Journal of Experimental Botany, Vol. 53, No. 369, pp. 591-607,
April 1, 2002
© 2002 Oxford University Press
Review Article |
Overexpression of C4-cycle enzymes in transgenic C3 plants: a biotechnological approach to improve C3-photosynthesis
1 Botanik II, Botanisches Institut der Universität zu Köln, Gyrhofstrasse 15, D-50931 Cologne, Germany
2 Institut für Biologie I, RWTH, Worringer Weg 1, D-52074 Aachen, Germany
Received 8 November 2001; Accepted 18 December 2001
 |
Abstract |
|---|
 Top Abstract IntroductionThe role of C4-cycle...Progress to date in...Future perspectives and possible...References |
|---|
The process of photorespiration diminishes the efficiency of CO2 assimilation and yield of C3-crops such as wheat, rice, soybean or potato, which are important for feeding the growing world population. Photorespiration starts with the competitive inhibition of CO2 fixation by O2 at the active site of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) and can result in a loss of up to 50% of the CO2 fixed in ambient air. By contrast, C4 plants, such as maize, sugar cane and Sorghum, possess a CO2 concentrating mechanism, by which atmospheric CO2 is bound to C4-carbon compounds and shuttled from the mesophyll cells where the prefixation of bicarbonate occurs via phosphoenolpyruvate carboxylase (PEPC) into the gas-tight bundle-sheath cells, where the bound carbon is released again as CO2 and enters the Calvin cycle. However, the anatomical division into mesophyll and bundle-sheaths cells (‘Kranz’-anatomy) appears not to be a prerequisite for the operation of a CO2 concentrating mechanism. Submerged aquatic macrophytes, for instance, can induce a C4-like CO2 concentrating mechanism in only one cell type when CO2 becomes limiting. A single cell C4-mechanism has also been reported recently for a terrestrial chenopod. For over 10 years researchers in laboratories around the world have attempted to improve photosynthesis and crop yield by introducing a single cell C4-cycle in C3 plants by a transgenic approach. In the meantime, there has been substantial progress in overexpressing the key enzymes of the C4 cycle in rice, potato, and tobacco. In this review there will be a focus on biochemical and physiological consequences of the overexpression of C4-cycle genes in C3 plants. Bearing in mind that C4-cycle enzymes are also present in C3 plants, the pitfalls encountered when C3 metabolism is perturbed by the overexpression of individual C4 genes will also be discussed.
Key words: Biotechnology, crop yield, transgenic C3 plants.
|
Introduction |
|---|
|
TopAbstractIntroductionThe role of C4-cycle...Progress to date in...Future perspectives and possible...References |
|---|
Photorespiration decreases the efficiency of CO2 assimilation in C3 plants
Most of our crops, such as wheat, rice, soybean or potato are classified as C3 plants as the first product of atmospheric CO2 fixation is the 3-carbon compound 3-phosphoglycerate (3-PGA), which is produced in the Calvin cycle by Rubisco (the only enzyme capable of net carbon assimilation) in the chloroplast stroma. However, competition of O2 with CO2 at the active site of Rubisco (Chen and Spreitzer, 1992
; Jordan and Ogren, 1984) results in a loss of up to 50% of the carbon fixed in a process known as photorespiration (Ogren, 1984). Oxygenation of ribulose-1,5-bisphosphate (RubP) severely diminishes the efficiency of CO2 assimilation in C3 plants in ambient air and results in the formation of 3-PGA as well as 2-phosphoglycollate (2-PG). The latter is metabolized in three compartments of the leaf cell, the chloroplast, the peroxysomes and the mitochondria, involving numerous enzymatic reactions and transport processes (Fig. 1). The overall photorespiratory cycle is also linked to amino acid metabolism in that glycine, serine, glutamate, and glutamine are metabolized at high rates (Keys et al., 1978). Both CO2 and ammonia are released at equal rates in the reaction catalysed by the mitochondrial glycine decarboxylase complex (Oliver, 1994). The loss of CO2 during photorespiration is reflected in a CO2 compensation point (
) of CO2 assimilation of between 40–60 µl l-1 CO2 in the intercellular air space. At the CO2 compensation point, net CO2 assimilation equals CO2 release through photorespiration and mitochondrial respiration in the light. In high CO2 and/or low O2 the oxygenase activity of Rubisco is virtually absent, the flux through the photorespiratory carbon cycle negligible and the CO2 compensation point close to zero.
|
There are numerous reports on the improvement of growth and crop yield of C3 plants in an atmosphere containing elevated CO2 (Arp et al., 1998
; Besford, 1990; Chen et al., 1997; Teramura et al., 1990). This is mainly based on a faster biomass production due to an increase in CO2 assimilation rates and a suppression of photorespiration.
C4 plants have developed strategies to concentrate CO2 in the vicinity of Rubisco
During the evolution of higher plants, adaptations to low water supply and/or hot environments have developed independently several times (Edwards et al., 2001; Kellogg, 1999; Ku et al., 1996). Plants, which display C4-metabolism, release CO2 at high rates in the vicinity of Rubisco and thereby increase the ratio of RubP carboxylation/oxygenation substantially (Leegood, 1997, 2002). This strategy prevents major losses of CO2 by photorespiration and is accompanied by an increase in the water and nitrogen use efficiency compared to C3 plants (Sage and Pearcy, 1987). Common C4 plants are characterized by the so-called ‘Kranz’-anatomy with mesophyll cells surrounded by relatively thin cell walls and bundle-sheath cells surrounded by thick cell walls. This anatomical separation into different cell types is accompanied by a spatial separation of the prefixation of atmospheric CO2 in the mesophyll cells followed by the release of CO2 and its refixation via the C3 (Calvin) cycle in the bundle-sheath cells. Moreover, in order to keep intercellular diffusion ways short, mesophyll cells are in close proximity to the bundle-sheath (Dengler and Nelson, 1999). The first step in the C4-cycle is the carboxylation of PEP by phosphoenolpyruvate carboxylase (PEPC) in the cytosol of the mesophyll cells using HCO-3 as the inorganic carbon substrate (Cooper and Wood, 1971; O'Leary, 1982). This yields the C4 dicarboxylic acid oxaloacetate (OAA), which is either reduced to malate or transaminated to aspartate and is then transported into the bundle-sheath cells, where CO2 is released at high rates by decarboxylating enzymes. Depending on the type of C4 plants, CO2 is released either by chloroplastic NADP malic enzyme (ME), mitochondrial NAD-ME, or cytosolic phosphoenolpyruvate carboxykinase (PEPCK) (Hatch and Osmond, 1976; Leegood, 2002) (Fig. 2). Since bundle-sheath cells have low gas permeability, the CO2 concentration in solution is drastically increased, which causes a suppression of the Rubisco oxygenase activity and, consequently, photorespiration (Leegood, 1997). Pyruvate released by NADP-ME or NAD-ME is transferred back into the mesophyll chloroplasts where the inorganic carbon acceptor PEP is regenerated by pyruvate, Pi dikinase (PPDK). The reaction, catalysed by PPDK, consumes in addition two ATP per CO2 assimilated. One ATP is directly required by PPDK, the second ATP is needed for the conversion of the reaction product AMP into ADP catalysed by adenylate kinase (Fig. 2). The C4 strategy permits high rates of CO2 assimilation at a relatively small stomatal aperture and thereby increases the water use efficiency. Moreover, due to an almost complete suppression of RubP oxygenation the CO2 compensation point is close to zero or very low as compared to C3 plants.
|
C4 plants lacking ‘Kranz’ anatomy
Since the discovery of the C4 cycle some 36 years ago the spatial separation into mesophyll cells and bundle sheath cells was thought to be a prerequisite for an efficient CO2-concentrating mechanism. It is therefore quite amazing that the submerged aquatic plant Hydrilla verticillata was identified as being capable of inducing a C4-like metabolism (Fig. 3A), but lacking the ‘Kranz’-anatomy typical for C4 plants (Holaday and Bowes, 1980; Magnin et al., 1997; Salvucci and Bowes, 1981, 1983; Spencer et al., 1996). The switch from C3- to C4-like metabolism in H. verticillata is triggered by low CO2 concentrations (i.e. at high water temperatures) resulting in an increase in PEPC, ME and PPDK activities, which causes a substantial drop in the CO2 compensation point from around 40 µl l-1 in the C3-state down to below 10 µl l-1 in the C4-induced state (Spencer et al., 1996; Reiskind et al., 1997) (Fig. 3A). Further evidence for a C4-like metabolism without ‘Kranz’ anatomy emerged for other aquatic species such as Egeria densa (Casati et al., 2000) and (probably) Elodea canadensis (de Groote and Kennedy, 1977). Moreover, apart from the inducible C4-photosynthesis these aquatic species exhibit a pH-polarity of their leaf surfaces. The steady-state pH of the adaxial surface is acidic (pH 4.0), whereas the abaxial side of the leaf is alkaline (pH 10.0) (van Ginkel et al., 2001). An acidic environment allows higher CO2 concentrations at equilibrium despite a drop in the total inorganic carbon in the surrounding water. This supplementary mechanism to increase the availability of external CO2 is also apparent in the C4-induced state of H. verticillata.
|
Interestingly, single cell C4 photosynthesis appears not to be restricted to aquatic macrophytes. Recent determinations of
13C values (a measure for the activity of PEPC compared to Rubisco, which discriminates 13CO2 in favour of 12CO2) combined with anatomical studies revealed that the terrestrial chenopod Borszczowia aralocaspica (a halophyte with succulent leaves adapted to a semi-dry environment) is likely to carry out C4 metabolism in only one cell type (Freitag and Stichler, 2000). Moreover, the individual chlorenchyma cells of B. aralocaspica carry two types of chloroplasts with a distinct enzyme equipment and capability to produce starch. Immunolocalization of enzyme proteins revealed that Rubisco (as well as starch) is localized in chloroplasts of the basal part of the chlorenchyma cells (i.e. closer to the vascular bundles), whereas PPDK is localized in the chloroplasts of the distal parts of the cells (i.e. closer to the intercellular air space). Like Rubisco, NAD malic enzyme, which appears to act as the decarboxylating enzyme is also more abundant in the mitochondria of the basal parts of the chlorenchyma cells. PEPC is equally distributed throughout the cytosol (Vozesenskaya et al., 2001).
Crassulacean acid metabolism (CAM) plants also exhibit a single cell CO2-concentrating mechanism
A different strategy of a single cell CO2-concentrating mechanism is realized in Crassulacean acid metabolism (CAM) plants, which also show an adaptation to a minimized water supply. In CAM plants both the prefixation of HCO-3 by PEPC and the release of CO2 by decarboxylating enzymes is temporally separated (Osmond, 1978; Cockburn et al., 1979; Spalding et al., 1979). HCO-3 is prefixed in the dark by PEPC at low temperatures and open stomata and malic acid, which accumulates in the vacuole during the night is decarboxylated during the next day, when stomata are closed, either by cytosolic NADP-ME in concert with mitochondrial NAD-ME or by cytosolic PEPCK. This mechanism efficiently reduces the loss of water and also results in a high CO2 environment in the mesophyll during the day and thereby suppresses photorespiration.
Attempts to introduce a single cell CO2-concentrating mechanism into terrestrial crops
In laboratories around the world attempts to introduce single cell C4-like features into terrestrial C3 plants by a transgenic approach are in progress (Matsuoka et al., 2001). It is believed that the introduction of an intracellular CO2 pump might improve the efficiency of C3 photosynthesis by a substantial suppression of photorespiration. There has been progress in single, double and multiple overexpressions of C4-cycle enzymes in C3 crops. In particular, very high expression levels could be achieved in transgenic rice (Ku et al., 1999). However, physiological alterations caused by this approach have been addressed only in a few reports. As C4-cycle enzymes are also common in C3 plants, although with much smaller activities, the introduction of individual C4-cycle genes might perturb not only primary metabolism, but in certain cases also secondary metabolism (Häusler et al., 2001).
Although a single cell C4-like system appears to operate in aquatic macrophytes and also in a terrestrial chenopod adapted to semi-dry climates, it is still an open question whether a biotechnologically inserted CO2 pump in terrestrial C3 crops can efficiently increase the carboxylation/oxygenation ratio of Rubisco. It is questionable whether the additional CO2 released from the products of PEP carboxylation can be retained within the mesophyll or whether it is just lost to the atmosphere, because of high diffusion rates of CO2 in the airspace of the leaves (Leegood, 2002). In order to match the requirements of the metabolic environment of individual C3 crop species, future research might also focus more on genetically engineered C4-cycle enzymes driven by more specific promoters. In this review there will be a focus on biochemical and physiological consequences of the overexpression of C4-cycle genes in the C3 plants Solanum tuberosum and Nicotiana tabacum. The progress in the genetic manipulation of rice plants, particularly the achievement of extremely high expression levels will be dealt with in a second review.
|
The role of C4-cycle enzymes in C3 plants |
|---|
|
TopAbstractIntroductionThe role of C4-cycle...Progress to date in...Future perspectives and possible...References |
|---|
Exceptionally, all enzymes and metabolite transporters involved in the C4 pathway also occur in C3 plants, although at much lower activities and different tissue specificities. Before the question as to the feasibility of the approach to establish a C4-like cycle in C3 plants is tackled, one has to be aware of the physiological functions of the C3 forms of the individual enzymes and transporters summarized in Table 1
.
|
PEPC [EC 4.1.1.31]
As in C4 plants, PEPC in C3 plants is subject to complex regulation by metabolites and covalent modification by reversible phosphorylation (Andreo et al., 1987
; Van Quy et al., 1991; Duff and Chollet, 1995; Zhang et al., 1995; Chollet et al., 1996). In certain types of heterotrophic tissues from developing fruits and seeds of C3 plants, the CO2 respired is recaptured by PEPC. These tissues contain relatively high PEPC activities (reviewed by Latzko and Kelly, 1983). In leaves of C3 plants one role of PEPC is the supply of carbon skeletons for amino acid biosynthesis following nitrate assimilation (Andrews, 1986; Melzer and O'Leary, 1987) (Fig. 4). During nitrate reduction in the cytosol OH- ions are generated. As both processes, nitrate reduction and malate formation, are closely correlated the production of OAA via PEPC and the consecutive formation of malic acid may buffer cytosolic pH (Martinoia and Rentsch, 1994). Moreover, PEPC also plays an important role in stomatal movement. An increased influx of potassium ions into the guard cells during stomatal opening and the concomitant efflux of protons via the plasma membrane-bound ATP-dependent proton pump would result in a rapid alkalization of the cytosol (Raschke et al., 1988). This is counteracted by the synthesis of malic acid from neutral sugars or transitory starch via PEPC. For instance, a substantial decrease in the stomatal aperture has been reported for Vicia faba after the application of DCDP (3,3-dichloro-2-dihydroxyphosphinoyl-2-propeoate) a specific inhibitor of PEPC (Asai et al., 2000).
|
NADP-ME [EC 1.1.1.40]
NADP-ME contributes to a huge variety of metabolic pathways in green and non-green tissues of C3 plants as reviewed previously (Edwards and Andreo, 1992). It is found in leaves, etiolated tissues, seeds, roots, fruits, and tubers (potato) in the chloroplast as well as in the cytosol. In fruits, NADP-ME is involved in ripening and in the de-acidification of the vacuole as well as in the provision of reducing equivalents and carbon skeletons for sucrose biosynthesis via gluconeogenesis. This is comparable to the situation in CAM plants in the light. Together with PEPC, NADP-ME might also serve as a pH-stat and in combination with NAD malate dehydrogenase (MDH) could be involved in the conversion of NADH to NADPH. In oil-storing tissues, such as seeds of rapeseed or wheat germ, NADP-ME is engaged in the proliferation of reducing power and carbon skeletons for fatty acid biosynthesis.
Leaves of solanaceous species, such as potato and tobacco, contain substantial activities of cytosolic NADP-ME (Knee et al.,1996). There are indications that the cytosolic enzyme is associated with vascular bundles, particularly with developing xylem and internal phloem (Schaaf et al., 1995) suggesting that NADPH produced by this reaction is required for lignin biosynthesis. Moreover, NADP-ME appears to be involved in stress responses (Casati et al., 1999). In leaf tissues, the expression of NADP-ME is increased severely by wounding in combination with glutathionine treatment (Schaaf et al., 1995).
NAD-ME [EC 1.1.1.39]
NAD-ME is localized in the mitochondria and in C3 plants is involved in the anaplerotic carbon supply for amino acid biosynthesis (Fig. 4
). Malate formed by PEPC can either be used in the citric acid cycle or it can be decarboxylated to provide pyruvate and subsequently acetyl-CoA (Douce and Neuburger, 1989).
NADP-MDH [EC 1.1.1.82]
In leaves of C3 plants, NADP-MDH plays a central role in shuttling excessive redox equivalents from the chloroplast into the cytosol (Scheibe, 1987). This so-called malate valve operates at high light intensities, when electron transport rates and NADPH generation exceed the demands for CO2 assimilation and thereby prevents overreduction of the stroma and/or increased Mehler-peroxidase reaction (Polle, 1996). NADP-MDH is subjected to reversible activation by thioredoxin-m and thereby responds to the redox state of the stroma.
PEPCK [EC 4.1.1.39]
As reviewed recently (Leegood et al., 1999) plant PEPCK is a cytosolic enzyme and is basically involved in ATP-dependent gluconeogenetic PEP production from OAA. In C3 plants, high PEPCK activities were detected during germination of oil-storing seeds, during fruit ripening, in phloem-associated cells, and in developing seeds. PEPCK is also found in trichomes of tobacco and cucumber leaves and is believed to be involved in the production of secondary metabolites. There also appears to be a function of PEPCK in plant defence reactions.
PPDK [EC 2.7.9.1]
In C3 plants, low PPDK activities were reported for a number of different tissue types and, depending on the species, it is found only in chloroplasts or in both chloroplasts and the cytosol (Aoyagi and Bassham, 1984; Nomura et al., 2000). The function of PPDK in C3 plants is less clear. However, for guard cells of Vicia faba it has been proposed that chloroplastic PPDK in concert with cytosolic NADP-ME could play a role in gluconeogenetic PEP generation from pyruvate during stomatal closure (Schnabl, 1981).
Metabolite transporters
High fluxes of C4 photosynthesis are accompanied by transport processes across membranes, particularly the inner chloroplast envelope. In the mesophyll cells, pyruvate enters the chloroplast via a pyruvate H+ symporter as a substrate for PPDK (Flügge et al., 1985). Likewise, PEP generated within the chloroplasts has to get access to the cytosol via a PEP/phosphate translocator (PPT). Furthermore, there are significantly high rates of OAA/malate exchange required for a fast production of malate inside the mesophyll chloroplasts. Moreover, in NADP-ME C4 plants, malate must enter the bundle-sheath chloroplasts. Following oxidative decarboxylation of malate by NADP-ME, pyruvate formed by this reaction has to be exported from the chloroplast.
Pyruvate imported into chloroplasts of C3 plants can be used for fatty acid biosynthesis or the production of branched chain amino acids. A rapid malate/OAA exchange across the inner envelope is required for the malate valve in C3 plants. However, the PPT is the only metabolite transporter also involved in C4 metabolism, which has been cloned from a variety of non-C4-tissues and is found at low activities in the inner envelope membrane of chloroplasts and non-green plastids of C3 plants as well as in maize roots (Fischer et al., 1997). Most likely, the PPT provides plastids with PEP as a substrate for the shikimate pathway (Streatfield et al., 1999). Import of PEP (rather than export) is a prerequisite for aromatic amino acid biosynthesis, which is localized entirely within this organelle and results in a huge variety of secondary products downstream of phenylalanine (Schmid and Amrhein, 1995).
The need for an accelerated metabolite exchange across the chloroplast envelope as observed in C4 plants ought to be considered as well in attempts to introduce C4-like features into C3 plants. Therefore genes encoding the relevant plastidial metabolite transporters, apart from the PPT, should be identified and isolated.
|
Progress to date in the physiological and biochemical characterization of C3 plants overexpressing C4-cycle genes |
|---|
|
TopAbstractIntroductionThe role of C4-cycle...Progress to date in...Future perspectives and possible...References |
|---|
As C4-cycle enzymes (and metabolite transporters) have distinct functions in C3 plants (Table 1
), an increase in their activities by individual overexpression is likely to perturb metabolism or to trigger compensational changes in metabolic fluxes. Compensational changes in metabolism as a response to the overexpression of C4-cycle enzymes are interesting to study, as these changes may point to not yet unravelled regulatory interrelationships between different metabolic pathways. It is conceivable that a fully operational single cell C4-cycle circumvents such metabolic perturbations.
Overexpression of PEPC is the first step for establishing a C4 cycle in C3 plants
The efficient fixation of atmospheric CO2 by PEPC is a prerequisite for the insertion of a single cell C4 cycle in C3 plants. The first cDNAs of the C4-type PEPC sequences for cloning were obtained from maize (Izui et al., 1986) and Flaveria trinervia (Poetsch et al., 1991). A successful introduction of the maize PEPC into transgenic tobacco plants under the control of its own promoter or the mesophyll-specific promoter of the chlorophyll a/b binding protein gene (cab) was achieved (Hudspeth et al., 1992) and later on by using the constitutively expressing cauliflower mosaic virus (CaMV) 35S promoter (Kogami et al., 1994; Benfey and Chua, 1990). In these types of transgenic plants PEPC activity was increased slightly more than 2-fold. Apart from an increase in the malate contents, there were no apparent effects on the rate of CO2 assimilation. However, in contrast to the wild type, the quantum yield for CO2 assimilation appeared to be unaffected by increasing temperatures in one transgenic line (Kogami et al., 1994). The lack of decrease in the quantum yield with increasing temperature suggested a refixation of respired CO2 by PEPC in the transgenics. However, the CO2 compensation point measured in a sealed chamber remained unaffected in transgenic tobacco plants compared to the control plants (Hudspeth et al., 1992; Kogami et al., 1994).
In rice, high expression of the maize PEPC was achieved recently using the complete maize PEPC gene including exons and introns and its own promoter for transformation and resulted in a 110-fold increase in PEPC activity measured in vitro (Ku et al., 1999). The reported decline in O2 inhibition of CO2 assimilation would have been consistent with an attenuation of photorespiration. However, a more detailed analysis of these plants suggested that this effect was likely to be based on phosphate limitation of photosynthesis under conditions that promote high photosynthetic fluxes (Fukayama et al., 2001; Matsuoka et al., 2000). Still, it is hard to conceive as to why phosphate limitation of photosynthesis should occur with an increase in PEPC activitiy. Provided that PEP is generated via glycolysis starting from triose phosphates exported from the chloroplast in the light, the phosphate stoichiometry would be balanced as for each triose phosphate exported one phosphate is released from PEP by PEPC and could serve as a counter exchange substrate regardless of whether the flux into sucrose biosynthesis was slowed down or not (Fig. 4). If phosphate limitation was the cause for lower CO2 assimilation rates under optimum conditions, it is therefore more likely that it occurs at the site of cytoplasmic glyceraldehyde-3-phosphate dehydrogenase, leading to the subsequent formation of ATP by phosphoglycerate kinase. This reaction sequence could (at least temporarily) deplete cytoplasmic phosphate pools required for triose phosphate counter exchange. It is also conceivable that additional OAA formed by PEPC is reduced to malate in the stroma and hence competes with reducing equivalents needed for 3-PGA reduction. However, to the knowledge of the authors, neither of these alternative explanations has been addressed experimentally.
As potato was chosen as a model plant for one laboratory study, cDNAs and genomic clones of the endogenous C3-type PEPC were isolated (Merkelbach et al., 1993). However, it was then decided to use PEPC from a bacterial source rather than the endogenous enzyme from potato or other plants. Bacterial enzymes lack, for instance, the regulatory properties of plant enzymes, such as the covalent modification by phosphorylation. Overexpression of two bacterial PEPC genes from Escherichia coli and Corynebacterium glutamicum under the control of the CaMV 35S promoter resulted eventually in a 5-fold increase in PEPC activity in potato leaves with the enzyme from C. glutamicum (Gehlen et al., 1996). Depending on the composition of the in vitro assay an increase of up to 20-fold could be determined (Häusler et al., 2001). Similar to the report on the PEPC overexpressors of tobacco (Hudspeth et al., 1992; Kogami et al., 1994), the absolute rates of photosynthetic CO2 assimilation and electron transport were not severely affected in potato plants overexpressing the bacterial PEPC or in plants with an antisense repression of the endogenous enzyme. Likewise, the CO2 compensation point (), determined in a closed chamber was unaffected in the trangenics. However, dark CO2 release after illumination was considerably increased in PEPC overexpressors and diminished in the antisense plants (Gehlen et al., 1996). This suggested that PEPC affects the rate of respiratory CO2 release. More recently it could be demonstrated that the CO2 compensation point (*) independent of dark respiration in the light (Rd) (determined according to Brooks and Farquhar, 1985), was appreciably diminished in PEPC overexpressors and slightly increased in potato plants with an antisense repression of the endogenous PEPCs (Fig. 5) (Häusler et al., 1999). Rd (the rate of dark respiration in the light) was increased in a range of PEPC overexpressors and slightly decreased in the antisense plants. As * solely reflects the kinetic properties of Rubisco (Brooks and Farquhar, 1985), but not accounts for carboxylation of PEP by O2-insensitive PEPC, the changes in * were interpreted as changes in the CO2/O2 ratio in the vicinity of Rubisco, which was most likely due to an increased release of CO2 from products of PEP carboxylation. In theory, even more CO2 can be deliberated from PEP carboxylation products than is actually fixed via PEPC in the form of HCO-3. If, for instance, malate were completely oxidized in the citric acid cycle (Fig. 4), this would result in the release of four CO2 for each carboxylation of PEP on the expense of photoassimilates. However, all intermediate stoichiometry of HCO-3 uptake and CO2 release might occur as well. An enhanced CO2 release by decarboxylating enzymes would result in an efficient HCO-3/CO2 pump (provided that the activity of cytosolic carbonic anhydrase is absent or low), which could increase the CO2 concentration in the mesophyll cells as well as in the vicinity of Rubisco leading to a slight inhibition of photorespiration. These theoretical considerations were reinforced by the observation that overexpression of PEPC induced the endogenous cytosolic NADP-ME of potato plants by a factor of 4–6 (Häusler et al., 2001). This induction was accompanied by increased activities of a putative cytosolic pyruvate kinase and a presumably cytosolic NADP isocitrate dehydrogenase (Chen, 1998) suggesting that an increased activity of cytosolic NADP-ME in potato leaves initiates CO2 release from PEP carboxylation products (Häusler et al., 2001).
|
In another approach to increase PEPC activity, the endogenous potato enzyme was modified in a way that the phosphorylation site was mutated, which yielded an enzyme with a higher affinity for PEP and lowered sensitivity towards malate inhibition. A detailed report on the analysis of these plants is in preparation by Rademacher et al. Transgenic potato plants with a 4–5-fold increase in the activity of the modified version of potato PEPC also exhibited a pleiotropic increase in endogenous cytosolic NADP-ME (Häusler et al., 2001
). The sole increase in PEPC activity therefore appears to perturb metabolic fluxes and can lead to a waste of photoassimilates (T Rademacher, personal communication).
Stomatal movement is affected in PEPC transgenics of potato
PEPC from C. glutamicum was expressed under the control of the CaMV 35S promoter. Hence the enzyme is constitutively expressed in most tissues and it was very likely to be also present in stomatal guard cells. Interestingly, stomatal opening was accelerated in PEPC overexpressors and delayed in PEPC antisense plants compared to the wild type (Gehlen et al., 1996). The average half time for stomatal opening was 3 min, 9 min, and 6 min for PEPC overexpressors, the antisense plants and the wild type, respectively. Furthermore, oscillations in stomatal conductance were also observed in PEPC overexpressors. These side-effects, underlining the role of PEPC during stomatal movement (see also Asai et al., 2000), deserve a more thorough investigation.
Pleiotropic effects on secondary metabolism
Overexpression of PEPC from bacterial and plant sources not only affects primary metabolism, but also appears to have an effect on the abundance of secondary plant products. The contents of UV protectants (flavonoids), for instance, were significantly smaller in potato plants overexpressing PEPC from C. glutamicum or the mutated version of the endogenous enzyme (Häusler et al., 2001) and were also lower in analogous transgenic tobacco plants (Jun Li, unpublished data). This might be explained in terms of a limitation on PEP import into the chloroplast. PEPC might compete with the initial steps of the shikimate pathway for PEP, which serves as one of the precursors, and thereby diminishes the flux through the shikimate pathway. This is an hypothesis which needs to be proven, particularly in comparison with plants overexpressing PPDK, which are capable of producing additional PEP within the chloroplasts.
Overexpression of decarboxylating enzymes
The introduction of PEPC activity is only the first step in establishing an intracellular CO2 pump in C3 plants. Depending on the desired mode of CO2 release either PEPCK or NADP malic enzyme could serve as decarboxylases within the chloroplasts. The simplest way to release CO2 would be to decarboxylate OAA by PEPCK (Fig. 3B). This would have the advantage of generating PEP as the substrate for PEPC without any further enzymatic steps, but would also require (i) a unidirectional import of OAA into the chloroplasts and an efficient export of PEP from the chloroplasts.
For the second alternative, the decarboxylation of malate by NADP-ME, either the unidirectional transport of OAA or of malate would be required as malate could either be formed from OAA by stromal NADP-MDH or by extraplastidial NAD-MDH isozymes. Furthermore, as NADP-ME produces pyruvate within the chloroplast, an additional conversion into PEP by PPDK or an alternative enzyme is necessary.
Overexpression of PEPCK
PEPCK is a cytosolic enzyme. It is therefore essential to target it to the chloroplast by fusion to an appropriate chloroplast targeting peptide. A cytosolic overexpression of PEPCK together with PEPC would result in a futile cycle. Recently, Susuki et al. have reported that an overexpression of PEPCK in rice can lead to a substantial activity of 3 U mg-1 chlorophyll (Suzuki et al., 2000). The PEPCK gene from Urochloa panicoides was fused to the Rubisco small subunit (rbcS) transit sequence and expressed under the control of the maize PEPC or PPDK promoter. Elegant carbon isotope feeding experiments revealed that a higher flux of 14CO2 into C4 compounds was presumably caused by increased cytosolic PEP levels resulting in a higher flux through PEPC. These findings suggest that the PPT allows a significant efflux of PEP from the chloroplast in rice plants. It also implies that OAA can enter the chloroplast without any known counter exchange substrate. Moreover, by feeding radiolabelled malate the same authors could show that CO2 released by the action of PEPCK enters the Calvin cycle and causes a 3-fold increase in the labelling of 3-PGA and sucrose compared to the controls. The rates of CO2 assimilation and the CO2 compensation point remained unaffected. The lack of effect on photosynthesis is consistent with data obtained for transgenic tobacco plants (Häusler et al., 2001) overexpressing the PEPCK gene from Sinorhizobium meliloti (Osteras et al., 1995) fused to the rbcS stromal transit sequence. However, unlike in rice, PEPCK activity could only be detected reliably in isolated chloroplasts of transgenic tobacco plants (Häusler et al., 2001). The introduction of the same construct in potato plants was not successful.
Overexpression of chloroplastic NADP-ME
For the second approach, the increase in NADP-ME, the cDNA of chloroplastic NADP-ME from the C3 plant Flaveria pringlei (Lipka et al., 1994) was used to transform potato and tobacco plants under the control of the double CaMV 35S pomoter (Lipka et al., 1999; Häusler et al., 2001). However, due to high activities of endogenous cytosolic NADP-ME in solanaceous species combined with relatively poor expression levels, the presence of the introduced chloroplastic enzyme could only be confirmed in chloroplast extracts (Lipka et al., 1999). It was increased 5-fold with reference to chloroplast protein. Overexpression of the C3 chloroplast enzyme had no large effects on photosynthetic performance of transgenic potato plants. Single NADP-ME overexpressors of tobacco were also generated, but have not been analysed on a physiological level.
High expression of the maize C4-type NADP-ME in rice under the control of either the cab or CaMV 35S promoter have been reported independently by two Japanese groups (Takeuchi et al., 2000; Tsuchida et al., 2001). The transgenic rice plants were severely compromized in growth and exhibit photoinhibition and photodamage combined with a decline in chlorophyll contents probably due to an excessive NADPH production within the stroma in the light. As an indicator for the stromal NADPH/NADP ratio the activation state of stromal NADP-MDH (i.e. the in vivo activity as a fraction of the activity of the fully activated enzyme) was doubled in the transgenic lines compared to the wild type (Tsuchida et al., 2001). Moreover, in one approach the ultrastructure of the chloroplasts appeared to be agranal, similar to bundle-sheath chloroplasts of NADP-ME C4 plants (Takeuchi et al., 2000). From attempts to determine *, a perturbation of the stromal redox states was also proposed for potato plants overexpressing NADP-ME (Häusler et al., 2001). However, due to the relatively low activity of the introduced NADP-ME, the transgenic potato lines lacked any visible phenotype.
Overexpression of enzymes that regenerate PEP from pyruvate
If CO2 is released by overexpressed NADP-malic enzyme, the pyruvate formed has to be converted to PEP as the substrate for PEPC. This could be achieved by overexpressing PPDK or bacterial PEP synthetase (PEPS), an enzyme catalysing a similar reaction as PPDK but without PPi formation. As outlined earlier, PEP also serves as a substrate for the shikimate pathway, which is localized entirely in the chloroplast. Chloroplasts and most non-green plastids lack the ability to produce PEP via glycolysis. Thus, PEP has to be imported from the cytosol by the PPT. It is therefore conceivable that the additional formation of PEP by overexpressed PPDK or PEPS could stimulate the flux through the shikimate pathway. In order to prevent these unpredictable side-effects, PPDK may be overexpressed in the cytosol (Sheriff et al., 1998).
Overexpression of PPDK
A successful overexpression of PPDK in C3 plants has been reported for tobacco (Sheriff et al., 1998), potato (Ishimaru et al., 1998) Arabidopsis (Ishimaru et al., 1997), and rice (Nomura et al., 2000; Fukuyama et al., 2001). Overexpression of the C4 maize PPDK gene in potato (Ishimaru et al., 1998) resulted in about a 5-fold increase in activity and caused a depletion in pyruvate content and a slight elevation of PEP content. Malate content was also considerably increased, presumably because of higher fluxes through PEPC and subsequent malic acid formation. For rice plants, a drastic increase in PPDK activity was achieved when the maize PPDK gene, including introns and exons as well as its own promoter and terminator, was used for transformation. Despite the fact that approximately 35% of the total leaf protein consisted of introduced PPDK, the transformants lacked a visible phenotype or changes in growth or fertility (Fukayama et al., 2001). Transgenic potato and tobacco plants overexpressing the C4-type PPDK gene from Flaveria trinervia under the control of the 35S promoter have also been generated in the laboratory and await a thorough analysis.
Overexpression of PEPS
As plant PPDK is highly regulated (i.e. by reversible phosphorylation), the PEPS gene from E. coli was isolated and characterized (Niersbach et al., 1992). Potato plants were generated overexpressing PEPS targeted to the chloroplasts by fusion to the rbcS transit sequence (Panstruga et al., 1997). There were no large effects on photosynthetic CO2 assimilation and electron transport. However, transpiration rates were slightly higher compared to the wild type and stomatal closure was delayed appreciably indicating a perturbation in the metabolism of stomatal guard cells (Panstruga et al., 1997). There were some specific effects on the contents of certain amino acids in the transgenics. The leaf contents of glutamate were lowered by 25% whereas aspartate and asparagine contents were increased by more than 30% compared to the wild type. Contents of branched chain amino acids deriving from pyruvate were also altered. The reason for these changes has not been further elucidated (R Panstruga et al., unpublished information). PEPS overexpressors were compromised in shoot fresh weight, final heights of the plants and tuber yield, and leaves turned slightly yellowish and appeared wilted when the plants were grown in a non-air-conditioned greenhouse during temporary hot periods in the summer (R Panstruga et al., unpublished observations). However, these effects were absent when plants were grown under a controlled temperature and light regime in a growth cabinet.
Overexpression of the PPT
One prerequisite for an operational C4-cycle in C3 plants is the efficient export of PEP formed in the chloroplast by PPDK or PEPCK targeted to the chloroplast. In C3 plants there is only low activity of a chloroplastic PPT (Fischer et al., 1997; however see also Suzuki et al., 2000). Tobacco transformants overexpressing the heterologous PPT from cauliflower buds under the control of the CaMV 35S promoter have been generated and characterized in the laboratory of UI Flügge. In the transformants, transport rates for PEP were increased 10-fold compared to the wild type (L Voll, P Nicolay, K Fischer, RE Häusler, UI Flügge, unpublished results).
Double transformants
The introduction of individual C4-cycle genes in C3 crops can be regarded as a test for their functionality in the transformed plant species. A combined overexpression of C4-cycle genes could provide additional information on whether or not these enzymes may work in concert. As outlined above, single overexpression of C4-cycle genes might perturb metabolism, as all C4 enzymes also have distinct functions in C3 plants. The analysis of double and multiple transformants might help to underline the feasibility of the approach to establish a single cell C4 cycle in C3 crops.
Combined overexpression of PEPC and chloroplastic NADP-ME
In order to release CO2 directly at the site of Rubisco, potato double transformants with increased activities of PEPC and chloroplastic NADP-ME from Flaveria pringlei were generated (Lipka et al., 1999). NADP-ME activity in isolated chloroplasts of the transformants was increased about 5-fold compared to the wild type. These double transformants exhibited a less pronounced temperature-dependent decrease of CO2 assimilation combined with a considerable decline in the electron requirement for CO2 assimilation (i.e. the electron transport rate over apparent CO2 assimilation) at elevated temperatures. Moreover, O2 inhibition of CO2 assimilation was attenuated significantly both at 25 °C and 35 °C (Häusler et al., 2001). Lower rates of electron transport for similar or even higher rates of apparent CO2 assimilation as well as an attenuation of O2 inhibition of CO2 assimilation may point at an appreciable decline in photorespiration rates. However, determinations of * and Rd produced equivocal results for double PEPC/NADP-ME overexpressors (Häusler et al., 2001). * appeared to be decreased from 45 µl l-1 in the wild type to about 20 µl l-1 in the double transformants only at an intermediate photon flux density (PFD), but not at high light or very low PFDs. This suggests that the NADPH produced by chloroplastic NADP-ME might perturb stromal redox potentials, particularly at high PFDs (compare Tsuchida et al., 2001).
From the potato system another interesting aspect emerged. The induction of the endogenous cytosolic NADP-ME as a result of PEPC overexpression was attenuated in double transformants containing PEPC and chloroplastic NADP-ME in combination despite the identical activities of overexpressed PEPC as compared to the single transformants (Häusler et al., 2001). This suggests that a redirection of the metabolic flux into the chloroplast circumvents pleiotropic changes observed in the single transformants and leads to the assumption that a fully operational C4 cycle could be decoupled from the residual metabolism in C3 plants provided that all of its components are expressed in a well-balanced way.
Combined overexpression of PEPC with NADP-ME or PEPCK in tobacco
A moderate overexpression of PEPC from C. glutamicum combined with NADP-ME from F. pringlei or with PEPCK from S. meliloti in tobacco plants had no substantial effects on the photosynthetic performance of the plants (Häusler et al., 2001).
Combined overexpression of PEPC with PPDK in rice plants
Recently Ku et al. reported a 35% increase in the photosynthetic capacity and a 22% increase in grain yield in rice plants overexpressing PEPC and PPDK in combination (Ku et al., 2001). A detailed analysis of these plants is awaited. It is conceivable that this observation is linked to an increased provision of PEP by PPDK for the shikimate pathway inside the chloroplast (Streatfield et al., 1999; Matsuoka et al., 2000), an assumption, which requires experimental support.
Multiple overexpression of C4-cycle enzymes
The introduction of a fully operational C4 cycle might ultimately give the clue as to whether or not C3 photosynthesis can be improved. Multiple overexpressors of potato have been obtained by stepwise transformation with plasmids containing different resistance genes (kanamycin, hygromycin, sulphonylamide, and BASTA® [bar]) for selection. Transgenic lines containing PEPC from C. glutamicum or the mutated endogenous PEPC, combined with NADP-ME, PPDK and the PPT have been generated and await a thorough analysis. For tobacco, transgenic plants overexpressing single C4-cycle genes were genetically crossed and a number of heterozygous lines were obtained that overexpress multiple genes (PEPC, NADP-ME, PPDK, PEPS, PEPCK, as well as the PPT [Fig. 3B]) in various combinations (Jun Li, unpublished results). The progenies of these plants are being screened at the moment for better growth in an atmosphere containing a lowered CO2 concentration.
Species-dependent metabolic responses towards overexpression of C4-cycle genes
Recent data (Häusler et al., 2001) show that even closely related species such as potato and tobacco respond differentially towards an overexpression of C4-cycle genes. For instance, the induction of cytosolic NADP-ME in PEPC overexpressors was less apparent in tobacco transformants compared to transgenic potato plants. Moreover, the most pronounced attenuating effects on photorespiration were observed with PEPC/NADP-ME double transformants of potato plants, but for single PEPC overexpressors of tobacco plants. It is likely that these species-dependent differences in the acceptance for redirecting metabolism by introduced C4-cycle enzymes might hinder a successful introduction of an operational C4-cycle in all of the crop plants intended to be used in this approach.
|
Future perspectives and possible field applications |
|---|
|
TopAbstractIntroductionThe role of C4-cycle...Progress to date in...Future perspectives and possible...References |
|---|
There has been significant progress in the overexpression of the key enzymes of C4-type biochemistry in transgenic C3 plants. Nevertheless, it is still uncertain whether this approach will be sufficient to suppress photorespiration. The following section will discuss possible alternative and/or supplementary approaches to improve a C4-type photosynthesis in C3 plants.
Is a C4-like CO2 pump in C3 plants realistic on the basis of current experimental data?
So far only a limited number of transgenic plants containing a maximum of two C4-cycle enzymes have been investigated physiologically. Most of the observed effects were based on pleiotropic changes in metabolism. The efficiency by which aquatic plants are capable of switching from a C3-type to a C4-like metabolism in single cells when the availability of CO2 declines is encouraging in the prospect that it might be possible to establish a similar system in terrestrial C3 crops by a transgenic approach. Moreover, the terrestrial chenopod B. aralocaspica appears to contain a single cell CO2-concentrating mechanism, distributed between the cytosol, the mitochondria and two types of chloroplasts in only one type of chlorenchyma cells (Freitag and Stichler, 2000; Vozesenskaya et al., 2001). Although this is a very exciting discovery, it ought to be considered that the anatomy of this halophytic chenopod is adapted to semi-dry environments with succulent leaves, central vascular bundles, a hypodermis, large chlorenchyma cells, water storing cells, and, most relevantly, a small intercellular air space. These features of leaf anatomy are particularly different from those of the C3 crops used in this study's transgenic approaches.
In an aquatic system, only diffusion of CO2 through the water and the cell wall resistance for CO2 have to be considered, whereas in terrestrial C3 plants diffusion through the surrounding air, stomatal conductance for CO2, and diffusion through the air space within the leaves have to be taken into account as well (Farquhar et al., 1980; Laisk and Loreto, 1996). In C4 plants the gas-tight bundle-sheath prevents losses of CO2 from the cells and allows high CO2 environment in the vicinity of Rubisco. It is still an open question, whether or not the CO2 released by the introduced HCO-3/CO2 pump can be efficiently retained within the leaf, or whether a good part of it diffuses out of the cells into the intercellular space and would then just result in a waste of energy (see also Leegood, 2002). It might, for instance, become necessary to decrease the free air space in leaves of C3 crops and/or to minimize stomatal aperture.
What about carbonic anhydrase?
The initial step of C4 photosynthesis is the rapid conversion of CO2 into HCO-3 catalysed by carbonic anhydrase (CA) to provide the substrate for carbon fixation by PEPC (Hatch and Burnell, 1990). It is estimated that photosynthesis in C4 plants would be slowed down by a factor of 104 in the absence of cytoplasmic CA (Badger and Price, 1994). However, this aspect has not been addressed so far in biotechnological approaches to transfer C4-like features into C3 plants. Similar considerations as for CA might apply for pyrophosphatase and adenylate kinase activities, which are essential for the metabolism of the PPDK reaction products AMP and PPi in the chloroplast. However, both enzymes are highly active in C3 chloroplasts (Gould and Winget, 1973; Schlattner et al., 1996).
Engineered C4 enzymes
Most approaches so far have been aimed at the introduction of the highest possible activities of the respective enzymes in transgenic C3 plants. This is reasonable because these enzymes are expressed to very high levels in leaves of C4 plants. However, a high expression of the introduced enzyme is not necessarily linked with high in vivo activities. Modulation of enzyme activity by covalent modification (i.e. PEPC, PPDK) and the availability of co-factors and substrates (for review see Leegood, 1997) ought to be considered as well. For instance, C4-specific PEPC is phosphorylated in the light to reduce its sensitivity towards malate inhibition. It is questionable whether the protein kinase required for phosphorylation is expressed in the correct temporal and spatial pattern in C3 plants. Furthermore, C4 PEPC has a low substrate affinity for PEP, which might be unfavourable in the C3 environment. C3 isoforms of PEPC show higher affinities to their respective substrates, but are very sensitive to inhibition by malate (Svensson et al., 1997). The catalytic properties of PEPC from potato have been modified by genetic engineering (T Rademacher et al., unpublished data) in the way that the enzyme possesses both C3 (high affinity towards PEP) and C4 (reduced sensitivity towards product inhibition) features and is not subjected to covalent modification. It appears to be useful and perhaps even necessary to engineer more of the C4-cycle enzymes and/or the respective promoters to be better adapted to the specific requirements of a C4-like cycle in a ‘C3 environment’.
Morphological adaptations
The introduction of a ‘true C4 cycle’ into C3 plants would ultimately require alterations in the leaf anatomy. There are numerous deviations from the classical C3 pattern of leaf anatomy associated with the C4 syndrome in both mono- and dicotyledonous plants (Nelson and Langdale, 1992). However, most of these features are not absolutely essential to perform C4 photosynthesis (Dengler and Nelson, 1999). For example, the suberin lamella around the bundle-sheath cells and the agranal ultrastructure of bundle-sheath chloroplasts observed in NADP-ME type C4 plants like maize are absent in other C4 species. Eventually, it will be crucial to separate primary carbon fixation and photosynthetic carbon reduction in two different tissue types or in distinct parts of one cell (as realized in B. aralocaspica) which are in intimate contact, and to minimize the leakage of CO2 from the tissue or the part of the cell where Rubisco is active. Bundle-sheath cells with distinct morphology compared to mesophyll cells are also present in C3 plants. In Arabidopsis, reticulate mutants have been isolated with disturbed chloroplast development in mesophyll cells, but intact chloroplasts in the bundle-sheath indicating that the different cell types follow a separated developmental programme (Kinsman and Pyke, 1998). Differences in developmental programmes of mesophyll and bundle-sheath in C3 plants might be utilized to target distinct components of the C4 cycle into the respective cell types. Moreover, the efficiency of metabolite exchange between the different cell types would require closer vein spacing inside the leaf. The signals determining the density of veins are not definitely identified, but there is significant variability within a single organism. For instance, maize foliar leaves show a high vein density significant for C4 plants, whereas the leaf sheath or the husk leaf exhibit distances between the veins similar to C3 plants (Langdale et al., 1988). Refined mutant screens and comparative gene expression analyses will be necessary to identify the factors responsible for changes in leaf anatomy.
Promoters
One very successful strategy is the transfer of complete genes from maize into rice (Ku et al., 1999). This approach leads to very high expression levels and resembles, at least in part, the spatial and light-dependent expression patterns found in C4 plants. Overexpressed C4-cycle enzymes might perturb the metabolism of whole plants if transcription is not restricted to the photosynthetic tissues. The use of endogenous promoters from C3 plants with properties similar to the C4 promoters would be advantageous. Recently, Tsuchida et al. used the rice cab promoter in rice for the overexpression of maize NADP-ME, which led to a strong accumulation of the protein in chloroplasts (Tsuchida et al., 2001). Alternatives for leaf-specific and light-induced gene expression include the use of other photosynthetic promoters like the lhcb promoter (Cerdan et al., 2000) or the rbcS promoter (Kyozuka et al., 1993).
Furthermore, inducible promoters could be used (Gatz and Lenk, 1998; Zuo and Chua, 2000). A comparison of the induced with the non-induced state of expression would aid physiological analysis under controlled conditions.
Field applications
C4 plants are more productive than C3 plants when they are grown under their respective optimum conditions (Brown, 1999). C4 plants exhibit higher water and nitrogen use efficiencies compared to C3 plants, which results in an increased dry matter production (reviewed in Brown, 1999). Concentrating CO2 at the site of Rubisco should allow engineered C3 plants to reduce stomatal conductance under drought conditions without a dramatic decline in the rate of CO2 assimilation (Drake et al., 1997). This would allow both the use of new areas for crop production required for feeding the growing world population and the reduction of inputs into the system (such as fertilizers) and thus conserving natural resources. Both factors are much more relevant to today's necessities than the mere increase in biomass production.
|
Acknowledgments |
|---|
The authors would like to thank Dr Mitsue Miyoa and Hiroshi Fukayama for the kind provision of as yet unpublished data.
|
Notes |
|---|
3 To whom correspondence should be addressed. Fax: +492214705039. E-mail: rainer.haeusler{at}uni\|[hyphen]\|koeln.de
|
Abbreviations |
|---|
*, CO2 compensation point in the absence of dark respiration in the light; CA, carbonic anhydrase; NAD(P)-MDH, NAD(P)-dependent malate dehydrogenase; NAD(P)-ME, NAD(P)-dependent malic enzyme; OAA, oxaloacetate; PEP, phosphoenolpyruvate; PEPC, phosphoenolpyruvate carboxylase; PEPCK, phosphoenolpyruvate carboxykinase; PEPS, phosphoenolpyruvate synthetase; 3-PGA, 3-phosphoglycerate; 2-PG, 2-phosphoglycollate; PFD, photon flux density; PPDK, pyruvate, orthophosphate dikinase; PPT, phosphoenolpyruvate/phosphate translocator; Rd, dark respiration in the light; Rubisco, ribulose 1,5-bisphosphate carboxylase/oxygenase; TPT, triose phosphate/phosphate translocator..