Planta

Planta (1992)187:511-516

9 Springer-Verlag1992

Photosynthesis and photorespiration in a mutant of the cyanobacterium Synechocystis PCC 6803 lacking carboxysomes Yehouda Marcus 1.' **, Joseph A. Berry 1, and John Pierce 2

1 Department of Plant Biology,CarnegieInstitution of Washington, 290 Panama St., Stanford, CA 94305, USA 2 AgriculturalProducts Departments E.I. Du Pont de Nemours and CompanyInc., Wilmington,DE 19880, USA Received 6 February 1991; accepted 24 February 1992

Abstract. A mutant of the cyanobacterium Synechocystis PCC 6803 was obtained by replacing the gene of the carboxylation enzyme ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) with that of the photosynthetic bacterium Rhodospirillum rubrum. This mutant consequently lacks carboxysomes - the protein complexes in which the original enzyme is packed. It is incapable of growing at atmospheric CO2 levels and has an apparent photosynthetic affinity for inorganic carbon (Ci) which is 1 000 times lower than that of the wild type, yet it accumulates more Ci than the wild type. The mutant appears to be defective in its ability to utilize the intracellular Ci pool for photosynthesis. Unlike the carboxysomal carboxylase activity of Rubisco, which is almost insensitive to inhibition by 02 in vitro, the soluble enzyme is competitively inhibited by 02. The photosynthetic rate and Ci compensation point of the wild type were hardly affected by low 02 levels. Above 100 gM 02, however, both parameters became inhibited. The CO2 compensation point of the mutant was linearly dependent on 02 concentration. The higher sensitivity of the mutant to 02 inhibition than that expected from in-vitro kinetics parameters of Rubisco, indicates a low capacity to recycle photorespiratory metabolites to Calvin-cycle intermediates. Key words: Carbonate transport Carboxysome- Inorganic carbon compensation point - Photorespiration Ribulose-l,5-bisphosphate carboxylase/oxygenase Synechocystis PCC 6803 (mutant)

Introduction

Cyanobacterial ribulose-l,5-bisphosphate carboxylase/ oxygenase (Rubisco) has a lower specificity factor ( = * Present address: Otto Warburg Center of Biotechnology,The

Faculty of Agriculture, The Hebrew University of Jerusalem, P.O.B. 12, 76100 Rehovot, Israel; FAX (0)8/462181 ** To whom correspondenceshould be addressed Abbreviations: c~= inorganic carbon; RuBP= ribulose-1,5-bisphosphate; Rubisco= ribulose-l,5-bisphosphate carboxylase/oxygenase

carboxylation rate/oxygenation rate) than that of photosynthetic eukaryotic organisms in vitro (for a review, see Badger and Andrews 1987). This is expected to lead to 02 inhibition of photosynthesis, and to high rates of photorespiration and glycolate excretion. However, 02 inhibition of wild-type cyanobacterial photosynthesis is less than that expected from the kinetics parameters of Rubisco (Kaplan et al. 1980). The entire photorespiratory pathway has not been well clarified (Codd and Stewart 1973; Tolbert et al. 1985) and its capacity to recycle the phosphoglycolate produced in each oxygenation cycle to Calvin-pathway intermediates is questionable (Tolbert et al. 1985). There is solid evidence for the presence of phosphoglycolate phosphatase and a low activity of glycolate dehydrogenase (Codd and Stewart 1973; Tolbert et al. 1985); however, the rest of the metabolic pathway is still a matter for discussion between two groups (Codd and Stewart 1973; Tolbert etal. 1985). Glycolate is an end-product photorespiratory metabolite of green algae grown in high CO2 and then transferred to air (Kaplan and Berry 1981; for a review, see Tolbert et al. 1985). Its excretion is not observed in cyanobacteria (Tolbert et al. 1985). This could imply a reduction in photorespiration, which is attributed, at least partially, to the operation of an inorganic-carbon (Ci)concentrating mechanism (for a review, see Kaplan et al. 1988). However, it is an oversimplification to assume that this is the only reason. The elevated CO2 concentration within the cell enhances the rate of 02 evolution which may in turn cause 02 accumulation within the cell (Badger et al. 1985). Most of the cyanobacterial Rubisco is confined to polyhedral bodies (carboxysomes; Coleman et al. 1982; Marcus et al. 1986). Reinhold et al. (1987) suggested that carbonic anhydrase is localized in the carboxysome near the carboxylation site and catalyzes the conversion of HCO 3 to CO2. The latter is required to sustain the observed photosynthetic rate. In the noncarboxysomal cytoplasmic space the Ci species (CO2 and HCO~) are presumably in disequilibrium. Recently, this model was experimentally supported by Price and Badger (1989) who showed that expression of a foreign carbonic-an-

512

Y. Marcus et al. : Photosynthesis in a carboxysome-deficient cyanobacterium mutant

hydrase gene in the c y a n o b a c t e r i u m Synechococcus increases the effiux o f Ci. A chemically induced high-COz-requiring m u t a n t unable to utilize the Ci p o o l was suggested to have impaired c a r b o x y s o m a l function (Marcus et al. 1986). " C y a n o r u b r u m " is a chimeric m u t a n t o f the c y a n o b a c t e r i u m Synechocystis P C C 6803, in which the original gene for R u b isco has been replaced by the c o r r e s p o n d i n g gene f r o m the p h o t o s y n t h e t i c bacterium Rhodospirillum rubrum (Pierce et al. 1989). Replacement o f the gene causes the disappearance o f the c a r b o x y s o m e s , allowing the m u t a n t to g r o w only at high CO2 concentrations and m a k i n g its g r o w t h extremely sensitive to 0 2 (Pierce et al. 1989). In this study the c y a n o r u b r u m m u t a n t was used to investigate the role o f c a r b o x y s o m e s in cyanobacterial photosynthesis and photorespiration.

--

140

~c ~ 7o

_~5o m

.~ 3o O pH

Q. ~0 0 0

Materials and methods

Synechocystis PCC 6803 and the cyanorubrum mutant were provided by one of us (J.P.; Pierce et al. 1989). Cells were grown at 30~ C in a modified BG! 1 medium (Stanier et al. 1971) containing 10mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid (Hepes; pH 8.0). Cultures were continuously illuminated at 25 lamol photons.m 2-s-~ and aerated with 3% CO2 in air. Uptake of Ca was measured using the filtering centrifugation method as previously described (Kaplan et al. 1988). Phosphoglycolate levels were determined as previously described by Marcus et al. (1983). Exchange of 02, C~ uptake and accumulation of photosynthetic products were determined as described elsewhere (Kaplan et al. 1980). The C~ compensation point was determined as follows: a given amount of bicarbonate was added to the culture at the compensation point (i.e. Ci concentration at which net 02 exchange = 0). The cells consumed CO2 till they reached a new compensation point at a higher O2 level. The amount of CO2 consumed was calculated from the amount of 02 evolved. The amount left over between the given and consumed C divided by the volume of the culture medium was defined as the Ci compensation point. Spheroplasts were prepared as previously described by Marcus et al. (1986). Carboxysomes were obtained by lysing spheroplasts into reaction medium containing 1% Triton x 100. Soluble Rubisco was obtained by short sonication of the particulated fraction. The solution was centrifuged for 10 min at 10000.g and the supernatant was collected. The carboxylase activity of Rubisco was measured, as previously described (Marcus et al. 1986), in stoppered vials pre-aerated with the appropriate atmosphere. Results and discussion The a p p a r e n t p h o t o s y n t h e t i c affinity for extracellular CO2 o f the c y a n o r u b r u m mutant ( K 1 / 2 ( C O 2 ) = 2 m M C O 2 ; Fig. 1) was approx. 1000 times lower than that o f the wild type (data n o t shown). This m a y explain w h y the the m u t a n t grows only in a CO2-enriched a t m o sphere (Pierce et al. 1989). T h e replacement o f the Rubisco gene alone c a n n o t explain this low p h o t o s y n t h e t i c affinity since the specificity factor o f R. rubrum is only threefold lower than that o f the c y a n o b a c t e r i a l Rubisco (Badger and A n d r e w s 1987; Pierce et al. 1989) and its Km (CO2) is only slightly lower than that o f the cyanobacterial Rubisco. It is w o r t h m e n t i o n i n g that Pierce et al. (1989) accurately in-

8.0

t~ pH 7 . 4 5

I

i

i

[

I

1

2

3

/-,

5

I

9 pH

7.0

[3pH

6.0

I

I

I

I

6 7 CO 2 ( r a M )

8

9

10

..',

I 15

Fig. l. Carbon-dioxide-dependent 02 evolution in the cyanorubrum mutant of Syneehoeystis PCC 6803 at various pHs. HighCO2-grown cells were resuspended in 50 mM BTP buffer at the pH indicated. The photon fluence rate was 500 ~tmol.m2-s -1, 25~ C. The level of CO2 was calculated from the equilibrium equation

60

,,

50

...y

/o

4O ra//~A

30

A

b

o-

[]

pH 8 . 4 pH 8 . 0 pH 7 . 6

2o

10 0

i

0

I

20

i

I

40

~

I

60

f

I

~

I

80

100

CO~ 2

(~uM)

i

I

120

i

I

140

~

I

160

i

180

Fig. 2. Inorganic-carbon pool of the cyanorubrum mutant as a function of CO~- concentration at various pHs. The accumulation of Ci was determined after 60 s incubation with t4Ci, by the filtering-centrifugation method (see Materials and methods). Levels of CO ] - were calculated from the equilibrium equation. The rest of the experimental conditions are the same as in Fig. 1

serted the R. rubrum gene in the place which was occupied by the original gene and its expression is therefore the same as that o f the cyanobacterial gene. The K1/2 (CO2) o f cyanobacteria is strongly affected by the activity o f the CO2-concentrating m e c h a n i s m (Kaplan et al. 1980 and for reviews see Badger 1987; K a p l a n et al. 1988). A high K1/2 (CO2) could therefore be the consequence o f a decrease in the capability to accumulate Ci. However, this was not f o u n d to be the case (Figs. 2, 3): the m u t a n t accumulated m o r e Ca than the wild type, b o t h when the cells were g r o w n under

Y. M a r c u s et al. : P h o t o s y n t h e s i s in a c a r b o x y s o m e - d e f i c i e n t c y a n o b a c t e r i u m m u t a n t

r

E 1.s

/./"

E

*cv

,, L-CRB

1

/~'"

o

~,_

/

j.j

.,,,, .....

..-'""

.............. H-CRB

L-6803~ ~

................. *

..'

............

.s

0

1

I

I

[

1

[

10

20

30

40

50

60

Time (s) Fig. 3. Transport of Ci in high-CO2-grown (H) or ]ow-CO2-adapted (12 h) (L) Synechocystis PCC 6803 (5803) and cyanorubrum mutant (CRB) cells as a function of time. Cells were exposed to

0.7 mM Ci at the beginning of the experiment high COz concentrations and when high-CO2-grown cells were transferred to l o w - C O 2 conditions. The high ability of the mutant to accumulate C~ could be explained in a number of ways: (i) The Ci-concentrating mechanism is inducible, i.e. its activity increases during adaptation to low CO2 (Fig. 3; Kaplan et al. 1980; Marcus et al. 1982; Marcus et al. 1983). It has been previously suggested that this system is induced by a product of photorespiration (Marcus et al. 1983). Since the mutant has a high rate of photorespiration (see below) it presumably produces more of the inducing photorespiratory product than the wild type, which may lead to higher Ci-transport capability. (ii) The lack of carboxysomes in the mutant (Pierce et al. 1989) may result in the disappearance or dilution of a factor with carbonicanhydrase-like activity - presumably residing in the carboxysome (Marcus et al. 1986; Reinhold et al. 1987) thereby rendering it ineffective. Consequently, the conversion of bicarbonate - the main C~ species in the cytoplasm - to CO2 could be limited, thereby limiting the dissipation of the Ci pool, either by diffusion or by COz fixation. (iii) The R. rubrum Rubisco may not be activated appropriately in the cyanobacterium, rendering it less active in low external CO2 concentrations. This would also limit the dissipation of the accumulated C~. Although the latter is feasible (Somerville et al. 1982), it does not in itself explain why Ci in the mutant is accumulated at up to 10-fold higher concentrations than in the wild type (data not shown). The effect of extracellular pH on apparent photosynthetic affinity for Ci and Ci uptake was analyzed to elucidate the roles of CO2, HCO~- and CO~- in the supply of Ci for these processes (Raven 1970; Marcus et al. 1986; Kaplan et al. 1988). Assuming that pH alters only the concentration ratios of the C~ species, then the cells should respond only to changes in the concentration of the species in use and a single correlation curve between the response and the calculated concentration of this species should be obtained.

70

513

If the internal C i pool, which is thought to be made up mainly of bicarbonate (Reinhold et al. 1987), contributes significantly to CO2 fixation, then the same external Ci species should serve for both Ci uptake and photosynthesis. We found that although CO 2- was taken up for the internal Ci pool, CO2 in high external concentrations served as the substrate for photosynthesis. The apparent Km values of these processes were 10 gM CO 2- and 2 mM CO2 (Figs. 1, 2). Vma. was hardly affected by the change in pH (data not shown), indicating that the response does not result from titration of the cells. Previous investigations have shown HCO~ and CO2 to be the species which are actively taken up (for a review, see Kaplan et al. 1988). These conclusions were based primarily on disequilibrium experiments in which either CO2 or HCO~ was supplied; HCO~ and CO 2are indistinguishable in such experiments. It is likely, however that if CO 2-, and not HCO~ is transported (Fig. 2), then CO2 will also be taken up because, at the more acidic pHs in which the cells concentrate Ci, the CO 2 - concentration will be negligible and unable to sustain the observed CO2-fixation rate. Addition of H 14CO 3 to cyanorubrum preloaded with H12COf did not result in an appreciable momentary decrease in the specific activity of the CO2-fixation rate due to dilution of the 14CO2 by the enormous accumulated Ci pool (data not shown). This indicates that the pool is either not available or only slowly exchangeable for fixation by the carboxylation enzyme. The phenotype of the cyanorubrum mutant is very similar to that of the chemically induced cyanobacterium mutant Anacystis nidulans E1 (Marcus et al. 1986). Recently, Friedberg et al. (1989) showed that the latter has morphologically aberrant carboxysomes, supporting the notion that disorganization of the carboxysomes may indeed result in a high-CO2-requiring phenotype. Of the in-vitro Rubisco activity in A. nidulans spheroplasts lysed into carboxylase reaction medium, 80-90% can be spun down by a low, short-term centrifugal force 10000.g for 10 min (data not shown). This heavy fraction contains Rubisco packed in large polyhedral protein complexes which are visible by electron microscopy (for a review, see Codd and Marsden 1984). The saturation curve of the carboxysomal carboxylase reaction was not pure Michaelis-Menten (Fig. 4a), probably as a result of the compact packing of the carboxylase in the carboxysomes, which may limit the carboxylase reaction by limiting the diffusion of CO2. In order to estimate the extent of 02 inhibition of the carboxylase reaction, the ratio of Rubisco activities under 100% N2 and 100% 02 was calculated and plotted against CO2 concentration (Fig. 4b). The carboxysomal carboxylase was almost insensitive to 02 inhibition. Solubilized carboxylase originating from carboxysomes was competitively inhibited by 02 in a crude extract (Fig. 4 b). Similar results were obtained with purified soluble Rubisco (data not shown). Phosphoglycolate, the primary oxygenation product of ribulose-l,5-bisphosphate in the reaction catalyzed by Rubisco was not observed under limiting CO2 concentrations and N2 atmosphere, whether the reaction was

514

Y. Marcus et al. : Photosynthesis in a carboxysome-deficient cyanobacterium mutant #'

a

.--. 8O

T t'r

3-

3O ......-*

E r

o

20

.......... .....y"

E

,j

9

y

~ r

10

. ."..o~. ' J/ o

% ~.x"

E >..

' ....

S

O O r (3.

o

i

i

i

20 HC03 (mM)

/,

30

i

b g

isoloted

Rubisco

2

n-

Q corboxysomes .13

p.c__ cI.--

0

0

I

I

1

10

20

30

I 40

I

1

50

60

/x

5g 02

.............

o

I lOO C~ ( p , M )

02 .~oo

Fig. 5. Photosynthesis as a function of external Ci concentration in various 02 levels. High-CO2-grown Synechocystis PCC 6803 cells were resuspended in 50 mM Hepes-NaOH (pH 7.8). The remaining experimental conditions were as in Fig. 1

60

4-, 13, 0 m u o ~E

14% 02

8~ o~

--43-- 21% 0

rr 10

9

o

70

HCO~ (raM)

Fig. 4. a Carboxysomal RuBP carboxylase of Rubisco of A. nidulans as a function of Ci concentration at ambient 02 concentrations, b Sensitivity of carboxysomal and soluble RuBP carboxylase reactions of Rubisco of A. nidulans to 02 inhibition as a function of Ci concentration. The sensitivity was defined as the ratio between RuBP carboxylase activity under an atmosphere of 100% N2 or 100% 02. See Materials and methods for details

catalyzed by carboxysomes or by the soluble enzyme. Only trace amounts of phosphoglycolate were observed in the reaction catalyzed by carboxysomes under an 02 atmosphere. Phosphoglycolate at up to 5% o f the amount of 3-phosphoglycerate was observed under an 02 atmosphere in the reaction catalyzed by soluble Rubisco. These results confirm a previous report of Coleman et al. (1982). In order to determine whether the observed lack o f 02 inhibition also occurs in vivo, we tested the effect o f O2 on photosynthesis (Fig. 5) and C~ compensation point (Fig. 6a, b). At 02 concentrations below 100 ~tM, the change in the Ci compensation point of the wild

type was negligible9 At concentrations above 100 ~tM, 02 was inhibitory (Fig. 6a). The photosynthetic response of the wild type to Oz similarly showed an inhibition at levels of 21% 02 (Fig. 5). The biphasic response of the Ci compensation point to 02 (Fig. 6a) may indicate (i) that the carboxylase is partially protected against 02 inhibition; consequently, oxygenation is reduced and the Ci compensation point is insensitive to O2 within a certain range; or (ii) that the capacity to recycle photorespiratory metabolites to Calvin-cycle intermediates is limited and therefore more CO2 is required to compensate for the loss of those intermediates. Unlike the wild type, cyanorubrum uses only external CO2 (Fig. 1). Thus only that hat was considered in determining the compensation point. The O2-dependent Ci compensation point function of cyanorubrum differed from that of the wild type in two respects: (i) its values were much higher than those of the wild type; and (ii) the CO2 compensation point of the mutant was almost linearly dependent on 02, even at low 02 concentrations where the compensation point of the wild type was insensitive to O2 (Fig. 6a, b). At 02 concentrations above 240 pM the slope of the function was steeper. Thus the CO2 concentration which saturated photosynthesis at low 02 levels became the CO2 compensation point at 300 ~tM 02 (Figs. 1, 6b), clearly indicating that photosynthesis in the mutant is more sensitive to O2 inhibition than that in the wild type. This is in accordance with previous findings of Pierce et al. (1989) who showed its growth to be extremely sensitive to 02. The net 02 exchange, A, can be expressed as: A=E-t.Vo-R1

(Eq. 1)

where E is the gross 02-evolution rate, t is the stoichiometric coefficient of 02 consumption due to photorespir-

Y. Marcus et al. : Photosynthesis in a carboxysome-deficient cyanobacterium mutant

5O

a

v

515

of 02 evolution is equal to the rate of carboxylation (Vc) then at the CO2 compensation point (F): Vc(1 - t/S. [O1]/[CO21) = 0

(Eq. 2)

~- 40 or

, ~

0 Q_

F = t. [O21/s

tO

3O

, ~

U t-

2O

O_

E 0

10

oa2

0

0

100

200

0 2 (ffM) 700

I

1

1

::L 600 --I--

so

500

C

.2 r-

400 OJ

[]

300 -

0_

E o O

200

[]

c,,I

0 0

r_2

100 -

t/s = 0.57

o 0

I

I

I

100

200

300

400

02 ( f f M ) Fig. 6. a Ci compensation point of Synechocystis PCC 6803 cells. b CO2 compensation point of the cyanorubrum mutant as a function of 02 concentration

atory metabolism, Vo is the rate of Rubisco oxygenation and R1 is the respiration rate in the light. Assuming that the specificity factor S (the ratio of Rubisco carboxylase to oxygenase reactions; Jordan and Ogren 1981) is the same in vivo as in vitro, the rate of oxygenation can be obtained from the specificity factor of the isolated enzyme in vitro at given CO2 and 02 concentrations, and the rate of carboxylation. If the rate of respiration in the light is negligible and the rate

(Eq. 3)

This equation is essentially similar to the one developed for CO2 exchange (Jordan and Ogren 1981; or see Woodrow and Berry 1988 for a review). From Eq. 3 we can compare the in-vitro and in-vivo kinetics parameters of Rubisco, as has been previously done for higher plants (Woodrow and Berry 1988). The COz compensation point increased more steeply than expected with O2 concentration. The slope between 0 and 240 ~tM 02 was 0.57 (Fig. 6b). Assuming that t = 1, namely oxygen is only consumed in the oxygenase reaction of Rubisco, and that the specificity factor is 15 (Badger and Andrews 1974), the slope should have been 0.066. This discrepancy may have resulted in various ways. (i) The CO2 compensation point measured by net 02 exchange is not inevitably identical to the compensation point measured by net CO2 exchange, because of the different stoichiometries of 02 consumption and CO2 evolution in photorespiration. Hence, while 02 is in flux equilibrium, there might be a net flux of COg resulting in different CO2 concentrations on each side of the membrane. The CO2 gradient would be dependent on resistance to diffusion through the cell wall and plasma membrane. Evidence has previously been provided of a large resistance to COz diffusion between the external medium and the carboxylation site (Marcus et al. 1982; Marcus etal. 1986; Badger 1987). If CO2 influx is required to maintain zero 02 exchange and there is considerable resistance of the cell wall and plasma membrane to diffusion of CO2 then a CO2 gradient is created, i.e. the concentration of external CO2 will be higher than that at the carboxylation site. However, it should be noted that Price and Badger (1989) recently calculated a higher permeability of the cyanobacterial cell wall and plasma membrane to CO2 diffusion than had previously been assumed. (ii) A low recycling rate of photorespiratory metabolites to pentose-phosphate pathway intermediates may cause limitation of ribulose1,5-bisphosphate, thereby amplifying the Oz effect. (iii) High photorespiration may lead to an accumulation of metabolites which may inhibit carbon fixation (Chastain and Ogren 1985). (iv) Cyanobacteria may oxidize the primary photorespiratory product phosphoglycolate to CO2. The reducing power - six electrons - produced in these reactions is probably transferred through the respiratory electron-transport chain to oxygen as the terminal electron acceptor. Since some of the electron carriers are common to the respiratory and photosynthetic electron-transport chain (Scherer etal. 1988), an increase in photorespiration may cause inhibition of linear electron transport from H20 to NADP + and a consequent decrease in 02 evolution. The stoichiometry of 02 consumption would depend on whether 02 is reduced to H 2 0 o r to H 2 0 2.

516

Y. Marcus et al.: Photosynthesis in a carboxysome-deficient cyanobacterium mutant

Such a metabolic pathway could involve the oxidation of glyoxylate to CO2 through the dicarboxylic-acid cycle, in which acetyl-CoA, malate, oxaloacetate and pyruvate play a role (Kornberg and Gotto 1961). Codd and Stewart (1973) presented evidence of such a pathway in the cyanobacterium Anabaena cylindrica. Alternatively, glyoxylate could be partially oxidized via the tricarboxylic-acid cycle, although Pearce et al. (1969) have pointed out that this cycle is incomplete in cyanobacteria. Yehouda Marcus was supported by a United States-Israel Binational Agricultural Research and Development Fund (BARD) postdoctoral fellowship and, during the writing of this manuscript, by the Alexander von Humboldt foundation and the Otto Warburg Center for Biotechnology in Agriculture. We thank Professor Nora Reinhold and Professor Aaron Kaplan (Department of Botany, The Hebrew University of Jerusalem, Israel), and Dr. Wil Terazaghi (Department of Plant Biology, Stanford, Calif., USA) for many exciting discussions. We also thank Mrs. Camille Vainstein for her help in preparing the manuscript.

References Badger, M.R. (1987) The CO2 concentrating mechanism in aquatic phototrophs. In: The biochemistry of plants, vol. 10, pp. 219273, Hatch, H.D., Boardman, N.R., eds. Academic Press, San Diego Badger, M.R., Andrews, T.J. (1987) Co-evolution of Rubisco and CO2 concentrating mechanisms. Progr. Photosynth. Res. 9, 601 609 Badger, M.R., Bassett, M., Comins, H.N. (1985) A model for HCO 3 accumulation and photosynthesis in the cyanobacterium Synechococcus sp. : Theoretical prediction and experimental observation. Plant Physiol. 77, 465 471 Chastain, C.J., Ogren, W.L. (1985) Photorespiration induced reduction of ribulose bisphosphate carboxylase activation level. Plant Physiol. 77, 851-856 Codd, G.A., Marsden, W.J.N. (1984) The carboxysomes (polyhedral bodies) of autotrophic prokaryotes. Biol. Rev. 59, 389-422 Codd, G.A., Stewart, W.D.P. (1973) Pathways of glycolate metabolism in the blue green alga Anabaena cylindrica. Arch. Microbiol. 94, 11-28 Coleman, J.R., Seeman, J.R., Berry, J.A. (1982) RuBPcarboxylase in carboxysomes of the blue green algae. Carnegie Inst. Washington Yearb. 81, 83-87 Friedberg, D., Kaplan, A., Ariel, R., Kessel M., Seijffers, J. (1989) The 5' flanking region of the gene encoding the large subunit of Ribulose 1,5 bisphosphate carboxylase/oxygenase is crucial for growth of the cyanobacterium Synechococcus sp. strain PCC 7942 at the level of CO2 in air. J. Bacteriol. 171, 6069-6076 Jordan, D.B., Ogren, W.L. (1981) Species variation in the specificity of ribulose bisphosphate carboxylase/oxygenase. Nature 291, 513-515

Kaplan, A., Berry, J.A. (1981) Glycolate excretion and the oxygen to carbon dioxide net exchange ratio during photosynthesis in Chlamydomonas reinhardtii. Plant Physiol. 67, 229 232 Kaplan, A., Badger, M.R., Berry, J.A. (1980) Photosynthesis and the intracellular inorganic carbon pool in the blue green alga Anabaena variabilis: response to external COz concentration. Planta 149, 219-226 Kaplan, A., Marcus, Y., Reinhold, L. (1988) Inorganic carbon uptake by cyanobacteria. Methods Enzymol. 167, 534-538 Kornberg, H.L., Gotto, A.M. (1961) The metabolism of C2 compounds in microorganismes. Synthesis of cell constituents from glycolate by Pseudomonas sp. Biochem. J. 78, 69 81 Marcus, Y., Zenvirth, D., Harel, E., Kaplan, A. (1982) Induction of HCO 3 transporting capability and high photosynthetic affinity in inorganic carbon by low concentration of CO2 in Anabaena variabilis. Plant Physiol. 69, 1008-1012 Marcus, Y., Harel, E., Kaplan, A. (1983) Adaptation of the cyanobacterium Anabaena variabilis to low CO2 concentration in their environment. Plant Physiol. 71,208-210 Marcus, Y., Schwarz, R., Friedberg, D., Kaplan, A. (1986) High CO2 requiring mutant of Anacystis nidulans R2. Plant Physiol. 82, 610-612 Pearce, J., Leach, C.K., Carr, N.G. (1969) The incomplete tricarboxylic acid cycle in the blue green alga Anabaena variabilis. J. Gen. Microbiol. 55, 371-378 Pierce, J., Carlson, T.J., Williams, J.G.K. (1989) A cyanobacterial mutant requiring the expression of ribulose bisphosphate carboxylase from photosynthetic anaerobe. Proc. Natl. Acad. Sci. USA 86, 5753-5757 Price, G.D., Badger, M.R. (1989) Expression of human carbonic anhydrase in the cyanobacterium Synechococcus PCC 7942 creates a high CO2 requiring mutant phenotype. Plant Physiol. 91,505-513 Raven, J. (1970) Exogenous inorganic sources in plant photosynthesis. Biol. Rev. 45, 167-221 Reinhold, L., Zviman, M., Kaplan, A. (1987) Inorganic carbon fluxes and photosynthesis in cyanobacteria: a quantitative model. Progr. Photosynth. Res. 6, 289-300 Scherer, S., Almon, H., Boger, P. (1988) Interaction of photosynthesis, respiration and nitrogen fixation in cyanobacteria. Photosynthesis Res. 15, 95-114 Somerville, C.R., Portis, A.R., Ogren, W.L. (1982) A mutant of Arabidopsis thaliana which lacks activation of RuBP carboxylase in vivo. Plant Physiol. 70, 381-387 Stanier, R.Y., Kunisawa, K., Mandel, M., Cohen Bazir, G. (1971) Purification and properties of unicellular blue green algae (order Chroococales). Bacteriol. Rev. 35, 171 205 Tolbert, N.E., Husig, D.H., Husig, D.W., Moroney, J.M., Wilson, B.J. (1985) Relation of glycolate excretion to the DIC pool in microalgae. In: Inorganic carbon uptake by aquatic photosynthetic organisms, pp. 211 227, Lucas, W.L., Berry, J.A., eds. American Society of Plant Physiologists, Rockville Woodrow, I.E., Berry, J.A. (1988) Enzymatic regulation of photosynthetic CO2 fixation in C3 plants. Annu. Rev. Plant Physiol. 39, 533-593