This is an open access article published under a Creative Commons Attribution (CC-BY) License, which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited.
Article pubs.acs.org/biochemistry
Cross-Species Analysis of Protein Dynamics Associated with Hydride and Proton Transfer in the Catalytic Cycle of the Light-Driven Enzyme Protochlorophyllide Oxidoreductase Robin Hoeven, Samantha J. O. Hardman, Derren J. Heyes,* and Nigel S. Scrutton* Centre for Synthetic Biology of Fine and Speciality Chemicals, Manchester Institute of Biotechnology, The University of Manchester, 131 Princess Street, Manchester M1 7DN, U.K. S Supporting Information *
ABSTRACT: Experimental interrogation of the relationship between protein dynamics and enzyme catalysis is challenging. Light-activated protochlorophyllide oxidoreductase (POR) is an excellent model for investigating this relationship because photoinitiation of the reaction cycle enables coordinated turnover in a “dark-assembled” ternary enzyme−substrate complex. The catalytic cycle involves sequential hydride and proton transfers (from NADPH and an active site tyrosine residue, respectively) to the substrate protochlorophyllide. Studies with a limited cross-species subset of POR enzymes (n = 4) have suggested that protein dynamics associated with hydride and proton transfer are distinct [Heyes, D. J., Levy, C., Sakuma, M., Robertson, D. L., and Scrutton, N. S. (2011) J. Biol. Chem. 286, 11849−11854]. Here, we use steady-state assays and single-turnover laser flash spectroscopy to analyze hydride and proton transfer dynamics in an extended series of POR enzymes taken from many species, including cyanobacteria, algae, embryophytes, and angiosperms. Hydride/proton transfer in all eukaryotic PORs is faster compared to prokaryotic PORs, suggesting active site architecture has been optimized in eukaryotic PORs following endosymbiosis. Visible pump−probe spectroscopy was also used to demonstrate a common photoexcitation mechanism for representative POR enzymes from different branches of the phylogenetic tree. Dynamics associated with hydride transfer are localized to the active site of all POR enzymes and are conserved. However, dynamics associated with proton transfer are variable. Protein dynamics associated with proton transfer are also coupled to solvent dynamics in cyanobacterial PORs, and these networks are likely required to optimize (shorten) the donor−acceptor distance for proton transfer. These extended networks are absent in algal and plant PORs. Our analysis suggests that extended networks of dynamics are disfavored, possibly through natural selection. Implications for the evolution of POR and more generally for other enzyme catalysts are discussed.
E
Here a cross-species analysis of protein dynamics that impact hydride/proton transfer has been undertaken using the lightactivated enzyme protochlorophyllide oxidoreductase (POR; EC 1.3.1.33). POR catalyzes reduction of the C17−C18 double bond of the substrate protochlorophyllide (Pchlide) to form chlorophyllide (Chlide); NADPH is the reducing coenzyme. The catalyzed reaction is an essential step in the chlorophyll biosynthesis pathway and is required for subsequent assembly of the photosynthetic apparatus.18,19 Enzyme-bound Pchlide acts as a photoreceptor. Upon absorption of light, an excitedstate intramolecular charge transfer (ICT) complex involving excited-state interactions between the carboxyl group at the C17 position of Pchlide and a conserved tyrosine residue is formed.20 This creates a transient electron-deficient site at the C17 position, which enables nucleophilic attack of hydride from the pro-S face of NADPH on the microsecond time scale.21−23 The resulting negative charge at the C18 position enables
nzymes are dynamic molecules, and understanding the role of protein dynamics associated with bond breaking and/or making is important.2−9 Answers to a number of key questions are needed, including the following: (i) Do conformational fluctuations within an enzyme active site couple to longer-range structural variations?10−13 (ii) Are the dynamic profiles of enzyme homologues conserved alongside reaction chemistry, or is there variability?4 (iii) Are the dynamic profiles of enzyme homologues influenced by natural selection? The solvent environment is also coupled to the enzyme, and any influence of solvent dynamics on reaction rate can be probed by varying solution viscosity.7,14−17 Solution visocity effects, on either enzyme turnover (kcat) or the rate constants for individual steps of the catalytic cycle, should therefore inform on the conservation (or otherwise) of dynamics in enzymes that catalyze the same chemistry but are from different species. Moreover, any altered dynamic profiles might arise through natural selection if they give rise to a phenotypic advantage (e.g., through enhanced catalytic rate, enzyme stability, or similar). © 2016 American Chemical Society
Received: December 16, 2015 Revised: January 22, 2016 Published: January 25, 2016 903
DOI: 10.1021/acs.biochem.5b01355 Biochemistry 2016, 55, 903−913
Article
Biochemistry
Figure 1. Scheme for the reaction mechanism of protochlorophyllide oxidoreductase (POR). The first step involves the light-activated hydride transfer from NADPH (red) to the C17 position of Pchlide, followed by a proton transfer from a conserved Tyr residue (purple) in the active site to the C18 position.
Chlamydomonas reinhardtii, Cyanidioschyzon merolae, Daucus carota, Lyngbya majuscule, Nicotiana tabacum, Nostoc punctiforme, Pinus mugo, Physcomitrella patens, and Zea mays were synthesized by GenScript (GenScript Inc.) and digested with NdeI and BamHI HF restriction enzymes (New England Biolabs Ltd.). The genes were cloned into an NdeI- and BamHI-digested pET9-His vector, which is a derivative of pET9a from Novagen described previously,34 and confirmed by sequencing (Eurofins Scientific). The full protein sequences encoded by each gene are found in Figure S1. The constructs were transformed into chemically competent Escherichia coli BL21(DE3) pLysS cells (Agilent Technologies) using the heat-shock method recommended by the commercial supplier. Transformed cells (pET9-His-POR) were grown as described previously.35 Details of protein purification are given in the Supporting Information. Steady-State Enzyme Assays. UV−visible absorption data were acquired using a Varian Cary 50 UV−vis spectrophotometer with samples contained in a 100 μL quartz cuvette (1 cm path length). Illumination of the samples was conducted with a 455 nm LED (Thorlabs, Inc.) fitted with a converging lens and held in place to focus the light onto the sample from above. Chlide formation was followed by measuring the time-dependent absorbance change at 670 nm. The slope of the absorbance change per unit time was used to calculate the initial rate of turnover. The concentrations of the substrates and products were calculated using the following extinction coefficients in aqueous solution: NADPH, 6.22 mM−1 cm−1 at 340 nm; NADP+, 17.8 mM−1 cm−1 at 260 nm; Pchlide, 23.95 mM−1 cm−1 at 630 nm; and Chlide, 69.96 mM−1 cm−1 at 670 nm. Protein concentrations were determined using the Bio-Rad DC assay kit (Bio-Rad Laboratories Inc.) with bovine serum albumin as a standard. Laser Flash Photolysis. A neodymium-doped yttrium aluminum garnet (Nd:YAG) laser (Brilliant B, Quantel) was used to generate a 1064 nm pulse (∼8 ns), which is guided through two harmonics to triple the frequency (to 355 nm) and finally tuned by an optical parametric oscillator (OPO) to achieve a 450 nm pulse. Upon excitation of the sample, spectral changes were monitored by measuring the absorption at a
subsequent transfer of a proton from a conserved tyrosine residue to form Chlide product (Figure 1).24,25 Both the hydride and proton transfer reactions proceed by quantum mechanical tunneling.23,26 A role for fast (so-called promoting) motions has been inferred for both hydride and proton transfer reactions catalyzed by POR.7,23 A series of ordered product release and coenzyme binding steps have been shown to follow hydride and proton transfer, and these are required to complete the catalytic cycle. These binding/release steps are also linked to major conformational fluctuations in the enzyme.27 Studies relating the effects of solvent properties on protein dynamics are often compromised when using thermally activated enzymes because of the limitations of rapid mixing methods that are required to initiate catalysis. These limitations are overcome using light-activated POR. Here, synchronous triggering of catalysis in a “dark-assembled” enzyme−substrate complex is enabled using a laser pulse.18 POR is found in a wide range of organisms enabling cross-species comparison of the hydride and proton transfer chemistry and associated solvent/ protein dynamics.1,28 POR likely arose ca. 2.4 billion years ago with the rise in atmospheric oxygen levels and was introduced into plant cells during the primary symbiosis of an ancestral cyanobacterium.29−33 Apart from angiosperms, most oxygenic phototrophs still have light-independent PORs to reduce Pchlide, but among prokaryotic phototrophs, POR is found only in cyanobacteria, suggesting that light-dependent chlorophyll biosynthesis was established prior to eukaryotic photosynthesis.28,30 We previously suggested that dynamics associated with the POR hydride and proton transfers have distinctive profiles across prokaryotic and eukaryotic organisms. This was inferred from data acquired for only four POR enzymes (three cyanobacterial PORs and one plant POR).1 Here we expand on this analysis by investigating the dynamics of hydride and proton transfer in PORs from a wide range of cyanobacteria, algae, liverwort, moss, and higher plants.
■
MATERIALS AND METHODS Expression and Purification of POR Homologues. Genes encoding the light-activated POR enzymes of 904
DOI: 10.1021/acs.biochem.5b01355 Biochemistry 2016, 55, 903−913
Article
Biochemistry
Figure 2. Phylogenetic tree of POR genes. The cyanobacterium and prochlorophyte group (red) is located at the top, while the plants (dark green) are at the bottom, with the branches between them being algal species (gold). The bryophytes (liverwort and moss) are highlighted by the dashed light green box. The scale bar indicates how many amino acids are replacements per site.
for two or three independent samples. For each sample, measurements were performed several times and averaged prior to data fitting. The cuvette used had a 2 mm path length in the pump direction and 1 cm path length in the probe direction. Samples contained 50 μM enzyme, 250 μM NADPH (except C. reinhardtii, Cy. merolae, L. majuscula, and D. carota, where a 2 mM NADPH concentration was used due to the reduced affinity of the enzyme for NADPH), and 12 μM Pchlide in
single wavelength using an Applied Photophysics LKS spectrometer. A pulsed 150 W xenon arc flash lamp was used to produce white light that passes through a monochromator to generate a single-wavelength probe beam. The scattered laser light was eliminated using a second monochromator positioned after the sample. Laser photoexcitation studies to determine rates of hydride and proton transfer were as described previously.23 Typically, each data point is an average collected 905
DOI: 10.1021/acs.biochem.5b01355 Biochemistry 2016, 55, 903−913
Article
Biochemistry Table 1. Summary of Km, Ki, Vmax, and kcat Values for Each of the POR Homologuesa
a
Kinetic parameters for POR from cyanobacteria (red), algae (gold), non-angiosperm land plants (green), and angiosperms (blue) are shown.
with a power of 0.5 μJ and a beam diameter of ∼200 μm. Samples were flowed at a rate of approximately 30 mL/min through a 0.2 mm path length quartz cell (at room temperature) to ensure that a different area of the sample is excited with each pump laser pulse. Samples containing 500 μM POR, 200 μM Pchlide, and 4 mM NADPH in activity buffer with 10% glycerol, 0.1% 2-mercaptoethanol, and 0.5% Triton X-100 were prepared in the dark. Global Analysis. The data sets were analyzed globally using the open-source software Glotaran.37 This procedure reduces the matrix of change in absorbance as a function of time and wavelength, to a model of one or more exponentially decaying time components, as described in the main manuscript, each with a corresponding difference spectrum [evolution-associated difference spectra (EADS)]. Errors quoted with the lifetime values are the standard errors calculated during the global analysis. For the analysis, the pre-excitation background was subtracted and both the “fast” and “slow” data sets were fitted to a simple sequential model in which one species converts to another, which then persists for the lifetime of the experiment. The “fast” data sets were the result of one scan with a
