Balancing carboxylation and regeneration of ribulose-1,5- bisphosphate in leaf photosynthesis: temperature acclimation of an evergreen tree, Quercus myrsinaefolia

Authors


K. Hikosaka Fax: 81-22-2176699; e-mail: hikosaka@mail.cc.tohoku.ac.jp

Abstract

Changes in the temperature dependence of the photosynthetic rate depending on growth temperature were investigated for a temperate evergreen tree, Quercus myrsinaefolia. Plants were grown at 250 μmol quanta m–2 s–1 under two temperature conditions, 15 and 30 °C. The optimal temperature that maximizes the light-saturated rate of photosynthesis at 350 μL L–1 CO2 was found to be 20–25 and 30–35 °C for leaves grown at 15 and 30 °C, respectively. We focused on two processes, carboxylation and regeneration of ribulose-1,5-bisphosphate (RuBP), which potentially limit photosynthetic rates. Because the former process is known to limit photosynthesis at lower CO2 concentrations while the latter limits it at higher CO2 concentrations, we determined the temperature dependence of the photosynthetic rate at 200 and 1000 μL L–1 CO2 under saturated light. It was revealed that the temperature dependence of both processes varied depending on the growth temperature. Using a biochemical model, we estimated the capacity of the two processes at various temperatures under ambient CO2 concentration. It was suggested that, in leaves grown at low temperature (15 °C), the photosynthetic rate was limited solely by RuBP carboxylation under any temperature. On the other hand, it was suggested that, in leaves grown at high temperature (30 °C), the photosynthetic rate was limited by RuBP regeneration below 22 °C, but limited by RuBP carboxylation above 22 °C. We concluded that: (1) the changes in the temperature dependence of carboxylation and regeneration of RuBP and (2) the changes in the balance of these two processes altered the temperature dependence of the photosynthetic rate.

INTRODUCTION

The temperature dependence of photosynthesis is known to change even in the same individual plant subjected to changing temperature regimes (Berry & Björkman 1980). In many species, when plants are grown under lower temperature conditions, the temperature optimum of photosynthesis shifts to lower temperatures (Lange et al. 1974; Slatyer 1977; Mooney, Björkman & Collatz 1978; Berry & Björkman 1980; Badger, Björkman & Armond 1982; Ferrar, Slatyer & Vranjic 1989). Such acclimation to temperature has been considered to contribute to adaptation to seasonal fluctuations of temperature (Berry & Björkman 1980).

Despite the rather large number of studies conducted to clarify the mechanisms of temperature acclimation of photosynthesis, various hypotheses have been proposed. (1) Changes in heat stability of photosynthetic enzymes in vivo: Investigating a desert shrub, Nerium oleander, Badger et al. (1982) showed that plants grown at low temperature had a higher activity of several photosynthetic enzymes at low temperatures but had lower in vivo heat stability relative to plants grown at high temperature. Consequently, photosynthesis at low temperatures was higher in leaves grown at low temperature, while the inverse was the case for photosynthesis at higher temperatures. (2) Stomatal conductance: Based on a theoretical model, Kirschbaum & Farquhar (1984) suggested that lower stomatal conductance causes the temperature optimum of photosynthesis to decrease. (3) Temperature acclimation in the electron transport system: Several studies have shown that the temperature dependence of electron transport capacity changes with growth temperature (Armond, Schreiber & Björkman 1978; Badger et al. 1982; Mitchell & Barber 1986). (4) CO2 conductance within a leaf: Makino, Nakano & Mae (1994) examined rice leaves and showed that the photosynthetic rate per unit RuBPCase (ribulose-1,5-bisphosphate carboxylase) content decreased with decreasing growth temperature. They suggested that CO2 conductance from the intercellular space to chloroplasts decreases with decreasing growth temperature. A decrease in CO2 concentration in chloroplasts would cause the temperature optimum of photosynthesis to decrease. (5) Balancing carboxylation and regeneration of RuBP: According to the model of Farquhar, von Caemmerer & Berry (1980), the photosynthetic rate is a minimum of capacities of carboxylation and regeneration of RuBP. If the temperature dependence of the two processes is different, when the two processes colimit the photosynthetic rate at a certain temperature, the photosynthetic rate at the other temperatures is limited by either of the two processes. Farquhar & von Caemmerer (1982) suggested that a change in the balance between the two processes can alter the temperature dependence of photosynthesis. Hikosaka (1997) predicted that, because the colimitation of photosynthesis leads to a minimization of nitrogen cost for photosynthesis (von Caemmerer & Farquhar 1981; Farquhar 1989; Hikosaka & Terashima 1995), the investment of protein in each process should be adjusted to realize colimitation at the growth temperature.

Although these five hypotheses have been suggested, our knowledge of the mechanism of change in the temperature dependence of photosynthesis is still limited. For example, the difference in stomatal conductance between leaves grown at different temperatures is generally too small to explain a change in the temperature dependence of photosynthesis (Ferrar et al. 1989). In the study of Badger et al. (1982), the temperature dependence of photosynthesis was different between leaves grown at contrasting temperatures even at the temperatures where the enzymes of both leaves were stable. They also suggested that the temperature dependence of photosynthesis does not necessarily reflect that of electron transport. Because Makino et al. (1994) investigated gas exchange characteristics only at one temperature, it is not clear how the changes observed in their study affect the temperature dependence of photosynthesis. On the other hand, because the above-mentioned hypotheses are not exclusive of each other, it is possible that all of them occur simultaneously.

To clarify the mechanisms of changes in the temperature dependence of the photosynthetic rate which depend on growth temperature, it is indispensable to know the limiting step of photosynthesis at a given temperature. In the present study, we focused on two processes of photosynthesis, carboxylation and regeneration of RuBP. According to the model of Farquhar et al. (1980), the former tends to limit photosynthesis at low CO2 concentrations while the latter does so at high CO2 concentrations. Therefore, from the measurement of photosynthesis at high and low CO2 concentrations, the temperature dependence of the two processes may be clarified. Following the prediction by Hikosaka (1997), we investigated how the balance between the two processes differs among leaves grown at contrasting temperatures.

MATERIALS AND METHODS

Seedlings of a broad-leaved evergreen tree, Quercus myrsinaefolia Bloom (Fagaceae), were used in the present study. This species is distributed in warm-temperate regions and is known to exhibit seasonal changes in the temperature dependence of photosynthesis (Takenaka 1986). Seedlings in pots (one plant in a pot) filled with soil were grown in growth chambers. The temperature in the two chambers was controlled at 15 °C or 30 °C throughout the day. The plants were illuminated by 400 W halide lamps (metal halide lamp, Mitsubishi, Tokyo, Japan) and the photoperiod was 14/10 h (light/dark). The photon flux density (PFD) on the plants was 250–300 μmol m–2 s–1 throughout daytime. The relative humidity was held constant at 80%. The plants were watered every day and 100 mL of nutrient solution [12 mM N; see Hikosaka, Terashima & Katoh (1994) for details] was supplied once a week. Young leaves that emerged after transfer of the pots to the growth chambers were used for the study.

For photosynthetic measurements, an open gas exchange system was used [for details, see Hikosaka et al. (1998)]. The water vapour pressure deficit was less than 1·2 kPa in most cases and was less than 1·6 kPa even when some exceptional cases were included. For one leaf, either CO2 dependence of photosynthesis at a given temperature or the temperature dependence of photosynthesis at atmospheric CO2 concentration of 200, 350 or 1000 μL L–1 was determined under saturating light conditions (800–1000 μmol m–2 s–1), and the respiration rate at a given CO2 and temperature was determined.

MODEL

In the present study, the model of Farquhar et al. (1980) was used with some modifications. The gross photosynthetic rate (net photosynthesis + day respiration, P, μmol m–2 s–1) is given by a minimum of the RuBP-saturated rate of photosynthesis (Pc, photosynthesis limited by RuBP carboxylation, μmol m–2 s–1) and the RuBP-limited rate of photosynthesis (Pr, photosynthesis limited by RuBP regeneration, μmol m–2 s–1):

image

Pc is expressed as:

image

where R (μmol m–2 s–1) is the day respiration but assumed to be the same as that in the dark, Vcmax (μmol m–2 s–1) is the maximum velocity of RuBP carboxylation per leaf area, Ci (μL L–1) is the CO2 concentration at intercellular spaces, Kc (μL L–1) and Ko (mL L–1) are the Michaelis constants for CO2 and O2, respectively, and O (mL L–1) is the O2 concentration. Γ* (μL L–1) is the CO2 compensation point in the absence of respiration defined as follows:

image

where Vomax (μmol m–2 s–1) is the maximum velocity of RuBP oxygenation per leaf area and τ is a specificity factor of RuBPCase. Pr is expressed as follows:

image

where J (μmol m–2 s–1) is the electron transport rate.

RESULTS

The temperature dependence of the respiration rate determined in the dark was independent of the growth temperature (Fig. 1). Therefore, respiration rates at various temperatures were estimated from the rate obtained at a certain temperature using the equation given in Fig. 1. The mean and standard deviation (SD) of the respiration rate were 0·54 ± 0·07 μmol m–2 s–1 (n = 6) at 15 °C for 15 °C-grown leaves and 1·07 ± 0·21 μmol m–2 s–1 (n = 5) at 30 °C for 30 °C-grown leaves (these values include the values that are not shown in the figure).

Figure 1.

. Temperature dependence of the dark respiration rate. Values are normalized at 25 °C. Closed and open symbols denote leaves grown at 15 and 30 °C, respectively. Different symbols denote data obtained from different leaves. The fitted curve is Y = 1 + 0·074915(X– 25) + 0·0019336(X– 25)2 (r = 0·98).

Figure 2(a) shows the temperature dependence of the light-saturated rate of gross photosynthesis (net photosynthesis + dark respiration) at 350 μL L–1 of air CO2 concentration (Ca). The mean and SD of the gross photosynthetic rate were 5·68 ± 1·09 μmol m–2 s–1 (n = 9) at 15 °C for 15 °C-grown leaves and 7·17 ± 1·61 μmol m–2 s–1 (n = 8) at 30 °C for 30 °C-grown leaves (these values include the values that are not shown in the figure). Because the absolute values varied from leaf to leaf, relative values of gross photosynthesis normalized at a certain growth temperature are shown in Fig. 2(b). It is clear that the temperature dependence of the photosynthetic rate was different depending on the growth temperature. The optimal temperature for photosynthesis was 20–25 and 30–35 °C for leaves grown at 15 °C and 30 °C, respectively. The intercellular concentration of CO2 was markedly lower in leaves grown at a low temperature due to low stomatal conductance but was independent of the temperature for photosynthetic measurement. The mean and SD were 228 ± 14 and 298 ± 24 μL L–1 for leaves grown at low and high temperatures, respectively.

Figure 2.

. Temperature dependence of the gross photosynthetic rate (net photosynthesis + dark respiration) at 350 μL L–1. (a) Absolute values of photosynthesis of each leaf; (b) relative values normalized at growth temperature (15 and 30 °C). Closed and open symbols denote leaves grown at 15 and 30 °C, respectively. Fitted curves are two-dimensional polynomials.

To investigate the temperature dependence of the carboxylation and regeneration of RuBP, the temperature dependences of the light-saturated rates of photosynthesis were determined at 200 and 1000 μL L–1Ca, respectively. The relative values of gross photosynthesis where the rate at 20 °C was adjusted to 1 are shown in Fig. 3(a & b). The temperature dependence of the photosynthetic rates at 200 and 1000 μL L–1Ca changed depending on growth temperatures but it should be noted that the different Cis between the two growth temperatures would affect this difference (Kirschbaum & Farquhar 1984).

Figure 3.

. Temperature dependence of the gross photosynthetic rate at 200 (a) and 1000 μL L–1 (b). Relative values normalized at 20 °C are shown. Closed and open symbols denote leaves grown at 15 and 30 °C. Different symbols denote data obtained from different leaves. Fitted curves are two-dimensional polynomials.

Next, we estimated the capacity of the two processes, the carboxylation and regeneration of RuBP, to determine which limits photosynthesis at 350 μL L–1Ca. However, because photosynthetic capacity varied from leaf to leaf (Fig. 2a), the capacities had to be estimated as relative values. The calculations were as follows: we first obtained the temperature dependence of photosynthesis at 200, 350 and 1000 μL L–1Ca (Figs 2 & 3). We regard the gross photosynthesis (net photosynthesis + respiration) at 200, 350 and 1000 μL L–1Ca as P200, P350 and P1000, respectively. Then the CO2 dependence of photosynthesis at the growth temperature was determined to obtain the ratios of P200 to P350 (R200) and P1000 to P350 (R1000) (Table 1). Using the model of Farquhar et al. (1980), the capacities of carboxylation or regeneration of RuBP were estimated from P200 or P1000, respectively. From Eqn 2, Vcmax was obtained as follows:

Table 1.  . Gross photosynthetic rates (net photosynthesis + dark respiration) at 200 and 1000 μL L–1 of ambient CO2 concentration relative to those at 350 μL L–1 (R200 and R1000, respectively). Mean values of R200 or R1000 obtained for each leaf are shown. The rates were obtained at growth temperature. The number of samples was 4–6 Thumbnail image of
image

Similarly, J was obtained as:

image

Because Ci when P350 was determined is known as 228 and 298 μL L–1 for leaves grown at 15 and 30 °C, respectively, Eqns 2 and 4 give the capacities of carboxylation (Mc350) and regeneration (Mr350) of RuBP at 350 μL L–1Ca. For example, Mc350 of 15 °C-grown leaves was calculated as follows:

{Ci+Kc(1+O/Ko)}(228–Γ*)

Mc350 = P200––––––––––––––––––––––––

{228+Kc(1+O/Ko)}(Ci–Γ*)

Then, the capacity of RuBP carboxylation at 350 μL L–1Ca relative to P350 at the growth temperature (MRc350) was calculated as follows:

image

where P200G is the photosynthetic rate at a certain growth temperature when the temperature dependence was determined at 200 μL L–1. Similarly, the capacity of RuBP regeneration at 350 μL L–1 relative to P350 (MRr350) was obtained as:

image

where P1000G is the photosynthetic rate at a certain growth temperature when the temperature dependence was determined at 1000 μL L–1. Because P200G/R200 and P1000G/R1000 should equal P350 at the growth temperature, if P350 is limited only by RuBP carboxylation, MRc350 at the growth temperature is expected to be 1 and MRr350 at the growth temperature would be higher than 1.

The following temperature dependence of Kc, Ko, and τ was tentatively assumed so that the model fits the actual data well. We tried to apply several curves presented by previous works or those modified by us (judged by eye) and selected the following curves:

image
image
image

where T is the leaf temperature (K) and Rg is the gas constant (8·3 J K–1 mol–1). These equations are after Harley & Tenhunen (1991) except for Kc, which is double the value used in their study. At 25 °C, Kc, Ko and τ are 610 μL L–1, 161 mL L–1, and 2325, respectively; 210 mL L–1 was used for O. Figure 4 shows examples of model application. The solid and dotted curves shown are calculated from P200 and P1000 using Eqns 5 and 6, respectively.

Figure 4.

. Examples of CO2 dependence of photosynthetic rates (net photosynthesis) at different temperatures. Solid and dotted lines are CO2 dependence of carboxylation and regeneration of ribulose bisphosphate (RuBP), respectively, estimated as mentioned in the text.

Figure 5 shows the temperature dependence of Vcmax and that of J, which were calculated from photosynthetic rates at 200 and 1000 μL L–1Ca, respectively. The temperature dependence of Vcmax in 15 °C-grown leaves was similar to that of Vcmax in 30 °C-grown leaves below 20 °C but was different at higher temperatures (Fig. 5a). The temperature dependence of J was quite different between the two leaves (Fig. 5b). The apparent activation energy was lower in 15 °C-grown leaves.

Figure 5.

. Temperature dependence of Vcmax (a) and J (b) estimated as mentioned in the text. Relative values normalized at 20 °C are shown. Closed and open symbols denote leaves grown at 15 and 30 °C. Different symbols denote data obtained from different leaves. Fitted curves are two-dimensional polynomials.

In Fig. 6, the capacities of carboxylation and regeneration of RuBP at 350 μL L–1Ca are compared with the photosynthetic rate at 350 μL L–1Ca to assess which of the two processes limit photosynthesis at a given temperature. The capacities and photosynthetic rates are expressed as relative values where the photosynthetic rate at 350 μL L–1Ca under growth temperature is 1. For 15 °C-grown leaves (Fig. 6a), the capacity of carboxylation of RuBP was always lower than that of regeneration of RuBP. The photosynthetic rate at 350 μL L–1Ca nearly corresponded to the capacity of RuBP carboxylation at all temperatures. This suggests that the photosynthetic rate is limited by RuBP carboxylation at any temperature. For 30 °C-grown leaves (Fig. 6b), the capacity of RuBP carboxylation was higher than that of RuBP regeneration below 22 °C but was lower above this temperature. The photosynthetic rate nearly corresponded to the lower capacity of carboxylation and regeneration of RuBP, suggesting that the photosynthetic rate is limited by RuBP regeneration below 22 °C and is limited by RuBP carboxylation above 22 °C.

Figure 6.

. Comparison of the gross photosynthetic rate (net photosynthesis + dark respiration) and the potential rates of carboxylation and regeneration of ribulose bisphosphate (RuBP) at 350 μL L–1 for leaves grown at 15 (a) and 30 °C (b). Closed circles, the photosynthetic rate at 350 μL L–1; open circles, the potential rate of RuBP carboxylation at 350 μL L–1; squares, the potential rate of RuBP regeneration at 350 μL L–1. Relative values normalized at growth temperature (15 and 30 °C) are shown (see text for details). Fitted curves are two-dimensional polynomials.

Because Ci was quite different between growth temperatures as mentioned above, we evaluated the effect of the difference in Ci. Figure 7(a & b) shows the capacities of carboxylation and regeneration of RuBP at Ci = 298 (observed for 30 °C-grown leaves) estimated from 15 °C-grown leaves and those at Ci = 228 (observed for 15 °C-grown leaves) estimated from 30 °C-grown leaves. As indicated by Kirschbaum & Farquhar (1984), the change in Ci caused a shift of the optimal temperature of photosynthesis, but that change was small (1–2 °C). The decrease in Ci by 70 μL L–1 in 30 °C-grown leaves resulted in a 5 °C decrease in the temperature at which carboxylation and regeneration of RuBP colimit the photosynthetic rate.

Figure 7.

. (a) Comparison of the capacities of carboxylation and regeneration of ribulose bisphosphate (RuBP) at Ci = 298 μL L–1 for leaves grown at 15 °C. (b) Comparison of the potential rates of carboxylation and regeneration of RuBP at Ci = 228 μL L–1 for leaves grown at 30 °C. Open circles, the potential rate of RuBP carboxylation at 350 μL L–1; squares, the potential rate of RuBP regeneration at 350 μL L–1. Relative values where photosynthesis at 228 and 298 μL L–1 of Ci for 15 and 30 °C-grown leaves are scaled to 1 are shown, respectively. Fitted curves are two-dimensional polynomials.

In Fig. 8, CO2 dependence of photosynthesis at 15 °C is compared between leaves grown at 15 and 30 °C. Relative values normalized at Ci = 200 μL L–1 are shown. Leaves grown at 15 °C had a higher rate of photosynthesis at high CO2 concentrations, suggesting a higher capacity of RuBP regeneration in 15 °C-grown leaves.

Figure 8.

. CO2 dependence of gross photosynthesis at 15 °C of leaves grown at 15 and 30 °C. Relative values normalized at Ci = 200 μL L–1 are shown. Closed and open symbols denote leaves grown at 15 and 30 °C. Different symbols denote data obtained from different leaves. Solid line: photosynthesis limited by RuBP carboxylation according to the model of Farquhar et al. (1980); dotted lines: photosynthesis limited by RuBP regeneration according to the model, which was separately fitted for each set of data of leaves grown at different temperatures.

DISCUSSION

The present study clearly shows how the temperature dependence of the photosynthetic rate differs between leaves grown at contrasting temperatures. One of the factors responsible for the difference was the change in the temperature dependence of the two processes, carboxylation and regeneration of RuBP. Both Vcmax and J in 30 °C-grown leaves showed greater temperature dependence relative to those in 15 °C-grown leaves (Fig. 5a & b). The change in the balance between the two processes was also the responsible factor. In 15 °C-grown leaves, the capacity of RuBP regeneration at 350 μL L–1Ca was always higher than that of RuBP carboxylation at 350 μL L–1Ca (Fig. 6a), while in 30 °C-grown leaves the capacity of RuBP regeneration at 350 μL L–1Ca crossed over that of RuBP carboxylation at 350 μL L–1Ca (Fig. 6b). Consequently, the temperature dependence of the photosynthetic rate of 15 °C-grown leaves was determined solely by the temperature dependence of RuBP carboxylation, while in 30°C-grown leaves, RuBP regeneration determined photosynthetic rates at lower temperatures and RuBP carboxylation determined the rates at higher temperatures. Such a change in the temperature dependence of the photosynthetic rate due to balancing the two processes is consistent with the prediction by Hikosaka (1997).

Although Hikosaka (1997) predicted that the photosynthetic rate should be colimited by carboxylation and regeneration of RuBP at the growth temperature for efficient nitrogen economy of photosynthesis, the temperature where the two processes colimit photosynthesis was different from the growth temperatures (Fig. 6a & b). However, the ratio of the capacity of RuBP regeneration to that of RuBP carboxylation at 350 μL L–1Ca under growth temperature was 1·3–1·4, which was similar between the leaves. This fact implies that there is some physiological or ecological significance to maintenance of this ratio at a certain growth temperature. We present several possibilities. First, the PFD used when photosynthesis was measured was higher than that for growth. If J is lower at lower PFD, the capacities of regeneration and carboxylation of RuBP conceivably colimit the photosynthetic rate. However, because the photosynthetic capacity of Q. myrsinaefolia leaves was low (5–7 μmol m–2 s–1), the difference between the photosynthetic rate at saturated light and that at 250 μmol m–2 s–1 would be small. Second, RuBP regeneration capacity may be increased as a strategy to avoid photoinhibition of photosynthesis. Photoinhibition reduces the electron transport capacity of photosystem II (Aro, Virgin & Andersson 1993) and lowers J. However, if the capacity of RuBP regeneration at 350 μL L–1Ca was higher than that of RuBP carboxylation at 350 μL L–1Ca, a small decrease in J would not affect the photosynthetic rate. Third, a higher capacity of RuBP regeneration may be a preadaptation to efficient use of sunflecks. When leaves that have been under dim light, such as those found in the understorey of forests, are suddenly illuminated by strong light, such as sunflecks, oxygen evolution is quickly increased but the increase of uptake of CO2 is slow (Kirschbaum & Pearcy 1988). This may be because conversion of light energy to chemical energy is rapid while enzymatic processes are more gradual. In this case, a high capacity of electron transport is advantageous in utilizing light energy but the capacities of enzymatic processes are not necessarily high, which may be consistent with lower capacities of RuBP carboxylation at 350 μL L–1Ca.

In the present study, we used the model of Farquhar et al. (1980) to predict the capacities of carboxylation and regeneration of RuBP at 350 μL L–1Ca. However, Kc used in the present study (610 μL L–1) was higher than the value used in previous studies. For example, Kc values at 25 °C in previous studies were 310 (Kirschbaum & Farquhar 1984) and 305 μL L–1 (Harley & Tenhunen 1991). This difference may probably be due to the difference in CO2 conductance from intercellular spaces to chloroplasts. In the present study, we used a model where photosynthesis is assumed to respond to Ci. However, now it is known that the CO2 level at chloroplasts (Cc) is lower than Ci (von Caemmerer & Evans 1991). Several works have suggested that species with lower photosynthetic capacity have a lower level of Cc (Lloyd et al. 1992; Epron et al. 1995). From a determination of 13C composition, we also suggested that Q. myrsinaefolia leaves have a lower Cc level than leaves of an annual, Chenopodium album (Hikosaka et al. 1998). If CO2 conductance from intercellular spaces to chloroplasts differs, CO2 dependence of RuBP carboxylation was theoretically predicted to change (Harley et al. 1992). Therefore, the Kc value is apparently larger in such leaves.

The temperature dependence of RuBP regeneration changed depending on the growth temperature (Fig. 5b). Kirschbaum & Farquhar (1984) suggested that the temperature dependence of RuBP regeneration reflects that of the electron transport rate. It is well known that the apparent activation energy of electron transport increases with increasing growth temperature (Armond et al. 1978; Badger et al. 1982; Mitchell & Barber 1986). This is consistent with the present results. However, Badger et al. (1982) have shown that the temperature dependence of electron transport does not necessarily correspond to that of the photosynthetic rate at high CO2 concentration. Therefore, other factors may be involved in the change in the temperature dependence of RuBP regeneration. It is also possible that the difference in in vivo heat stability of some photosynthetic proteins (Badger et al. 1982) caused a difference in the temperature dependence of RuBP regeneration.

Several studies have shown that photosynthesis at low temperatures is often limited by triose-phosphate utilization (Labate & Leegood 1988; Sage, Sharkey & Pearcy 1990, Huner et al. 1993). For example, in C. album leaves grown at 27/23 °C (day/night), the photosynthetic rate at 15 °C under normal CO2 concentration was independent of CO2 concentration, suggesting that photosynthesis is limited by triose-phosphate utilization (Sage et al. 1990). However, in 30 °C-grown leaves of Q. myrsinaefolia, the photosynthetic rate measured at 30 °C showed CO2 dependence under 600 μL L–1 of Ci (Fig. 4b). This suggests that triose-phosphate utilization did not limit photosynthesis at 350 μL L–1Ca in the present study.

The temperature dependence of RuBP carboxylation also changed depending on the growth temperature (Fig. 5a). When compared at Ci = 298 μL L–1, the temperature optima of RuBP carboxylation were 20–23 °C and 28–30 °C for leaves grown at low and high temperatures, respectively (Figs 7a & 6b). Such a change was also observed by Ferrar et al. (1989) who investigated several species of Eucalyptus grown at contrasting temperatures and determined the initial slope of the CO2 response curve; different initial slopes resulted from the difference in the potential of RuBP carboxylation. In leaves grown at a high temperature, the initial slope increased with the temperature, but in leaves grown at a low temperature, the initial slope did not increase as the temperature increased. They considered that RuBPCase may be inactivated or damaged at temperatures higher than the growth temperature. Badger et al. (1982) showed that the in vivo thermal stability of RuBPCase differed depending on the growth temperature. There are also several studies that suggested a decline in the RuBPCase activity at higher temperatures (Machler & Nosberger 1980; Weiss 1981; Monson et al. 1982; Kirschbaum & Farquhar 1984; Kobza & Edwards 1987).

The balance between carboxylation and regeneration of RuBP changed with growth temperature (Figs 6a, b & 8). Three explanations are possible for this change. One is a change in nitrogen partitioning among photosynthetic components. Hikosaka (1997) predicted that colimitation of photosynthesis can be achieved by investments of proteins in the limiting processes. If this hypothesis is true, 30 °C-grown leaves invested more nitrogen into RuBPCase relative to 15 °C-grown leaves. Second, low Cc in leaves grown at a low temperature may result in a low level of carboxylation capacity, which was suggested for rice plants by Makino et al. (1994). The third possibility is changes in the activity of photosynthetic components due to changes in the thermal stability of enzymes or acclimation of thylakoid reactions as mentioned above. Further studies are necessary to clarify how the balance between carboxylation and regeneration of RuBP changes.

Acknowledgements

The authors would like to thank M. C. Kato for technical assistance and I. Terashima and A. Makino for helpful comments. This work was supported in part by the Japanese Ministry of Education, Science and Culture (no. 09740574).

Ancillary