Rubisco, Rubisco activase, and global climate change
Journal of Experimental Botany, Vol. 59, No. 7, pp. 1581–1595, 2008 doi:10.1093/jxb/ern053 Advance Access publication 23 April, 2008
SPECIAL ISSUE REVIEW PAPER
Rubisco, Rubisco activase, and global climate change Rowan F. Sage1,*, Danielle A. Way1 and David S. Kubien2 1 2
Department of Ecology and Evolutionary Biology, University of Toronto, Toronto, ON, Canada M5S 3B2 Department of Biology, University of New Brunswick, Fredericton, NB, Canada E3B 5A3
Received 11 September 2007; Revised 28 January 2008; Accepted 5 February 2008
Abstract
Key words: Black spruce, climate change, C3 photosynthesis, Picea mariana, Rubisco, Rubisco activase, temperature.
Introduction The earth’s climate is predicted to warm by an average of 2–4 C during the next century as a result of increased greenhouse gases in the atmosphere (IPCC, 2007). Much of the climate warming associated with greenhouse gas enrichment will occur at higher latitudes, higher elevations, and during the winter. In Canada, for example, the boreal region is predicted to warm by 3–10 C (IPCC, 2007). Also higher atmospheric CO2 levels tend to reduce stomatal conductance and transpiration, thereby lowering latent heat loss and causing higher leaf temperatures (Kimball and Bernacchi, 2006; Bernacchi et al., 2007). Change in the atmospheric composition and climate will therefore increase the temperature of photosynthetic tissues, particularly in high latitude ecosystems such as the boreal forest. During periods of the year when plants normally experience optimal temperatures, it can also be expected that the new climate regimes will warm leaves well above their thermal optimum, potentially to a point where electron transport and Rubisco activase are heat labile. Such warming will reduce photosynthetic capacity, possibly offsetting gains in carbon assimilation associated with elevated CO2. Because climate warming will increase the frequency of supra-optimal temperatures, it is important to understand the limitations on photosynthesis above the photosynthetic thermal optimum, both in the short-term (minutes to hours), as may occur during a heat wave, and over the long term (days to years). Long-term photosynthetic
* To whom correspondence should be addressed. E-mail: [email protected] Abbreviations: A, net CO2 assimilation rate; Ci, the intercellular partial pressure of CO2; PSII, photosystem II; VPD, vapour pressure difference. ª The Author [2008]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
Global warming and the rise in atmospheric CO2 will increase the operating temperature of leaves in coming decades, often well above the thermal optimum for photosynthesis. Presently, there is controversy over the limiting processes controlling photosynthesis at elevated temperature. Leading models propose that the reduction in photosynthesis at elevated temperature is a function of either declining capacity of electron transport to regenerate RuBP, or reductions in the capacity of Rubisco activase to maintain Rubisco in an active configuration. Identifying which of these processes is the principal limitation at elevated temperature is complicated because each may be regulated in response to a limitation in the other. Biochemical and gas exchange assessments can disentangle these photosynthetic limitations; however, comprehensive assessments are often difficult and, for many species, virtually impossible. It is proposed that measurement of the initial slope of the CO2 response of photosynthesis (the A/Ci response) can be a useful means to screen for Rubisco activase limitations. This is because a reduction in the Rubisco activation state should be most apparent at low CO2 when Rubisco capacity is generally limiting. In sweet potato, spinach, and tobacco, the initial slope of the A/ Ci response shows no evidence of activase limitations at high temperature, as the slope can be accurately modelled using the kinetic parameters of fully activated Rubisco. In black spruce (Picea mariana), a reduction in the initial slope above 30 C cannot be explained by the known kinetics of fully activated Rubisco, indicating that activase may be limiting at high temperatures. Because black spruce is the dominant species in the boreal forest of North America, Rubisco activase may be an unusually important
factor determining the response of the boreal biome to climate change.
1582 Sage et al.
processes, and uses theoretical models of photosynthesis to explain observed gas exchange patterns in terms of the biochemical controls. Such approaches have provided clarity in the understanding of the biochemical controls over the light response to photosynthesis, and the acclimation of C3 photosynthesis to elevated CO2 (von Caemmerer and Farquhar, 1981; Long et al., 2004).
Patterns of the C3 photosynthetic response to temperature C3 photosynthesis typically exhibits a thermal optimum between 20 C and 35 C. The optimum is broad at low CO2, with a shallow peak usually centred in the mid20 C range (Fig. 1a). As CO2 levels increase, the optimum shifts to higher temperatures, and the peak sharpens, such that at high CO2, photosynthesis exhibits a very sharp peak centred around 30–35 C in most species (Fig. 1e; Berry and Raison, 1981; Sage, 2002). Species acclimated and/or adapted to cooler temperatures exhibit lower thermal optima, while high temperature-adapted species can exhibit thermal optima exceeding 35 C at elevated CO2 (Berry and Raison, 1981; Sage and Kubien, 2007). A change in the shape of the thermal response of photosynthesis depends upon the individual temperature responses of the diffusion and biochemical limitations controlling the rate of photosynthesis. Also, the temperature responses of photorespiration and dark respiration contribute to the thermal response of A. As temperatures increase, the relative rates of photorespiration and dark respiration increase relative to the in vivo capacity for Rubisco carboxylation. This pattern is demonstrated in Fig. 1, which shows a modelled temperature response of the net CO2 assimilation rate (A), the gross CO2 assimilation rate, and the rate of RuBP consumption at three different CO2 levels: 180 lbar (the atmospheric CO2 level during the last ice age 18 000 years ago), the current CO2 level of 380 lbar, and a future, high CO2 level of 700 lbar. The RuBP consumption rate is the sum of the carboxylation and oxygenation rates of Rubisco. The difference between gross photosynthesis and RuBP consumption reflects the inhibition associated with photorespiration (arrow 1 in Fig. 1b). The difference between A and gross CO2 assimilation indicates the limitation on carbon gain associated with day respiration in the mitochondria (arrow 2 in Fig. 1b). At low CO2, the rise in day respiration with temperature has a proportionally larger effect due to the relatively low photosynthetic rate (Fig. 1a, b). Also, the relative effect of photorespiration is much greater at low CO2 due to the relative lack of CO2 to inhibit RuBP oxygenation (compare parts b and f in Fig. 1). Inhibition of photorespiration becomes substantial at elevated CO2 levels such as 700 lbar. Under these conditions, the effect of the underlying biochemical
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
responses can be divided into two categories. First, for individuals, long-term exposure to new thermal regimes can lead to acclimation of the photosynthetic apparatus, allowing for stable or even increased rates of photosynthesis in the new, warmer growth regimes. Secondly, long-term exposure to elevated temperatures over multiple generations leads to natural selection and adaptation of populations to higher temperatures. A major concern for the future is that species will not be adapted to the new climate regimes, leading to widespread die-offs and range shifts (Sykes and Prentice, 1995; Jump et al., 2006). Dire scenarios include biome collapses, crop failure, a loss of carbon sequestration, and widespread weed invasions (Patterson et al., 1999; Peng et al., 2004). Weedy species from warmer regions are capable of rapid migration and can quickly invade ecosystems where native dominants are stressed, thereby aggravating problems arising from climate change. Acclimation could allow natives to adjust to climate warming and maintain themselves, and high genetic diversity in key traits could allow for rapid adaptation of resident populations to climate change (Bradshaw and Holzapfel, 2001; van Dijk and Hautekeete, 2007). While there is uncertainty and controversy regarding the future (e.g. Xu et al., 2007), there is little controversy over the need to understand the mechanisms controlling biotic responses to high temperature. Such understanding will improve predictive power while providing directions for adaptation and mitigation. At the current time, the mechanisms controlling the response of net CO2 assimilation rate (A) to elevated temperature remain unclear and controversial (CraftsBrandner and Salvucci, 2004; Schrader et al., 2004). As a result, it is difficult to interpret patterns of acclimation and adaptation to elevated temperature and to screen for variation in the critical traits that limit high temperature carbon gain. Numerous hypotheses have been proposed to explain the nature of the photosynthetic controls at high temperature, and each has substantial experimental support. The leading hypotheses for photosynthetic limitation above the thermal optimum are heat lability of Rubisco activase on the one hand, and a limitation in electron transport on the other (Salvucci and Crafts-Brandner, 2004a, b; Cen and Sage, 2005; Sharkey, 2005; Sage and Kubien, 2007). Other possibilities may be present but have not been sufficiently evaluated; for example, the contribution of the mesophyll diffusive conductance to photosynthetic limitation remains uncertain (Yamori et al., 2006b; Flexas et al., 2007). This review addresses the leading mechanisms proposed to limit photosynthesis above the thermal optimum, and the difficulties in identifying a specific process as a major control over carbon gain. To overcome these difficulties, it is argued that high temperature research should emphasize a holistic approach that simultaneously assesses whole leaf gas exchange and the biochemical capacity of the major
Rubisco and climate change 1583
limitations, rather than photo- and day respiration, dominates the thermal response of photosynthesis (Fig. 1f; see also Sharkey, 1988; von Caemmerer, 2000). Diffusion limitations
Two categories of diffusion limitation on A are also recognized. The first is diffusion of CO2 from the atmosphere and through the boundary layer and stomata
to the intercellular spaces. The conductance to CO2 diffusion by the stomata is the major regulatory control over this limitation, and hence it is emphasized in gas exchange analyses. Stomatal limitations tend to be greater at elevated temperature for two reasons. First, the increase in the CO2 response slope of Rubisco-limited A and RuBPregeneration-limited A at elevated temperatures enhances the degree to which a given stomatal conductance limits A (Farquhar and Sharkey, 1982; Sage and Sharkey,
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
Fig. 1. The modelled temperature response of the net (Anet, continuous lines) and gross (Agross, dotted lines) CO2 assimilation rates, and the RuBP consumption rate at three CO2 levels. (a), (c), and (e) are absolute responses, and (b), (d), and (f) are the same responses divided by the net CO2 assimilation rate at the thermal optimum. At 25 C the parameters for each curve were Vcmax¼80 lmol m2 s1, Jmax¼150 lmol m2 s1, Tp¼10 lmol m2 s1, and the day respiration rate (Rd)¼1 lmol m2 s1. Temperature corrections follow Bernacchi et al. (2001) for Rubisco-limited A, Bernacchi (2003) for RuBPregeneration-limited A, and Hendrickson et al. (2004) for triose-phosphate use-limited A. The rate of RuBP consumption is the sum of the Rubisco carboxylation and oxygenation rates, where the oxygenation rate, Vo, was calculated as: Vo¼(2C*/Ci)3[min(Wc, Wj,Wp)] where Wj is the electron transport-limited carboxylation rate and Wp is the triose-phosphate use-limited rate of carboxylation. Model details are described in Appendix 1.
1584 Sage et al.
Biochemical limitations The predominant controls over the photosynthetic thermal response are the biochemical components of the photosynthetic apparatus. For clarity, the main biochemical processes proposed to limit A at elevated temperatures are segregated into three general categories. The first category represents limitations predicted by the family of models derived from the theoretical interpretations of Farquhar et al. (1980) as modified by Sharkey (1985) and coworkers (Harley and Sharkey, 1991) to account for feedback effects from starch and sucrose synthesis (for various versions of the Farquhar et al. temperature model, see also Kirschbaum and Farquhar, 1984; von Caemmerer, 2000, Bernacchi et al., 2001, 2003; Medlyn et al., 2002; Cen and Sage, 2005; Hikosaka et al., 2006). These models universally assume constant and high activation states of Rubisco, and divide the limitation on A into the capacity of Rubisco to consume RuBP, the capacity of light harvesting and electron transport to regenerate RuBP, and the capacity of starch and sucrose synthesis to regenerate Pi for photophosphorylation. The second category represents control over photosynthesis by the heat lability of Rubisco activase. Heat lability of Rubisco activase has not been widely incorporated into the Farquhar et al. series of models, but can easily be by adjusting the fully activated carboxylation capacity (Wc) to account for the number of deactivated catalytic sites of Rubisco, as shown in equation (1) (Sage, 1990):
Wc # ¼ ½Vcmax ðactÞ ðCÞ=½CþKc ð1 þ O=Ko Þ
ð1Þ
In equation (1), the in vivo carboxylation rate of Rubisco (Wc#) is a function of the maximum catalytic capacity of Rubisco (Vcmax), the activation state of Rubisco (act, in ratio form, not as a percentage), the stromal CO2 concentration (C), the Michaelis constants for carboxylation (Kc) and oxygenation (Ko), and the O2 concentration in the stroma (O). The value of Wc# is then used to model A as shown by Farquhar et al. (1980) in place of the fully activated carboxylation capacity. The activation state of Rubisco is easily determined by rapidly extracting Rubisco and measuring its initial activity relative to the activity of the same extract after incubation with saturating levels of CO2 and magnesium (Sage et al., 1993). There is as yet no means to model the relationship between the activation state of Rubisco and the capacity of Rubisco activase. Temperature and the limitations arising from the capacities of Rubisco versus RuBP regeneration
The family of models derived from the two-limitation model of Farquhar et al. (1980) has provided a comprehensive ability to explain the biochemical controls over the photosynthetic temperature response—assuming the underlying assumptions are met. As such, they need to be
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
1987). To avoid an increase in stomatal limitation, the stomatal conductance would generally have to increase with temperature. Secondly, as temperatures increase, the vapour pressure difference (VPD) between leaf and air rises exponentially. This rise in VPD can reduce stomatal conductance, causing a reduction in intercellular CO2 and A as temperatures increase (Berry and Bjorkman, 1980; Fredeen and Sage, 1999). Declines in A due to high VPDinduced stomatal closure are common on hot afternoons, and explain much of the phenomenon known as midday stomatal closure (Pons and Welschen, 2003; Tay et al., 2007). At relatively low VPD (40 C) than those where electron transport and A are inhibited by rising temperature (>35 C; Berry and Bkorkman, 1980; Yamasaki et al., 2002; Sharkey and Schrader, 2006). Instead, the mechanisms are proposed to arise from increased proton leakiness across the thylakoid membrane, or altered interactions between membranes and the thylakoid protein complexes (Bukhov et al., 1999; Sharkey and Schrader, 2006). In response to these immediate effects, cyclic electron flow is proposed to be up-regulated at the
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
considered first in order to provide context for deviations that may occur should Rubisco activase become limiting. At low CO2, the Rubisco capacity to consume RuBP is the predominant limitation on photosynthesis across a wide range of temperatures (Kirschbaum and Farquhar, 1984; von Caemmerer, 2000; Hikosaka et al., 2006). The Rubisco limitation generally corresponds to CO2 levels that are below the Rubisco Km for CO2. When an enzyme experiences substrate levels below the Km, it typically has a low thermal dependence because the Km and Vcmax have similar temperature responses (Berry and Raison, 1981). The Q10 values for the Km and Vcmax of Rubisco are near 2, except at low temperature (
SPECIAL ISSUE REVIEW PAPER
Rubisco, Rubisco activase, and global climate change Rowan F. Sage1,*, Danielle A. Way1 and David S. Kubien2 1 2
Department of Ecology and Evolutionary Biology, University of Toronto, Toronto, ON, Canada M5S 3B2 Department of Biology, University of New Brunswick, Fredericton, NB, Canada E3B 5A3
Received 11 September 2007; Revised 28 January 2008; Accepted 5 February 2008
Abstract
Key words: Black spruce, climate change, C3 photosynthesis, Picea mariana, Rubisco, Rubisco activase, temperature.
Introduction The earth’s climate is predicted to warm by an average of 2–4 C during the next century as a result of increased greenhouse gases in the atmosphere (IPCC, 2007). Much of the climate warming associated with greenhouse gas enrichment will occur at higher latitudes, higher elevations, and during the winter. In Canada, for example, the boreal region is predicted to warm by 3–10 C (IPCC, 2007). Also higher atmospheric CO2 levels tend to reduce stomatal conductance and transpiration, thereby lowering latent heat loss and causing higher leaf temperatures (Kimball and Bernacchi, 2006; Bernacchi et al., 2007). Change in the atmospheric composition and climate will therefore increase the temperature of photosynthetic tissues, particularly in high latitude ecosystems such as the boreal forest. During periods of the year when plants normally experience optimal temperatures, it can also be expected that the new climate regimes will warm leaves well above their thermal optimum, potentially to a point where electron transport and Rubisco activase are heat labile. Such warming will reduce photosynthetic capacity, possibly offsetting gains in carbon assimilation associated with elevated CO2. Because climate warming will increase the frequency of supra-optimal temperatures, it is important to understand the limitations on photosynthesis above the photosynthetic thermal optimum, both in the short-term (minutes to hours), as may occur during a heat wave, and over the long term (days to years). Long-term photosynthetic
* To whom correspondence should be addressed. E-mail: [email protected] Abbreviations: A, net CO2 assimilation rate; Ci, the intercellular partial pressure of CO2; PSII, photosystem II; VPD, vapour pressure difference. ª The Author [2008]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
Global warming and the rise in atmospheric CO2 will increase the operating temperature of leaves in coming decades, often well above the thermal optimum for photosynthesis. Presently, there is controversy over the limiting processes controlling photosynthesis at elevated temperature. Leading models propose that the reduction in photosynthesis at elevated temperature is a function of either declining capacity of electron transport to regenerate RuBP, or reductions in the capacity of Rubisco activase to maintain Rubisco in an active configuration. Identifying which of these processes is the principal limitation at elevated temperature is complicated because each may be regulated in response to a limitation in the other. Biochemical and gas exchange assessments can disentangle these photosynthetic limitations; however, comprehensive assessments are often difficult and, for many species, virtually impossible. It is proposed that measurement of the initial slope of the CO2 response of photosynthesis (the A/Ci response) can be a useful means to screen for Rubisco activase limitations. This is because a reduction in the Rubisco activation state should be most apparent at low CO2 when Rubisco capacity is generally limiting. In sweet potato, spinach, and tobacco, the initial slope of the A/ Ci response shows no evidence of activase limitations at high temperature, as the slope can be accurately modelled using the kinetic parameters of fully activated Rubisco. In black spruce (Picea mariana), a reduction in the initial slope above 30 C cannot be explained by the known kinetics of fully activated Rubisco, indicating that activase may be limiting at high temperatures. Because black spruce is the dominant species in the boreal forest of North America, Rubisco activase may be an unusually important
factor determining the response of the boreal biome to climate change.
1582 Sage et al.
processes, and uses theoretical models of photosynthesis to explain observed gas exchange patterns in terms of the biochemical controls. Such approaches have provided clarity in the understanding of the biochemical controls over the light response to photosynthesis, and the acclimation of C3 photosynthesis to elevated CO2 (von Caemmerer and Farquhar, 1981; Long et al., 2004).
Patterns of the C3 photosynthetic response to temperature C3 photosynthesis typically exhibits a thermal optimum between 20 C and 35 C. The optimum is broad at low CO2, with a shallow peak usually centred in the mid20 C range (Fig. 1a). As CO2 levels increase, the optimum shifts to higher temperatures, and the peak sharpens, such that at high CO2, photosynthesis exhibits a very sharp peak centred around 30–35 C in most species (Fig. 1e; Berry and Raison, 1981; Sage, 2002). Species acclimated and/or adapted to cooler temperatures exhibit lower thermal optima, while high temperature-adapted species can exhibit thermal optima exceeding 35 C at elevated CO2 (Berry and Raison, 1981; Sage and Kubien, 2007). A change in the shape of the thermal response of photosynthesis depends upon the individual temperature responses of the diffusion and biochemical limitations controlling the rate of photosynthesis. Also, the temperature responses of photorespiration and dark respiration contribute to the thermal response of A. As temperatures increase, the relative rates of photorespiration and dark respiration increase relative to the in vivo capacity for Rubisco carboxylation. This pattern is demonstrated in Fig. 1, which shows a modelled temperature response of the net CO2 assimilation rate (A), the gross CO2 assimilation rate, and the rate of RuBP consumption at three different CO2 levels: 180 lbar (the atmospheric CO2 level during the last ice age 18 000 years ago), the current CO2 level of 380 lbar, and a future, high CO2 level of 700 lbar. The RuBP consumption rate is the sum of the carboxylation and oxygenation rates of Rubisco. The difference between gross photosynthesis and RuBP consumption reflects the inhibition associated with photorespiration (arrow 1 in Fig. 1b). The difference between A and gross CO2 assimilation indicates the limitation on carbon gain associated with day respiration in the mitochondria (arrow 2 in Fig. 1b). At low CO2, the rise in day respiration with temperature has a proportionally larger effect due to the relatively low photosynthetic rate (Fig. 1a, b). Also, the relative effect of photorespiration is much greater at low CO2 due to the relative lack of CO2 to inhibit RuBP oxygenation (compare parts b and f in Fig. 1). Inhibition of photorespiration becomes substantial at elevated CO2 levels such as 700 lbar. Under these conditions, the effect of the underlying biochemical
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
responses can be divided into two categories. First, for individuals, long-term exposure to new thermal regimes can lead to acclimation of the photosynthetic apparatus, allowing for stable or even increased rates of photosynthesis in the new, warmer growth regimes. Secondly, long-term exposure to elevated temperatures over multiple generations leads to natural selection and adaptation of populations to higher temperatures. A major concern for the future is that species will not be adapted to the new climate regimes, leading to widespread die-offs and range shifts (Sykes and Prentice, 1995; Jump et al., 2006). Dire scenarios include biome collapses, crop failure, a loss of carbon sequestration, and widespread weed invasions (Patterson et al., 1999; Peng et al., 2004). Weedy species from warmer regions are capable of rapid migration and can quickly invade ecosystems where native dominants are stressed, thereby aggravating problems arising from climate change. Acclimation could allow natives to adjust to climate warming and maintain themselves, and high genetic diversity in key traits could allow for rapid adaptation of resident populations to climate change (Bradshaw and Holzapfel, 2001; van Dijk and Hautekeete, 2007). While there is uncertainty and controversy regarding the future (e.g. Xu et al., 2007), there is little controversy over the need to understand the mechanisms controlling biotic responses to high temperature. Such understanding will improve predictive power while providing directions for adaptation and mitigation. At the current time, the mechanisms controlling the response of net CO2 assimilation rate (A) to elevated temperature remain unclear and controversial (CraftsBrandner and Salvucci, 2004; Schrader et al., 2004). As a result, it is difficult to interpret patterns of acclimation and adaptation to elevated temperature and to screen for variation in the critical traits that limit high temperature carbon gain. Numerous hypotheses have been proposed to explain the nature of the photosynthetic controls at high temperature, and each has substantial experimental support. The leading hypotheses for photosynthetic limitation above the thermal optimum are heat lability of Rubisco activase on the one hand, and a limitation in electron transport on the other (Salvucci and Crafts-Brandner, 2004a, b; Cen and Sage, 2005; Sharkey, 2005; Sage and Kubien, 2007). Other possibilities may be present but have not been sufficiently evaluated; for example, the contribution of the mesophyll diffusive conductance to photosynthetic limitation remains uncertain (Yamori et al., 2006b; Flexas et al., 2007). This review addresses the leading mechanisms proposed to limit photosynthesis above the thermal optimum, and the difficulties in identifying a specific process as a major control over carbon gain. To overcome these difficulties, it is argued that high temperature research should emphasize a holistic approach that simultaneously assesses whole leaf gas exchange and the biochemical capacity of the major
Rubisco and climate change 1583
limitations, rather than photo- and day respiration, dominates the thermal response of photosynthesis (Fig. 1f; see also Sharkey, 1988; von Caemmerer, 2000). Diffusion limitations
Two categories of diffusion limitation on A are also recognized. The first is diffusion of CO2 from the atmosphere and through the boundary layer and stomata
to the intercellular spaces. The conductance to CO2 diffusion by the stomata is the major regulatory control over this limitation, and hence it is emphasized in gas exchange analyses. Stomatal limitations tend to be greater at elevated temperature for two reasons. First, the increase in the CO2 response slope of Rubisco-limited A and RuBPregeneration-limited A at elevated temperatures enhances the degree to which a given stomatal conductance limits A (Farquhar and Sharkey, 1982; Sage and Sharkey,
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
Fig. 1. The modelled temperature response of the net (Anet, continuous lines) and gross (Agross, dotted lines) CO2 assimilation rates, and the RuBP consumption rate at three CO2 levels. (a), (c), and (e) are absolute responses, and (b), (d), and (f) are the same responses divided by the net CO2 assimilation rate at the thermal optimum. At 25 C the parameters for each curve were Vcmax¼80 lmol m2 s1, Jmax¼150 lmol m2 s1, Tp¼10 lmol m2 s1, and the day respiration rate (Rd)¼1 lmol m2 s1. Temperature corrections follow Bernacchi et al. (2001) for Rubisco-limited A, Bernacchi (2003) for RuBPregeneration-limited A, and Hendrickson et al. (2004) for triose-phosphate use-limited A. The rate of RuBP consumption is the sum of the Rubisco carboxylation and oxygenation rates, where the oxygenation rate, Vo, was calculated as: Vo¼(2C*/Ci)3[min(Wc, Wj,Wp)] where Wj is the electron transport-limited carboxylation rate and Wp is the triose-phosphate use-limited rate of carboxylation. Model details are described in Appendix 1.
1584 Sage et al.
Biochemical limitations The predominant controls over the photosynthetic thermal response are the biochemical components of the photosynthetic apparatus. For clarity, the main biochemical processes proposed to limit A at elevated temperatures are segregated into three general categories. The first category represents limitations predicted by the family of models derived from the theoretical interpretations of Farquhar et al. (1980) as modified by Sharkey (1985) and coworkers (Harley and Sharkey, 1991) to account for feedback effects from starch and sucrose synthesis (for various versions of the Farquhar et al. temperature model, see also Kirschbaum and Farquhar, 1984; von Caemmerer, 2000, Bernacchi et al., 2001, 2003; Medlyn et al., 2002; Cen and Sage, 2005; Hikosaka et al., 2006). These models universally assume constant and high activation states of Rubisco, and divide the limitation on A into the capacity of Rubisco to consume RuBP, the capacity of light harvesting and electron transport to regenerate RuBP, and the capacity of starch and sucrose synthesis to regenerate Pi for photophosphorylation. The second category represents control over photosynthesis by the heat lability of Rubisco activase. Heat lability of Rubisco activase has not been widely incorporated into the Farquhar et al. series of models, but can easily be by adjusting the fully activated carboxylation capacity (Wc) to account for the number of deactivated catalytic sites of Rubisco, as shown in equation (1) (Sage, 1990):
Wc # ¼ ½Vcmax ðactÞ ðCÞ=½CþKc ð1 þ O=Ko Þ
ð1Þ
In equation (1), the in vivo carboxylation rate of Rubisco (Wc#) is a function of the maximum catalytic capacity of Rubisco (Vcmax), the activation state of Rubisco (act, in ratio form, not as a percentage), the stromal CO2 concentration (C), the Michaelis constants for carboxylation (Kc) and oxygenation (Ko), and the O2 concentration in the stroma (O). The value of Wc# is then used to model A as shown by Farquhar et al. (1980) in place of the fully activated carboxylation capacity. The activation state of Rubisco is easily determined by rapidly extracting Rubisco and measuring its initial activity relative to the activity of the same extract after incubation with saturating levels of CO2 and magnesium (Sage et al., 1993). There is as yet no means to model the relationship between the activation state of Rubisco and the capacity of Rubisco activase. Temperature and the limitations arising from the capacities of Rubisco versus RuBP regeneration
The family of models derived from the two-limitation model of Farquhar et al. (1980) has provided a comprehensive ability to explain the biochemical controls over the photosynthetic temperature response—assuming the underlying assumptions are met. As such, they need to be
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
1987). To avoid an increase in stomatal limitation, the stomatal conductance would generally have to increase with temperature. Secondly, as temperatures increase, the vapour pressure difference (VPD) between leaf and air rises exponentially. This rise in VPD can reduce stomatal conductance, causing a reduction in intercellular CO2 and A as temperatures increase (Berry and Bjorkman, 1980; Fredeen and Sage, 1999). Declines in A due to high VPDinduced stomatal closure are common on hot afternoons, and explain much of the phenomenon known as midday stomatal closure (Pons and Welschen, 2003; Tay et al., 2007). At relatively low VPD (40 C) than those where electron transport and A are inhibited by rising temperature (>35 C; Berry and Bkorkman, 1980; Yamasaki et al., 2002; Sharkey and Schrader, 2006). Instead, the mechanisms are proposed to arise from increased proton leakiness across the thylakoid membrane, or altered interactions between membranes and the thylakoid protein complexes (Bukhov et al., 1999; Sharkey and Schrader, 2006). In response to these immediate effects, cyclic electron flow is proposed to be up-regulated at the
Downloaded from jxb.oxfordjournals.org by guest on September 24, 2010
considered first in order to provide context for deviations that may occur should Rubisco activase become limiting. At low CO2, the Rubisco capacity to consume RuBP is the predominant limitation on photosynthesis across a wide range of temperatures (Kirschbaum and Farquhar, 1984; von Caemmerer, 2000; Hikosaka et al., 2006). The Rubisco limitation generally corresponds to CO2 levels that are below the Rubisco Km for CO2. When an enzyme experiences substrate levels below the Km, it typically has a low thermal dependence because the Km and Vcmax have similar temperature responses (Berry and Raison, 1981). The Q10 values for the Km and Vcmax of Rubisco are near 2, except at low temperature (
