1961-1964, Department of Biochemistry, University of Cambridge
1964-1965, Department of Biochemistry, University of Bristol
Ph.D. degree, University of Cambridge, Biochemistry, 1965
Research for Ph.D. degree under Professor J. B. Chappell.
The work was predominantly on ion transport in mitochondria, and was of importance in providing early experimental evidence for the chemiosmotic hypothesis. The work led to several new conclusions about the mechanism of transport of ions and metabolites across the mitochondrial membrane, and the mechanism of ionophores, including:
Research during this period was in collaboration with Professor L. Packer, of U.C., Berkeley. The work was on:
Ion transport in chloroplasts, and its relation to light-induced light scattering and volume changes. This led to the formulation of a unified hypothesis relating ion movements in chloroplasts to light-scattering and volume changes.
More significant was the subsequent discovery and investigation of NH4+-uptake by chloroplasts which led to the elucidation of the mechanism of uncoupling by ammonium salts and other amines.
During this period research interests diversified to include many aspects of energy conservation in photosynthesis.
This work led to a demonstration of the electrogenic nature of H+ uptake by these organelles, an explanation of the mechanism of action of nigericin and other protonophores acting through neutral exchange mechanisms, an estimation of the potential of the H+-gradient, and the development of methods using pH indicators to measure rapid H+-uptake.
Previous work had shown that a class of absorbance changes in photosynthetic systems could be interpreted as shifts to the red of the absorbance spectrum of the carotenoids of the bulk ligh-harvesting antenna. We showed that similar absorbance changes could be induced by ionic gardients, and introduced the use of ionic gradients (of K+ in the presence of valinomycin, or of H+ in the presence of protonophores), to calibrate the response as a function of membrane potential. Our work demonstrated that in chromatophores these electrochromic shifts were a linear function of the electrical potential across the membrane, and could be used as a membrane voltmeter with nanosecond time resolution. It was shown that excitation by light induces electrogenic electron flow across the chromatophore membrane both in the photochemical reaction and in subsequent dark electron flow. We also looked at local electrochromic changes of carotenois in isolated reaction centers. Using a mutant of Rps. sphaeroides , G1C, which we showed to contain only one major carotenoid (>98% neurosporene), we were able to measure the spectrum of the change for a single known species. This made possible for the first time a realistic theoretical treatment, and a less ambiguous interpretation of the carotenoid change.
Quenching of chlorophyll fluorescence that is dependent on the high energy state was characterised in isolated chloroplasts, to determine which component(s) of the proton gradient controlled the phenomenon. The quenching (subsequently known as qE) was shown to be dependent on the chemical H+-gradient, and we suggested that onset of the pH gradient leads to an increase in rate of thermal deactivation of excited chlorophyll, controlled by the membrane environment. The physiological importance of this mechanism is now recognized in the extensive literature on nonphotochemical quenching (or qE quenching), which has shown that this DpH-dependent excitation dissipation is the mechanism by which plants protect themselves against the harmful effects of light in excess of that needed to meet metabolic demand.
The energy dependence of delayed fluorescence of chloroplasts and chromatophores from photosynthetic bacteria was investigated and further characterized. It was shown that the reversal of the photochemical reactions leading to emission of delayed fluorescence is dependent on the electrochemical H+-gradient, but with contributions of chemical and electrical components that depended on the system under study. A mechanism relating the intensity of delayed fluorescence to both the redox free-energy stored in the photoproducts, and the proton gradient, was proposed to explain this effect. This is now generally accepted, and provides a framework for understanding the work terms of the photochemical reactions in situ, and the mechanism of thermoluminescence.
Redox titrations of chromatophores in the dark were used to characterise the spectrophotometrically identifiable components of the electron transport chains of Rps. sphaeroides (with Dr. J. B. Jackson, in collaboration with Professor P. L. Dutton, The Johnson Foundation, University of Pennsylvania), and of Rps. capsulata. From the dependence on ambient redox potential of the flash induced changes, it was possible to identify the components involved in photosynthetic electron transport. Similarly, the dependence on redox potential of other spectrophotometric changes (carotenoid change, and H+-uptake (see also ref. 43)) made it possible to show the association between these reactions and the reactions of electron flow. These studies were extended (in collaboration with Professor B. A. Melandri and Dr. D. Zannoni, University of Bologna) to include the similar electron transport and energy coupling reactions in respiratory particles from Rps. capsulata grown aerobically in the dark, which identified the cytochrome oxidase.
A study of amine distribution with respect to pH gradients artificially induced across liposome membrane showed that quenching of the fluorescence of the monoamine, 9-amino acridine can be used to measure the gradients. The technique was used to follow transport of ions and redox dyes across liposome membranes, and to measure pH gradients across chromatophore and chloroplast membranes.
The use of pH indicators in a spectrophotometric method for the rapid measurement of small light-induced pH changes led to the discovery and characterisation of rapid H+ binding by chromatophores. Rapid H+-uptake was dependent on the reduction of a H-carrying secondary acceptor by an electron from the primary acceptor and a H+ from the external medium. Successive flashes allowed the extent of H+-uptake to approach that induced by continuous illumination. Saturation was reached by a series of discrete rapid reactions, indicating that this reaction of the electron transport pathway represents an elementary and integral part of the mechanism of the light driven H+-pump of the chromatophores.
Using reaction centres from Rps. sphaeroides, and Rps. capsulata, the flash induced reaction of secondary donors and acceptors with the purified preparations were studied. It was shown that cytochrome c2, the native secondary donor, reacts with second-order kinetics to reduce the photooxidised reaction centre, while the primary acceptor reduces endogenous ubiquinone in a reaction which involves the uptake of a hydrogen ion from the aqueous phase, and which has many of the features of the rapid H+-uptake observed on flash illumination of chromatophores.
Antibodies were prepared (in collaboration with Dr. G. Hauska, University of Bochum, West Germany) against purified cytochromes and reaction centres from the photosynthetic bacteria. These were used to identify the location of the antigens in situ using preparations of both chromatophores and protoplasts. The results showed unambiguously that cytochrome c2 is a periplasmic protein, situated on the inner side of the chromatophore membrane, which reacts from the aqueous phase on the opposite side of the membrane to that on which the reduction of the secondary acceptor occurs. This was the first work to demonstrate that the anisotropic arrangement of an electron transfer chain expected from the chemiosmotic hypothesis, since the the photochemical reaction centre was shown to span the membrane, and the photoreaction to involve the flow of electronic charge across the membrane.
Computer-linked systems were developed to perform automated redox titrations. From a small amount of material (5 ml suspension), it was possible to obtain spectra and mid-point redox potentials of all the spectrophotometrically identifiable components, or to follow automatically the dependence on redox potential of flash induced kinetic changes, in a day's work. This new facility made possible a number of projects, for instance in the characterisation of mutant bacterial strains, where the small amount of material available precludes attack by conventional techniques.
An apparatus was developed for measuring delayed fluorescence in the microsecond range following a single laser flash. The kinetics reflect the rate of electron flow to the photooxidant of photosystem 2 both from the water-splitting reactions and by reversal of the photoact. These kinetics were compared with the microsecond induction phase of fluorescence (which reflects similar events), and were used to observe effects of pH, preillumination and flash sequence, with a view to identifying the involvement of protons in the oxygen-evolving reactions. We showed by direct measurement of H+-release using indicators that when chloroplasts are illuminated from the dark adapted state, protons are released to the inner thylakoid space in synchrony with the transitions of the oxygen evolving complex, giving a quartenary pattern (starting in the S0-state) of approximately 1, 0, 1, 2 . We interpreted this pattern as indicating that the elements of water were incorporated into the enzymic cycle at transitions earlier than those leading to release of oxygen.
We incorporated purified reaction centres from Rps. sphaeroides into liposomes, and subsequently into solution in hexane. These preparations were used to reconstitute the proton pumping activity associated with the reaction centre complex.
The following items of apparatus were designed and constructed:
Dual wavelength spectrophotometer (1968), scanning dual wavelength and split beam (dual purpose) spectrophotometer (1969) (in association with Professor Chappell); Phosphoroscope (1970); Rapid response spectrophotometer for laser flash kinetics (1970); Rapid response dual wavelength spectrophotometer (1971); Computer-linked scanning spectrophotometer (1973); Computer-based signal averaging facility (1973); Automated spectral scanning and flash kinetic redox titration facilities (1974-75); Microsecond fluorescence induction fluorimeter (1975); Microsecond delayed fluorescence photometer (1976); Microprocessor-based base-line corrector for scanning spectrophotometer (1976); Microprocessor-based signal averager (1976); Microprocessor-based scanning and kinetic spectrophotometer (1976); Various other smaller pieces of equipment (recording electrodes, fluorimeters, light scattering, photometer, etc.) were also constructed over this period.
We have continued our studies of Rps. sphaeroides and Rps. capsulata and characterised the involvement of three new components in the chain,- the Rieske-type iron sulfur center, a bound cytochrome c (cyt c1), and a low potential cytochrome b (cyt b566 or cyt bL, Em,7 = -90 mV),-and have been able to account for their role in cyclic photosynthetic electron transfer in terms of a modified Q-cycle. These components are analogous to the similar components of the mitochondrial ubiquinol: cytochrome c oxidoreductase, and perform the same function. The techniques developed in this work have made it possible to assay many of the important physico-chemical parameters describing turn-over of the complex using the enzyme in situ, thus by-passing the requirement for an isolated enzyme for kinetic analysis. This greatly simplifies analysis of functional changes in mutant strains.
In chloroplasts, we have re-examined the redox potentials of the cytochromes, and demonstrated that the b6 component can be resolved into high and low potential species. In collaboration with Drs. Pierre and Anne Joliot, we have investigated the kinetic behavior of the cytochrome b6f complex. Our own conclusion is that the general behavior is well understood in terms of the modified Q-cycle mechanism; the reduction of plastoquinone is anomalous, and may provide important clues to the mechanism of the Qi-site.
We have further characterised the electrochromic changes of both the carotenoid and chlorophyll bands (in collaboration with Dr. R. Niedermann), and shown that the carotenoid population undergoing change in response is that associated with the B800-B850 pigment-protein complex of Rps. sphaeroides. In contrast, bacteriochlorophyll from both pigment protein complexes shows an electrochromic change. In collaboration with Dr. B. Honig and Drs. T. and H. Kakitani, we developed a quantum mechanical explanation of the electrochromic response, which accounts for the different behaviour of the pigments in different complexes, and extends classical electrochromic theory to the local molecular domain.
We used the electrochromic change to demonstrate that, in addition to the well characterized electrogenic oxidation of cyt bH, reduction of cyt bH by cyt bL is also electrogenic. By deconvoluting the contributions of the electrochromic change, we have been able to analyze separately the changes in poise of the redox reactions and the proton gradient on illumination, and to measure the thermodynamic gradients in the system as the electron transfer chain approaches static head against the proton gradient in the coupled steady state.
In chloroplasts, we showed that the two steps by which electrons reduce the secondary quinone occur with different kinetics. We showed that the electron transfer rates are differentially sensitive to a variety of inhibitors and herbicides, and are modified in plants which develop a resistance to herbicides. We characterized the kinetics and thermodynamics of the PS II two-electron gate, and the role of protons in stabilizing semiquinones. We have made a detailed analysis of the changes in kinetic and thermodynamic parameters determining turn-over of the two-electron gate in wild-type and herbicide resistant strains.
In Rps. sphaeroides, we have shown that in chromatophores the two-electron gate operates so as to reduce ubiquinone to ubiquinol, which migrates by diffusion to its site of oxidation by the ubiquinol:cyt c2 oxidoreductase. The behavior of the two-electron gate is modulated by the redox poise of the system, and by pre-illumination. Even with a weak measuring beam, the characteristic binary pattern is observed only when cyt c2 is oxidized. If care is taken to prevent any prior illumination, the binary pattern is seen at lower potentials, suggesting that the loss of binary oscillations on reduction of cytochrome c2 reflects an inhibition of the back-reaction of QB.- produced by the weak background illumination.
We have characterised a number of new inhibitors, and studied the mechanism of action of these and some other established inhibitors. In Rps. sphaeroides, we demonstrated that a new class of quinone-analogue inhibitor, exemplified by UHDBT (5-n-undecyl-6-hydroxy-4,7-dioxo-benzothiazole) functions by binding at or close to the Rieske-type 2Fe.2S center, and appears to block its reactions with the chain.We have shown that myxothiazol competes with UHDBT or quinol for a common site. However, the differential effects on the Rieske 2Fe.2S center, and differential resistance in different mutant strains show that the two inhibitors bind differentially to the subunits which make up the catalytic site.
In chloroplasts, UHDBT also inhibits the oxidation of the primary quinone acceptor of photosystem II. We have shown that DBMIB, and 3-undecyl-2-hydroxy-l,4-naphthoquinone have a similar inhibitory effect, and that the inhibition can be accounted for by competition of the quinone analogue with the endogenous quinone or quinol for the binding site at which the quinone is reduced to a stable bound semiquinone during the operation of the two-electron gate. We have studied a number of other quinone analogues and other inhibitory sites to see if a similar mechanism can be shown to operate elsewhere.
We have used a regenerative stopped flow apparatus linked to a fluorescence photometer to assay the kinetics of binding of inhibitors at the two-electron gate. Binding can be assayed under conditions in which the quinone is oxidized, or when the site is occupied by a semiquinone. The results are consistent with a detailed model for the site previously developed, and allow us to describe many of the properties of inhibitor binding at the two-electron gate in terms of a relatively simple model.
We have pursued our interest in the involvement of protons by using other techniques for studying these reactions, including a detailed study of the relation between the proton gradient and the partial reactions of the donor-side, through studies of delayed fluorescence. We have investigated inhibitory treatments to characterize the role of the different subunits of the oxygen evolving complex, the role of cyt b-559, the mechanism of the ADRY effect, and the role of Cl-.
In collaboration with Drs. R. Gennis, S. Kaplan and T. Donohue we have been investigating the role of specific amino acids in the structure and catalytic mechanisms of this enzyme. We are using a combination of techniques from molecular genetics, enzymology and biophysics in a protocol involving prediction and modelling of the protein structure, molecular engineering of specific residues, biophysical assay of the function to assess the consequence of mutation, and adjustment to the structural model. The fbc operon encoding the three major subunits of the UQH2:cyt c2 oxidoreductase has been cloned and sequenced, a battery of molecular engineering techniques has been developed to allow specific mutation at any site. A large number of site-directed mutants have been produced, and functional modifications explored using previously developed kinetic and spectrometric techniques. We have used specific mutation to explore prosthetic group ligands, and the role of residues at putative catalytic sites in binding of substrates and inhibitors.
Protein structural prediction has had a limited success. We have identified two factors contributing to this lack of success: a) the forces determining the partition of particular residues to specific secondary structures are weak; b) the targets chosen for probability studies have been inappropriate. More specifically, by lumping all helices (or all sheets, turns, etc.) into a common category, previous workers have failed to recognize that the physico-chemical parameters determining partition into different secondary structures are different in the protein exterior and interior. In order to overcome these difficulties, we have a) reclassified secondary structure in the protein database to include subclasses according to location (buried or surface) in the protein, b) we are using neural networks to assess the importance of long-range interactions (as probed by hydropathy, amphipathy, acrophilicity, etc.) in stabilization of secondary structure; and c) we are using sequences comparison and information theory to assess the importance of conserved features, and the "mutability moment" as an indicator of tertiary structure.
We have used these approaches to predict a folding pattern for the subunits of the cytochrome b subunit. We were able to suggest a change from the established 9-helix model to one with only 8 transmembrane helices. Subsequent mapping of inhibitor resistant mutants, and characterization of specific mutant strains, has lent strong support to the model. In collaboration with Dr. R. Gennis, we have also use molecular genetics techniques to explore the topology of the cytochrome b subunit through fusion of partial sequences with the phoA gene of alkaline phosphatase. The results strongly support the new 8-helix model. We have used computer model building to suggest a tertiary structure for the cytochrome b subunit, showing the location of catalytic sites for oxidation and reduction of quinone, and are using energy minimization and molecular dynamics to generate a plausible physical model.
We have been using inhibitor titrations to explore the organization of the electron transfer chain in the membrane, and have demonstrated that in chromatophores of Rb. sphaeroides the electron transfer chain is delocalized. We have developed a computer model in which the effects of delocalization on the distribution of electrons in the chain can be statistically modelled. We have shown that the apparent low equilibrium constant between the donor chain (cytochromes c1 and c2) and the reaction center, previously thought to indicate the presence of supercomplexes, can be explained by heterogeneity in the distribution of the bc-complex, cytochrome c2 and the reaction center in a population of chromatophores.
The UQH2:cyt c2 oxidoreductase has been purified as a highly active enzyme in high yield. The isolated enzyme has the same thermodynamic and spectroscopic properties as the enzyme in situ, contains the same prosthetic groups, and has an activity is close to that measured in situ. The enzyme has four subunits, three of which, cytochrome b, Reiske-type 2Fe.2S protein, and cytochrome c1, contain the centers associated with redox active prosthetic groups. N-terminal sequencing data indicate that these three subunits are the proteins coded by the fbc operon. More recently we have constructed an fbc gene with a coding sequence for a (His)6 tag at the C-terminal end of cytochrome b, and expressed this in Rb. sphaeroides. The complex synthesized is fully active, and can be rapidly isolated and purified using a Ni-affinity column. The isolated complex is obtained in high yield, and has a similar activity and properties as the native complex. We are using this preparation to synthesize sufficient protein for structural studies using X-ray crystallography, and detailed spectroscopic investigations. The construct will provide a vehicle for future molecular engineering studies where isolation of the complex is needed.
We made the first computer model of the D1 and D2 proteins which form the core of photosystem II, by using the X-ray crystallographic structure of Rps. viridis as a template. We have used these to predict sites for modification of the catalytic sites by site-directed mutagenesis. We have developed a new protocol for molecular engineering in the D1 protein of Chlamydamonas reinhardtii by construction of a plasmid containing an intron-free psbA gene in tandem with a aadA spectinomycin resistance cassette. We have used the model to construct specific mutations in residues thought to contribute to the QB-site, and on the oxygen-evolving apparatus on the donor side.
Studies of the two-electron gate in mutant strains.
In collaboration with Dr. Sue Golden (Texas A & M), we have studied the kinetic consequences of specific mutations in the secondary quinone binding site of photosystem II of Anacystis, and are now extending this work to C. reinhardtii. Some mutations occur naturally, giving rise to herbicide resistant strains, and we are constructing others using PCR-based techniques for site directed mutagenesis. We have developed a set of PCR-based tools for amplification and sequencing of the psbA gene which codes for the D1 protein of Photosystem II. We have sequenced wild-type and herbicide resistant strains of Amaranthus hybridus, and shown the codon change leading to modification of the QB-site. We are using our new protocols to explore mutations in C. reinhardtii.
Studies of the donor-side
The D1 protein is thought to contain the primary ligands for the Mn-center which catalyses the oxidation of water. We are investigating the structure-function relationships in these spans, by using biophysical analysis of the kinetic and thermodynamic parameters to assay the consequences of mutation, in order to identify contributions of specific residues to the mechanism of catalysis. We have extended our structural models to include the parts of the protein thought to be involved in the reactions of oxygen evolution, and have generated a large number of mutations to test potential ligands, and the role of tyrosine 161 as secondary donor.
We have shown that inhibition of photosynthesis by exposure to UV-irradiation leads to a loss of components on the donor side of photosystem II, including Mn in oxygen evolving preparations, and the EPR signals due to the Tyr+ radicals of Z and D. Kinetics of electron transfer on the acceptor side of PS II were unaffected in centers which retained a normal fluorescence yield. However, a loss of fluorescence accompanied loss of activity and damage to the donor side. We are further investigating the mechanism of UV photoinhibition, and comparing it with photoinhibition by visible light of high intensity, or in systems with restricted donation to photosystem II. We have explored the role of the donor side reactions in the down-regulation of photosynthesis at high light intensity, associated with loss of fluorescence yield, and demonstrated that qE-quenching does not require an active photosystem II, or the presence of the donor-side Mn-complex. We found that although the donor-side reactions were not needed for qE-quenching, an inpaired donor side led to photoinhibition at relatively low light intensity, and suggested that this might indicate the importance of the oxidized donor-side in generation of damage to photosystem II.
With the demonstration that turn-over of photosystem II is not required for qE-quenching, we have turned our attention to the role of light-harvesting complexes and the xanthophyll cycle in these quenching processes, and the protection against photoinhibition. We have proposed a hypothesis for the mechanism of qE-quenching which explains the loss of fluorescence following formation of a low lumenal pH in terms of a change in liganding of chromophores in the minor chlorophyll protein (CP) complexes. The minor CPs provide the interface between the bulk antenna and the reaction center, and are the main site of the xanthophyll cycle changes. Sequence comparison has shown that in the minor CPs, glutamate residues substitute for glutamine chlorophyll ligands found in the bulk LHCII. We are currently setting up to test this hypothesis using synthesis of the apoproteins of the light-harvesting complexes, and specific mutagenesis in E. coli, reconstitution with pigments, and assay of the quenching mechanism using in vitro protocols.
In the intact plant, the photochemical reactions, and the reactions of electron transfer and proton transport which they drive, are part of an integrated mechanism which responds to the physiological state of the plant as determined by environmental factors. Rates of electron transfer, and the control of cyclic and non-cyclic pathways, can be assayed directly by following the changes on illumination of redox components of the chain by using spectrophotometry, or indirectly by using fluorescence techniques. The generation and utilization of the proton gradient can be assayed by following the 515 nm electrochromic change. One aim of our research is to develop portable instrumentation to facilitate such measurements. Several instruments have been completed: i) a portable kinetic fluorimeter with measuring-flash optics for use in the field; ii) a portable kinetic spectrophotometer with measuring-flash optics, which allows good kinetic resolution and high sensitivity. The two instruments are portable versions of laboratory instruments developed in parallel, but are set up for use with intact leaves, and have been used in field studies. We have also developed a photometer for measuring glow curves, and a video imaging apparatus for measurement of fluorescence, through which kinetics of fluorescence induction can be imaged, or induction curves measured at selected points in the field of interest. The longer term aim of the research is to understand how the physiological state of the intact system is determined by environmental factors, and to relate these to the decreased yields of photosynthesis found under adverse conditions. The down-regulation of photosynthesis under conditions of high light, especially when the stomata are closed, involves a feed-back through the pH gradient. This regulates exciton delivery to the photosystems through qE-quenching, and is part of a finely tuned control which matches excitation delivery to flux through the photosynthetic chain to meet metabolic demand. We are currently developing second generation versions of the above instruments to explore these control mechanisms in intact plants.
Flux through the ATP-synthase in intact leaves can be followed by measurement of the decay of the electrochromic changes following flash activation. The slow decay observed in dark-adapted leaves after a single flash, became rapid (half-time < 5 ms) after pre-illumination at low intensity, due to activation of the ATP-synthase by reduction of thiol groups. The light requirement for activation, and time course of reoxidation, have been examined under laboratory and field conditions. The kinetics of deactivation show a diurnal pattern which seems to be controlled by a circadian clock rather than directly by light. Reduction of the ATP-ase is through the thioredoxin system, which also regulates several other enzymes in the Calvin cycle. Activation shows a rapid onset and slow decay, and requires very low light intensities, suggesting a modest degree of reduction of thioredoxin. We proposed a general hypothesis for control of those enzymes of metabolism linked to the thioredoxin system through redox poise which has since been support by further work on the redox potentials and kinetics of these enzymes.
The following new instruments have been built:
Computer linked scanning split beam spectrophotometer; computer-based flash kinetic fluorimeter (several versions); computer-based flash kinetic spectrophotometer (Joliot-type)(several versions); conventional kinetic spectrophotometer; pseudo-steady state kinetic spectrophotometer; automatic base line correcting amplifier; anaerobic redox titration vat with automatic sample delivery to flow cuvette; regenerative stopped-flow mixing apparatus linked to fluorescence photometer; glow-curve photometer; numerous smaller items (flash lamps, power supplies, amplifiers, timers, interfacing modules, etc). Most recently we have developed an apparatus for computer aided fluorescence video imaging for use in assaying photosynthetic activity in intact leaves, and for screening for mutant bacterial strains on culture dishes. We have also constructed a number of portable field instruments, as detailed above.