Microcalorimetric determination of the reaction enthalpy changes associated with the carboxylase and oxygenase reactions catalysed by ribulose 1,5-bisphosphate carboxylase/oxygenase (RUBISCO)

Abstract

The total reaction enthalpy change Δ_rH associated with the carboxylation and oxygenation of D-ribulose 1,5-bisphosphate (RuBP) catalysed by RUBISCO from spinach and Rhodospirillum rubrum was measured for the first time with sensitive microcalorimetry. The enthalpy changes Δ_r,cH and Δ_r,oH associated with the carboxylase and oxygenase reactions were determined separately using the known ratio of these reactions. Δ_r,cH and Δ_r,oH corrected for the heat of neutralisation due to proton release during the overall carboxylation or oxygenation processes. For the corrected enthalpy change Δ_r,cH′ of the carboxylase reaction catalysed by RUBISCO of spinach a value of −21.34±0.4 kJ mol⁻¹ was determined. In the case of the oxygenase reaction for RUBISCO of spinach a reaction enthalpy change Δ_r,oH′ of −319.1±9.6 kJ mol⁻¹ which is typical for energy rich reactions, was found. This value is comparable with Δ_r,oH′=−291.4±5.8 kJ mol⁻¹ associated with the oxygenase reaction catalysed by Co²⁺-activated RUBISCO from R. rubrum.

---

Introduction

The carbon dioxide fixing enzyme D-ribulose 1,5-bisphosphate carboxylase/oxygenase (RUBISCO) of higher plants has a molecular weight of 550000 Da and consists of eight large catalytic (L) and eight small regulatory (S) subunits. It is the key enzyme for the carbon reductive pathway (Calvin cycle) and for photorespiration. The core of the hexadecameric RUBISCO from higher plants comprises four pairs of L₂-dimers, which are similar in structure to the dimeric bacterial RUBISCO (L₂) from Rhodospirillum rubrum.¹ RUBISCO from higher plants show more complex regulatory properties,^2,3 a higher carboxylase/oxygenase activity ratio⁴ and in some cases an apparent negative co-operative behaviour compared with the bacterial enzyme.^5,6 RUBISCO is activated in the presence of magnesium ions and carbon dioxide.⁷ In the presence of Mg²⁺-ions and HCO₃⁻-ions all RUBISCO enzymes catalyse both the carboxylation and the oxygenation of the substrate D-ribulose 1,5-bisphosphate (RuBP).⁸ Partitioning between the carboxylase and oxygenase reaction is strongly influenced by the presence of metal ion cofactors.⁸ Transition metal ions, like Mn²⁺, Co²⁺ and Ni²⁺, favour the oxygenase reaction, while in the presence of Mg²⁺-ions the carboxylase reaction dominates.⁸ In the presence of Co²⁺-ions the dimeric RUBISCO from R. rubrum exclusively catalyses the oxygenase reaction.⁹ Here we report for the first time on microcalorimetric measurements of the reaction enthalpies Δ_r,oH and Δ_r,oH for the carboxylase and oxygenase reaction, respectively. It is demonstrated that the ratio of the reaction rates of these processes can be measured in a single isothermal titration calorimetry (ITC) experiment, if it is possible to determine the values of the reaction enthalpies Δ_r,cH and Δ_r,oH independently. Therefore ITC allows the screening of photosynthetic organisms very efficiently to select RUBISCO enzymes in respect of high carboxylase/oxygenase activity ratios.

Materials and methods

D-Ribulose 1,5-bisphosphate (tetrasodium salt, hydrate) was obtained from Sigma (Deisenhofen, Germany). RUBISCO from spinach was purified to electrophoretic homogeneity as described by Vater and Salnikow.¹⁰ The enzyme preparations showed specific activities between 1 and 1.5 units mg⁻¹. RUBISCO from spinach was dialysed at 4°C prior to use against 100 mM Tris-HCl buffer of pH 8.0 (1 mM dithioerythritol, 0.1 mM EDTA) for the determination of reaction enthalpies. RUBISCO from R. rubrum was a kind gift from G. H. Lorimer (formerly at DuPont, USA). After long term storage at −20°C the bacterial enzyme was treated with 50 mM dithioerythritol and 5 mM EDTA according to Tabita and McFadden¹¹ and dialysed prior to use against 100 mM Tris-HCl buffer of pH 8.0 (1 mM dithioerythritol, 0.1 mM EDTA) at 4°C. The protein concentration was calculated from the extinction at 280 nm applying an extinction coefficient ε_280nm of 1.664 mg⁻¹ cm⁻¹ and 0.974 mg⁻¹ cm⁻¹ for RUBISCO from spinach¹² and RUBISCO from R. rubrum,¹³ respectively.

Isothermal titration calorimetry (ITC) experiments to monitor the time course of the carboxylation and oxygenation of RuBP were performed using the OMEGA highly sensitive microcalorimeter manufactured by MicroCal Inc. (Northampton, Mass., USA). A detailed description of the design and operation of this instrument was published previously.¹⁴ In our ITC measurements it is not necessary to de-gas solutions to prevent baseline noise and artefacts due to bubble formation. The solutions were equilibrated with air to control the oxygen concentration; the CO₂ concentration was adjusted via the NaHCO₃ concentration and the pH of the solutions. We checked experimentally that there is no measurable exchange of gaseous substrates between the cell of the ITC instrument and the outside environment during an experiment. This is due to a narrow long tube connecting the cell and the outside, which is practically closed during ITC action. For measurements of the heat production accompanying the overall enzyme catalysed reaction 5 mg ml⁻¹ RUBISCO from spinach or 0.9 mg ml⁻¹ RUBISCO from R. rubrum in 100 mM Tris-HCl buffer of pH 8.0 (10 mM MgCl₂ , 40 mM NaHCO₃, 1 mM dithioerythritol and 0.1 mM EDTA) equilibrated with air were loaded into the sample cell of the calorimeter (volume=1.3592 ml). The reference cell was filled with distilled water. Solutions of 10 mM and 2.75 mM RuBP in 100 mM Tris-HCl buffer of pH 8.0 (10 mM MgCl₂ , 40 mM NaHCO₃, 1 mM dithioerythritol and 0.1 mM EDTA) equilibrated with air were filled into 100 μl syringes for ITC experiments with RUBISCO of spinach and R. rubrum, respectively. The system was allowed to equilibrate and a stable baseline had to be established before the automated injection procedure was initiated. A typical experiment involved three injections of 4, 10 or 25 μl of the RuBP solution into the sample cell. During the reaction the cell was stirred continuously at 400 rpm. The overall reaction enthalpy change Δ_rH which comprises both the carboxylase and oxygenase reaction catalysed by RUBISCO from spinach was also measured replacing MgCl₂ as cofactor by the transition metal ions Mn²⁺ (2 mM), Ni²⁺ (1 mM) and Co²⁺ (1 mM). The data obtained were evaluated to determine the enthalpy change Δ_rH. Δ_rH was estimated by integration of the whole heat production curve. The heat obtained was then divided by the injected molar amount of RuBP. The reaction enthalpy change Δ_r,oH of the oxygenase reaction of RUBISCO from R. rubrum was determined in the presence of 1 mM CoCl₂ and 40 mM NaHCO₃ in the absence of MgCl₂ .

The measured reaction enthalpy change Δ_rH is the sum of the evolved heat per mole of RuBP produced in the carboxylase and oxygenase reactions which contribute in a different manner to the overall reaction. In the presence of Mg²⁺ as cofactor the carboxylase reaction is favoured in comparison to the oxygenase reaction at pH 8.0 (10 mM MgCl₂ and 40 mM NaHCO₃). In contrast the transition metal ions Mn²⁺, Co²⁺ and Ni²⁺ promote the oxygenase reaction.⁸ The contributions of both reactions under our experimental conditions in the presence of Mg²⁺, Mn²⁺, Co²⁺ and Ni²⁺ were calculated based on previously published data.^4,15 The ratio of the rates of carboxylase and oxygenase reaction is given by eqn. (I):⁴

where v_c=reaction rate of the carboxylase reaction, v_o=reaction rate of the oxygenase reaction, V_c=maximal reaction rate of the carboxylase reaction, V_o=maximal reaction rate of the oxygenase reaction, K_c=Michaelis–Menten constant for CO₂, and, K_o=Michaelis–Menten constant for O₂. The ratio of the kinetic constants is defined as the specificity constant Γ, which is species characteristic and depends only on the chemical nature of the cofactor, pH, ionic strength and temperature.

There exists a linear dependence between the ratio of the reaction rates of the carboxylase and oxygenase reaction and the ratio of the concentrations of the gaseous substrates CO₂ and O₂.^4,15

The ratio of the change of the substrate concentration RuBP (turnover) due to the carboxylase (Δ[RuBP]_C) and oxygenase reaction (Δ[RuBP]_O) is equivalent to the ratio of the reaction rates of both processes under the condition that the concentrations of the gaseous substrates CO₂ and O₂ are constant during the reaction:

For a given ratio of the concentrations of the gaseous substrates CO₂ and O₂ the specificity constant Γ is a measure for the ratio of the turnover of the substrate RuBP in the carboxylase and oxygenase reaction. The measured reaction enthalpy change Δ_rH is the sum of the reaction enthalpies for the carboxylase and oxygenase reaction multiplied with their contributions to the total turnover of the substrate RuBP.

where x_c=RuBP turnover in the carboxylase reaction, x_o=RuBP turnover in the oxygenase reaction, x_total=total RuBP turnover in the carboxylase and oxygenase reaction. The ratio of the reaction rates for the carboxylase and oxygenase reaction v_c/v_o was calculated according to eqn. (I). The concentration of CO₂ in a solution of 40 mM NaHCO₃ at pH 8.0 was calculated to be 0.6 mM.¹⁶ The concentration of O₂ in an aqueous solution was measured with an oxygen electrode to be 0.3 mM at 25°C. Therefore, the ratio of CO₂ to O₂ was 2 under the conditions of our ITC experiments. For Mg²⁺-activated RUBISCO from spinach and R. rubrum specificity constants Γ of 80 and 15 were applied.^4,15 The Mn²⁺-activated enzymes from spinach and R. rubrum possess specificity constants Γ of 3 and 1.5, respectively.^4,15

Results

The sum of the reaction enthalpies of the carboxylation and oxygenation of RuBP by RUBISCO from spinach was measured using different metal ions as cofactor (Table 1). For the fully activated enzyme in the presence of 40 mM NaHCO₃ and 10 mM MgCl₂ in Tris-HCl buffer pH=8.0 an enthalpy change Δ_rH=−62.3±0.6 kJ mol⁻¹ was determined (Fig. 1a). Under these conditions the carboxylase reaction is favoured in comparison with the oxygenation process. The transition metal ions Mn²⁺, Co²⁺ and Ni²⁺ support the oxygenase reaction and increase the measured molar enthalpy change Δ_rH compared to the experiments with Mg²⁺-ions as cofactor.

Table 1 Reaction enthalpy change Δ_rH per mole RuBP for the sum of carboxylase and oxygenase reaction catalysed by RUBISCO from spinach^a

pH	Cofactor	ΔrH/ kJ mol(RuBP)^−1
a The enzyme was incubated in the presence of 40 mM NaHCO3 in 100 mM Tris-HCl buffer (1 mM dithioerythritol and 0.1 mM EDTA) at pH 8.0 (pH 7.0) and 25°C with the metal ions indicated in the table.
7.0	10 mM Mg^2+	−61.4±0.6
8.0	10 mM Mg^2+	−62.3±0.6
8.0	2 mM Mn^2+	−110.8±1.2
8.0	1 mM Co^2+	−107.8±1.2
8.0	1 mM Ni^2+	−124.1±1.3

---

Fig. 1 (a) Diagram of the enthalpy in kJ mol⁻¹ of injectant (lower part) and of the heat produced (upper part) in each of three injections of 4 μl of 10 mM RuBP (with 10 mM MgCl₂ and 40 mM NaHCO₃) into a solution of 9.3 μM (5 mg ml⁻¹) Mg²⁺-activated RUBISCO from spinach in 100 mM Tris-HCl buffer of pH 8.0 (1 mM dithioerythritol and 0.1 mM EDTA) at 25°C measured by ITC. (b) Diagram of the enthalpy in kJ mol⁻¹ of injectant (lower part) and of the heat produced (upper part) in each of three injections of 4 μl of 2.75 mM RuBP (with 1 mM CoCl₂ and 40 mM NaHCO₃) into a solution of 7.9 μM (0.9 mg ml⁻¹) Co²⁺-activated RUBISCO from R. rubrum in 100 mM Tris-HCl buffer of pH 8.0 (1 mM dithioerythritol and 0.1 mM EDTA) at 25°C measured by ITC.

The oxygenation of RuBP shows a much higher exothermic reaction enthalpy change Δ_r,oH than the carboxylase reaction. Ni²⁺-ions support the oxygenase reaction more strongly than Mn²⁺ and Co²⁺ resulting in the highest negative values of the measured reaction enthalpy change Δ_rH. The reaction enth alpy changes Δ_r,cH and Δ_r,oH of the carboxylase and oxygenase reactions were determined by using the calculated ratio of both reactions (eqn. (IV)) which was estimated with the aid of the previously published specificity constants Γ for RUBISCO from spinach and R. rubrum^4,15 (eqn. (I) and (II)).

According to eqn. (I) and (III) the part of RuBP converted in the carboxylase reaction (x=Δ[RuBP]_c/Δ[RuBP]_total) was calculated with the aid of the specificity constant Γ and a ratio of 2 of the concentrations of the gaseous substrates CO₂ and O₂ in aqueous solution at atmospheric pressure:

Γ=80 (for RUBISCO from spinach and Mg²⁺-ions as cofactor)

In the presence of 10 mM Mg²⁺ and 40 mM HCO₃⁻-ions 99.4% of the total RuBP was converted in the carboxylase reaction. If one neglects the 0.6% oxygenase reaction, then Δ_rH measured in the presence of Mg²⁺-ions corresponds to a reaction enthalpy change Δ_r,cH of −62.3±0.6 kJ mol⁻¹ attributed to the carboxylase reaction. For Mn²⁺-ions (Γ=3) only 85.7% of the total RuBP was converted in the carboxylase reaction. Under these conditions a higher Δ_rH of −110.8±1.2 kJ mol⁻¹ was measured because of 14.3% oxygenase reaction. According to eqn. (IV) a reaction enthalpy change Δ_r,oH=−401±8 kJ mol⁻¹ was calculated for the oxygenase reaction.

A reaction enthalpy change Δ_rH=−120.4±1.3 kJ mol⁻¹ was determined for the sum of the carboxylase and oxygenase reaction for the Mg²⁺-activated RUBISCO from R. rubrum (with 10 mM MgCl₂ and 40 mM NaHCO₃) at pH 8.0. In the presence of Co²⁺ the enzyme exclusively catalyses the oxygenase reaction.⁹

For the Co²⁺-activated bacterial RUBISCO (with 1 mM CoCl₂ and 40 mM NaHCO₃) a Δ_rH of −373.3±2.5 kJ mol⁻¹ was measured which corresponds to the reaction enthalpy change Δ_r,oH of the oxygenase reaction (100% oxygenase reaction). This result agrees quite well with the value of Δ_r,oH=−401±8 kJ mol⁻¹ calculated for the oxygenase reaction catalysed by RUBISCO of spinach.

The reaction enthalpies for the carboxylase and oxygenase reaction have to be corrected for the heat of neutralisation due to proton release associated with both processes.

For the carboxylation of RuBP the following stoichiometry was applied:¹⁷

The molar reaction enthalpy change Δ_r,cH of the carboxylase reaction includes the heat of neutralisation of the released protons by the buffer which must be measured separately and subtracted. Therefore, in a second ITC experiment, an equivalent amount of hydrochloric acid was neutralised in a mixture of RUBISCO, 10 mM MgCl₂ and 40 mM NaHCO₃ in 100 mM Tris-HCl pH 8.0 yielding a molar heat of neutralisation of −40.96±0.4 kJ mol⁻¹. The molar enthalpy change Δ_r,cH′ of the carboxylase reaction is therefore Δ_r,cH′=−62.3(−40.96)=−21.34±0.4 kJ mol⁻¹.

For the oxygenase reaction the following stoichiometry was applied:¹⁷

The reaction enthalpy change Δ_r,oH′ of the oxygenase reaction of RUBISCO from spinach is therefore Δ_r,oH′=−401 −2(−40.96)=−319.1±9.6 kJ mol⁻¹. For the oxygenase reaction of R. rubrum RUBISCO a corrected molar reaction enthalpy change (Δ_r,oH′ of −373.3−2(−40.96)=−291.4±5.8 kJ mol⁻¹ was calculated.

Discussion

In previous ITC studies we reported binding constants of sugar phosphates to RUBISCO including their binding enthalpies.^18,19 In this report we determined the total enthalpy change Δ_rH associated with the carboxylase and oxygenase reaction catalysed by RUBISCO from spinach or R. rubrum. If the ratio of carboxylation and oxygenation is known, the enthalpy changes Δ_r,cH and Δ_r,oH for the carboxylase and oxygenase reaction can be determined separately from Δ_rH.

The data obtained for Δ_r,cH and Δ_r,oH were corrected for the heat of neutralisation due to the release of protons during the enzymatic reaction and their reaction with the applied buffer. For the corrected reaction enthalpy change Δ_r,cH′ of the carboxylase reaction predominantly catalysed by Mg²⁺-activated RUBISCO from spinach a value of −21.34±0.4 kJ mol⁻¹ was measured which is in excellent agreement with the value of −20.06±2.1 kJ mol⁻¹ determined by Kitzinger et al.²⁰ For the oxygenase reaction which is catalysed more efficiently by Mn²⁺-activated RUBISCO from spinach than by the Mg²⁺-activated enzyme a much higher reaction enthalpy change Δ_r,oH′ of −319.1±9.6 kJ mol⁻¹ was determined. This value is comparable with the reaction enthalpy change Δ_r,oH′=−291.4±5.8 kJ mol⁻¹ associated with the oxygenase reaction exclusively catalysed by the Co²⁺-activated RUBISCO from R. rubrum. In spite of the different nature of the metal cofactors Mn²⁺ and Co²⁺ the corresponding enthalpy changes associated with the oxygenase reaction showed comparable results.

Obviously, the oxygenase reaction is an energy rich reaction which probably proceeds via peroxide states of the substrate RuBP.¹⁷ This is consistent with the finding that in the presence of Mn²⁺-ions RUBISCO from spinach shows chemiluminescence during catalysis²¹ which is a good indicator for an energy rich process.

If Δ_r,cH and Δ_r,oH are determined precisely the ratio of the reaction rates of the carboxylation and oxygenation of RuBP can be measured in a single ITC experiment. Measurements of this ratio as a function of the concentrations of CO₂ and O₂ require a more detailed treatment than is given here and will be presented in a forthcoming publication. The determination of reaction rates as a function of the gaseous substrates with high accuracy in microcalorimetric experiments should enable us to calculate the specificity constant Γ of RUBISCO enzymes of photosynthetic organisms without the necessity to use radiochemical assays. As a prospect, this is important for a screening of RUBISCO enzymes of various origin for high carboxylase/oxygenase ratios which may be useful to select and exploit plants with highly efficient CO₂-reduction pathway. Analysis of their protein structure, in particular of their reaction centres, will be of high value to engineer plants with enhanced crop yields.

Acknowledgements

Acknowledgement We would like to thank Dr G. H. Lorimer (until 1997 at Du Pont, USA) for providing us with the RUBISCO from R. rubrum used in this study.

References

1. S. Knight, I. Andersson and C.-I. Bränden, J. Mol. Biol., 1990, 215, 113.
2. G. H. Lorimer, Annu. Rev. Plant Physiol., 1981, 32, 349.
3. A. R. Portis Jr., Annu. Rev. Plant Physiol., 1992, 43, 415.
4. D. B. Jordan and W. L. Ogren, Nature (London), 1981, 291, 513.
5. R. G. Jensen and G. Zhu, in Research in Photosynthesis, ed. N. Murata, Kluwer Academic Publishers, Dordrecht, 1992, vol. 3, p. 617..
6. S. Johal, B. E. Partridge and R. J. Chollet, J. Biol. Chem., 1985, 260, 9894.
7. G. H. Lorimer, M. R. Badger and T. J. Andrews, Biochemistry, 1976, 15, 529.
8. J. T. Christeller, Biochem. J., 1981, 193, 839.
9. P. D. Robinson, M. N. Martin and F. R. Tabita, Biochemistry, 1979, 18, 4453.
10. J. Vater and J. Salnikow, Arch. Biochem. Biophys., 1979, 194, 190.
11. F. R. Tabita and B. A. McFadden, J. Biol. Chem., 1974, 249, 3453.
12. J. M. Paulsen and M. D. Lane, Biochemistry, 1966, 5, 2350.
13. F. R. Tabita and B. A. McFadden, J. Biol. Chem., 1974, 249, 3459.
14. T. Wiseman, S. Willistone, J. F. Brandts and L.-N. Lin, Anal. Biochem., 1989, 179, 131.
15. D. B. Jordan and W. L. Ogren, Arch. Biochem. Biophys., 1983, 227, 425.
16. N. P. Hall, M. J. Cornelius and A. J. Keys, Anal. Biochem., 1983, 132, 152.
17. H. M. Miziorko and G. H. Lorimer, Annu. Rev. Biochem., 1983, 52, 507.
18. J. Frank, J. Vater and J. F. Holzwarth, J. Chem. Soc., Faraday Trans., 1998, 94, 2127.
19. J. Frank, J. Vater and J. F. Holzwarth, Phys. Chem. Chem. Phys., 1999, 1, 455.
20. C. Kitzinger, B. L. Horecker and A. Weisbach, in 20th Int. Congr. Physiol. Sci. Brussels, 1956, Catherine Press, Bruges, Belgium, p. 502..
21. S. N. Mogel and B. A. McFadden, Biochemistry, 1990, 29, 8333.

---