Enzymology below 200 K: The kinetics and thermodynamics of the photochemistry catalyzed by protochlorophyllide oxidoreductase

Abstract

The chlorophyll biosynthesis enzyme protochlorophyllide reductase (POR) catalyzes the light-dependent reduction of protochlorophyllide (Pchlide) into chlorophyllide in the presence of NADPH. As POR is light-dependent, catalysis can be initiated by illumination of the enzyme-substrate complex at low temperatures, making it an attractive model for studying aspects of biological proton and hydride transfers. The early stages in the photoreduction, involving Pchlide binding and an initial photochemical reaction, have been studied in vitro by using low-temperature fluorescence and absorbance measurements. Formation of the ternary POR-NADPH-Pchlide complex produces red shifts in the fluorescence and absorbance maxima of Pchlide, allowing the dissociation constant for Pchlide binding to be measured. We demonstrate that the product of an initial photochemical reaction, which can occur below 200 K, is a nonfluorescent intermediate with a broad absorbance band at 696 nm (A696) that is suggested to represent an ion radical complex. The temperature dependence of the rate of A696 formation has allowed the activation energy for the photochemical step to be calculated and has shown that POR catalysis can proceed at much lower temperatures than previously thought. Calculations of differences in free energy between various reaction intermediates have been calculated; these, together with the quantum efficiency for Pchlide conversion, suggest a quantitative model for the thermodynamics of the light-driven step of Pchlide reduction.

Chlorophyll is the most abundant pigment on Earth and is essential for life as the cofactor for the photosynthetic proteins that harvest sunlight and convert it to photochemical energy. Within the chlorophyll biosynthetic pathway NADPH:protochlorophyllide oxidoreductase (POR; EC 1.3.1.33) catalyzes the light-dependent trans addition of hydrogen across the C17–C18 double bond of the D-ring of protochlorophyllide (Pchlide) to produce chlorophyllide (Chlide) (1). It is one of only two enzymes known to require light for catalysis; the other is DNA photolyase (2). As a result of this requirement for light, the reaction is an important regulatory step in the chlorophyll biosynthetic pathway and subsequent assembly of the photosynthetic apparatus (3).

Comparisons of the amino acid sequence of POR with other sequences in the database have indicated that it is a member of the short-chain dehydrogenase family of enzymes (4, 5). All of the enzymes in this family catalyze NADP(H)- or NAD(H)-dependent reactions involving hydride and proton transfers. They have many structural features in common, including conserved cofactor binding domains and highly conserved Tyr and Lys residues, which have been shown to be essential for POR activity (5–7). Indeed, a homology model of POR has been constructed by using this family as a template (6). As POR is light activated, the enzyme:substrate complex can be formed in the dark, hence removing the diffusion-associated components from analyses of the kinetics of Pchlide photoreduction. When combined with a cryoenzymology approach to study steps in an enzyme-catalyzed reaction (8, 9), POR provides a unique opportunity to trap intermediates in the reaction pathway by initiating catalysis with illumination at low temperatures. Therefore, POR may serve as an important generic model for studying the mechanism of catalysis by this family of enzymes.

The reaction catalyzed by POR involves hydride transfer from the pro-S face of NADPH to the C-17 position of the Pchlide molecule (10, 11). It has been proposed that the conserved Tyr donates a proton to the C18 position with the Lys residue thought to be important for lowering the apparent pK_a of the Tyr, allowing deprotonation to occur (5). A number of spectroscopic forms of Pchlide and various reaction intermediates have been identified in etioplast membranes (12–16). However, several processes, such as POR aggregation and prolamellar body formation (17), can complicate Pchlide photoreduction in etioplasts and, as a result, the catalytic mechanism of POR remains largely unresolved.

The recent use of heterologously expressed POR protein has provided an excellent opportunity to study the reaction in greater detail (7, 18–21). As a result, an intermediate at 682 nm has been observed by using low-temperature fluorescence measurements (7, 20), although careful inspection of the spectra reveals that only low levels of conversion (<10%) could be measured. In the present work His-tagged POR from the cyanobacterium Synechocystis (21) has been analyzed by low-temperature fluorescence and absorbance measurements. High levels of photoconversion (>95% of bound Pchlide) were obtained by using this enzyme, providing the opportunity to produce difference spectra with a very high signal-to-noise ratio. Consequently, we have been able to analyze the Pchlide-binding properties of POR, detect a previously uncharacterized intermediate after the initial photochemical step in the reduction, estimate the quantum yield for Pchlide reduction, and carry out a thorough examination of the temperature dependence of the light-driven step in the reaction pathway.

Materials and Methods

Expression and Purification of Synechocystis POR.

His-tagged POR from Synechocystis sp. PCC6803 was overproduced in Escherichia coli and purified, as described (21). The enzyme was purified further on a second column (2.5 cm × 10 cm) containing red Sepharose CL-6B (Amersham Pharmacia) equilibrated with 50 mM Tris⋅HCl, pH 7.5/1 mM DTT/20 mM NaCl. The resin was washed with 10–20 column volumes of this buffer, and the His-tagged POR was eluted with 50 mM Tris⋅HCl, pH 7.5/1 mM DTT/1 M NaCl.

Pchlide Preparation.

Pchlide was isolated from R. capsulatus ZY5 cultures grown in RCV⁺ medium (22) containing 25 μg/μl rifampicin in the dark at 32°C. This strain has a mutation in one of the three subunits required for the light-independent reduction of Pchlide, which results in the accumulation of the pigment. During growth, Pchlide was released into the medium and adsorbed on to polyurethane foam bungs. The pigment was extracted from the bungs into 100% ice-cold acetone and passed down a CM Sepharose column (2.5 cm × 10 cm, Sigma) equilibrated with 100% acetone. Contaminating pigments were removed with 5% methanol/95% acetone, and Pchlide was eluted with 25% methanol/75% acetone. The pure Pchlide was then dried under nitrogen and stored in the dark at −20°C until required. Before use, the pigment was redissolved in 100% methanol.

Low-Temperature Fluorescence and Absorbance Assays.

POR assays were carried out between 100 K and 200 K in activity buffer [60% sucrose/50 mM Tris⋅HCl, pH 7.5/100 mM NaCl/0.1% (vol/vol) Genapol X-080/0.1% (vol/vol) β-mercaptoethanol] using an OpstistatDN nitrogen bath cryostat (Oxford Instruments, Oxford, U.K.). The temperature of the sample was monitored with a thermocouple sensor (Comark, Stevenage, U.K.). To initiate POR activity, illumination was provided by a Schott KL1500 electronic cold-light source (1500 μmol m⁻² s⁻¹ white light) for 10 min. Fluorescence-emission spectra were recorded by using a SPEX FluoroLog spectrofluorimeter (SPEX Industries, Metuchen, NJ) at 77 K. The exciting light was provided from a xenon light source, excitation monochromator slit widths were 4.5 nm, and emission monochromator slit widths were 3.6 nm. Absorbance spectra were recorded with a Cary 500 Scan UV-visible-NIR spectrophotometer (Varian) at 77 K. Normalization of the spectra, spectral deconvolutions, and difference spectra were performed by using the GALACTICS software.

The apparent K_d values were obtained by fitting the fluorescence changes against the concentration of POR by using the following equation:

ΔF = [(ΔF_max [POR])/(K_d + [POR])] + F₀, [1]

where ΔF is the change in the ratio of fluorescence emission at 631 nm:644 nm upon binding to Pchlide, ΔF_max is the apparent maximum change in the fluorescence ratio, K_d is the apparent dissociation constant for Pchlide binding to POR, and F₀ is the initial ratio of fluorescence emission at 631 nm:644 nm.

The data were fitted and standard errors were calculated by nonlinear regression analysis by using the SIGMA PLOT program (SPSS, Chicago).

Quantum Efficiency for Pchlide Photoreduction.

The quantum efficiency for the reaction was calculated by measuring the amount of Chlide produced after samples containing 2.34 μM POR, 19.4 μM Pchlide, and 250 μM NADPH in 50 mM Tris⋅HCl (pH 7.5), 0.1% (vol/vol) Genapol X-080, 0.1% (vol/vol) β-mercaptoethanol were illuminated for varying lengths of time at 25°C. The weak exciting light (4 μmol m⁻² s⁻¹) was provided at 642 nm with a 3-nm bandwidth by using the SPEX FluoroLog spectrofluorimeter. The exact amount of light absorbed by enzyme-bound Pchlide could be determined by using the absorbance of the sample and the calculated K_d for Pchlide binding.

Activation Energy of the Photochemical Step.

To calculate the activation energy for the light-driven step of Pchlide reduction samples containing 35 μM POR, 1.07 μM Pchlide and 250 μM NADPH in activity buffer were maintained at different temperatures by using the OpstistatDN nitrogen bath cryostat. Photochemistry was initiated by illuminating the samples with monochromatic light at 450 nm (200 μmol m⁻² s⁻¹) by using the SPEX FluoroLog spectrofluorimeter. This actinic light was simultaneously used to monitor the fluorescence emission at 644 nm. All spectra were normalized, and the initial rate of fluorescence decrease was used to determine the rate constant at various temperatures between 100 K and 220 K. The activation energy was calculated by using the Arrhenius equation:

ln k = (−E_act/RT) + c, [2]

where k is the rate constant, E_act is the activation energy, R is the gas constant (8.314 kJ mol⁻¹ K⁻¹), T is the temperature (K), and c is a constant.

Results

Binding of Pchlide to POR.

It has been reported recently that Pchlide binding can be measured by studying changes in the fluorescence yield of the pigment (7). However, such measurements can prove to be misleading; in these experiments the Pchlide was insoluble, as no detergent was present in the buffer used. Therefore, it has to be considered that this enhancement of Pchlide fluorescence is simply a result of the partitioning of Pchlide from an aqueous environment onto the protein, rather than specific Pchlide binding. To test this idea, we showed that a similar enhancement also could be observed in the presence of other proteins, for example BSA (Fig. 1B Inset).

Fig 1.

The binding of Pchlide to POR measured by low-temperature fluorescence. (A) Fluorescence emission spectra (77 K) of free Pchlide (1.1 μM), Pchlide + POR (30 μM), Pchlide + POR + NADP⁺ (500 μM), and Pchlide + POR + NADPH (200 μM). All spectra were recorded with an excitation wavelength of 450 nm and were normalized to the fluorescence of 0.5 μM fluorescein at 500 nm. The spectra are offset from each other for ease of clarification. Difference spectra are shown in Inset, using the Pchlide-only sample as a blank. (B) Fluorescence titration of 0.13 μM Pchlide with POR in the presence of 250 μM NADPH. The ratio of fluorescence emission peak heights at 631 nm and 644 nm was measured after excitation at 450 nm at 77 K. The points represent the experimental data, and the line represents the predicted values obtained by fitting the data points to Eq. 1 by nonlinear regression analysis. (Inset) Fluorescence enhancement of Pchlide at increasing concentrations of BSA. Fluorescence emission was measured at 631 nm after excitation at 440 nm.

We have monitored the binding of Pchlide to POR in a different way, by measuring Pchlide fluorescence emission spectra at 77 K in the presence and absence of POR (Fig. 1A). Free or unbound Pchlide yielded a single band at 631 nm (F631) with a full-width half-maximum (FWHM) of 15 nm. The fluorescence maximum and bandwidth remained unchanged in the presence of POR alone. However, the inclusion of either of the cofactors NADPH or NADP⁺ results in the detection of enzyme-bound Pchlide species with characteristic red-shifted fluorescence maxima, arising from the formation of a ternary complex. To ensure that all of the enzyme was in a nucleotide-bound form, the concentrations of NADPH (200 μM) and NADP⁺ (500 μM) used in these experiments were set much higher than the K_d reported for cofactor binding (21). In the presence of NADP⁺, a shoulder appears at the red edge of the main Pchlide band. Difference spectra (Fig. 1A Inset) show that this red shift arises from a second bound Pchlide species with a fluorescence maximum at 641 nm (F641). This species also was investigated by spectral deconvolution (data not shown), which yielded a band with an FWHM of 13 nm. When NADPH is used, a band with a fluorescence maximum at 644 nm (F644) and an FWHM of 10 nm is observed, representing bound photoactive Pchlide. As a result of showing that free and bound Pchlide are spectrally distinct species, the equilibrium-binding constant for Pchlide could be determined by measuring the ratio of the 644 nm:631 nm peaks at a range of enzyme concentrations (Fig. 1B). Under these conditions, the apparent dissociation constant for Pchlide binding in the presence of NADPH was calculated to be 7.7 ± 0.7 μM.

Low-temperature absorbance measurements at 77 K show that a similar red shift is observed upon Pchlide binding to the enzyme (Fig. 2A). The 630-nm absorbance band (A630), observed in the absence of POR, is accompanied by an extra band at 642 nm (A642) in the presence of POR and NADPH. In the Soret region, the absorbance maximum shifts from 443 nm to 446 nm. Inspection of the difference spectra (Fig. 2A Inset) reveals that this shift is caused by increases in absorbance at 454 nm and 465 nm and a simultaneous decrease in absorbance at 438 nm upon binding.

Fig 2.

Low-temperature absorbance and fluorescence excitation spectra of free and bound Pchlide. (A) Absorbance spectra (77 K) of free Pchlide (3.5 μM) and Pchlide in the presence of 80 μM POR and 1 mM NADPH. A difference spectrum is shown (Inset), using the Pchlide-only sample as a blank. (B) Fluorescence excitation spectra (77 K) of the 631-nm and 644-nm emission bands of a sample containing 1.1 μM Pchlide, 30 μM POR, and 200 μM NADPH.

Fig. 2B shows the excitation spectra of the 631-nm and 644-nm emitting bands, which further emphasizes the distinctive properties of the POR-NADPH-Pchlide ternary complex. The F631 band, arising from free Pchlide, has two main excitation peaks at 445 nm and 452 nm, whereas the spectrum of the F644 band reveals a considerable red shift in the excitation maxima and the detection of two new excitation peaks at 454 nm and 465 nm, similar to the features observed in the absorbance spectra.

Pchlide Reduction Involves the Formation of a Nonfluorescent Intermediate.

The first photochemical step in the POR-catalyzed reduction of Pchlide was detected by illuminating samples between 100 K and 200 K. Importantly, a large proportion of the Pchlide (>60%) used in these experiments was in the photoactive form, which has not been the case for previous in vitro low-temperature studies (7, 20).

Fluorescence-emission spectra, each taken at 77 K, show that an initial light-dependent reaction can be observed at temperatures below 200 K (Fig. 3A). This step involves the disappearance of the fluorescence band at 644 nm, representing the photoactive POR-NADPH-Pchlide ternary complex, upon illumination. Emission spectra and difference spectra (Fig. 3A and Inset) show that the intermediate that is subsequently formed is nonfluorescent, as no other fluorescence bands were detectable. There are no observable changes in the intensity of the peak at 631 nm, confirming that Pchlide F644 is the only Pchlide form responsible for POR activity. Furthermore, the 641-nm band formed in the presence of NADP⁺ is also unaffected by illumination at these temperatures (data not shown). A temperature dependence of this photochemical step reveals that it has gone to completion at temperatures above 180 K (Fig. 3B).

Fig 3.

Detection of an initial nonfluorescent intermediate upon illumination at low temperatures. (A) Fluorescence emission spectra (77 K) of samples containing 1.1 μM Pchlide, 30 μM POR, and 200 μM NADPH after illumination for 10 min at varying temperatures. An excitation wavelength of 470 nm was used to ensure that the 644-nm band is the major fluorescent species. All spectra were normalized to the fluorescence of 0.5 μM fluorescein at 500 nm and are offset for ease of clarification. Difference spectra (Inset) were measured by using a nonilluminated sample as a blank. (B) Temperature dependence of the initial photochemical step in the POR-catalyzed reaction. The relative decrease in fluorescence at 644 nm was calculated by using the fluorescence spectra from A. (C) Fluorescence emission spectra (77 K) of a sample containing 1.1 μM Pchlide, 30 μM POR, and 200 μM NADPH after illumination at 180 K for 10 min (dashed line) and after subsequently warming to room temperature for 10 min in the dark (solid line). Spectra were recorded with an excitation wavelength of 430 nm and were normalized to the fluorescence of 0.5 μM fluorescein at 500 nm. A difference spectrum is shown in Inset, using the sample after illumination at 180 K as a blank.

To confirm that formation of the nonfluorescent intermediate is the only photochemical event required for the reduction of Pchlide, samples that had been preilluminated at 180 K were warmed to room temperature in the dark, and low-temperature fluorescence spectra were retaken at 77 K (Fig. 3C). The appearance of a band at 674 nm (F674) was detected, confirming that formation of the Chlide product had occurred. After warming in the dark, the intensity of the 644-nm fluorescence band also increased, whereas the intensity of the 631-nm band decreased. Hence, free Pchlide, emitting at 631 nm, is able to rebind to the enzyme at room temperature to form F644 following a single turnover at low temperatures.

The Nonfluorescent Intermediate Has an Absorbance Band at 696 nm.

Low-temperature absorbance spectra were also used to monitor this initial photochemical step in the reaction (Fig. 4A). Difference spectra (Fig. 4B) show the disappearance of the photoactive Pchlide band at 642 nm upon illumination. However, there is a simultaneous appearance of a new broad absorbance band with a maximum at 696 nm (A696) and an FWHM of 34 nm. A temperature dependence of the formation of this 696-nm band (Fig. 4B Inset) showed that it mirrored the F644 disappearance.

Fig 4.

Low-temperature absorbance measurements reveal the formation of a band at 696 nm. (A) Absorbance spectra (77 K) of samples containing 3.5 μM Pchlide, 80 μM POR, and 1 mM NADPH after illumination for 10 min at various temperatures. The spectra are offset successively by 0.04 absorbance units for ease of identification. (B) The difference spectra of samples from A using a nonilluminated sample as a blank clearly show the formation of a band at 696 nm. (Inset) Relative increase in the 696-nm absorbance band (black circles) after illumination for 10 min at a range of temperatures. The temperature dependence of the F644 decrease from Fig. 3B is also overlaid (white circles).

Quantitative Analysis of the Temperature Dependence and Quantum Efficiency of the Light-Activated Step Catalytic Step.

The quantum efficiency for the photoreduction of Pchlide to Chlide can be used to estimate a maximum value for the activation energy of the reaction. This quantum efficiency, estimated as described in Materials and Methods, was calculated to be 21%.

The initial light-driven step catalyzed by POR converts the POR-NADPH-Pchlide ternary complex (F644; A642) to the nonfluorescent intermediate A696. The initial rates of fluorescence decrease at 644 nm upon illumination were used to calculate rate constants for this step at various temperatures (Fig. 5A Inset). Subsequently, the activation energy for the photoconversion of the ternary complex to A696 was determined by using the Arrhenius plot of ln k vs. 1/T to be 18.8 kJ mol⁻¹ (Fig. 5A).

Fig 5.

Thermodynamics of the photochemical step of Pchlide reduction. (A) An Arrhenius plot of ln k vs. 1/T for the initial photochemical step. The straight line has a slope of −E_act/R, allowing the activation energy of 18.8 kJ mol⁻¹ to be calculated. (Inset) Rates of F644 decrease at various temperatures that were used to obtain the rate constants for the Arrhenius plot. (B) An overall energy diagram for the photoreduction of Pchlide. The free energy difference between free and bound Pchlide was calculated by using the apparent K_d for Pchlide binding of 7.7 μM (−ln K_d = ΔG/RT). The first state detectable after the light step is A696; the free energy difference between the POR-NADPH-Pchlide ternary complex and A696 was estimated to be 14.8 kJ mol⁻¹ using the red shift in absorbance from 642 to 696 nm. The dotted line for the last step denotes the fact that the details of this conversion and the number of steps involved have not been determined.

Discussion

The photoreduction of Pchlide to Chlide, catalyzed by POR, is a key reaction in the chlorophyll biosynthetic pathway. Moreover, the unique light dependence of the enzyme makes it an attractive model for studying the mechanism of other oxidoreductases and, more generally, biological proton and hydride transfers. In the present work, the Pchlide-binding properties of POR and the initial photochemical step in the reduction have been analyzed in detail by using a combination of low-temperature fluorescence and absorbance techniques. Heterologously expressed POR from Synechocystis (21) was used in the experiments, allowing the detection of previously uncharacterized Pchlide species before and after illumination.

Previous studies on etioplast membranes have revealed the identification of several spectrally distinct forms of Pchlide (12–14). The use of heterologously expressed POR proteins has shown that two of these forms, Pchlide F633 and Pchlide F645, represent free Pchlide and photoactive Pchlide, respectively (7, 20). It is generally assumed that the photoactive state arises simply from Pchlide and NADPH binding to the enzyme (12–15, 23). Contrary reports have recently claimed that the binding of Pchlide and NADPH alone are not sufficient for the formation of the photoactive state (7). A change in conformation of the POR protein was invoked to explain a 6-fold discrepancy between K_d and K_m values. However, the K_d value was based on observed changes in the fluorescence yield of the pigment, which we have shown can be caused by nonspecific pigment–protein interactions and cannot be used safely as an accurate measure of Pchlide binding. Hence, it is highly probable that photoactive Pchlide does indeed result from the two substrates binding to the enzyme, and any conformational changes that occur would have to be measured separately.

We measured the binding of Pchlide to POR by using low-temperature fluorescence experiments to monitor the formation of photoactive (enzyme-bound) Pchlide at increasing POR concentrations. Upon binding to the enzyme in the presence of NADPH, the emission maximum of Pchlide red shifts from 631 nm to 644 nm, and, as a result, the apparent dissociation constant for Pchlide binding to POR has been calculated to be 7.7 μM. This result is in very close agreement with the apparent K_m for the same enzyme, which was recently calculated to be 8.6 μM (21).

The exact cause of the red shift upon binding is unclear but may result from pigment dimerization or from interactions with the protein or dinucleotide cofactor (15). However, the latter is a more likely explanation, as the small FWHM (10 nm) and Stokes shift (2 nm) of the new band, A642-F644, are inconsistent with pigment dimerization (24, 25). Interestingly, we have observed different fluorescent species in the presence of NADPH (F644) and NADP⁺ (F641), suggesting that interactions with the cofactor may play a significant role in the shift. Similar differences also have been measured in etioplast membranes, where a POR-NADP⁺-Pchlide complex had a fluorescence band at 637 nm and a POR-NADPH-Pchlide complex emitted at 644 nm (15). Previous experiments with etioplast membranes have also revealed the existence of a photoactive Pchlide state with a fluorescence band at 655 nm (12–15). This species was not detected in our experiments and can be attributed to large POR-NADPH-Pchlide aggregates that form only in these membranes (15, 23).

Intermediates in enzyme reactions have often been identified by working at low temperatures to slow down the kinetics of their formation (8, 9). In the present work we have combined this approach with the use of light to trigger the catalytic reaction. Therefore, because POR is light-dependent, it provides a unique opportunity to form an active enzyme:substrate complex without triggering any catalytic activity. A single turnover can then be initiated by illumination, using low temperatures to trap intermediates in the reaction pathway. This method was recently used to observe the in vitro formation of an intermediate with a fluorescence band at 682 nm at temperatures above 220 K (7, 20). We have now initiated catalysis in a previously unexplored temperature region and have shown that photochemistry can occur at temperatures well below 220 K. Hence, formation of the F682 intermediate that was previously identified (7, 20) is not the first step in the photoreduction of Pchlide and must be preceded by an initial light-dependent step.

Our results have shown that the first step in the reaction upon illumination is the formation of a nonfluorescent intermediate. The intermediate does not form when NADP⁺ is included in place of NADPH, confirming that this initial photochemical step will only proceed if Pchlide is in a ternary complex with POR and NADPH. The appearance of Chlide was observed after samples were warmed to room temperature in the dark, demonstrating that it is the only photochemical process required for the reduction of Pchlide. Furthermore, one or more dark steps must be required before Chlide is produced.

The disappearance of the fluorescence band at 644 nm is accompanied by the appearance of an absorbance band with a broad maximum at 696 nm. A matching temperature-dependence for each of these processes provides thermodynamic evidence that both spectral changes represent an identical event. We report here on the formation of such an intermediate using a purified enzyme, although a similar nonfluorescent intermediate with a broad absorption band at 690–705 has been detected as the first product of a POR-catalyzed reaction in etioplast membranes (16). Previous experiments using purified pigments to generate chlorophyll cation and anion radicals have shown that such species are nonfluorescent with red-shifted absorbance maxima (26, 27). Because A696 is red-shifted with respect to A642, we suggest that it is likely to be a Pchlide radical. This intermediate should receive further characterization in detail by using EPR and ENDOR spectroscopy.

Calculations of the temperature dependence and quantum efficiency of the POR-catalyzed reaction help to provide a picture of the thermodynamics of the light-driven step. Also helpful are the K_d for Pchlide binding and the optical properties of the Pchlide substrate, which allow estimates to be made of the ground-state free energies of the free and enzyme-bound Pchlide molecules. All of these thermodynamic properties can be summarized in an energy diagram for the reaction (Fig. 5B).

The free energy for the reaction: Pchlide_free + POR-NADPH → E-NADPH-Pchlide can be estimated from the K_d to be −29 kJ mol⁻¹. It would seem that this ternary complex can be cooled down to a state that is photochemically active even at 120 K, which suggests that a “near attack” conformation (28) is achieved either before freezing or during the cooling process. Once in this state, the energy of a single photon is sufficient to surmount an energy barrier for the reaction, which we estimate as 18.8 kJ mol⁻¹ from measurements of the temperature dependence of the initial photochemical step. On the other hand, the quantum efficiency, which takes into account all of the light and dark steps through to Chlide, was estimated as 21%. Assuming a single-photon process for the light-dependent step and calculating the energy of a photon at 642 nm to be 191 kJ mol⁻¹, the overall conversion to the product apparently requires ≈40 kJ mol⁻¹. In a theoretical analysis of proton and hydride transfer in alcohol dehydrogenase (29), which is a structural homologue of POR (6), barriers for proton and hydride transfers have been calculated to be ≈85 and 39 kJ mol⁻¹, respectively. We conclude that some of the free energy of the absorbed photon will be required to drive the subsequent dark reactions. It will be important to study the dark reactions in detail to obtain a complete picture of the thermodynamics of catalysis by POR.

Further considerations of the mechanistic details of this enzyme will require analysis of A696 by EPR techniques. In addition, the application of ultrafast time-resolved techniques should allow measurement of the rate constants for the light-driven step at room temperature, and allow us a glimpse of how this step is coupled to the excited states of the enzyme-bound Pchlide molecule.

Abbreviations

• POR, protochlorophyllide oxidoreductase

• Pchlide, protochlorophyllide

• Chlide, chlorophyllide

• FWHM, full-width half-maximum