition of thylakoid membranes (Öquist 1982; Ottander, Campbell & Öquist 1995; Falk et al. 1996; Vogg et al. 1998; Mikami & Murata 2003). These changes could explain the time difference in acclimation. Many papers showed the shift of the optimum temperature for the electron transport rate measured in thylakoids depending on growth temperature (Tieszen & Helgager 1968; Armond et al. 1978; Badger et al. 1982; Mitchell & Barber 1986; Mawson & Cummins 1989; Yamasaki et al. 2002). Recently, Yamasaki et al. (2002) showed that, in winter wheat, large changes in the temperature dependence of the electron transport on both sides of the PSII reaction centre (water oxidation side and plastoquinone side) were major factors contributing to the temperature acclimation of photosynthesis. Such a detail analyses will elucidate the differential acclimation between the absolute rate and the optimum temperature of the electron transport.
Temperature dependence of RuBP carboxylation
The temperature dependence of RuBP carboxylation has not been considered to vary much, because the Rubisco kinetic parameters are relatively constant across different species and growth conditions (Badger et al. 1982; Brooks & Farquhar 1985; von Caemmerer 2000). In the present study, however, we found that the temperature dependence of IS, the index of RuBP carboxylation, drastically changed depending on the growth temperature (Fig. 5). Moreover, the temperature dependencies of HL leaves showed shoulders near the peak for HT leaves (Fig. 5). The latter suggests that HL leaves would have two populations of Rubisco, the low-temperature and high-temperature types. Such changes in Rubisco populations probably contributed to the downward shift in the optimum temperature of photosynthesis.
We also confirmed that the Rubisco content markedly increased in response to LT (Table 1). The large amounts of Rubisco in LT and HL leaves contributed to maintaining the substantial photosynthetic rates at low temperatures (Berry & Björkman 1980; Badger et al. 1982), because the activity of Rubisco decreases with the decrease in temperature. However, the increase in Rubisco content did not lead to the proportional increase in IS (Fig. 5) or Vcmax (Fig. 7). This discrepancy may be due to the decrease in the CO2 concentration in the chloroplast stroma (Cc) in the LT and HL leaves. Evans & Terashima (1988) found that the ratio of IS to Rubisco content per leaf area decreased with the increase in Rubisco content. Because the chloroplasts in the leaves having high Rubisco contents were thick, they argued that Cc decreases with fattening of the chloroplasts. Given that there is a finite resistance to CO2 diffusion from the intercellular airspace to the chloroplast per unit chloroplast surface area facing the intercellular air space (Sc), Cc decreases with the increase in Rubisco content per Sc. Rubisco contents in LT and HL leaves were markedly high (more than 3 µmol m−2). Thus, Cc would decrease considerably.
On the other hand, Makino et al. (1994) suggested that deformation of chloroplasts in the low-temperature grown rice might increase the resistance per se. Growth at low temperature generally causes an increase in leaf thickness (Boese & Huner 1990). This was also the case in the present study (data not shown, see the data of LMA in Table 1). CO2 diffusion in the intercellular spaces is slower in thick leaves, which could also lower Cc in the LT and HL leaves slightly. It is also probable that the discrepancy was due to existence of two Rubisco populations. If the two populations of Rubisco have different optimum temperatures, and that compositions of these populations differ depending on growth conditions, then, the maximum rate will not be proportional to the content of Rubisco. Another possibility is the decrease in Rubisco activation state, as previous studies showed the Rubisco activation state decreased with increasing leaf N (Lawlor 1987; Mächler et al. 1988; Cheng & Fuchigami 2000). However, Evans & Terashima (1988) did not observe the decrease in the activation state with the increase in leaf N in spinach leaves. Our study also failed to detect marked decrease in the activation state with the increase in leaf N (data not shown).
Relationships between values of A1500 (Jmax) and IS (Vcmax)
The balance between A1500 and IS differed among the leaves, and A1500/IS ratios in HL leaves were between those of HT and LT leaves (Fig. 6). Jmax/Vcmax also differed in a similar manner (Fig. 7). In the present estimation of Jmax and Vcmax, we used the same kinetic parameters for Rubisco irrespective of the growth temperatures. Because the kinetic parameters of Rubisco would possibly differ depending on the growth temperature, in particular in spinach, and we ignored the resistance to CO2 diffusion from the intercellular spaces to chloroplast stroma (Evans & Loreto 2000), we can only discuss large differences in the balance between these temperature dependencies. However, the variation of Jmax/Vcmax depending on growth temperature was marked (Fig. 7) and Jmax/Vcmax in LT leaves was clearly higher than that in HT leaves at any temperature (Fig. 6). Moreover, Jmax/Vcmax of HL leaves were intermediate between those of HT and LT leaves.
Farquhar & von Caemmerer (1982) suggested that the shift in the optimum temperature of photosynthesis is due to the change in the ratio of Jmax and Vcmax. Hikosaka (1997) hypothesized that an increase in the photosynthetic rate at growth temperature can be achieved by an investment of proteins to either of the two partial reactions that is limiting photosynthesis at the growth temperature. If his hypothesis is valid, LT leaves should invest more nitrogen into the components to increase Jmax having the higher optimum temperature than Vcmax, and this was actually the case (Fig. 7). We analysed the extents to which the change in Jmax/Vcmax contributes to the shift of the optimum temperature for photosynthesis, based on C3 photosynthetic models (see Appendix). For example, we calculated the optimum temperature in the hypothetical HT leaves which have the same temperature dependencies of Jmax and Vcmax with HT leaves, and Jmax/Vcmax values for LT leaves or HL leaves. Then, we compared the optimum temperatures in such leaves with that in HT leaves. The results indicate that the changes in Jmax/Vcmax actually caused shifts in the optimum temperature for photosynthesis. However, the extents of the shifts were at most 3 °C. In HL leaves, the balance change would explain more than half the actual shift of the optimum temperature of A360 (3–6 °C). In LT leaves, however, it was much less than the actual shift (9 °C). Therefore, we conclude that the shift in the optimum temperature of photosynthesis was caused by both of the changes in Jmax/Vcmax and the temperature dependencies of the capacity for RuBP regeneration and carboxylation. There are reports indicating that the growth temperature does not affect Jmax/Vcmax (Leuning 1997; Bunce 2000; Medlyn et al. 2002). Further studies are necessary to clarify whether the growth temperature affects the balance between Jmax and Vcmax, and to what extent the changes of Jmax/Vcmax affect the shift of the optimum temperature.
The ratio of the content of Cyt f to that of Rubisco (Cyt f/Rubisco), the ratio of limiting components of Jmax and Vcmax, changed with the growth temperature (Fig. 8). Cyt f/Rubisco increased after the transfer from HT to LT (Table 1). Moreover, the increase in these components after the transfer did not occur synchronously. After the transfer to LT, the increase in Cyt f content was more rapid than that in Rubisco.
In the present study, we showed that the shift in the optimum temperature of photosynthesis was ascribed not only to the changes in the balance between the two partial reactions but also to the changes in the temperature dependence of the two partial reactions, in particular, RuBP carboxylation. We confirmed that the balance between RuBP regeneration and RuBP carboxylation changed with the changes in Cyt f/Rubisco. In contrast to the widely held view, we found the marked change in the temperature dependence of RuBP carboxylation, and the changes depended on the duration at LT. We also found that the absolute rate of RuBP regeneration promptly increased with lowering the growth temperature, while the change in the optimum temperature was small and occurred slower than that of RuBP carboxylation. Biochemical characterizations of these changes are in progress.
We are grateful to Dr S-I. Miyazawa for instruction of various techniques and to Dr K. Hikosaka (Tohoku University) for generous advice. This work is supported by Japan Society for the Promotion of Science and the Ministry of Agriculture, Forestry and Fishery, Japan (Bio-Design Program).
Vcmax and Jmax
Farquhar et al. (1980) proposed that net leaf photosynthesis, A, can be modelled as the minimum of two limiting rates:
Pc is the RuBP-saturated rate of photosynthesis (µmol m−2 s−1) and Pr is the RuBP-limited rate of photosynthesis (µmol m−2 s−1). Rd (µmol m−2 s−1) is the day respiration rate. Pc is expressed as a function of the CO2 concentration at the intercellular spaces (Ci, µL L−1):
where Vcmax (µmol m−2 s−1) is the maximum rate of RuBP carboxylation on the leaf area basis, Kc (µL L−1) and Ko (mL L−1) are the Michaelis constants for CO2 and O2, respectively, O (mL L−1) is the O2 concentration, Γ* (µL L−1) is the CO2 compensation point in the absence of day respiration. Because IS should be identical to dPc(Ci)/dCi at Ci = Vcmax is expressed as
On the other hand, Pr is expressed as
where Jmax (µmol m−2 s−1) is the rate of electron transport. Using A1500, Jmax is expressed as
These equations of the model assume that the internal conductance, the conductance for CO2 diffusion from the intercellular space to the carboxylation site, is negligible.
Temperature dependencies of Kc, Ko and Γ*
For the temperature dependencies of the kinetic constants (Kc and Ko), we used equations of Harley & Tenhunen (1991):
where R is the universal gas constant (R = 8.314 J K−1 mol−1) and Tl is leaf temperature (K).
For the temperature dependence of the CO2 compensation point in the absence of day respiration, we used the equation of Brooks & Farquhar (1985), who estimated the CO2 compensation point (Γ*) of spinach leaves in vivo by the gas-exchange technique: