Energy Sources for HCO₃⁻ and CO₂ Transport in Air-Grown Cells of Synechococcus UTEX 625

Abstract

Light-dependent inorganic C (C_i) transport and accumulation in air-grown cells of Synechococcus UTEX 625 were examined with a mass spectrometer in the presence of inhibitors or artificial electron acceptors of photosynthesis in an attempt to drive CO₂ or HCO₃⁻ uptake separately by the cyclic or linear electron transport chains. In the presence of 3-(3,4-dichlorophenyl)-1,1-dimethylurea, the cells were able to accumulate an intracellular C_i pool of 20 mm, even though CO₂ fixation was completely inhibited, indicating that cyclic electron flow was involved in the C_i-concentrating mechanism. When 200 μm N,N-dimethyl-p-nitrosoaniline was used to drain electrons from ferredoxin, a similar C_i accumulation was observed, suggesting that linear electron flow could support the transport of C_i. When carbonic anhydrase was not present, initial CO₂ uptake was greatly reduced and the extracellular [CO₂] eventually increased to a level higher than equilibrium, strongly suggesting that CO₂ transport was inhibited and that C_i accumulation was the result of active HCO₃⁻ transport. With 3-(3,4-dichlorophenyl)-1,1-dimethylurea-treated cells, C_i transport and accumulation were inhibited by inhibitors of CO₂ transport, such as COS and Na₂S, whereas Li⁺, an HCO₃⁻-transport inhibitor, had little effect. In the presence of N,N-dimethyl-p-nitrosoaniline, C_i transport and accumulation were not inhibited by COS and Na₂S but were inhibited by Li⁺. These results suggest that CO₂ transport is supported by cyclic electron transport and that HCO₃⁻ transport is supported by linear electron transport.

Cyanobacteria grown under limited C conditions develop a mechanism for the active transport of C_i, which can lead to intracellular C_i concentrations in excess of 1000 times the extracellular C_i concentration (Kaplan et al., 1980; Shelp and Canvin, 1984; Espie and Canvin, 1987; Espie et al., 1988a; Miller et al., 1990; Badger and Price, 1992). CO₂ and HCO₃⁻ are transported by separate and independent systems (Espie et al., 1988a, 1988b; Miller et al., 1988b; Price et al., 1992; Salon et al., 1996a).

Light supplies the energy for active C_i transport, because no transport occurs in the dark (Shelp and Canvin, 1984; Ogawa et al., 1985b; Kaplan et al., 1987; Sültemeyer et al., 1993). Action spectrum studies suggest that PSI is primarily responsible for the supply of energy for transport, but a low level of PSII activity is apparently required for activation of C_i uptake (Ogawa et al., 1985a; Kaplan et al., 1987).

Somewhat puzzling was the observation that DCMU completely inhibited C_i fixation but did not completely inhibit C_i transport and the formation of the intracellular C_i pool (Ogawa et al., 1985a; Kaplan et al., 1987; Miller et al., 1988c; Ritchie et al., 1996). CO₂ transport but not HCO₃⁻ transport was inhibited in a mutant with a defect in cyclic electron flow (Ogawa, 1993). We have also observed significant C_i transport and accumulation when PSI electron acceptors such as PNDA and methyl viologen are supplied (Li and Canvin, 1997b), suggesting that some transport can be supported by linear electron transport.

To investigate the photosynthetic reactions that may be involved in providing energy for CO₂ and HCO₃⁻ transport and accumulation, we examined C_i uptake in the presence of inhibitors and artificial electron acceptors of photosynthesis. Our results suggest that CO₂ transport is supported by cyclic electron transport and HCO₃⁻ transport is supported by linear electron transport.

MATERIALS AND METHODS

Organisms and Growth

The unicellular cyanobacterium Synechococcus leopoliensis UTEX 625 (University of Texas Culture Collection, Austin) was grown at 30°C in modified, unbuffered Allen's medium bubbled with air, as previously described (Espie and Canvin, 1987; Li and Canvin, 1997a). Cells were harvested 48 h after inoculation. At that time, the Chl a concentration was between 4 and 5 μg Chl mL⁻¹, the [C_i] was 50 μm, and the pH of the medium was 9.6.

Experimental Procedure

Prior to experiments, cells were washed three times in 25 mm BTP-HCl, pH 8.0, as previously described (Li and Canvin, 1997a). The resuspended cells were transferred into a glass chamber (40 mm in diameter) at 30°C and 80 μmol m⁻² s⁻¹ light and allowed to fix contaminant C_i in the buffer and reach the CO₂ compensation point (Miller and Canvin, 1989). The chamber was flushed with CO₂-free air to facilitate the removal of the C_i. The Chl content was determined spectrophotometrically at 665 nm after extraction in methanol (MacKinney, 1941).

All measurements were performed by MS. When required, 6 mL of cell suspension was transferred to the reaction chamber (20 mm in diameter) connected via a membrane inlet to the mass spectrometer and allowed to reach the CO₂ compensation point at 30°C and pH 8.0. White light to drive C_i transport and photosynthesis was provided by a tungsten halogen projection lamp at 400 μmol m⁻² s⁻¹. The [O₂] during the experiments was maintained between 80 and 100 μm.

MS

A membrane inlet mass spectrometer (model MM 14–80 SC, VG Gas Analysis Systems, Middlewich, UK) was used in these experiments. Concentrations of ¹⁶O₂, ¹²CO₂, and ¹³CO₂ (m/z 32, 44, and 45, respectively), in the 6-mL reaction chamber were measured simultaneously. The output signal was directed to a computer (486, Altair, Kingston, Canada), and the mass spectrometer was controlled using software (Spectrascan 5.3.1, Petrasoft 5.3.2) from the same company as the spectrometer (VG Gas Analysis Systems). The suspension was stirred with a rotating magnetic bar. Gases and chemicals were introduced into the reaction chamber through a small port. A response time of 2.3 s was obtained with two masses, and 1.1 s was obtained with one mass.

Calibration for CO₂ was performed by injecting known amounts of K₂¹³CO₃ into the stoppered cuvette in 6 mL of BTP-HCl, pH 8.0, and calculation of the equilibrium concentration of CO₂ at 30°C and pH 8.0 was as described by Miller et al. (1988a). Calibration for ¹⁶O₂ was made as described previously (Miller et al., 1988a; Li and Canvin, 1997a). The value of ¹⁶O₂ concentration in equilibrium with air was taken as 240 μm. Rates of ¹⁶O₂ evolution were calculated from the slope of the 32 signal after the appropriate corrections were made for leak rates (2% h⁻¹) during the course of the experiment (Miller et al., 1988a).

C_i Uptake and Accumulation

The activity of CO₂ uptake and the intracellular C_i pool size were measured using the mass spectrometer by following the [CO₂] after the addition of ¹³C_i to the medium both in the presence and in the absence of CA. In the absence of added CA, the intracellular C_i pool was determined either by taking a sample or by injecting CA into the medium (Salon et al., 1996a). When CA is present the [CO₂] represents total C_i (Badger and Price, 1994). In the presence of IAC, the C_i transported into the cells was not fixed and leaked back into the medium when the light was turned off. The intracellular C_i was determined as the difference between the external [C_i] in the light and the [C_i] that occurred at equilibrium upon darkening the cells. Similar estimates were made when CO₂ fixation was allowed. The internal [C_i] was calculated using an intracellular volume of 75 μL mg⁻¹ Chl (Miller et al., 1988c).

Chemicals

PNDA, Na₂S, IAC, CA, and BTP were obtained from Sigma. K₂¹³CO₃ (99 atom % ¹³C) was obtained from MSD Isotopes (Montreal, Canada). The K₂¹³CO₃ was dissolved in BTP-HCl, pH 8.0, as a concentrated solution for addition to the reaction medium (Miller et al., 1988b). PNDA was dissolved in DMSO, and DCMU was dissolved in methanol before use. COS was obtained in lecture bottles from Matheson Gas Products (Ontario, Canada). Solutions of COS were prepared as described previously (Miller et al., 1989; Salon et al., 1996b).

RESULTS

C_i Uptake and Accumulation and Effect of DCMU and PNDA

For air-grown cells of Synechococcus UTEX 625 we showed previously that light-dependent C_i uptake and accumulation can be determined using a mass spectrometer equipped with a membrane inlet system that measures the extracellular concentration of dissolved CO₂ (Miller et al., 1988c; Espie et al., 1989; Salon et al., 1996b). Figure 1, A and F, shows the time course of CO₂ depletion following illumination of the suspensions after they had been incubated in the dark with 100 μm C_i in the presence of the CO₂-fixation inhibitor IAC. When measurements were made with added CA in the reaction mixture (Fig. 1A), the CO₂ signal was directly proportional to the total [C_i]. The disappearance of CO₂ from the medium represented transport and accumulation of total C_i within the cells.

Figure 1.

The effect of 5 μm DCMU (B, C, G, and H) and 200 μm PNDA (D, E, I, and J) on changes in the [CO₂] (A and F) of the medium in air-grown cells of Synechococcus UTEX 625. Cells were preincubated in 25 mm BTP-HCl buffer, pH 8.0, in the light in the presence of 25 mm NaCl and 25 μg mL⁻¹ CA (A–E) or 25 mm NaCl (F–J). After reaching the CO₂-compensation point, the cells were darkened and IAC (3.3 mm) was added 5 min before C_i injection (A–G and I). DCMU or PNDA was injected either 10 s before the addition of C_i (B, D, and G–J) or during C_i uptake, as indicated (C and E). C_i was added to yield a final [C_i] of 100 μm, as indicated (↑). C_i uptake was initiated by turning on the light (L), and cells were darkened (D↑) to allow intracellular C_i to leak back into the medium. The size of the intracellular C_i pool (↕) is shown. When CA was not present initially, it was added before darkening (F–J). In D, E, I, and J, simultaneous measurements of O₂ evolution were made, and numbers by the O₂ traces represent the rates of O₂ evolution in micromoles per milligram of Chl per hour. The addition of CA, even in the absence of cells, caused a 10% increase in the CO₂ signal, and this enhancement has been subtracted from the actual tracing record (F–J). [Chl] (in micrograms per milliliter) was 6.3 in A; 6.5 in B; 9.2 in C, D, and E; 4.57 in F and G; 7.2 in H; 5.2 in I; and 5.75 in J. Chemicals other than NaCl added before the C_i are shown in the figures without regard to time of addition.

Upon darkening, the intracellular C_i that had been accumulated into the cells by the C_i-concentrating mechanism leaked out, as shown by the increase in the extracellular CO₂ trace (Fig. 1A). When the light was switched on in the absence of CA, there was a rapid decline in [CO₂], followed by a gradual increase to a steady-state level (Fig. 1F). After a period of CO₂ uptake, the addition of CA in the light resulted in an increase in the [CO₂] in the medium (Fig. 1F), indicating a disequilibrium of the HCO₃⁻-CO₂ system prior to the addition of CA. The C_i taken up by the cells was quantitatively released back into the medium when the actinic light was turned off (Fig. 1F).

The effect of DCMU (5 μm) on the ability of cells to take up and accumulate C_i is shown in Figure 1 (B, C, G, and H). The experiments were carried out in the presence of 25 mm Na⁺ and in the presence (Fig. 1, B and C) or absence (Fig. 1, G and H) of added CA in cells in which CO₂ fixation was either inhibited with IAC (Fig. 1, B, C, and G) or not inhibited (Fig. 1H). In the presence of added DCMU, upon illumination there was substantial disappearance of CO₂ from the medium (Fig. 1B). Since the cells were incubated with added CA to maintain the CO₂-HCO₃⁻ equilibrium, the disappearance of CO₂ from the medium represented transport and accumulation of total C_i. The CO₂ taken up by the cells was quantitatively released back into the medium when the actinic light was turned off, and an intracellular C_i pool of 19.7 mm was observed (Fig. 1B). A similar intracellular C_i pool remained when DCMU was added to the cells during C_i uptake (Fig. 1C).

In the absence of CA, a rapid uptake of CO₂ occurred and the extracellular [CO₂] was maintained below the initial level whether CO₂ fixation was inhibited (Fig. 1G) or not inhibited (Fig. 1H; compare with Fig. 1F). The CO₂ and HCO₃⁻ were not in equilibrium, because after a steady extracellular CO₂ level was observed, the addition of CA caused an increase in [CO₂]. When the lights were turned off, the [CO₂] returned to the initial level. The calculated C_i pool sizes were 18.8 mm (Fig. 1G) for IAC-inhibited cells and 15.6 mm (Fig. 1H) for noninhibited cells. In the latter case, the addition of DCMU prevented any CO₂ fixation.

The changes in extracellular [CO₂] with 200 μm PNDA were also monitored in cells in which CO₂ fixation was inhibited with IAC (Fig. 1, D, E, and I) or not inhibited (Fig. 1J). O₂ evolution was also recorded. With added CA, [CO₂] and hence [C_i] decreased upon illumination. The C_i transported into the cells was not fixed and leaked back into the medium when the lights were turned off (Fig. 1, D and E). The calculated intracellular C_i pool size was 22.6 mm (Fig. 1D) when PNDA was added at the beginning of the experiment and 20.7 mm when the same concentration of PNDA was added during C_i uptake (Fig. 1E).

When the experiments with PNDA were performed in the absence of CA, different results were observed (Fig. 1, I and J). After the equilibrium of CO₂ and HCO₃⁻ was established and the light was turned on, the [CO₂] initially decreased. After a short time, however, the [CO₂] in the medium increased whether CO₂ fixation was inhibited with IAC (Fig. 1I) or not inhibited (Fig. 1J). When CO₂ fixation was inhibited, the [CO₂] increased to a constant level that was much higher than the initial equilibrium concentration. The addition of CA resulted in a decrease in the [CO₂], showing that the [CO₂] had been above the equilibrium value. The extracellular [CO₂] strongly suggests that C_i accumulation was the result of active HCO₃⁻ transport and that CO₂ transport was inhibited (Badger, 1985; Salon et al., 1996b, 1997). When the cells were darkened, the intracellular C_i pool appeared in the medium and the extracellular [CO₂] returned to its initial equilibrium value (Fig. 1I). The intracellular C_i pool was 17.5 mm (Fig. 1I).

When CO₂ fixation was not inhibited with IAC (Fig. 1J), PNDA caused similar changes in the extracellular [CO₂], but the [CO₂] never increased above the initial value. Again, the addition of CA showed that the [CO₂] was above the equilibrium value. In this case a much larger intracellular C_i pool (37.5 mm) was observed.

When CO₂ fixation was inhibited (Fig. 1, D, E, and I), PNDA supported substantial rates of O₂ evolution. When CO₂ fixation was allowed (Fig. 1J), the rate of O₂ evolution was higher, possibly reflecting both CO₂ and PNDA reduction. As shown by the initial and final [CO₂] about one-half of the C_i was fixed during the experiment. PNDA does not completely inhibit CO₂ fixation (Li and Canvin, 1997b).

The effect of [DCMU] or [PNDA] is shown in Figure 2. At 0.5 μm DCMU, O₂ evolution was completely inhibited but C_i accumulation was still more than 45% of that obtained in control cells (Fig. 2A). Even at 10 μm DCMU, C_i accumulation was 28% of the control. Significant C_i accumulation occurred even upon addition of 500 μm PNDA, and O₂ evolution as high as 240 μmol mg⁻¹ Chl h⁻¹ was observed (Fig. 2B).

Figure 2.

Effect of [DCMU] (A) and [PNDA] (B) on the rate of O₂ evolution (▵) and the intracellular C_i pool (○). Cells (6–9 μg Chl mL⁻¹) were incubated in the presence of 25 mm NaCl and 25 μg mL⁻¹ CA and allowed to reach the CO₂-compensation point. IAC (3.3 mm) was added; 5 min later the cells were darkened and 100 μm C_i was injected. After about 1 min, C_i transport and accumulation were initiated by turning on the light. DCMU or PNDA was added to the cells after the internal C_i pool had reached its maximum size. The C_i pool remaining after DCMU and PNDA addition was calculated upon darkening the cells. The control rate of O₂ evolution was 256 μmol mg⁻¹ Chl h⁻¹. The maximum size of the internal pool was 56 mm. The data are given as means ± sd (n = 2–5).

Cyclic Electron Flow-Dependent Active CO₂ Transport and Linear Electron Flow-Dependent HCO₃⁻ Transport

It has been well established that active transport of CO₂ and HCO₃⁻ proceeds via separate and independent transporters (Espie et al., 1989; Miller and Canvin, 1989; Salon et al., 1996a). CO₂ transport is inhibited by COS or Na₂S (Espie et al., 1988b, 1989; Miller et al., 1989), and HCO₃⁻ transport is inhibited by Li⁺ in air-grown cells of Synechococcus UTEX 625 (Espie et al., 1988b; Miller et al., 1988b, 1990; Espie and Kandasamy, 1994). Therefore, it was of interest to determine the effects of these inhibitors on C_i uptake and accumulation in DCMU- and PNDA-treated cells.

When 30 μm COS (Fig. 3A) or 200 μm Na₂S (Fig. 3B) was added to cell suspensions that had been allowed to concentrate C_i internally in the presence of 5 μm DCMU, there was an immediate increase in the [CO₂]. It has been well documented that there is always a burst of CO₂ in the light when CO₂ transport is quenched by various compounds (Miller et al., 1989; Salon et al., 1996b). The effects of COS and Na₂S were rapid, and we interpret the fast increase in the extracellular [CO₂] as being due to rapid leakage of the CO₂ that had been accumulated within cells. Darkening the cells after CA equilibration did not result in any increase in the [CO₂], indicating that the previously accumulated C_i pool had leaked from cells upon injection of COS or Na₂S (Fig. 3, A and B).

Figure 3.

Effects of COS (30 μm, A and E), Na₂S (200 μm, B and F), DES (10 μm, C and G), and Li⁺ (D and H) on C_i uptake in the presence of DCMU (A–D) or PNDA (E–H). Cell suspensions were incubated in the presence of 25 mm (A–G) or 5 mm NaCl (H) and allowed to reach the CO₂-compensation point. IAC (3.3 mm) was added, and 5 min later cells were darkened and either 5 μm DCMU (A-D) or 200 μm PNDA (E–H) and 100 μm C_i was injected in the dark. After C_i equilibration, the light was switched on as indicated (L) and C_i was allowed to accumulate. The inhibitors were injected either during C_i uptake, as indicated (A–C, G, and H) or before the addition of C_i (D, E, and F). The effect of the inhibitors was rapid, and there were no difference in terms of order of addition. After CA addition the cells were darkened and the size of the intracellular C_i pool (↕) was determined. Results were corrected for changes due to CA addition. The dashed lines indicate the changes in m/z = 45 signal seen upon the addition of the inhibitor in the absence of cells. [Chl] (in micrograms per milliliter) was 10.4 in A and C; 4.57 in B; 5 in D and G; 9 in H; and 5.75 in E and F. Chemicals other than NaCl added before the C_i are shown in the figures without regard to time of addition.

The effect of COS and Na₂S on C_i uptake and accumulation with PNDA-treated cells was also determined (Fig. 3, E and F). The inhibitors were incubated with cells in the dark and then 100 μm C_i was injected. The formation of significant intracellular C_i pools occurred in the presence of added 30 μm COS (Fig. 3E) or 200 μm Na₂S (Fig. 3F) in the light. The pattern of C_i uptake and accumulation was typical of that seen when HCO₃⁻ transport occurs in the absence of CO₂ transport (Salon et al., 1996b). The results, therefore, indicated that CO₂ inhibitors had little effect on C_i transport and accumulation in PNDA-treated cells.

When cells were allowed to accumulate C_i internally in the presence of 5 μm DCMU, the addition of DES (Fig. 3C), an inhibitor of membrane-bound ATPase (McEnery and Pederson, 1986), resulted in a rapid increase in the extracellular [CO₂] to above the initial equilibrium level. The addition of CA to cells in the light caused a decrease in [CO₂] and a rapid re-establishment of the CO₂-HCO₃⁻ equilibrium. As shown by the absence of an increase in the extracellular [CO₂] upon darkening, all or most of the intracellular C_i had leaked from the cells upon the addition of DES (Fig. 3C). When the same concentration of DES was added to PNDA-treated cells, there was no burst of CO₂ and the cells still retained an intracellular pool of 16.9 mm (Fig. 3G).

When 5 mm Li⁺ was added to a cell suspension that had been allowed to concentrate C_i internally in the presence of 200 μm PNDA and 5 mm Na⁺ (Fig. 3H), there was a slow increase in [CO₂]. Upon darkening, there was no further increase in the [CO₂], indicating that the accumulated C_i had leaked out upon the addition of Li⁺. The degree of Li⁺ inhibition on C_i accumulation was largely reversed by increasing the [Na⁺]. In the presence of 25 mm Na⁺, 5 mm Li⁺ inhibited C_i accumulation by only 15%, whereas 50 mm Li⁺ reduced the accumulation by 70% (data not shown). For cells treated with DCMU, however, the uptake of CO₂ proceeded normally in the presence of 50 mm Li⁺ and cells were able to accumulate an intracellular C_i pool of 28.7 mm (Fig. 3D).

The addition of 1 μm DCMU results in the discharge of the PNDA-supported intracellular C_i pool (Fig. 4A), and PNDA addition results in the discharge of the intracellular C_i pool in the presence of DCMU.

Figure 4.

The effects of 1 μm DCMU (A) on C_i uptake and accumulation with PNDA and 100 μm PNDA (B) on C_i uptake and accumulation with DCMU in the absence of CA. Cell suspensions containing 25 mm NaCl with 100 μm C_i and 3.3 mm IAC were incubated as described in Figure 1, with 5 μm DCMU (B) or 200 μm PNDA (A). After C_i equilibration, the light was switched on as indicated (L) and C_i was allowed to accumulate. DCMU and PNDA were injected as indicated, and CA was added either before the addition of DCMU (A) or after the addition of PNDA (B) to maintain the CO₂-HCO₃⁻ equilibrium. The light was turned off as indicated (D). [Chl] was 5.75 (A) and 3.9 μg mL⁻¹ (B).

DISCUSSION

The use of MS to monitor the free [CO₂] in the medium provides a direct measurement of the extent of C_i uptake and accumulation (Badger, 1985; Miller et al., 1988c; Salon et al., 1996a, 1996b). It has also allowed active CO₂ transport to be distinguished from HCO₃⁻ transport in air-grown cyanobacteria (Espie et al., 1988a, 1989; Salon et al., 1996a, 1996b), in which only Na⁺-dependent HCO₃⁻ transport occurs. In the present study the MS technique was used to determine the capacities to which C_i uptake could occur in the presence of inhibitors or artificial electron acceptors of photosynthesis to identify the role of photosynthetic electron transport in the operation of the C_i-concentrating mechanism.

It is well established that cells of cyanobacteria, when grown at air levels of CO₂, have developed the capacity to accumulate high intracellular [C_i]. The accumulation of C_i is an active transport process, since it occurs only in the light and against concentration and electrochemical gradients (Fig. 1, A and F; Badger et al., 1980; Shelp and Canvin, 1984; Miller et al., 1990). Photosynthesis supplies energy for the concentrating mechanism (Badger et al., 1980; Ogawa et al., 1985b; Kaplan et al., 1987), although the reactions involved in providing the energy for this process are not known. It has been suggested that PSI-driven cyclic electron flow is the source of energy (Ogawa and Inoue, 1983; Ogawa et al., 1985a, 1985b; Ogawa, 1993). However, the fact that DCMU inhibits the accumulation of C_i and the enhanced Mehler reaction in low-CO₂ cells suggests the involvement of linear electron transport (pseudocyclic photophosphorylation) in the operation of the C_i-concentrating mechanism (Miller et al., 1988a, 1991; Sültemeyer et al., 1993).

Data presented in Figure 1 clearly demonstrate that the ability to concentrate C_i is reduced by the presence of higher concentrations of DCMU and PNDA, but the cells still retained some concentrating capacity. It can be seen that the intracellular [C_i] was 34.5% of that in the absence of DCMU (Fig. 1, F and G) and 32.1% of that in the absence of PNDA (Fig. 1, F and I). The intracellular C_i accumulation occurred only in the light, suggesting that cyclic or linear electron flow can at least partially drive C_i uptake. PNDA accepts electrons from PSI (Elstner and Zeller, 1978), and the high rate of O₂ evolution that was observed even in the presence of IAC indicates that it is an efficient acceptor of PSI-derived electrons (Fig. 1, D, E, I, and J). The rate of O₂ evolution was higher in noninhibited than in inhibited cells. Significant C_i accumulation occurred even upon the addition of a high concentration of PNDA. These observations are consistent with the results obtained by Li and Canvin (1997b).

A similar concentrating capacity was observed when cells were treated with methyl viologen, which also accepts electrons from PSI (data not shown). However, when the electrons were drained to the PSII acceptors 2,6-dimethylbenzoquinone or oxidized diaminodurene, no intracellular C_i accumulation was observed (data not shown). The inhibition of the intracellular C_i accumulation in PNDA-treated cells by 1 μm DCMU is also supportive of the concept that the electron transport from water to Fd (PNDA) is involved in the transport of C_i (Fig. 4A). Evidently, the transport of C_i is driven only by PSI in DCMU-treated cells. In this case, 100 μm PNDA totally suppressed the ability of cells to accumulate C_i internally (Fig. 4B).

In the absence of added CA, much larger differences between DCMU- and PNDA-treated cells were observed in the extracellular [CO₂] traces. When cells were incubated with DCMU, the [CO₂] in the medium was rapidly reduced and maintained this low level during the light. Darkening the cells resulted in only a transient excess of the expected equilibrium level whether or not CO₂ fixation was allowed (Fig. 1, G and H). In contrast, when cells were incubated with PNDA, a rapid uptake of CO₂ was initially observed, but then the extracellular [CO₂] increased much higher than the expected equilibrium value (Fig. 1, I and J). The addition of CA restored the extracellular HCO₃⁻-CO₂ equilibrium, and when the cells were darkened the intracellular C_i pool appeared in the medium (Fig. 1, I and J).

Salon et al. (1996b) also observed this CO₂ release in cells treated with CO₂ transport inhibitors. In this case, cells continued to transport HCO₃⁻, which was converted to CO₂, which leaked from the cells because it could not be retransported. The differences in the patterns between DCMU- and PNDA-treated cells in the absence of added CA can be explained. In the presence of PNDA, active HCO₃⁻ transport was responsible for the accumulation of the intracellular C_i pool and the increase in the extracellular [CO₂] was due to rapid leakage of CO₂ from cells. The observation that C_i uptake and accumulation in DCMU-treated cells were inhibited by CO₂-transport inhibitors such as COS and Na₂S, whereas the HCO₃⁻-transport inhibitor Li⁺ inhibited C_i accumulation in PNDA-treated cells supports this interpretation (Fig. 3, A, B, E, and F).

As discussed earlier, the accumulation of C_i proceeds against a concentration gradient, which implies that it must be an energy-requiring process. In an earlier work, DES inhibited the transport of CO₂ (Miller et al., 1988c). In the present study, 10 μm DES completely inhibited CO₂ accumulation in DCMU-treated cells, indicating that CO₂ transport was fueled by ATP (Fig. 3C). HCO₃⁻ transport and accumulation, however, could proceed in the presence of the same concentration of DES in PNDA-treated cells (Fig. 3H). These results suggest that PNDA-dependent linear electron flow was responsible for the C_i accumulation, and the linear electron flow may be more important than its coupling to ATP formation. We do not know how linear electron transport supplies the energy for HCO₃⁻ transport or why CO₂ transport appears to be driven only by ATP generated through cyclic phosphorylation (Ogawa, 1993). Linear electron transport should also generate ATP, but it does not appear to be available to drive CO₂ transport. If it drives HCO₃⁻ transport, the ATPase involved would not be sensitive to DES. Further speculation on energy coupling is possible, but any firm conclusions concerning the subject cannot be made at the present time.

Abbreviations:

BTP

1,3-bis[tris(hydroxymethyl)methylamino] propane

CA

carbonic anhydrase (carbonate dehydratase, EC 4.2.1.1)

Chl

chlorophyll

C_i

dissolved inorganic carbon (CO₂ + HCO₃⁻ + CO₃²⁻)

DES

diethylstilbestrol

IAC

iodoacetamide

PNDA

N,N-dimethyl-p-nitrosoaniline