Plant Physiol. (1998) 116: 823-832

Oxygen Requirement and Inhibition of C₄ Photosynthesis¹
An Analysis of C₄ Plants Deficient in the C₃ and C₄ Cycles

João P. Maroco, Maurice S.B. Ku, Peter J. Lea, Louisa V. Dever, Richard C. Leegood, Robert T. Furbank, and Gerald E. Edwards^*

Department of Botany, Washington State University, Pullman, Washington 99164 (J.P.M., M.S.B.K., G.E.E.); Division of Biological Sciences, Lancaster University, Lancaster LA1 4YQ, United Kingdom (P.J.L., L.V.D.); Robert Hill Institute, Department of Animal and Plant Sciences, University of Sheffield, Sheffield S10 2TN, United Kingdom (R.C.L.); and Division of Plant Industry, Commonwealth Scientific and Industrial Research Organization, P.O. Box 1600, Canberra ACT 2601, Australia (R.T.F.)

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

The basis for O₂ sensitivity of C₄ photosynthesis was evaluated using a C₄-cycle-limited mutant of Amaranthus edulis (a phosphoenolpyruvate carboxylase-deficient mutant), and a C₃-cycle-limited transformant of Flaveria bidentis (an antisense ribulose-1,5-bisphosphate carboxylase/oxygenase [Rubisco] small subunit transformant). Data obtained with the C₄-cycle-limited mutant showed that atmospheric levels of O₂ (20 kPa) caused increased inhibition of photosynthesis as a result of higher levels of photorespiration. The optimal O₂ partial pressure for photosynthesis was reduced from approximately 5 kPa O₂ to 1 to 2 kPa O₂, becoming similar to that of C₃ plants. Therefore, the higher O₂ requirement for optimal C₄ photosynthesis is specifically associated with the C₄ function. With the Rubisco-limited F. bidentis, there was less inhibition of photosynthesis by supraoptimal levels of O₂ than in the wild type. When CO₂ fixation by Rubisco is limited, an increase in the CO₂ concentration in bundle-sheath cells via the C₄ cycle may further reduce the oxygenase activity of Rubisco and decrease the inhibition of photosynthesis by high partial pressures of O₂ while increasing CO₂ leakage and overcycling of the C₄ pathway. These results indicate that in C₄ plants the investment in the C₃ and C₄ cycles must be balanced for maximum efficiency.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

Although in C₃ plants the decrease of the O₂ partial pressures from ambient levels (approximately 20 kPa) to approximately 2 kPa can increase the net rate of CO₂ fixation by up to 50% as a result of reduced photorespiration, in C₄ plants no significant effect is generally observed (Edwards and Walker, 1983). This apparent lack of response of C₄ photosynthesis to O₂ led to the early conclusion that C₄ plants are O₂ insensitive and that photorespiration is not apparent. C₄ plants are capable of concentrating CO₂ in the bundle-sheath cells (where Rubisco is exclusively localized) to levels that have been estimated to exceed 3 to 20 times the atmospheric CO₂ concentration (Jenkins et al., 1989; Dai et al., 1993; Hatch et al., 1995; He and Edwards, 1996). Therefore, the ratio of [CO₂] to [O₂] increases in the bundle-sheath cells, and photorespiration is considered insignificant because of the suppression of the oxygenase reaction of Rubisco (Edwards and Walker, 1983; Edwards et al., 1985; Hatch, 1987; Byrd et al., 1992; Dai et al., 1993; Hatch et al., 1995). Even so, measurable rates of photorespiration have been observed in C₄ plants: in maize, from studies of Gly metabolism in leaf discs (Marek and Stewart, 1983), ¹⁸O₂ incorporation in glycolate in intact leaves (deVeau and Burris, 1989), and ¹⁴C incorporation in Gly and Ser in isolated bundle-sheath cells (Farineau et al., 1984); and in Amaranthus edulis, from studies of NH₄⁺ production (Lacuesta et al., 1997). In other studies it may be partially responsible for ¹⁸O₂ uptake in C₄ plants (Furbank and Badger, 1982; Badger, 1985).

Rates of photorespiration in C₄ plants under ambient atmospheric conditions have been estimated at 3 to 7% of the rate of CO₂ fixation (Farineau et al., 1984; deVeau and Burris, 1989; Dever et al., 1995; Lacuesta et al., 1997), and even higher under low CO₂ and/or higher O₂ partial pressures (Farineau et al., 1984; Dai et al., 1993, 1995). Because of the high resistance of the bundle-sheath cells to gas diffusion (Furbank et al., 1989; Jenkins et al., 1989; Byrd et al., 1992; He and Edwards, 1996), it is generally accepted that CO₂ released during photorespiration will be partially refixed by Rubisco. However, estimates of leakage rates of CO₂ from the bundle sheath vary from 10 to 50% of the C₄ cycle flux, depending on the method of analysis or assumptions used in modeling (Farquhar, 1983; Evans et al., 1986; Henderson et al., 1992; Hatch et al., 1995; He and Edwards, 1996). The release of ¹⁴CO₂ from intact leaves of C₄ plants after a pulse with ¹⁴CO₂ was also shown to be consistently higher under 20 kPa O₂ than 2 kPa O₂ (about 8%; see fig. 4 in Hatch et al., 1995). Additionally, O₂ partial pressures in the bundle-sheath cells may be even higher than the atmospheric levels in C₄ plants having PSII activity in the bundle-sheath cells (Furbank et al., 1989), thus increasing the rate of photorespiration.

Generally, there are no significant differences in photosynthetic rates of C₄ plants at 2 versus 20 kPa O₂, even when CO₂ is limiting for photosynthesis (Dai et al., 1993, 1995; Maroco et al., 1997). Even if photorespired CO₂ is partially refixed by Rubisco in the bundle-sheath cells, or by PEPC in the mesophyll cells, when CO₂ is limiting, some inhibition of photosynthesis by O₂ should occur. Because that is not the case (Edwards and Walker, 1983; Edwards et al., 1985; Byrd et al., 1992), some other inhibitory mechanism must operate. Indeed, when the response of net CO₂ fixation is measured under different O₂ partial pressures from 20 kPa to 5 to 10 kPa, a measurable increase in net photosynthesis is observed. Below this O₂ partial pressure, net photosynthesis is then inhibited, with rates at 2 kPa being essentially the same as those at 20 kPa.

This phenomenon was first observed by Ku et al. (1983) in Flaveria trinervia and was then studied in some detail in maize, both NADP-ME species (Dai et al., 1993, 1995). Recently, we have shown that this dual response of O₂ is common to all C₄ photosynthetic plants, including both monocots and dicots (Maroco et al., 1997). Simultaneous gas-exchange and Chl fluorescence measurements under different CO₂ partial pressures suggested that above the optimal O₂ partial pressure, the inhibition of net photosynthesis is associated with photorespiration. Below the optimum, O₂ inhibition is associated with reduced PSII activity and efficiency of electron transport of open centers and possibly with a decrease in ATP supply to the C₄ cycle (Maroco et al., 1997).

Incorporation of ¹⁴CO₂ in C₄ acids in several C₄ species has previously been shown to be stimulated by increasing O₂ partial pressures (Glacoleva and Zalensky, 1978), and an O₂ requirement for maximum CO₂ assimilation has also been observed in C₃ species (Ziem-Hanck and Heber, 1980; Dietz et al., 1985). However, the optimal O₂ partial pressure for photosynthesis is lower in C₃ plants than for the C₃-C₄ intermediate and C₄ photosynthetic types: 1, 2, and 9 kPa, respectively (Dai et al., 1993, 1996). Taken together, these results suggest that compared with C₃ photosynthesis, C₄ photosynthesis requires a higher O₂ partial pressure for maximum photosynthetic CO₂ assimilation. However, it was not understood why C₄ plants have a higher O₂ requirement than C₃ plants (5-10 kPa versus 1-2 kPa), although we speculated that this could be because of the higher ATP demand for operating the C₄ cycle. Because pseudo-cyclic electron transport may at least in part provide extra ATP for the C₄ cycle (Edwards and Walker, 1983; Hatch, 1987; Furbank et al., 1990), a decrease of the O₂ partial pressure could impair this energy supply. Furthermore, increased reduction of electron carriers of the cyclic pathway may also be achieved under near-anaerobic conditions, limiting the production of ATP by cyclic electron transport (Ziem-Hanck and Heber, 1980; Suzuki and Ikawa, 1984a, 1984b, 1993).

To further understand the roles of the C₄ versus the C₃ cycle in the O₂ requirement and inhibition of C₄ photosynthesis, we used a mutant of the C₄ plant A. edulis (NAD-ME) that is deficient in PEPC activity (Dever et al., 1995), and the transgenic plant Flaveria bidentis (NADP-ME), which has reduced levels of Rubisco (Furbank et al., 1996). In this study we show that the higher O₂ requirement of C₄ photosynthesis is associated with the C₄ cycle, since plants deficient in the C₄ isoform of PEPC have O₂ requirements similar to those of C₃ plants (about 1 kPa). Results obtained with the two species also provide further evidence that the inhibition of C₄ photosynthesis by supraoptimal O₂ partial pressures is a result of photorespiration. Transgenic F. bidentis plants with reduced Rubisco activity and increased bundle-sheath CO₂ concentration (von Caemmerer et al., 1997) are less sensitive, whereas PEPC mutants are more sensitive to supraoptimal O₂ partial pressures.

	MATERIALS AND METHODS

Top Abstract Introduction Methods Results Discussion References

Plant Material and Growth Conditions

2

1

2

1

Plant Screening and Enzyme Activity

PEPC activity was determined at 30°C by following the carboxylation of PEP to oxaloacetate and its reduction to malate by malate dehydrogenase coupled with NADH oxidation. The assay medium (total volume of 1 mL) contained 50 mm Tris-HCl, pH 8.0, 10 mm NaHCO₃, 5 mm MgCl₂, 0.1 mm NADH, 2 units of malate dehydrogenase, and 25 µL of the enzyme extract. The reaction was initiated by the addition of 50 µL of 50 mm PEP (final concentration of 2.5 mm) (Sigma).

Rubisco activity was measured radiometrically by the incorporation of H¹⁴CO₃ into acid-stable products. The assay mixture (total volume of 150 µL) contained 50 mm Tris-HCl, pH 8.0, 10 mm MgCl₂, 5 mm DTT, 20 mm NaH¹⁴CO₃ (specific activity of 5.89 × 10⁵ cpm/µmol), and 15 µL of enzyme extract. The assay mixture was incubated in 20-mL glass scintillation vials for 2 min at 30°C, and the reaction was started by the addition of 20 µL of 10 mm ribulose bisphosphate (final concentration of 1.3 mm). After 1 min at 30°C the reaction was stopped with 50 µL of tricarboxylic acid (20%), and the samples were left at room temperature for 10 min and then thoroughly flushed with mild air for 10 min. Ten milliliters of scintillation liquid (Bio-Safe II, Research Products International, Mount Prospect, IL) was added to the samples and the activity counted in a liquid scintillation counter (model LS700, Beckman). Enzyme activity was calculated after correction for background counts and counting efficiency.

Total soluble protein was measured using Coomassie Plus reagent (Pierce) according to the method of Bradford (1976). PEPC and Rubisco (LSU) contents were estimated by densitometric analysis of SDS-PAGE gels of total soluble protein using National Institutes of Health imaging software (Scion, Marlboro, MA). Total Chl was determined by incubation of 40 µL of the crude sample supernatant in 960 µL of absolute ethanol for 2 h in the dark, and then measured according to Wintermans and de Motts (1965).

SDS-PAGE and Western-Blot Analysis

Photosynthetic enzymes, PEPC, Rubisco (LSU and SSU), and carbonic anhydrase were identified by western immunoblotting. Maize PEPC antibody was courtesy of R. Chollet (University of Nebraska, Lincoln), and barley Rubisco SSU and LSU antibodies were courtesy of N.H. Chua (Rockefeller University, New York, NY). After SDS-PAGE, protein was electrotransferred to a nitrocellulose membrane overnight in transfer buffer (150 mm Tris-HCl, pH 8.0, 20 mm Gly, 3 mm SDS, and 5% methanol) at 4°C and 250 mA, with final transfer for 1 h at 800 mA. The membrane was blocked with 5% fat-free dry milk in TBS buffer (20 mm Tris-HCl, pH 7.5, and 0.5 m NaCl) and incubated with shaking for 2 h at room temperature with the antibodies (1:6000 dilution) in the same solution. After washing with TBS buffer, the membrane was incubated with goat anti-rabbit IgG conjugated to alkaline phosphatase for 1 h at room temperature. The immunolocalized bands were then revealed by incubation of the membrane in alkaline phosphatase reaction medium containing 5 mm Tris-HCl, pH 9.5, 0.325 mg/mL nitroblue tetrazolium, and 0.165 mg/mL 5-bromo-4-chloro-3-indolyl phosphate (all reagents and alkaline phosphatase-conjugated secondary antibody were obtained from Bio-Rad).

Gas Exchange and Chl a Fluorescence

2

1

The quantum yield of PSII was calculated as _PSII = (F_m  F_s)/F_m (Genty et al., 1989), and the state of reduction of the Q_A pool was estimated as 1  q_P, where q_P = (F_m  F_s)/(F_m  F_o) is the photochemical quenching (Dietz et al., 1985). The efficiency of PSII open centers for electron transport was calculated as (F_m  F_o)/F_m (Öquist and Chow, 1992). The quantum yield of CO₂ fixation (_CO2 = A/absorbed PPFD) was calculated as the ratio of net CO₂ fixation to PPFD absorbed, assuming a leaf absorptivity of 85% for C₄ plants (Oberhuber et al., 1993; Oberhuber and Edwards, 1993). Dark-type mitochondrial respiration was not included in the calculation because it is not known how this changes in the light under varying O₂.

Statistical Analysis

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Enzyme Activity, SDS-PAGE, and Western Blotting

2

1

2

1

2

1

View this table: [in this window] [in a new window]   Table I. Total soluble protein, Chl, PEPC, and Rubisco content, and PEPC and Rubisco activity in wild type (WT), heterozygous (Pp), and PEPC homozygous mutants (pp) of A. edulis All values except the PEPC and Rubisco contents are the average of three or four replicates, with se values in parentheses. Rubisco and PEPC contents were estimated as described in ``Materials and Methods''. Means with different letter suffixes are statistically significantly different at  = 0.05.

When expressed on a leaf-area basis, the Rubisco content of heterozygous plants was about 10% lower than that in the wild-type plants, and the Rubisco content of the homozygous mutants was about 50% of that in the wild-type plants (P < 0.05). However, when these values were expressed as a percentage of the total soluble protein, no significant differences were found (P > 0.1). SDS-PAGE and analysis of total soluble leaf protein (Fig. 1) for these enzymes confirmed the pattern of enzyme activity, with estimates of PEPC and Rubisco contents within the ranges reported for other C₄ species (Table I) (Schmitt and Edwards, 1981; Sugiyama et al., 1984; Baer and Schrader, 1985).

View larger version (36K): [in this window] [in a new window]   Figure 1. A, Coomassie blue-stained SDS-PAGE gel of soluble leaf protein of A. edulis. WT, Wild type; Pp, heterozygous PEPC mutant; pp, homozygous PEPC mutant; MW, molecular mass in kilodaltons (kD). Thirty-five micrograms of protein was loaded per lane. Arrow indicates the PEPC band. B, Western blot of PEPC, LSU, carbonic anhydrase (CA), and SSU. Twenty-five micrograms of protein was loaded per lane.

The segregation of the SSU insert in F. bidentis was irregular, with a continuous range of Rubisco activity from less than 10% to 100% of that in the wild-type plants (55.0 ± 4.4 µmol m² s¹). This is consistent with a segregation of two independent antisense inserts in the T₁, giving a range of enzyme activities corresponding to 1, 2, 3, and 4 loci of the antisense insert. From this heterogeneous group, a subset of plants exhibiting normal growth and 33% of wild-type Rubisco activity was chosen for further studies. These SSU plants showed an approximately 34% reduction of total soluble protein (expressed on a leaf-area basis) relative to the wild-type plants (P = 0.03) (Table II). However, no statistically significant difference was observed in total Chl content among the segregates. Both Rubisco and PEPC contents were significantly lower in SSU plants than in the wild-type plants (P < 0.01). However, the Rubisco activity was 66% lower, whereas the PEPC activity was only 25% lower in the SSU relative to the wild-type plants (P < 0.001). SDS-PAGE separation of total soluble protein and identification with western-blot analysis confirmed that both LSU and SSU were the main polypeptides significantly reduced in the SSU plants used and that no significant changes were observed in carbonic anhydrase (Fig. 2).

View this table: [in this window] [in a new window]   Table II. Total soluble protein, Chl, PEPC, and Rubisco content, and PEPC and Rubisco activity in wild type (WT) and SSU plants of F. bidentis All values except the PEPC and Rubisco contents are the average of three or four replicates, with se values in parentheses. Rubisco and PEPC contents were estimated as described in ``Materials and Methods''. Means with different letter suffixes are statistically significantly different at  = 0.05.

View larger version (26K): [in this window] [in a new window]   Figure 2. A, Coomassie blue-stained SDS-PAGE gel of soluble leaf protein of F. bidentis. WT, Wild type; MW, molecular mass in kilodaltons (kD). Thirty-five micrograms of protein was loaded per lane. B, Western blot of PEPC, LSU, carbonic anhydrase (CA), and SSU. Twenty-five micrograms of protein was loaded per lane.

Gas Exchange and Chl a Fluorescence

PEPC-Deficient A. edulis

View larger version (33K): [in this window] [in a new window]   Figure 3. O2 effects on the net CO2 assimilation (a), net CO2 assimilation as a percentage of the maximum rates (b), quantum yield of PSII (c), electron use efficiency for CO2 assimilation (d), reduction state of the QA pool (e), and efficiency of PSII open centers (f) in A. edulis wild-type plants. Measurements were made at an ambient CO2 concentrations of 93 () and 33 Pa (), with corresponding intercellular CO2 values of 28.3 ± 3.7 and 15.8 ± 0.9 Pa, respectively. Error bars are the Fisher lsd values at  = 0.05. Error bar without symbol is the Fisher lsd value for the O2 × CO2 interaction.

For ambient CO₂ (33 Pa) and O₂ (20 kPa) partial pressures, inhibition of A by O₂ was approximately 13% of the maximum. Increasing the CO₂ partial pressures to approximately three times ambient levels (93 Pa) greatly reduced the O₂ inhibition to approximately 6% of the maximum (Fig. 3b). Below the optimal O₂ partial pressures, the reduction in A was associated with decreased efficiency of electron transport through PSII reaction centers (Fig. 3c). The increased reduction of the Q_A pool (Fig. 3e) and decreased efficiency of the remaining PSII open centers (Fig. 3f) can explain the observed reduction of the _PSII at suboptimal O₂ levels. The ratio of _CO2/_PSII, which reflects the efficiency of CO₂ fixation relative to PSII activity (Fig. 3d), decreased slightly at supraoptimal O₂ and increased exponentially at low O₂ partial pressures. Thus, the most efficient use of electron flow for CO₂ assimilation is at the lowest O₂ partial pressures.

The decrease of PEPC content and activity in the heterozygous A. edulis plants to about 50% of the wild-type levels (Table I) did not change the dual O₂ effect on A (Fig. 4a). Maximum net photosynthesis rates in the heterozygous plants were approximately 55% of those in the wild-type plants, both at 93 Pa CO₂ and at ambient CO₂ partial pressures (33 Pa). The optimal O₂ partial pressure for A was also shifted to 5 to 10 kPa (Fig. 4a), compared with 2.5 to 5 kPa in the wild type. The inhibition at supraoptimal O₂ partial pressures (20 kPa) and ambient CO₂ (33 Pa) was lower than the inhibition in the wild-type plants (11% versus 13%), but this difference was not statistically significant (P = 0.3) (Fig. 4a). No statistically significant difference was found at approximately three times ambient CO₂ (P = 0.2).

View larger version (33K): [in this window] [in a new window]   Figure 4. O2 effects on the net CO2 assimilation (a), net CO2 assimilation as a percentage of the maximum rates (b), quantum yield of PSII (c), electron use efficiency for CO2 assimilation (d), reduction state of the QA pool (e), and efficiency of PSII open centers (f) in the A. edulis PEPC heterozygous plants. Measurements were made at ambient CO2 concentrations of 93 () and 33 Pa (), with corresponding intercellular CO2 values of 20.5 ± 2.2 and 11.9 ± 0.7 Pa, respectively. Error bars are the Fisher lsd values at  = 0.05. Error bar without symbol is the Fisher lsd value for the O2 × CO2 interaction.

As described for the wild-type plants, a decrease of A at below-optimal O₂ partial pressures in this mutant was associated with the decrease in the _PSII (Fig. 4c). However, low O₂ was not as inhibitory to A and _PSII in the mutant as it was in the wild-type plants. The reduction state of the Q_A pool (Fig. 4e) was similar to the reduction state in the wild-type plants at three times ambient CO₂ partial pressures, but was higher at ambient CO₂ partial pressures. No statistically significant differences were observed in the _CO2/_PSII ratio (Fig. 4d) or in the efficiency of the PSII open centers (Fig. 4f) under varying O₂ at the two CO₂ partial pressures.

The almost total suppression of PEPC in the A. edulis homozygous mutant (Table I) resulted in negative A rates under ambient CO₂ partial pressures (Fig. 5a). At this CO₂ concentration, reducing the O₂ partial pressures from 20 to 10 kPa increased A by approximately 50% (Fig. 5b). However, at ambient CO₂, photorespiration was in excess of CO₂ fixation and so there was no net carbon gain at any O₂ partial pressure. At ambient CO₂ partial pressures the _CO2/_PSII ratio increased by more then 50% from ambient to 10 kPa O₂, and then decreased. A also decreased at lower O₂ partial pressure, possibly because of photoinhibition (Fig. 5d).

View larger version (35K): [in this window] [in a new window]   Figure 5. O2 effects on the net CO2 assimilation (a), net CO2 assimilation as percentage of the maximum rates (b), quantum yield of PSII (c), electron use efficiency for CO2 assimilation (d), reduction state of the QA pool (e), and efficiency of PSII open centers (f) in A. edulis PEPC homozygous mutants. Measurements were made at ambient CO2 concentrations of 93 () and 33 Pa (), with corresponding intercellular CO2 values of 74.0 ± 2.1 and 34.1 ± 0.2 Pa, respectively. Error bars are the Fisher lsd values at  = 0.05. Error bar without symbol is the Fisher lsd value for the O2 × CO2 interaction.

At 93 Pa CO₂, ambient O₂ partial pressures caused an inhibition of net photosynthesis of about 30% of the maximum rate (Fig. 5b). Optimal O₂ partial pressures occurred between 1 and 2 kPa, below which a large decrease in A was observed, as reported for C₃ species (Ziem-Hanck and Heber, 1980; Dietz et al., 1985; Dai et al., 1996). At 93 Pa CO₂, decreasing O₂ from ambient to approximately 1 kPa O₂ caused a statistically significant (P < 0.01), linear increase in A that was also followed by an approximately 2-fold increase in the ratio of _CO2 to _PSII (Fig. 5d). The trend observed in the _PSII response to low O₂ in the wild-type and heterozygous plants was also observed in the homozygous mutant (Fig. 5c). However, in the latter, the _PSII values were three and four times lower than the values in the wild-type and heterozygous plants, respectively. In contrast, the reduction state of the Q_A pool (Fig. 5e) was also much higher (up to four times) than that in the wild-type plants. No apparent effect of O₂ on the efficiency of open centers (Fig. 5f) was observed at 93 Pa CO₂, but a linear decrease was revealed at ambient CO₂ from 20 to about 0 kPa O₂.

SSU F. bidentis

View larger version (36K): [in this window] [in a new window]   Figure 6. O2 effects on the net CO2 assimilation rates (a), net CO2 assimilation as a percentage of the maximum rates (b), quantum yield of PSII (c), electron use efficiency for CO2 assimilation (d), reduction state of the QA pool (e), and efficiency of PSII open centers (f) in F. bidentis wild-type plants. Measurements were done at ambient CO2 concentrations of 9.3 (), 33 (), and 93 Pa (), with corresponding intercellular CO2 values of 2.4 ± 0.1, 11.0 ± 0.3, and 45.9 ± 1.4 Pa, respectively. Error bars are the Fisher lsd values at  = 0.05. Error bar without symbol is the Fisher lsd value for the O2 × CO2 interaction.

The ratio of _CO2 to _PSII increased linearly from ambient down to the optimal O₂ partial pressures and then exponentially for suboptimal O₂ partial pressures (Fig. 6d). Decrease of Rubisco activity to 33% of that of the wild type in the antisense plants (SSU) did not change the inhibition of net photosynthesis to below-optimal O₂ partial pressures (P = 0.08; P < 0.001 when the leaf-to-leaf variation is eliminated by expressing the rates in a relative term). Rather, it limits the effect of above-optimal O₂ partial pressures (Fig. 7a). At approximately one-third ambient CO₂ partial pressures, the inhibition of A by 20 kPa O₂ was about 7% of the maximum. However, at ambient CO₂ (32 Pa) this inhibition was only 2% (compared with 5% in the wild type), and at three times ambient CO₂ this inhibition was nonsignificantly reduced to 1% (Fig. 7b).

View larger version (35K): [in this window] [in a new window]   Figure 7. O2 effects on the net CO2 assimilation rates (a), net CO2 assimilation as a percentage of the maximum rates (b), quantum yield of PSII (c), electron use efficiency for CO2 assimilation (d), reduction state of the QA pool (e), and efficiency of PSII open centers (f) in F. bidentis SSU plants. Measurements were done at ambient CO2 concentrations of 9.3 (), 33 (), and 93 Pa (), with corresponding intercellular CO2 values of 3.1 ± 0.1, 15.3 ± 0.2, and 68.9 ± 2.0 Pa, respectively. Error bars are the Fisher lsd values at  = 0.05. Error bar without symbol is the Fisher lsd value for the O2 × CO2 interaction.

Contrary to what was observed in the wild-type plants, _PSII decreased linearly from high to low O₂ at low CO₂ (9.3 Pa) and decreased just below the optimal O₂ partial pressures for ambient (32 Pa) and high (93 Pa) CO₂ (Fig. 7c). The efficiency of PSII open centers (Fig. 7f) showed the same trend as that described for the _PSII, whereas the reduction state of the Q_A pool (Fig. 7e) was almost constant at ambient and high CO₂, but increased linearly with decreasing O₂ at low CO_2. At the lower CO₂ partial pressure, the ratio of _CO2 to _PSII increased linearly over the whole O₂ range_, whereas at ambient and high CO₂ partial pressures this ratio was almost constant (Fig. 7d).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Because the net rates of photosynthetic CO₂ assimilation are essentially the same at 20 and 2 kPa O₂, it has been generally accepted that C₄ plants are insensitive to O₂. However, we have shown recently that C₄ photosynthesis exhibits a dual response to O₂ from 20 to near 0 kPa, with an optimum around 5 kPa. Below the optimum, the decrease in photosynthesis is associated with decreased PSII activity, whereas above the optimum, photorespiration accounts for the inhibition of photosynthesis (Dai et al., 1995; Maroco et al., 1997). In this study, we evaluated the basis for the dual response of C₄ photosynthesis to O₂ using genetic modifications that limit either the C₃ or the C₄ cycle.

The O₂ Requirement of C₄ Photosynthesis and Its Association with the C₄ Cycle

In wild-type A. edulis plants the reduction state of the Q_A pool was low and essentially the same at the CO₂ partial pressures studied (33 and 93 Pa), with the ratio of _CO2 to _PSII being substantially higher at the higher CO₂ concentration. This suggests that O₂ does act as an alternative electron sink at 33 Pa CO₂, either as the final acceptor of the electron-transport carriers (the Mehler peroxidase reaction) or in photorespiration. In A. edulis heterozygous PEPC plants, the optimal O₂ level for maximum rates of net photosynthesis is slightly higher than that in the wild type (compare Fig. 3, a and b, with Fig. 4, a and b). A higher O₂ requirement for functioning of the electron transport chain in heterozygous plants was also suggested by the linear decrease in the efficiency of PSII open centers (Fig. 4f), with the increased reduction of the Q_A pool occurring only at ambient CO₂ and 0 kPa O₂; however, these differences are probably not significant.

Suppression of the C₄ cycle by a decrease of PEPC activity in A. edulis to levels found in C₃ plants greatly reduces A. Indeed, the A rates in the homozygous mutant are negative at ambient CO₂ partial pressures, and it requires up to three times ambient CO₂ partial pressures to maintain a net gain of carbon that is increased by up to 30% with decreasing O₂. Under ambient conditions, A in the mutant is limited by both photorespiration and bundle-sheath diffusive resistance (increasing the CO₂ concentration up to 30 times the ambient level, 930 Pa, led to photosynthetic rates close to 60% of those observed in the wild-type plants at ambient CO₂; data not shown). At approximately three times ambient CO₂ partial pressure (93 Pa), enough CO₂ apparently diffuses into the bundle-sheath cells to maintain a positive A. Under these conditions, i.e. in a C₃ photosynthetic mode, the O₂ requirement for maximum rates of photosynthesis is similar to that required by C₃ plants. In addition, changes in both A, the reduction state of Q_A, and _PSII in response to O₂ have the same form reported for the C₃ species spinach, sunflower, and Asarum europaeum (Dietz et al., 1985). Because mutant plants deficient in PEPC show O₂ requirements similar to those of C₃ plants, we conclude that the higher O₂ requirement of C₄ photosynthesis is specifically associated with the C₄ function.

Reduced CO₂ Fixation by Rubisco in C₄ Plants May Increase the CO₂ Concentration in the Bundle Sheath and Decrease Photorespiration

In the SSU F. bidentis plants, whereas suboptimal partial pressures of O₂ cause a similar response to that observed in wild-type plants, supraoptimal partial pressures are not so inhibitory to A as for the wild-type plants. Although at low CO₂ partial pressure, photorespiration apparently limits photosynthesis in the SSU plants, at ambient and approximately three times ambient CO₂ partial pressures, photorespiration seems to be suppressed. At 20 kPa O₂, photosynthetic rates are not statistically significantly different from the rates at 5 kPa, with the _CO2/_PSII ratio increasing only slightly from 20 to 5 kPa O₂ at 32 Pa CO₂.

If the rate of the C₄ cycle is not greatly affected in the SSU plants (PEPC activity is only 25% less; Table II) and CO₂ fixation in the bundle sheath is reduced, then a buildup of CO₂ should be expected (Furbank et al., 1996). Indeed, von Caemmerer et al. (1997) observed a higher carbon isotope discrimination in T₁ SSU F. bidentis plants with 40% less Rubisco, and concluded that the CO₂ concentration in the SSU plants was higher than that of the wild-type plants. In this scenario, photorespiration could indeed be reduced, as suggested by the current study. At the same time, the _CO2/_PSII response curves to O₂ are higher in the wild-type than in the SSU plants (Figs. 6d and 7d). This suggests that with a decrease of Rubisco capacity in SSU plants there may be some increase in other electron sinks. In part this could be linked to increased bundle-sheath leakage of CO₂ and overcycling of the C₄ cycle through pseudocyclic (the Mehler peroxidase reaction) ATP production.

In summary, the effect of O₂ on C₄ photosynthesis can be distinguished as two different components: (a) an O₂ requirement specifically associated with the C₄ cycle, and (b) an O₂ inhibition attributable to photorespiration. The strong requirement for O₂ in C₄ photosynthesis, which is apparent when the C₄ cycle is functional, provides support for the concept that this is linked to the O₂-dependent production of ATP by pseudocyclic/cyclic photophosphorylation. This O₂-dependent generation of ATP is probably associated with the extra energy required for regeneration of PEP, the primary substrate of the C₄ cycle. The inhibition of photosynthesis by supraoptimal partial pressures of O₂ may be accounted for largely, if not entirely, by photorespiration. The results of this study with two genetically modified C₄ plants indicate that when the C₄ cycle is deficient (i.e. ineffective in concentrating CO₂), there is an increase in photorespiration, and when the C₃ cycle is deficient, there is an increase in overcycling of the C₄ pathway and an increase in bundle-sheath CO₂ leakage. Thus, C₄ photosynthesis requires a coordinated function of the C₃ and C₄ cycles for maximum efficiency.

	FOOTNOTES

   Received September 4, 1997; accepted November 11, 1997.

	ABBREVIATIONS

Abbreviations: A, net CO₂ assimilation. SSU, antisense Rubisco small subunit. Chl, chlorophyll. F_m, maximum fluorescence level after a saturating light pulse on a dark-adapted leaf. F_m, maximum fluorescence after a saturating light pulse from a leaf during steady-state photosynthesis . F_o, basal fluorescence level on a dark-adapted leaf. F_o, minimum fluorescence from a leaf following steady-state illumination and quickly dark adapted under a pulse of far-red light to fully oxidize PSI. F_s, steady-state fluorescence on an illuminated leaf. LSU, Rubisco large subunit. ME, malic enzyme. PEPC, PEP carboxylase. SSU, Rubisco small subunit. _CO2, quantum yield of CO₂ fixation. _PSII, quantum yield of PSII activity.

	LITERATURE  CITED

Top Abstract Introduction Methods Results Discussion References

Badger MR (1985) Photosynthetic oxygen exchange. Annu Rev Plant Physiol 36: 27-53 [CrossRef][ISI]

Baer GR, Schrader LE (1985) Seasonal changes and genetic variability for DNA concentration, and cellular contents of soluble protein, chlorophyll, ribulose bisphosphate, and pyruvate Pi dikinase in maize leaves. Crop Sci 25: 909-916 [Abstract/Free Full Text]

Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254 [CrossRef][ISI][Medline]

Byrd GT, Brown H, Bouton JH, Basset CL, Black CC (1992) Degree of C₄ photosynthesis in C₄ and C₃-C₄ Flaveria species and their hybrids. I. CO₂ assimilation and metabolism and activities of phosphoenolpyruvate carboxylase and NADP-malic enzyme. Plant Physiol 100: 939-946 [Abstract/Free Full Text]

Cornic G, Briantais JM (1991) Partitioning of photosynthetic electron flow between CO₂ and O₂ reduction in a C₃ leaf (Phaseolus vulgaris L.) at different CO₂ concentrations and during drought stress. Planta 183: 178-184 [ISI]

Dai Z, Ku MSB, Edwards GE (1993) C₄ photosynthesis. The CO₂-concentrating mechanism and photorespiration. Plant Physiol 103: 83-90 [Abstract]

Dai Z, Ku MSB, Edwards GE (1995) C₄ photosynthesis. The effects of leaf development on the CO₂-concentrating mechanism and photorespiration in maize. Plant Physiol 107: 815-825 [Abstract]

Dai Z, Ku MSB, Edwards GE (1996) Oxygen sensitivity of photosynthesis and photorespiration in different photosynthetic types in the genus Flaveria. Planta 198: 563-571 [ISI]

deVeau EJ, Burris JE (1989) Photorespiratory rates in wheat and maize as determined by ¹⁸O-labeling. Plant Physiol 90: 500-511

Dever LV, Bailey KJ, Leegood RC, Lea PJ (1997) Control of photosynthesis in Amaranthus edulis mutants with reduced amounts of PEP carboxylase. Aust J Plant Physiol 24: 469-476 [ISI]

Dever LV, Blackwell RD, Fullwood NJ, Lacuesta M, Leegood RC, Onek LA, Pearson M, Lea PJ (1995) The isolation and characterization of mutants of the C₄ photosynthetic pathway. J Exp Bot 46: 1363-1376 [ISI]

Dietz K-J, Schreiber U, Heber U (1985) The relationship between the redox state of Q_A and photosynthesis in leaves at various carbon-dioxide, oxygen and light regimes. Planta 166: 219-226 [CrossRef]

Edwards G, Ku MSB, Monson RK (1985) C₄ photosynthesis and its regulation. In J Barber, NR Baker, eds, Photosynthetic Mechanisms and the Environment. Elsevier Science Publishers, Amsterdam, pp 287-328

Edwards GE, Walker GA (1983) C₃, C₄: Mechanisms, and Cellular and Environmental Regulation, of Photosynthesis. Blackwell Scientific, Oxford, UK

Evans JR, Sharkey TD, Berry JA, Farquhar GD (1986) Carbon isotope discrimination measured concurrently with gas exchange to investigate CO₂ diffusion in leaves of higher plants. Aust J Plant Physiol 13: 281-292

Farineau J, Lelandais M, Morot-Gaundry J-D (1984) Operation of the glycolate pathway in isolated bundle sheath strands of maize and Panicum maximum. Physiol Plant 60: 208-214

Farquhar GD (1983) On the nature of carbon isotope discrimination in C₄ species. Aust J Plant Physiol 10: 205-226

Furbank RT, Badger MR (1982) Photosynthetic oxygen exchange in attached leaves of C₄ monocotyledons. Aust J Plant Physiol 9: 553-558

Furbank RT, Chitty JA, von Caemmerer S, Jenkins CLD (1996) Antisense RNA inhibition of RbcS gene expression reduces Rubisco level and photosynthesis in the C₄ plant Flaveria bidentis. Plant Physiol 111: 725-734 [Abstract]

Furbank RT, Jenkins CL, Hatch MD (1989) CO₂ concentrating mechanism of C₄ photosynthesis. Permeability of isolated bundle sheath cells to inorganic carbon. Plant Physiol 91: 1364-1371 [Abstract/Free Full Text]

Furbank RT, Jenkins CLD, Hatch MD (1990) C₄ photosynthesis: quantum requirement, C₄ acid overcycling and Q-cycle involvement. Aust J Plant Physiol 17: 1-7

Genty B, Briantais J-M, Baker NR (1989) The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochim Biophys Acta 990: 87-92 [ISI]

Glacoleva TA, Zalensky OV (1978) Oxygen effects and ¹⁴C metabolism in desert plants. Plant Physiol 62: 204-209 [Abstract/Free Full Text]

Hatch MD (1987) C₄ photosynthesis: a unique blend of modified biochemistry, anatomy and ultrastructure. Biochim Biophys Acta 895: 81-106

Hatch MD, Agostino A, Jenkins CLD (1995) Measurement of the leakage of CI2 from bundle-sheath cells of leaves during C₄ photosynthesis. Plant Physiol 108: 173-181 [Abstract]

He D, Edwards GE (1996) Estimation of diffusive resistance of bundle sheath cells to CO₂ from modeling of C₄ photosynthesis. Photosynth Res 49: 195-208 [CrossRef]

Henderson SA, von Caemmerer S, Farquhar GD (1992) Short-term measurements of carbon isotope discrimination in several C₄ species. Aust J Plant Physiol 19: 263-285 [ISI]

Jenkins CLD, Furbank RT, Hatch MD (1989) Mechanism of C₄ photosynthesis. A model describing the inorganic carbon pool in bundle sheath cells. Plant Physiol 91: 1372-1381 [Abstract/Free Full Text]

Krall J, Edwards GE (1992) Quantum yield of photosystem II and of CO₂ fixation in higher plants. Physiol Plant 86: 180-187 [CrossRef]

Ku MSB, Monson RK, Littlejohn RO Jr, Nakamoto H, Fisher DB, Edwards GE (1983) Photosynthetic characteristics of C₃-C₄ intermediate Flaveria species. I. Leaf anatomy, photosynthetic responses to O₂ and CO₂, and activities of key enzymes in the C₃ and C₄ pathways. Plant Physiol 71: 944-948 [Abstract/Free Full Text]

Lacuesta M, Dever LV, Muñoz-Rueda A, Lea PJ (1997) A study of photorespiratory ammonia in the C₄ plant Amaranthus edulis, using mutants with altered photosynthetic capacities. Physiol Plant 99: 447-455 [CrossRef]

Marek LF, Stewart CR (1983) Photorespiratory glycine metabolism in corn leaf discs. Plant Physiol 73: 118-120 [Abstract/Free Full Text]

Maroco JP, Ku MSB, Edwards GE (1997) Oxygen sensitivity of C₄ photosynthesis: evidence from gas exchange and fluorescence analysis with different C₄ sub-types. Plant Cell Environ 20: 1525-1533 [CrossRef]

Oberhuber W, Dai Z, Edwards GE (1993) Light dependence of quantum yields of photosystem II and CO₂ fixation in C₃ and C₄ plants. Photosynth Res 35: 265-274

Oberhuber W, Edwards GE (1993) Temperature dependence of the linkage of quantum yield of photosystem II to CO₂ fixation in C₄ and C₃ plants. Plant Physiol 101: 507-512 [Abstract]

Öquist G, Chow WS (1992) On the relationship between the quantum yield of photosystem II electron transport, as determined by chlorophyll fluorescence and the quantum yield of CO₂-dependent O₂ evolution. Photosynth Res 33: 51-62 [CrossRef]

Schmitt MR, Edwards GE (1981) Photosynthetic capacity and nitrogen use efficiency of maize, wheat and rice: a comparison between C₃ and C₄ photosynthesis. J Exp Bot 32: 459-466 [Abstract/Free Full Text]

Sugiyama T, Mizuno M, Hayashi M (1984) Partitioning of nitrogen among ribulose-1,5-bisphosphate carboxylase/oxygenase, phosphoenolpyruvate carboxylase, and pyruvate orthophosphate dikinase as related to biomass productivity in maize seedlings. Plant Physiol 75: 665-669 [Abstract/Free Full Text]

Suzuki K, Ikawa T (1984a) Effect of oxygen on photosynthetic ¹⁴CO₂ fixation in Chroomonas sp. (Cryptophyta). I. Some characteristics of the oxygen effect. Plant Cell Physiol 25: 367-375 [Abstract/Free Full Text]

Suzuki K, Ikawa T (1984b) Effect of oxygen on photosynthetic ¹⁴CO₂ fixation in Chroomonas sp. (Cryptophyta). II. Effects of inhibitors, uncouplers and an artificial electron mediator on the inhibition of ¹⁴CO₂ fixation by anaerobiosis. Plant Cell Physiol 25: 377-384 [Abstract/Free Full Text]

Suzuki K, Ikawa T (1993) Oxygen enhancement of photosynthetic ¹⁴CO₂ fixation in a freshwater diatom Nitzschia ruttneri. Jpn J Phycol 41: 19-28

von Caemmerer S, Millgate A, Farquhar GD, Furbank RT (1997) Reduction of ribulose-1,5-bisphosphate carboxylase/oxygenase by antisense RNA in the C₄ plant Flaveria bidentis leads to reduced assimilation rates and increased carbon isotope discrimination. Plant Physiol 113: 469-477