During catalysis, all Rubisco (D-ribulose-1,5-bisphosphate carboxylase/oxygenase) enzymes produce traces of several by-products. Some of these by-products are released slowly from the active site of Rubisco from higher plants, thus progressively inhibiting turnover. Prompted by observations that Form I Rubisco enzymes from cyanobacteria and red algae, and the Form II Rubisco enzyme from bacteria, do not show inhibition over time, the production and binding of catalytic by-products was measured to ascertain the underlying differences. In the present study we show that the Form IB Rubisco from the cyanobacterium Synechococcus PCC6301, the Form ID enzyme from the red alga Galdieria sulfuraria and the low-specificity Form II type from the bacterium Rhodospirillum rubrum all catalyse formation of by-products to varying degrees; however, the by-products are not inhibitory under substrate-saturated conditions. Study of the binding and release of phosphorylated analogues of the substrate or reaction intermediates revealed diverse strategies for avoiding inhibition. Rubisco from Synechococcus and R. rubrum have an increased rate of inhibitor release. G. sulfuraria Rubisco releases inhibitors very slowly, but has an increased binding constant and maintains the enzyme in an activated state. These strategies may provide information about enzyme dynamics, and the degree of enzyme flexibility. Our observations also illustrate the phylogenetic diversity of mechanisms for regulating Rubisco and raise questions about whether an activase-like mechanism should be expected outside the green-algal/higher-plant lineage.

INTRODUCTION

The carboxylating enzyme Rubisco (D-ribulose-1,5-bisphosphate carboxylase/oxygenase, EC 4.1.1.39) is the central catalyst responsible for CO2 acquisition in all schemes of aerobic photosynthesis [13]. Despite the intense natural selection that presumably accompanies this critical role, all forms of Rubisco are characterized by low maximum catalytic rates (1–12 s−1) and a tendency to confuse the carboxylation substrate CO2 with O2, which, in plants, leads to the phenomenon known as photorespiration [1,2].

All forms of Rubisco have a common evolutionary origin, showing clear similarities in amino acid sequences, and the three-dimensional structures of the active sites are nearly superimposable [3,4]. This single phylogeny is, nevertheless, deeply branched and several major nodes may be discerned [3,5]. Form II Rubiscos, found in bacteria and some dinoflagellates, are homodimers of ∼52 kDa large subunits (L2) with active sites formed in the interfaces between the subunits such that there are two active sites per dimer. Sometimes these Form II dimers are assembled into higher oligomers [(L2)n, where n=1–5]. The more common Form I type is found in bacteria, algae and higher plants. This has the same fundamental L2 unit but it is assembled into a more elaborate quaternary structure where four such dimers are arranged like the staves of a barrel that is capped at each end with a tetramer of 12–18 kDa small subunits which assist in holding the structure together and massively stimulate catalytic activity [6] [(L2)4(S4)2=L8S8]. The Form I class is further subdivided into the ‘green’ subclass, found in bacteria and cyanobacteria (Form IA and Form IB) and green algae and higher plants (Form IB), and the ‘red’ subclass, found in bacteria (Form IC) and non-green algae (Form ID). The green and red subclasses are distinguished by a number of structural features, the most notable being C-terminal extensions of the small subunits that assemble together in each S4 cap to form a distinctive small barrel that is inserted into the lumen of the larger L8 barrel [7,8].

Structural diversity is paralleled by diversity in catalytic properties, most notably the specificity for CO2 relative to O2 which is generally lowest in the Form II class, intermediate in Forms IA, IB and IC and highest in Form ID [1,3]. As the structure of the active site is highly conserved, differences in catalysis must be due to changes in structure that are remote from the active site [3,4]. However, crystal structures do not give any information about the enzyme dynamics or the degree of enzyme flexibility. There is considerable evidence that enzyme activity is dependent on enzyme flexibility [9], and this may be responsible for the observed differences in Rubisco catalytic properties. It has recently been suggested that all Rubisco enzymes may be nearly perfectly optimized to the differing conditions in their subcellular environment, in which tight binding of the carboxylation intermediate reduces the oxygenation reaction, but is balanced by a reduced cleavage rate due to the tight binding [10].

Rubisco is activated by an unusual cofactor: CO2 (distinct from substrate CO2) in the form of a carbamate attached to a lysyl sidechain that completes the binding site for the catalytically essential Mg2+ ion and also extracts protons at several stages during catalysis [11]. If the substrate ribulose-P2 (D-ribulose 1,5-bisphosphate) binds to Form IB Rubisco from higher plants before carbamylation, the addition of CO2 and Mg2+ is greatly retarded. Plants and green algae have a motor protein, Rubisco activase, which disrupts the inactive enzyme–ribulose-P2 complex, with the aid of ATP hydrolysis, thus facilitating carbamylation and regulating Rubisco's activity according to the demands of photosynthetic metabolism [12].

The catalytic pathways of carboxylation and oxygenation involve a common enolization of ribulose-P2 to produce an enediol(ate) which is attacked at C-2 by CO2 or O2 (1). The reactions then pass through an analogous sequence of intermediates [1,11,13] to yield two molecules of P-glycerate or a molecule each of P-glycerate (3-phospho-D-glycerate) and P-glycolate (2-phosphoglycolate) respectively. Several of these intermediates, including the enediol, are not perfectly stabilized in the active site and decay to produce a range of non-phosphorylated, mono-phosphorylated and bisphosphorylated by-products [1,1418]. Two bisphosphorylated products, xylulose-P2 (D-xylulose 1,5-bisphosphate; derived by misprotonation of the enediol) and pentodiulose-P2 (D-glycero-2,3-pentodiulose 1,5-bisphosphate; arising from elimination of H2O2 from the peroxyketone intermediate of the oxygenase pathway and also by non-enzymatic oxidation of ribulose-P2) are particularly important because they are slow to dissociate from the higher-plant enzyme and are slow-binding inhibitors of catalysis [14,17,18]. Pentodiulose-P2 can be rearranged to from carboxytetritol-P2 (2′-carboxy-D-tetritol 1,5-bisphosphate) [16]. Xylulose-P2 also binds tightly to the uncarbamylated active site, promoting decarbamylation [19]. Inhibition by these compounds causes progressive loss of activity during catalysis in vitro, a characteristic of the higher-plant enzyme [17]. This self-inhibition is alleviated by a mutation which slightly alters the structure of the active site [18] or, in vivo, by the action of Rubisco activase [12].

Reactions catalysed by Rubisco

Scheme 1
Reactions catalysed by Rubisco

The predominant reactions carried out by Rubisco are the carboxylation or oxygenation of ribulose-P2, as indicated by the solid arrows. Intermediate compounds may also be partitioned to form other products, some of which are shown here indicated by the dotted arrows.

Scheme 1
Reactions catalysed by Rubisco

The predominant reactions carried out by Rubisco are the carboxylation or oxygenation of ribulose-P2, as indicated by the solid arrows. Intermediate compounds may also be partitioned to form other products, some of which are shown here indicated by the dotted arrows.

Prompted by recent better understanding of Rubisco self-inhibition [18] and observations that self-inhibition is only observed in higher-plant Rubiscos, with other forms of the enzyme showing varying degrees of inhibition following exposure of their uncarbamylated forms to ribulose-P2 [2022], in the present study we undertook a detailed investigation of the formation of catalytic by-products and inhibition by bisphosphates using Rubiscos drawn from the three major branches of Rubisco's phylogeny.

EXPERIMENTAL

Materials

Rubiscos were purified as described previously from tobacco [Nicotiana tabacum L. cv. Petit Havana (N,N)] [23] and spinach (Spinacia oleracea L.) [24,25]. The Synechococcus PCC6301 rbcLS operon in plasmid pSH1 [6] was expressed in Escherichia coli HB101 at 37 °C and the resultant Rubisco was purified as previously described [25]. The rbcM gene from Rhodospirillum rubrum, fused to a 5′ hexahistidyl-coding sequence in plasmid pRrHis [26], was expressed in E. coli BL21 (pLysS) at 25 °C in 2YT medium containing 30 μg·ml−1 kanamycin. The resultant Rubisco was purified by metal-affinity chromatography on Ni-NTA (Ni2+-nitrilotriacetate)–agarose (Qiagen) according to the manufacturer's instructions and stored snap-frozen at −80 °C in 9 mM sodium phosphate buffer (pH 7.6) containing 45 mM NaCl, 1 mM EDTA and 10% (v/v) glycerol.

The thermophilic rhodophyte Galdieria sulfuraria (strain 2393, UTEX Culture Collection Algae, University of Texas, Austin, TX, U.S.A.) was cultured in Cyanidium medium [27] in 10–20 litre glass containers at 30 °C under constant illumination (480–660 μmol of quanta·m−2·s−1 at the surface) and vigorous sparging with air. Cells were harvested with a continuous flow centrifuge, resuspended in 3 vol. of 100 mM Hepps/NaOH buffer (pH 8.0) containing 20 mM MgCl2, 1 mM EDTA, 1 mM PMSF and 1 mM dithiothreitol, lysed with a French Press (110 mPa) and clarified by centrifugation (13000 g for 10 min at 4 °C). The supernatant was fractionated with poly(ethylene glycol) 3350 (Sigma) to collect the protein precipitating between 10 and 20% (w/v), and the blue pellet was resuspended in 100 mM Hepps/NaOH buffer (pH 8.0) containing 20 mM MgCl2 and 1 mM EDTA. This preparation was used for measuring carbamylation and the binding of [14C]carboxypentitol-P2 (unresolved isomeric mixture of 2′-carboxy-D-arabinitol 1,5-bisphosphate and 2′-carboxy-D-ribitol 1,5-bisphosphate). For other experiments, Rubisco was further purified by chromatography on a MonoQ HR 5/5 column (GE Biosciences) equilibrated with 50 mM Tris/HCl buffer (pH 8.0) containing 1 mM EDTA and proteins were eluted as a 0–350 mM NaCl gradient over 28 column volumes. Rubisco activity eluted at 150 mM NaCl. Active fractions were dialysed and stored snap-frozen at −80 °C in 18 mM sodium phosphate buffer (pH 8.0) containing 1 mM EDTA and 10% (v/v) glycerol. SDS/PAGE analysis of this preparation revealed two dominant (>90%) protein bands at 54 and 17 kDa, corresponding to the Rubisco large and small subunits respectively.

Unlabelled and 1-3H-labelled ribulose-P2 were synthesized, purified, desalted and stored in liquid nitrogen as described previously [17]. Unlabelled and carboxyl-14C-labelled carboxypentitol-P2 were synthesized and carboxyarabinitol-P2 (2′-carboxy-D-arabinitol-1,5-bisphosphate) was purified from these preparations as described in [28]. Xylulose-P2 was synthesized and purified as described in [18]. Carboxyarabinitol-1-P (2′-carboxy-D-arabinitol-1-phosphate) was synthesized as described previously [29] and was purified by the same anion-exchange procedure used for carboxyarabinitol-P2 [28].

Measurement of activity

Prior to the assay, Rubisco was activated in a buffer containing 100 mM Hepps/NaOH (pH 8.0), 20 mM MgCl2, 1 mM EDTA and 20 mM NaHCO3. Tobacco Rubisco was also incubated at 50 °C for 10 min. Carboxylase activity was measured spectrophotometrically at 25 °C as described previously [18,30]. Assay mixtures contained 500 μM ribulose-P2, 5–20 nM active sites, and 5 mM (for G. sulfuraria), 20 mM (for tobacco) or 50 mM (for Synechococcus and R. rubrum) NaHCO3. Xylulose-P2 carboxylase activity was measured using the same method, but with 25–100 μM xylulose-P2 instead of ribulose-P2 and 0.5–2.0 μM active sites. Xylulose-P2 production was measured spectrophotometrically under CO2- and O2-free conditions as described previously [18]. Where non-linear time courses were observed, the data for product accumulation as a function of time were fitted to the following equation (Equation 1):

 
formula
(1)

to estimate vi and vf, the initial and final steady-state rates respectively, and kobs, the observed first-order rate constant.

Reaction products

Assays were carried out at 25 °C in buffer containing 100 mM Hepps/NaOH (pH 8.0), 20 mM MgCl2, 1 mM EDTA and 1 μM [1-3H]ribulose-P2 (3.2 MBq·nmol−1), as described previously [18]. Assays under carboxylating conditions used buffer that had been sparged with nitrogen before the addition of 20 mM NaHCO3, whereas assays under oxygenating conditions used buffer that had been sparged with oxygen, and had a final NaHCO3 concentration of 10 μM (carried over with the pre-activated enzyme solution). Assays were initiated with pre-activated enzyme (described earlier) to a final concentration of >3 μM active sites. The reaction was incubated at 25 °C for 30 min, before SDS was added to a final concentration of 2% (w/v) and borate was added to a final concentration of 10 mM. The denatured proteins were removed using an Ultrafree-MC filter unit (Mr cut-off 10000; Millipore), and the ribulose-P2 derivatives were separated by anion-exchange chromatography as described previously [18].

Competitive inhibition by xylulose-P2

Rapid-equilibrium inhibition by xylulose-P2 was measured as described in [18,31] by adding pre-activated enzyme (final concentration of 12 nM) to initiate spectrophotometric assays containing 2.5–500 μM ribulose-P2 and 0–75 μM xylulose-P2.

Release of inhibitors from tobacco, Synechococcus and R. rubrum Rubiscos

Tobacco, Synechococcus PCC6301 and R. rubrum Rubiscos were decarbamylated by incubating at 25 °C for 20 min in 100 mM Hepps/NaOH buffer (pH 8.0) containing 1 mM EDTA. These uncarbamylated, metal-free Rubiscos (2–7 μM) were incubated at 25 °C with 0.5 mM ribulose-P2 or 80 μM xylulose-P2 for 60 min. Carbamylated Rubiscos in the same buffer plus 20 mM MgCl2 and 5–50 mM NaHCO3 were similarly incubated with 20 μM carboxyarabinitol-1-P. Spectrophotometric carboxylase assays were initiated by adding the pre-incubated Rubisco preparations to otherwise complete assay mixtures (final Rubisco concentration of 150 nM). Xylulose-P2 and carboxyarabinitol-1-P concentrations in the assay mixtures were less than the Ki values and the ribulose-P2 concentration was 25 times the Km, thus minimizing rapid-equilibrium inhibition. Activity increased exponentially and the data were fitted to Equation 1 to estimate kobs.

To measure the release of carboxyarabinitol-P2, Rubisco from tobacco, Synechococcus or R. rubrum (1.5 μM) was carbamylated in 50 mM Hepps/NaOH buffer (pH 8.0) containing 10 mM MgCl2, 10 mM NaHCO3 and 1 mM EDTA, incubated at 25 °C for 10 min in the presence of 18 μM [carboxy-14C]carboxypentitol-P2, and separated from unbound radioactivity by gel filtration through Sephadex G-50 fine (GE Biosciences) that had been equilibrated with the same buffer. Unlabelled carboxypentitol-P2 was then added (200 μM) and, at intervals thereafter, samples were reapplied to the gel-filtration column to allow measurement of the radioactivity that remained bound. Data were fitted to the following exponential equation, derived from the differentiated form of Equation 1 (Equation 2):

 
formula
(2)

where I* is the amount of labelled inhibitor that has been released from the enzyme, It is the total amount of labelled inhibitor, and kobs is the observed first-order rate constant.

Release of inhibitors from G. sulfuraria Rubisco

G. sulfuraria Rubisco (3 μM) was decarbamylated by overnight dialysis at 4 °C against constantly nitrogen-sparged 100 mM Hepps/NaOH buffer (pH 8.0) containing 1 mM EDTA. Release of ribulose-P2 was measured by incubating the preparation at 25 °C for 10 min with 6 μM [1-3H]ribulose-P2. Unbound label was then removed by passing a sample through a 1 cm×20 cm column of Sephadex G-50 fine equilibrated with the dialysis buffer. Unlabelled ribulose-P2 (0.4 mM) was then added at zero time to the high-molecular-mass fraction and, after the stated times at 25 °C, samples were again gel-filtered as above and the protein-bound radioactivity remaining was measured. Data were fitted to Equation 2. To measure the release of xylulose-P2, the preparation was incubated with 200 μM xylulose-P2 for 10 min at 25 °C and then gel-filtered as above. The high-molecular-mass fraction was then added at zero time to the spectrophotometric carboxylase assay solution lacking ribulose-P2 (final Rubisco concentration of 150 nM). At various times, 0.5 mM ribulose-P2 was added to initiate the reaction and the initial rate (v) was recorded and expressed as the fraction of the similarly measured activity of a control without exposure to xylulose-P2 (vcon). The release of carboxyarabinitol-1-P and carboxyarabinitol-P2 were measured as described above.

Binding of carboxyarabinitol-P2

Pre-activated Rubisco (final concentration in assay of 5–30 nM) was added to the spectrophotometric assays as described above. Carboxyarabinitol-P2 was added to the assay (final concentration of 0.1–2.0 μM) and the mixture was incubated at 25 °C for 0–240 s, before ribulose-P2 (final concentration of 0.5 mM) was added to initiate the reaction. Alternatively, pre-activated Rubisco was incubated with carboxyarabinitol-P2 (5–600 μM) at 25 °C. At intervals thereafter, an aliquot was used to initiate an assay containing 0.5 mM ribulose-P2 (final concentration in assay of 5–30 nM active sites).

Analysis of carboxyarabinitol-P2 inhibition

Data were analysed in terms of a two-step model [18,28] where a rapid-equilibrium interaction of the inhibitor (I) with Rubisco (E) to form an initial complex (EI) is followed by a slow isomerization to a much tighter complex (EI*):

 
formula

where k1, k−1, k2 and k−2 are the rate constants for the individual forward and reverse reactions. When [I]≫[E] and k2k1:

 
formula
(3)

where Ki=k−1/k1.

RESULTS

Only higher-plant Rubiscos demonstrate self-inhibition

Of the four phylogenetically divergent Rubiscos included in this study, only that from the higher plant, tobacco, showed any sign of progressive loss of activity during catalysis at saturating CO2 and ribulose-P2 concentrations. The other three all had simple linear time courses (Figure 1).

Time courses of carboxylase assays

Figure 1
Time courses of carboxylase assays

To initiate the reactions, pre-activated Rubisco preparations were added to otherwise complete assay mixtures with saturating CO2 concentrations as described in the Experimental section. Data for product formation (a) were fitted to Equation 1 or to a linear regression, as appropriate, and the rate as a function of time (b) was calculated from the resultant parameter estimates.

Figure 1
Time courses of carboxylase assays

To initiate the reactions, pre-activated Rubisco preparations were added to otherwise complete assay mixtures with saturating CO2 concentrations as described in the Experimental section. Data for product formation (a) were fitted to Equation 1 or to a linear regression, as appropriate, and the rate as a function of time (b) was calculated from the resultant parameter estimates.

All Rubisco enzymes form by-products during catalysis

All Rubiscos catalysed the production of P-glycerate virtually exclusively under anaerobic, CO2-saturating conditions. Other phosphorylated products did not accumulate in amounts detectable chromatographically (Table 1). Progressive inactivation of the higher-plant enzyme is least under these conditions, and the residual inhibition is likely to be caused by chromatographically undetectable traces of pentodiulose-P2 present in the ribulose-P2 preparations [17]. However, when CO2 was scarce and O2 was present, the products of the side reactions became apparent. All Rubiscos partitioned 1–3% of their product to xylulose-P2. In addition, oxygenation by-products were produced by the tobacco and R. rubrum enzymes. The former converted 0.3% of the initial ribulose-P2 into pentodiulose-P2, whereas the latter produced 0.6% in the form of the rearranged product of pentodiulose-P2, carboxytetritol-P2 (Table 1). This is consistent with other reports of pentodiulose-P2 production by spinach and R. rubrum Rubisco [32].

Table 1
Products of Rubisco catalysis

Products derived from near-complete conversion of [1-3H]ribulose-P2 were analysed by anion-exchange chromatography after assays under carboxylating (saturating CO2, nitrogen-sparged) and predominantly oxygenating (CO2-free, air-sparged, 10 μM NaHCO3) conditions as described previously [18]. Although oxygenase activity was promoted under the latter conditions, the residual 10 μM NaHCO3 unavoidably carried over with the CO2/Mg2+, pre-activated Rubisco preparation allowed some carboxylation to occur, particularly with the high-specificity enzymes. Values shown are the amount of each radiolabelled product as a percentage of the ribulose-P2 consumed during the reaction (±S.D., n=2–4). Small amounts of some early eluting non-phosphorylated compounds, which include pyruvate and products of trace phosphatase contamination, did not resolve well, and are not shown. Xylulose-P2 was distinguished from co-chromatographing residual ribulose-P2 by further chromatography on a sugar-separating column following phosphatase treatment [18].

   [1-3H]ribulose-P2 consumed (%) 
Rubisco [HCO3] (mM) [O2] (mM) P-glycerate P-glycolate Xylulose-P2 Pentodiulose-P2 Carboxytetritol-P2 
Tobacco 20 98.6±0.4 <0.1 <0.1 <0.1 <0.1 
Tobacco 0.01 0.24 29.8±0.4 67.6±2.0 1.1±0.4 0.3±0.1 <0.1 
Synechococcus 50 98.8±0.1 <0.1 <0.1 <0.1 <0.1 
Synechococcus 0.01 0.24 17.4±2.5 77.4±4.2 2.8±0.5 <0.1 <0.1 
R. rubrum 50 98.0±0.2 <0.1 <0.1 <0.1 <0.1 
R. rubrum 0.01 0.24 5.6±2.5 92.0±2.2 0.9±0.4 <0.1 0.6±0.1 
G. sulfuraria 96.6±0.1 <0.1 <0.1 <0.1 <0.1 
G. sulfuraria 0.01 0.24 39.3±3.8 56.1±4.5 2.1±0.4 <0.1 <0.1 
   [1-3H]ribulose-P2 consumed (%) 
Rubisco [HCO3] (mM) [O2] (mM) P-glycerate P-glycolate Xylulose-P2 Pentodiulose-P2 Carboxytetritol-P2 
Tobacco 20 98.6±0.4 <0.1 <0.1 <0.1 <0.1 
Tobacco 0.01 0.24 29.8±0.4 67.6±2.0 1.1±0.4 0.3±0.1 <0.1 
Synechococcus 50 98.8±0.1 <0.1 <0.1 <0.1 <0.1 
Synechococcus 0.01 0.24 17.4±2.5 77.4±4.2 2.8±0.5 <0.1 <0.1 
R. rubrum 50 98.0±0.2 <0.1 <0.1 <0.1 <0.1 
R. rubrum 0.01 0.24 5.6±2.5 92.0±2.2 0.9±0.4 <0.1 0.6±0.1 
G. sulfuraria 96.6±0.1 <0.1 <0.1 <0.1 <0.1 
G. sulfuraria 0.01 0.24 39.3±3.8 56.1±4.5 2.1±0.4 <0.1 <0.1 

Although the instability of pentodiulose-P2 [17] limits the extent to which its interactions with Rubisco can be studied, some information could be obtained by deliberately oxidizing ribulose-P2 in the presence of Cu2+ ions [17,18] to produce a mixture that contained approx. 10% of the original ribulose-P2 in the form of the oxidation products, pentodiulose-P2 and carboxytetritol-P2 (Table 2). The oxidized mixture was exposed, under carboxylating conditions, to the four Rubiscos (with a molar excess of active sites over ribulose-P2 plus oxidation products). After the remaining ribulose-P2 was converted into P-glycerate, the distributions of products in the total reaction mixture and in the protein-bound fraction recoverable after rapid gel filtration were re-analysed. Tobacco, Synechococcus and G. sulfuraria Rubiscos did not materially alter the approx. 5:1 ratio between pentodiulose-P2 and carboxytetritol-P2 and the majority of both compounds was recovered bound to the Rubisco (Table 2). By contrast, R. rubrum Rubisco catalysed conversion of much of the pentodiulose-P2 into carboxytetritol-P2, with all of the former and most of the latter being released from the enzyme (Table 2). Xylulose-P2 production was also measured in continuous spectrophotometric assays under conditions designed to reduce the CO2 and O2 concentrations to zero while leaving an excess of ribulose-P2 remaining (Figure 2). Xylulose-P2 production accelerated in all assays as the residual CO2 was exhausted, eventually becoming linear at rates between 0.05% and 1.4% of the relevant CO2-saturated carboxylation rates (Table 3).

Production and carboxylation of xylulose-P2

Figure 2
Production and carboxylation of xylulose-P2

Xylulose-P2 formation (○) under CO2- and O2-free conditions, and xylulose-P2 carboxylation (■) at saturating CO2 and O2-free conditions with 50 μM xylulose-P2, were measured as described in the Experimental section. All assays showed initial lags (see text) and the data were fitted to Equation 1 to estimate the final, post-lag rate (vf). Xylulose-P2 formation by R. rubrum Rubisco continued to accelerate beyond the data shown and observations were extended to 40 min to allow estimation of vf.

Figure 2
Production and carboxylation of xylulose-P2

Xylulose-P2 formation (○) under CO2- and O2-free conditions, and xylulose-P2 carboxylation (■) at saturating CO2 and O2-free conditions with 50 μM xylulose-P2, were measured as described in the Experimental section. All assays showed initial lags (see text) and the data were fitted to Equation 1 to estimate the final, post-lag rate (vf). Xylulose-P2 formation by R. rubrum Rubisco continued to accelerate beyond the data shown and observations were extended to 40 min to allow estimation of vf.

Table 2
Binding and interconversion of ribulose-P2 oxidation products

As described previously [18], [1-3H]ribulose-P2 was oxidized in the presence of Cu2+ ions, reacted to near-completion under CO2-saturating conditions in the presence of excess amounts of the various Rubiscos, and the total products, and those remaining bound to the enzyme (isolated by rapid gel filtration), were analysed by anion-exchange chromatography. Values shown are the amount of radiolabel in pentodiulose-P2 and carboxytetritol-P2 as percentages of the total radioactivity in the starting ribulose-P2. Results for a single experiment are shown because the amounts of oxidation products produced by the oxidation procedure, and thus available for the enzymatic reaction (shown in the first row), varied somewhat between replicate oxidations, reflecting the instability of pentodiulose-P2 [17].

 [1-3H]ribulose-P2 consumed (%) 
 Pentodiulose-P2 Carboxytetritol-P2 
Rubisco Total Bound Total Bound 
Oxidized [1-3H]ribulose-P2 (no Rubisco) 8.9  1.8  
Tobacco 6.1 4.2 1.0 0.6 
Synechococcus 5.1 3.8 1.0 1.3 
R. rubrum 1.8 <0.1 5.4 1.0 
G. sulfuraria 4.7 5.2 1.1 1.2 
 [1-3H]ribulose-P2 consumed (%) 
 Pentodiulose-P2 Carboxytetritol-P2 
Rubisco Total Bound Total Bound 
Oxidized [1-3H]ribulose-P2 (no Rubisco) 8.9  1.8  
Tobacco 6.1 4.2 1.0 0.6 
Synechococcus 5.1 3.8 1.0 1.3 
R. rubrum 1.8 <0.1 5.4 1.0 
G. sulfuraria 4.7 5.2 1.1 1.2 
Table 3
Kinetic data
  Value 
Parameter Rubisco… Tobacco Synechococcus R. rubrum G. sulfuraria 
Catalytic parameters      
 CO2/O2 specificity  82a 43b 12c 166a 
Vmax (s−1)f  2.9±0.1e 13.9±0.1 4.2±0.1 1.2±0.1 
Km(CO2) (μM)  10.7a 280b 67d 3.3a 
Km(ribulose-P2) (μM)  58±4 54±3 3.9±1 376±42 
Ki(xylulose-P2) (μM)  4.8±0.4e 12.2±1.0 3.1±0.7 89±29 
 Self-inhibition during in vitro assays  Yes No No No 
Side reactions      
 Xylulose-P2 production (s−1)i  9.2±0.03×10−3e 4.92±0.06×10−3 56.9±6.5×10−3 10.5±0.08×10−3 
  (0.32%)g (0.04%)g (1.4%)g (0.88%)g 
 Xylulose-P2 carboxylation (s−1)l  0.46±0.02×10−3e 4.17±0.03×10−3 5.13±0.03×10−3 2.75±0.03×10−3 
  (0.02%)g (0.03%)g (0.12%)g (0.23%)g 
 Pyruvate synthesis (% of carboxylation)m  0.68% 0.73% 0.68% 1.96% 
Release from tightly bound complexes kobs (s−1)h      
 Ribulose-P2  2.09±0.01×10−3e 28.9±0.5×10−3 20.0±0.5×10−3 0.13±0.01×10−3 
 Xylulose-P2  0.60±0.01×10−3e 29.8±0.5×10−3 19.5±0.5×10−3 0.07±0.03×10−3 
 Carboxyarabinitol-1-P  2.8±0.1×10−3 8.3±0.1×10−3 19.4±0.2×10−3 0.8±0.1×10−3 
Carboxyarabinitol-P2 inhibitioni      
k2 (s−1 0.17±0.04e 0.28±0.09 >1.0j 0.086±0.016 
k−2 (s−1 <10−7k <10−7 k 4.1±0.2×10−6 <10−7k 
Ki (μM)  2.3±0.7e 1.72±1.10 >2h 220±101 
  Value 
Parameter Rubisco… Tobacco Synechococcus R. rubrum G. sulfuraria 
Catalytic parameters      
 CO2/O2 specificity  82a 43b 12c 166a 
Vmax (s−1)f  2.9±0.1e 13.9±0.1 4.2±0.1 1.2±0.1 
Km(CO2) (μM)  10.7a 280b 67d 3.3a 
Km(ribulose-P2) (μM)  58±4 54±3 3.9±1 376±42 
Ki(xylulose-P2) (μM)  4.8±0.4e 12.2±1.0 3.1±0.7 89±29 
 Self-inhibition during in vitro assays  Yes No No No 
Side reactions      
 Xylulose-P2 production (s−1)i  9.2±0.03×10−3e 4.92±0.06×10−3 56.9±6.5×10−3 10.5±0.08×10−3 
  (0.32%)g (0.04%)g (1.4%)g (0.88%)g 
 Xylulose-P2 carboxylation (s−1)l  0.46±0.02×10−3e 4.17±0.03×10−3 5.13±0.03×10−3 2.75±0.03×10−3 
  (0.02%)g (0.03%)g (0.12%)g (0.23%)g 
 Pyruvate synthesis (% of carboxylation)m  0.68% 0.73% 0.68% 1.96% 
Release from tightly bound complexes kobs (s−1)h      
 Ribulose-P2  2.09±0.01×10−3e 28.9±0.5×10−3 20.0±0.5×10−3 0.13±0.01×10−3 
 Xylulose-P2  0.60±0.01×10−3e 29.8±0.5×10−3 19.5±0.5×10−3 0.07±0.03×10−3 
 Carboxyarabinitol-1-P  2.8±0.1×10−3 8.3±0.1×10−3 19.4±0.2×10−3 0.8±0.1×10−3 
Carboxyarabinitol-P2 inhibitioni      
k2 (s−1 0.17±0.04e 0.28±0.09 >1.0j 0.086±0.016 
k−2 (s−1 <10−7k <10−7 k 4.1±0.2×10−6 <10−7k 
Ki (μM)  2.3±0.7e 1.72±1.10 >2h 220±101 

Data from a, [33]; b, [25]; c, [49]; d,[50]; e,[18]. Other data are from the present study [±S.E.M. derived from the curve-fitting procedures (Figures 2–5)].

f

Ribulose-P2 carboxylation at 500 μM ribulose-P2 and 5 mM (for G. sulfuraria), 20 mM (for tobacco) or 50 mM (for Synechococcus and R. rubrum) NaHCO3 at 25 °C.

g

Value as a percentage of the ribulose-P2 carboxylation rate.

h

Data from Figures 3 and 4.

i

Data from Figure 5.

j

Inhibition fast and linearly responsive to carboxyarabinitol-P2 concentration to 1.5 μM (Figure 5a).

k

No detectable release in 4 days (Figure 5c).

l

vf estimates from the data of Figure 2.

m

Measured spectrophotometrically as described previously [15].

Partitioning of the product of carboxylation to pyruvate was approx. 0.7% for tobacco, Synechococcus and R. rubrum Rubiscos (Table 3), in line with previous observations [15], but greatly increased to approx. 2% for the G. sulfuraria enzyme.

Inhibition by xylulose-P2

Rapid-equilibrium inhibition by xylulose-P2 was measured in initial-rate assays following simultaneous addition of xylulose-P2 and ribulose-P2. The Ki for xylulose-P2 varied from 3 to 90 μM, and the ratio to the Km for ribulose-P2 varying over an even wider range, with tobacco Rubisco being most potently inhibited by xylulose-P2 and R. rubrum Rubisco being inhibited the least (Table 3). Curiously, the Km values for ribulose-P2 determined in these experiments for the three Form I Rubiscos were consistently 3–4-fold greater than previous estimates [25,33]. This was not likely to be caused by competitive inhibition by additives necessarily present in the spectrophotometric assay, because recent experiments with the 14CO2-fixation assay, which lacks these additives, yielded a value (61 μM) for Synechococcus Rubisco, similar to that listed in Table 3 (S.M. Whitney and T.J. Andrews, personal communication).

Carboxylation of xylulose-P2

All four Rubiscos catalysed a low rate of carboxylation of xylulose-P2 (Figure 2). Whereas for tobacco the rate declined with time [18], in a similar way to that seen with ribulose-P2 as substrate, the time courses for the other three enzymes were linear after an initial lag while intermediates in the coupling system reached their steady-state levels. The maximum rates with xylulose-P2 varied over a 10-fold range, both in absolute terms and as a percentage of the rate of ribulose-P2 carboxylation (Table 3). Also notable was the variation in the ratio of xylulose-P2 carboxylation rate to the rate of xylulose-P2 production from ribulose-P2 under CO2- and O2-free conditions (Figure 2). Xylulose-P2 production by tobacco Rubisco was 20 times faster than xylulose-P2 carboxylation, whereas, for Synechococcus Rubisco, the two rates were similar (Table 3). This would prevent sequestration of significant amounts of the latter enzyme by xylulose-P2.

Release of inhibitors from tightly bound complexes

Uncarbamylated, metal-free Rubisco (E) can bind to ligands including xylulose-P2 and the substrate, ribulose-P2. The strength of this interaction can be measured by pre-incubating the E form of the enzyme with the ligand, and then measuring the rate at which activity is recovered when the treated enzyme is added to an assay mixture containing saturating concentrations of CO2, Mg2+ and ribulose-P2. The activity increases from an initial zero to a final steady state that matches the activity in controls containing the fully activated (ECM) form of the enzyme (Figure 3). The observed rate constant for this activation was increased for the Synechococcus and R. rubrum Rubisco enzymes compared with that of tobacco Rubisco, and was similar for both xylulose-P2 and ribulose-P2, with nearly full activity regained within 1 min. The activation rate of R. rubrum Rubisco was higher than that previously observed [34], but the previous work had been carried out at 2 °C, and it would be expected that the rate of inhibitor release would be higher at higher temperatures.

Release of inhibitors from tobacco, Synechococcus and R. rubrum Rubiscos

Figure 3
Release of inhibitors from tobacco, Synechococcus and R. rubrum Rubiscos

Uncarbamylated, metal-free Rubiscos were incubated at 25 °C with 0.5 mM ribulose-P2 (■) or 80 μM xylulose-P2 (▲) for 60 min. Carbamylated Rubiscos were similarly incubated with 20 μM carboxyarabinitol-1-P (○). Spectrophotometric carboxylase assays (see the Experimental section) were initiated by adding the pre-incubated Rubisco preparations to otherwise complete assay mixtures. Xylulose-P2 and carboxyarabinitol-1-P concentrations in the assay mixtures were less than the Ki values and the ribulose-P2 concentration was 25 times the Km, thus minimizing rapid-equilibrium inhibition. Activity increased exponentially and the data were fitted to Equation 1 to estimate kobs.

Figure 3
Release of inhibitors from tobacco, Synechococcus and R. rubrum Rubiscos

Uncarbamylated, metal-free Rubiscos were incubated at 25 °C with 0.5 mM ribulose-P2 (■) or 80 μM xylulose-P2 (▲) for 60 min. Carbamylated Rubiscos were similarly incubated with 20 μM carboxyarabinitol-1-P (○). Spectrophotometric carboxylase assays (see the Experimental section) were initiated by adding the pre-incubated Rubisco preparations to otherwise complete assay mixtures. Xylulose-P2 and carboxyarabinitol-1-P concentrations in the assay mixtures were less than the Ki values and the ribulose-P2 concentration was 25 times the Km, thus minimizing rapid-equilibrium inhibition. Activity increased exponentially and the data were fitted to Equation 1 to estimate kobs.

There were initial difficulties in obtaining the decarbamylated form of G. sulfuraria Rubisco, which necessitated overnight dialysis of the enzyme against nitrogen-sparged buffer. Release of ribulose-P2 and xylulose-P2 from decarbamylated G. sulfuraria Rubisco was too slow to be measured by the spectrophotometric assay. Release of ribulose-P2 was measured by the exchange of labelled ligands (Figure 4a). In this experiment, uncarbamylated Rubisco was incubated with [1-3H]ribulose-P2 to form the E–ribulose-P2 complex. Unbound inhibitors were removed by gel filtration, and an excess of unlabelled ribulose-P2 was added so that when a labelled ribulose-P2 molecule was released, an unlabelled ribulose-P2 molecule would replace it. Exchange of bound [1-3H]ribulose-P2 with [1-1H]ribulose-P2 was measured by gel filtration, allowing kobs to be measured using Equation 2. Release of ribulose-P2 was much slower than for other enzymes, with a half-time of nearly 90 min. Release of xylulose-P2 was measured by initially incubating the decarbamylated enzyme with xylulose-P2, and then removing excess inhibitors by gel filtration. Inhibited enzyme was then added to an assay mixture that contained high levels of Mg2+ and NaHCO3, allowing activation of the active sites once the xylulose-P2 was released. Samples were taken over time and ribulose-P2 was added to initiate the reaction, allowing measurement of the reaction rate (Figure 4b). The activity increased over time, and was fitted to the differentiated version of Equation 1 to calculate kobs, the first order rate constant. The initial activity was 20% of the fully activated rate, and was due to the activation occurring during the gel filtration, and incomplete initial decarbamylation of the enzyme. Release of xylulose-P2 was much slower than observed for other Rubisco enzymes, and also slower than the release of ribulose-P2 from the G. sulfuraria enzyme, with half of the activity regained after 160 min.

Release of inhibitors from G. sulfuraria Rubisco

Figure 4
Release of inhibitors from G. sulfuraria Rubisco

G. sulfuraria Rubisco was decarbamylated by overnight dialysis as described in the Experimental section. (a) The preparation was incubated for 10 min with 6 μM [1-3H]ribulose-P2. Unbound label was then removed by gel filtration. Unlabelled ribulose-P2 was added to the inhibited enzyme and, after the stated times, samples were again gel-filtered as above and the protein-bound radioactivity remaining was measured. (b) The preparation was incubated with 200 μM xylulose-P2 for 10 min and then gel-filtered. The inhibited enzyme was then added at zero time to the spectrophotometric carboxylase assay solution lacking ribulose-P2. At the times shown, ribulose-P2 was added to initiate the reaction and the initial rate (v) was recorded and expressed as the fraction of the similarly measured activity of a control without exposure to xylulose-P2 (vcon). (c) The preparation was carbamylated and incubated for 60 min with 20 μM carboxyarabinitol-1-P. A sample was then added to initiate otherwise complete carboxylase assays and the time course of product formation was recorded. The data for all three experiments were fitted to Equation 1 and kobs was estimated.

Figure 4
Release of inhibitors from G. sulfuraria Rubisco

G. sulfuraria Rubisco was decarbamylated by overnight dialysis as described in the Experimental section. (a) The preparation was incubated for 10 min with 6 μM [1-3H]ribulose-P2. Unbound label was then removed by gel filtration. Unlabelled ribulose-P2 was added to the inhibited enzyme and, after the stated times, samples were again gel-filtered as above and the protein-bound radioactivity remaining was measured. (b) The preparation was incubated with 200 μM xylulose-P2 for 10 min and then gel-filtered. The inhibited enzyme was then added at zero time to the spectrophotometric carboxylase assay solution lacking ribulose-P2. At the times shown, ribulose-P2 was added to initiate the reaction and the initial rate (v) was recorded and expressed as the fraction of the similarly measured activity of a control without exposure to xylulose-P2 (vcon). (c) The preparation was carbamylated and incubated for 60 min with 20 μM carboxyarabinitol-1-P. A sample was then added to initiate otherwise complete carboxylase assays and the time course of product formation was recorded. The data for all three experiments were fitted to Equation 1 and kobs was estimated.

Unlike ribulose-P2 and xylulose-P2, carboxyarabinitol-1-P binds most strongly to the carbamylated form of the Rubisco enzyme. The dissociation of carboxyarabinitol-1-P from this complex was measured by adding a small aliquot of inhibited enzyme to a spectrophotometric assay and monitoring the increase in activity. During the assay, the activity increased over time, after initially showing little activity (Figures 3 and 4c). Recovery of activity was less than complete, with most enzymes regaining 80% of the control activity. Tobacco and R. rubrum Rubisco released carboxyarabinitol-1-P at a similar rate to that for the release of ribulose-P2 from the decarbamylated enzyme, whereas G. sulfuraria Rubisco released carboxyarabinitol-1-P at a higher rate and Synechococcus Rubisco released carboxyarabinitol-1-P at a lower rate than ribulose-P2 release.

Inhibition by carboxyarabinitol-P2

Carboxyarabinitol-P2 binds to the carbamylated form of Rubisco in a two-step process involving a rapid-equilibrium phase followed by slow isomerization [28]. Data for Rubisco inhibition versus time were fitted to Equation 1 to calculate the binding constant (kobs) for a range of carboxyarabinitol-P2 concentrations. This was then plotted against the carboxyarabinitol-P2 concentration (Figure 5) and fitted to Equation 3 to calculate the binding rate (k2) and inhibition constant (Ki) (Table 3). Synechococcus Rubisco had a similar binding rate and inhibition constant to tobacco Rubisco, whereas G. sulfuraria had a similar binding rate, but the binding constant was 100 times greater. R. rubrum Rubisco gave a linear response to carboxyarabinitol-P2 concentrations up to 1.5 μM, and had a very high binding constant, with half of the sites occupied after 1 s when assayed at 1.6 μM carboxyarabinitol-P2. As a result, it was not possible to measure the binding constant at higher carboxyarabinitol-P2 concentrations.

Binding and release of carboxyarabinitol-P2

Figure 5
Binding and release of carboxyarabinitol-P2

(a) and (b) Pre-activated tobacco (○), Synechococcus (△), R. rubrum (●) and G. sulfuraria (□) Rubisco was incubated with carboxyarabinitol-P2 and the mixtures were incubated for 0–240 s before being used to initiate carboxylase assays. At each carboxyarabinitol-P2 concentration, initial rates recorded for each period of incubation were fitted to Equation 1 and kobs was estimated. Experiments were carried out in duplicate; the error bars show the range of observations. The resulting data were fitted to the hyperbolic Equation 3, allowing estimation of Ki and k2, except in the case of R. rubrum Rubisco where no saturation was apparent and kobs became too high to measure at carboxyarabinitol-P2 concentrations greater than 1.6 μM. In this case a linear regression fitted the data. (c) Release of carboxyarabinitol-P2 was measured by measuring the exchange of labelled inhibitor, as described in the Experimental section. Only R. rubrum Rubisco showed appreciable exchange of label during 4 days. These data were fitted to Equation 1, allowing estimation of the kobs for release.

Figure 5
Binding and release of carboxyarabinitol-P2

(a) and (b) Pre-activated tobacco (○), Synechococcus (△), R. rubrum (●) and G. sulfuraria (□) Rubisco was incubated with carboxyarabinitol-P2 and the mixtures were incubated for 0–240 s before being used to initiate carboxylase assays. At each carboxyarabinitol-P2 concentration, initial rates recorded for each period of incubation were fitted to Equation 1 and kobs was estimated. Experiments were carried out in duplicate; the error bars show the range of observations. The resulting data were fitted to the hyperbolic Equation 3, allowing estimation of Ki and k2, except in the case of R. rubrum Rubisco where no saturation was apparent and kobs became too high to measure at carboxyarabinitol-P2 concentrations greater than 1.6 μM. In this case a linear regression fitted the data. (c) Release of carboxyarabinitol-P2 was measured by measuring the exchange of labelled inhibitor, as described in the Experimental section. Only R. rubrum Rubisco showed appreciable exchange of label during 4 days. These data were fitted to Equation 1, allowing estimation of the kobs for release.

The model used for measuring the binding of carboxyarabinitol-P2 to Rubisco assumes that k−2=0, i.e. that binding was irreversible and the rate of inhibitor release was effectively zero. This was confirmed by measuring the exchange of bound [14C]carboxyarabinitol-P2 for unbound, unlabelled carboxyarabinitol-P2 (Figure 5). Tobacco, Synechococcus and G. sulfuraria Rubiscos showed negligible exchange of inhibitor over several days, which is consistent with an exchange rate for higher plants of approx. 1×10−8 s−1, it would take over 2 years for half of the inhibitor to be exchanged [35]. R. rubrum Rubisco showed slow exchange of bound inhibitor, with half of the inhibitor being exchanged after 2 days. These data could be fitted to Equation 2 to calculate kobs, and, due to the low [I*] in the assay, Equation 3 simplifies to kobs=k−2. The calculated value of k−2 is similar to other reported rates for R. rubrum Rubisco [36].

DISCUSSION

From the results in the present study, combined with information about the specificity factor [1,3], it appears that despite having a highly conserved active site [3,4] there is a range of activity for Rubisco. At one end of the spectrum, the Form II Rubisco from R. rubrum has a low specificity for CO2 over O2, produces several by-products during catalysis, and does not bind inhibitors very tightly. At the other extreme, Rubisco from the non-green alga G. sulfuraria has a high specificity for CO2 over O2, produces relatively few by-products during catalysis, has a high binding constant for inhibitors and a very low release rate of tight-binding inhibitors. Although a more extensive study would be required to confirm this, it is tempting to speculate that different Rubisco phylogenies have adopted alternative strategies for avoiding self-inhibition during catalysis.

As protein function depends on flexibility [9], the quaternary structure of Rubisco is almost certain to have an effect on the enzyme dynamics and flexibility. Form II Rubisco lacks small subunits and thus it would be expected that the R. rubrum enzyme may have increased flexibility, compared with the Form I enzymes, in which the small subunits form several interactions between the different subunits. This could account for the absence of slow, tight-binding inhibition, which requires the flexible loop six region to close tightly over the active site [37]. A flexible enzyme would be less prone to this inhibition, because the loop six region would not close over the active site, allowing the inhibitor to be released. Similarly, an enzyme with a high degree of flexibility would be less able to control the stability of intermediate compounds during catalysis, resulting in the increased formation of by-products such as xylulose-P2, pentodiulose-P2 and carboxytetritol-P2. Thus R. rubrum Rubisco does not show any decline in activity, despite producing several inhibitors during catalysis, as the inhibitors do not bind tightly to the enzyme.

In contrast, Form I Rubisco from G. sulfuraria had high specificity, and had a very low release rate of inhibitors, suggesting that the G. sulfuraria enzyme has less enzyme flexibility. G. sulfuraria does not show a decline in activity during catalysis as, although the enzyme can bind inhibitors very tightly, it has increased inhibition constants, and inhibition is unlikely to occur to a significant degree at the low physiological levels of inhibitor. The enzyme also shows an extraordinary avidity for the carbamylating CO2 molecule (S.M. Whitney, J. K. Janek and I. Saska, personal communication). Studies of enzyme dynamics have shown that, at a given temperature, thermostable enzymes are less flexible than thermolabile ones [38], and that, at lower temperatures, enzymes from extreme thermophiles are less flexible than those from mesophiles [39]. At the optimum growth temperature for an organism, the flexibility of enzymes from mesophiles and extreme thermophiles is similar [40]. This phenomenon is also observed for Rubisco specificity, since G. sulfuraria, which is a thermophilic red alga, and spinach Rubisco have similar specificity levels when measured at physiological temperature [41].

At higher temperatures, it would be expected that enzymes would have more flexibility, and carry out more side reactions. Cotton Rubisco has been observed to produce more xylulose-P2 when assayed at higher temperatures [42], and at higher temperatures spinach Rubisco has increased production of pentodiulose-P2 [32] and a higher rate of decline in activity [43]. At higher temperatures, tobacco Rubisco produces more inhibitors, but there is less decline in catalytic rate over time [44].

Interactions of the small subunits with other subunits in Form I Rubisco reduces the flexibility of the enzyme. This is shown by a reduction in the amount of side reactions catalysed by Form I Rubisco enzymes, such as those from tobacco and Synechococcus, when compared with Form II Rubisco, and the increased tendency for slow, tight-binding inhibition. Synechococcus Rubisco appears to lack self-inhibition during catalysis, due to an increased ability to carry out carboxylation of xylulose-P2, which would otherwise act as an inhibitor. The release of inhibitor from the decarbamylated enzyme is also faster compared with that of tobacco Rubisco.

Manipulation of the small subunits of Form I Rubisco influences the catalytic ability of the enzyme, demonstrating their importance in stabilizing the large subunit dimers. Removal of the small subunits from Synechococcus PCC6301 greatly reduced the rate of catalysis [6], and increased the tendency to carry out β-elimination of the enediol intermediate [45]. A hybrid enzyme consisting of Synechococcus PCC6301 large subunits with small subunits from Cylindrotheca (a diatom) had increased specificity relative to the native Synechococcus enzyme [46], showing that the small subunits can influence enzyme specificity. However, a hybrid enzyme consisting of sunflower Rubisco large subunits and tobacco Rubisco small subunits had a similar specificity to the native enzyme, but a reduced catalytic rate [47]. Recent studies with the Chlamydomonas Rubisco enzyme have shown that the interface between the large and small subunits contributes significantly to the differences in catalytic properties between algal and land-plant Rubisco enzymes [48]. Rubisco small subunits are likely to have a generic role in stabilizing the degree of enzyme flexibility, and reducing the ability of the enzyme to catalyse side reactions.

It has been proposed that all Rubisco enzymes may be nearly perfectly adapted to the differing conditions in the subcellular environment [10]. In order to improve the ability to differentiate between CO2 and O2 as substrates, the transition state for CO2 addition closely resembles the subsequent carboxyketone intermediate. However, the tight binding of this intermediate compound restricts the maximum catalytic throughput, and increases the fraction of the enzyme with enediol bound during catalysis. As a result, higher-plant Rubisco enzymes produce trace amounts of several by-products that can act as slow, tight-binding inhibitors. All forms of the Rubisco enzyme appear to produce these by-products; however, the by-products do not act as inhibitors, due to several different strategies. In higher plants, the protein Rubisco activase uses the energy from ATP hydrolysis to release these inhibitors, and thus regulate higher-plant Rubisco activity. No Rubisco activase proteins have been found in phylogenies other than the green algal/higher-plant lineage, and it is likely that these phylogenies do not utilize this method of Rubisco regulation. Enzyme dynamics influence the function of the enzyme, and, as such, Rubisco may provide an excellent model in extending the structure–function paradigm.

I acknowledge John Andrews, Heather Kane and Spencer Whitney (Molecular Plant Physiology Group, Research School of Biological Sciences, Australian National University, Australia) and Juliet Gerrard (Biochemistry Group, School of Biological Sciences, University of Canterbury, New Zealand) for useful discussions, comments and expertise.

Abbreviations

     
  • carboxyarabinitol-1-P

    2′-carboxy-D-arabinitol 1-phosphate

  •  
  • carboxyarabinitol-P2

    2′-carboxy-D-arabinitol 1,5-bisphosphate

  •  
  • carboxypentitol-P2

    unresolved isomeric mixture of carboxyarabinitol-P2 and 2′-carboxy-D-ribitol 1,5-bisphosphate

  •  
  • carboxytetritol-P2

    2′-carboxy-D-tetritol 1,5-bisphosphate

  •  
  • pentodiulose-P2

    D-glycero-2,3-pentodiulose 1,5-bisphosphate

  •  
  • P-glycerate

    3-phospho-D-glycerate

  •  
  • P-glycolate

    2-phosphoglycolate

  •  
  • ribulose-P2

    D-ribulose 1,5-bisphosphate

  •  
  • Rubisco

    D-ribulose-1,5-bisphosphate carboxylase/oxygenase

  •  
  • xylulose-P2

    D-xylulose 1,5-bisphosphate

References

References
1
Roy
H.
Andrews
T. J.
Leegood
R. C.
Sharkey
T. D.
von Caemmerer
S.
Rubisco: assembly and mechanism
Photosynthesis: Physiology and Metabolism
2000
Dordrecht
Kluwer Academic Publishers
(pg. 
53
-
83
)
2
Spreitzer
R. J.
Salvucci
M. E.
Rubisco: structure, regulatory interactions, and possibilities for a better enzyme
Annu. Rev. Plant Biol.
2002
, vol. 
53
 (pg. 
449
-
475
)
3
Tabita
F. R.
Microbial ribulose 1,5-bisphosphate carboxylase/oxygenase: a different perspective
Photosynth. Res.
1999
, vol. 
60
 (pg. 
1
-
28
)
4
Andersson
I.
Taylor
T. C.
Structural framework for catalysis and regulation in ribulose-1,5-bisphosphate carboxylase/oxygenase
Arch. Biochem. Biophys.
2003
, vol. 
414
 (pg. 
130
-
140
)
5
Delwiche
C. F.
Tracing the thread of plastid diversity through the tapestry of life
Am. Nat.
1999
, vol. 
154
 (pg. 
S164
-
S177
)
6
Andrews
T. J.
Catalysis by cyanobacterial ribulosebisphosphate carboxylase large subunits in the complete absence of small subunits
J. Biol. Chem.
1988
, vol. 
263
 (pg. 
12213
-
12220
)
7
Sugawara
H.
Yamamoto
H.
Shihata
N.
Inoue
T.
Okada
S.
Miyake
C.
Yokota
A.
Kai
Y.
Crystal structure of carboxylase reaction-oriented ribulose 1,5-bisphosphate carboxylase oxygenase from a thermophilic red alga, Galdieria partita
J. Biol. Chem.
1999
, vol. 
274
 (pg. 
15655
-
15661
)
8
Hansen
S.
Vollan
V. B.
Hough
E.
Andersen
K.
The crystal structure of Rubisco from Alcaligenes eutrophus reveals a novel central eight-stranded β-barrel formed by β-strands from four subunits
J. Mol. Biol.
1999
, vol. 
288
 (pg. 
609
-
621
)
9
Daniel
R. M.
Dunn
R. V.
Finney
J. L.
Smith
J. C.
The role of dynamics in enzyme activity
Annu. Rev. Biophys. Biomol. Struct.
2003
, vol. 
32
 (pg. 
69
-
92
)
10
Tcherkez
G. G. B.
Farquhar
G. D.
Andrews
T. J.
Despite slow catalysis and confused substrate specificity, all ribulose bisphosphate carboxylases may be nearly perfectly optimized
Proc. Natl. Acad. Sci. U.S.A.
2006
, vol. 
103
 (pg. 
7246
-
7251
)
11
Cleland
W. W.
Andrews
T. J.
Gutteridge
S.
Hartman
F. C.
Lorimer
G. H.
Mechanism of Rubisco: the carbamate as general base
Chem. Rev.
1998
, vol. 
98
 (pg. 
549
-
561
)
12
Portis
A. R.
Rubisco activase: Rubisco's catalytic chaperone
Photosynth. Res.
2003
, vol. 
75
 (pg. 
11
-
27
)
13
Taylor
T. C.
Andersson
I.
The structure of the complex between Rubisco and its natural substrate ribulose 1,5-bisphosphate
J. Mol. Biol.
1997
, vol. 
265
 (pg. 
432
-
444
)
14
Edmondson
D. L.
Kane
H. J.
Andrews
T. J.
Substrate isomerization inhibits ribulosebisphosphate carboxylase/oxygenase during catalysis
FEBS Lett.
1990
, vol. 
260
 (pg. 
62
-
66
)
15
Andrews
T. J.
Kane
H. J.
Pyruvate is a by-product of catalysis by ribulosebisphosphate carboxylase/oxygenase
J. Biol. Chem.
1991
, vol. 
266
 (pg. 
9447
-
9452
)
16
Harpel
M. R.
Serpersu
E. H.
Lamerdin
J. A.
Huang
Z. H.
Gage
D. A.
Hartman
F. C.
Oxygenation mechanism of ribulose-bisphosphate carboxylase/oxygenase. Structure and origin of 2-carboxytetritol 1,4-bisphosphate, a novel O2-dependent side product generated by a site-directed mutant
Biochemistry
1995
, vol. 
34
 (pg. 
11296
-
11306
)
17
Kane
H. J.
Wilkin
J. M.
Portis
A. R.
Andrews
T. J.
Potent inhibition of ribulose-bisphosphate carboxylase by an oxidized impurity in ribulose-1,5-bisphosphate
Plant Physiol.
1998
, vol. 
117
 (pg. 
1059
-
1069
)
18
Pearce
F. G.
Andrews
T. J.
The relationship between side reactions and slow inhibition of ribulose-bisphosphate carboxylase revealed by a loop 6 mutant of the tobacco enzyme
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
32526
-
32536
)
19
Zhu
G.
Jensen
R. G.
Xylulose 1,5-bisphosphate synthesized by ribulose 1,5-bisphosphate carboxylase/oxygenase during catalysis binds to decarbamylated enzyme
Plant Physiol.
1991
, vol. 
97
 (pg. 
1348
-
1353
)
20
Horken
K. M.
Tabita
F. R.
Closely related form I ribulose bisphosphate carboxylase/oxygenase molecules that possess different CO2/O2 substrate specificities
Arch. Biochem. Biophys.
1999
, vol. 
361
 (pg. 
183
-
194
)
21
Lee
E. H.
Harpel
M. R.
Chen
Y.-R.
Hartman
F. C.
Perturbation of reaction-intermediate partitioning by a site-directed mutant of ribulose-bisphosphate carboxylase/oxygenase
J. Biol. Chem.
1993
, vol. 
268
 (pg. 
26583
-
26591
)
22
Andrews
T. J.
Abel
K. M.
Kinetics and subunit interactions of ribulose bisphosphate carboxylase/oxygenase from the cyanobacterium, Synechococcus sp
J. Biol. Chem.
1981
, vol. 
256
 (pg. 
8445
-
8451
)
23
Servaites
J. C.
Crystalline ribulose bisphosphate carboxylase/oxygenase of high integrity and catalytic activity from Nicotiana tabacum
Arch. Biochem. Biophys.
1985
, vol. 
238
 (pg. 
154
-
160
)
24
Edmondson
D. L.
Badger
M. R.
Andrews
T. J.
A kinetic characterization of slow inactivation of ribulosebisphosphate carboxylase during catalysis
Plant Physiol.
1990
, vol. 
93
 (pg. 
1376
-
1382
)
25
Morell
M. K.
Paul
K.
O'Shea
N. J.
Kane
H. J.
Andrews
T. J.
Mutations of an active site threonyl residue promote β-elimination and other side reactions of the enediol intermediate of the ribulosebisphosphate carboxylase reaction
J. Biol. Chem.
1994
, vol. 
269
 (pg. 
8091
-
8098
)
26
Whitney
S. M.
Andrews
T. J.
Plastome-encoded bacterial ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) supports photosynthesis and growth in tobacco
Proc. Natl. Acad. Sci. U.S.A.
2001
, vol. 
98
 (pg. 
14738
-
14743
)
27
Allen
M. B.
Studies with Cyanidium caldarium, an anomalously pigmented chlorophyte
Arch. Microbiol.
1959
, vol. 
32
 (pg. 
270
-
277
)
28
Pierce
J.
Tolbert
N. E.
Barker
R.
Interaction of ribulosebisphosphate carboxylase/oxygenase with transition-state analogues
Biochemistry
1980
, vol. 
19
 (pg. 
934
-
942
)
29
Moore
B. D.
Kobza
J.
Seemann
J. R.
Measurement of 2-carboxyarabinitol 1-phosphate in plant leaves by isotope dilution
Plant Physiol.
1991
, vol. 
96
 (pg. 
208
-
213
)
30
Lilley
R. M.
Walker
D. A.
An improved spectrophotometric assay for ribulosebisphosphate carboxylase
Biochim. Biophys. Acta
1974
, vol. 
358
 (pg. 
226
-
229
)
31
McCurry
S. D.
Tolbert
N. E.
Inhibition of ribulose-1,5-bisphosphate carboxylase/oxygenase by xylulose 1,5-bisphosphate
J. Biol. Chem.
1977
, vol. 
252
 (pg. 
8344
-
8346
)
32
Kim
K.
Portis
A. R.
Oxygen-dependent H2O2 production by Rubisco
FEBS Lett.
2004
, vol. 
571
 (pg. 
124
-
128
)
33
Whitney
S. M.
Baldett
P.
Hudson
G. S.
Andrews
T. J.
Form I Rubiscos from non-green algae are expressed abundantly but not assembled in tobacco chloroplasts
Plant J.
2001
, vol. 
26
 (pg. 
535
-
547
)
34
Jordan
D. B.
Chollet
R.
Inhibition of ribulose bisphosphate carboxylase by substrate ribulose 1,5-bisphosphate
J. Biol. Chem.
1983
, vol. 
258
 (pg. 
13752
-
13758
)
35
Schloss
J. V.
Comparative affinities of the epimeric reaction-intermediate analogs 2-carboxy-D-arabinitol and 4-carboxy-D-arabinitol 1,5-bisphosphate for spinach ribulose 1,5-bisphosphate carboxylase
J. Biol. Chem.
1988
, vol. 
263
 (pg. 
4145
-
4150
)
36
Harpel
M. R.
Larimer
F. W.
Hartman
F. C.
Functional analysis of the putative catalytic bases His321 and Ser368 of Rhodospirillum rubrum ribulose bisphosphate carboxylase/oxygenase by site-directed mutagenesis
J. Biol. Chem.
1991
, vol. 
266
 (pg. 
24734
-
24740
)
37
Taylor
T. C.
Andersson
I.
Structural transitions during activation and ligand binding in hexadecameric Rubisco inferred from the crystal structure of the activated unliganded spinach enzyme
Nat. Struct. Biol.
1996
, vol. 
3
 (pg. 
95
-
101
)
38
Wagner
G.
Wüthrich
K.
Structural interpretation of the amide proton-exchange in the basic pancreatic trypsin-inhibitor and related proteins
J. Mol. Biol.
1979
, vol. 
134
 (pg. 
75
-
94
)
39
Wrba
A.
Schweiger
A.
Schultes
V.
Jaenicke
R.
Zavodszky
P.
Extremely thermostable D-glyceraldehyde-3-phosphate dehydrogenase from the Eubacterium Thermotoga maritima
Biochemistry
1990
, vol. 
29
 (pg. 
7584
-
7592
)
40
Daniel
R. M.
Dines
M.
Petach
H. H.
The denaturation and degradation of stable enzymes at high temperatures
Biochem. J.
1996
, vol. 
317
 (pg. 
1
-
11
)
41
Uemura
K.
Anwaruzzaman Miyachi
S.
Yokota
A.
Ribulose-1,5-bisphosphate carboxylase/oxygenase from thermophilic red algae with a strong specificity for CO2 fixation
Biochem. Biophys. Res. Commun.
1997
, vol. 
233
 (pg. 
568
-
571
)
42
Salvucci
M. E.
Crafts-Brandner
S. J.
Mechanism for deactivation of Rubisco under moderate heat stress
Physiol. Plant
2004
, vol. 
122
 (pg. 
513
-
519
)
43
Kim
K.
Portis
A.
Kinetic analysis of the slow inactivation of Rubisco during catalysis: effects of temperature, O2 and Mg2+
Photosynth. Res.
2006
, vol. 
87
 (pg. 
195
-
204
)
44
Schrader
S.
Kane
H. J.
Sharkey
T. D.
von Caemmerer
S.
High temperature enhances inhibitor production but reduces fallover in tobacco Rubisco
Funct. Plant Biol.
2006
 
in the press
45
Morell
M. K.
Wilkin
J.-M.
Kane
H. J.
Andrews
T. J.
Side reactions catalyzed by ribulose-bisphosphate carboxylase in the presence and absence of small subunits
J. Biol. Chem.
1997
, vol. 
272
 (pg. 
5445
-
5451
)
46
Lee
B.
Read
B. A.
Tabita
F. R.
Catalytic properties of recombinant octameric, hexadecameric, and heterologous cyanobacterial/bacterial ribulose-1,5-bisphosphate carboxylase/oxygenase
Arch. Biochem. Biophys.
1991
, vol. 
291
 (pg. 
263
-
269
)
47
Kanevski
I.
Maliga
P.
Rhoades
D. F.
Gutteridge
S.
Plastome engineering of ribulose-1,5-bisphosphate carboxylase/oxygenase in tobacco to form a sunflower large subunit and tobacco small subunit hybrid
Plant Physiol.
1999
, vol. 
119
 (pg. 
133
-
141
)
48
Spreitzer
R. J.
Peddi
S. R.
Satagopan
S.
Phylogenetic engineering at an interface between large and small subunits imparts land-plant kinetic properties to algal Rubisco
Proc. Natl. Acad. Sci. U.S.A.
2005
, vol. 
102
 (pg. 
17225
-
17230
)
49
Kane
H. J.
Viil
J.
Entsch
B.
Paul
K.
Morell
M. K.
Andrews
T. J.
An improved method for measuring the CO2/O2 specificity of ribulosebisphosphate carboxylase-oxygenase
Aust. J. Plant Physiol.
1994
, vol. 
21
 (pg. 
449
-
461
)
50
Morell
M. K.
Kane
H. J.
Hudson
G. S.
Andrews
T. J.
Effects of mutations at residue 309 of the large subunit of ribulosebisphosphate carboxylase from Synechococcus PCC6301
Arch. Biochem. Biophys.
1992
, vol. 
299
 (pg. 
295
-
301
)