@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Science, Faculty of"@en, "Microbiology and Immunology, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Tehrani, Ali"@en ; dcterms:issued "2009-12-02T00:00:00"@en, "2003"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description "The purple bacterial photosynthetic reaction centre (RC) contains three proteins called RC H, RC M and RC L. The RC H protein comprises three cellular domains: an 11 amino acid N-terminal sequence on the periplasmic side of the inner membrane; a single transmembrane ot-helix; a large C-terminal, globular cytoplasmic domain. The roles of these domains in Rhodobacter sphaeroides RC function and assembly was investigated, using a mutagenesis approach that included domain swapping with Blastochloris viridis RC H segments, periplasmic domain deletion and site-directed amino acid changes. Additionally, possible interactions between photosynthetic RC proteins that are thought to protect these membrane proteins from proteolytic digestion in RC complex assembly were evaluated by use of translationally in-frame (non-polar) RC gene-specific deletions. The RC H, RC M and RC L proteins were expressed from plasmids, either alone or in concert with one or both of the others, in a strain of R. sphaeroides that contained chromosomal deletions of all three RC genes. The RC H periplasmic domain was shown to be involved in the accumulation of the RC H protein in the cell membrane, while the transmembrane domain has an additional role in RC complex assembly, perhaps through interactions with RC M. The cytoplasmic domain functions in RC catalytic activity and complex assembly. The RC H cytoplasmic domain surface residues His-126 and His-128 jointly mediate proton transfer into the RC by acting as proton donors at the entrance of the proton transfer pathway. There is a correlation between the amounts of membrane-associated RC H and RC L, whereas RC M is found in the cell membrane independently of RC H and RC L. Furthermore, substantial amounts of RC M and RC L are found in the soluble fraction of cells only when RC H is present in the membrane. These data were used to propose models of RC catalytic mechanisms, and RC assembly in which the RC M protein accumulates in the cell membrane regardless of the presence of the RC H and RC L proteins, and the RC M protein is a nucleus for addition of RC L followed by RC H in assembly of the RC holocomplex."@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/16097?expand=metadata"@en ; dcterms:extent "7046687 bytes"@en ; dc:format "application/pdf"@en ; skos:note "Structure-Based Functional Studies of the Rhodobacter sphaeroides Reaction Centre H Protein by Ali Tehrani B.Sc, The University of Massachusetts, Amherst, 1994 M.Sc, The University of Massachusetts, Amherst, 1996 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Microbiology and Immunology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA November 2003 © Ali Tehrani, 2003 ABSTRACT The purple bacterial photosynthetic reaction centre (RC) contains three proteins called RC H , RC M and RC L. The RC H protein comprises three cellular domains: an 11 amino acid N-terminal sequence on the periplasmic side of the inner membrane; a single transmembrane ot-helix; a large C-terminal, globular cytoplasmic domain. The roles of these domains in Rhodobacter sphaeroides RC function and assembly was investigated, using a mutagenesis approach that included domain swapping with Blastochloris viridis RC H segments, periplasmic domain deletion and site-directed amino acid changes. Additionally, possible interactions between photosynthetic RC proteins that are thought to protect these membrane proteins from proteolytic digestion in RC complex assembly were evaluated by use of translationally in-frame (non-polar) RC gene-specific deletions. The RC H, RC M and RC L proteins were expressed from plasmids, either alone or in concert with one or both of the others, in a strain of R. sphaeroides that contained chromosomal deletions of all three RC genes. The RC H periplasmic domain was shown to be involved in the accumulation of the RC H protein in the cell membrane, while the transmembrane domain has an additional role in RC complex assembly, perhaps through interactions with RC M . The cytoplasmic domain functions in RC catalytic activity and complex assembly. The RC H cytoplasmic domain surface residues His-126 and His-128 jointly mediate proton transfer into the RC by acting as proton donors at the entrance of the proton transfer pathway. There is a correlation between the amounts of membrane-associated RC H and RC L , whereas RC M is found in the cell membrane independently of RC H and RC L. Furthermore, substantial amounts of RC M and RC L are found in the soluble fraction of cells only when RC H is present in the membrane. These data were used to propose models of RC catalytic mechanisms, and RC assembly in which the RC M protein accumulates in the cell membrane regardless of the presence of the RC H and RC L proteins, and the RC M protein is a nucleus for addition of RC L followed by RC H in assembly of the RC holocomplex. iv TABLE OF CONTENTS Abstract i i Table of contents iv List of Tables vii List of Figures viii Abbreviations x Acknowledgements xi 1. INTRODUCTION 1 1.1. Anoxygenic phototrophs 3 1.2. The anoxygenic purple non-sulfur phototrophic bacterium Rhodobacter sphaeroides 1 1.3. The photosynthetic apparatus 7 1.4. Organization and regulation of the R. sphaeroides photosynthesis genes 12 1.5. Overview of photosynthetic ATP synthesis in anoxygenic purple bacteria 14 1.6. A closer look at electron an proton transfer pathways in R. sphaeroides RC 16 1.7. The RC H protein function in photosynthesis and RC complex assembly 19 1.8. Thesis objectives and approach 19 2. MATERIALS AND METHODS 21 2.1. Bacterial strains and growth conditions 2.2. Plasmids utilized 23 2.2.1. Construction of plasmids used for RC H protein domain swap experiments 27 2.2.2. Construction of plasmids for site-specific mutagenesis of the R. sphaeroides RC H protein domains 33 2.2.3. Construction of the plasmids used in RC gene (protein) deletion studies 34 2.3. Construction of the R. sphaeroides APUHAAPUC strain 36 2.4. Construction of the R. sphaeroides A R C L H strain 36 2.5. In vitro D N A techniques 37 2.6. Bacterial conjugation 38 2.7. Absorption spectroscopy 39 2.8. Flash spectroscopy 39 2.9. Kinetics of RC formation 40 2.10. Electron transfer and proton uptake measurements 40 2.11. Chromatophore and soluble fraction preparation 40 2.12. SDS-Gel electrophoresis and western blotting 41 3. R E S U L T S 43 3.1. Analysis of RC H protein domain swaps on photosynthetic and RC assembly in R. sphaeroides 43 3.1.1. Summary 43 3.1.2. Development of the experimental system 43 3.1.3. Overview of the hybrid RC H proteins 46 3.1.4. Photosynthetic growth of LH2\" APUHAAPUC strains expressing hybrid RC H genes 48 3.1.5. Absorption spectroscopy of LH2\" APUHAAPUC strains expressing hybrid RC H genes 50 3.1.6. Photosynthetic growth and flash spectroscopy of APUHA strains expressing hybrid RC H genes 53 3.1.7. RC H , RC M , RC L proteins levels in membrane and soluble cell fractions 56 3.1.8. Kinetics of RC formation for the APUHAAPUC (pEVtrans) strain 59 3.2. Site specific mutagenesis and deletion of the R. sphaeroides RC H protein domains 61 3.2.1. Summary 61 3.2.2. Overview of the mutant RC proteins 61 3.2.3. Photosynthetic growth of APUHAAPUC strains expressing RC H periplasmic domain mutations 62 3.2.4. Absorption spectroscopy of LH2\" APUHAAPUC strains expressing RC H periplasmic domain mutations 63 3.2.5. Flash spectroscopy of APUHA strains expressing RC H periplasmic domain mutations 66 3.2.6. RC H, RC M , RC L proteins levels in membrane and soluble cell fractions of strains expressing RC H periplasmic domain mutations 67 3.2.7. Photosynthetic growth of APUHAAPUC strains expressing RC H cytoplasmic domain mutations 69 vi 3.2.1 Electron transfer and proton uptake measurements for APUHAAPUC strains expressing RC H cytoplasmic domain mutations 69 3.3 RC gene (protein) deletion studies 76 3.3.1 Summary 76 3.3.2 Overview of the R. sphaeroides A R C L H strain and the RC protein deletions 76 3.3.3 Photosynthetic growth of A R C L H strain expressing different combinations of RC genes 77 3.3.4 Absorption spectra of A R C L H strain expressing different combinations of RC genes 79 3.3.5 RC H, RC M , RC L proteins levels in membrane and soluble cell fractions 81 4. DISCUSSION 84 4.1. Functional analysis of R. sphaeroides RC H protein domains 84 4.2. The role of the RC H protein in R. sphaeroides RC complex assembly 93 4.3. RC-LH1 (core complex) interactions 98 4.4. Future directions 98 5. REFERENCES 101 APPENDIX 1 Amino acid alignment of R. sphaeroides and B. viridis RC H proteins 109 APPENDIX 2 Absorption spectra of APUHAAPUC (LH2\") intact cells containing plasmid pATSHR or pEVtrans 110 APPENDIX 3 Publications arising from this thesis research 111 LIST OF TABLES vii Table 2.1 Table 2.2 Table 3.1 Bacterial strains Plasmids used 22 23 Summary of RC levels and catalytic activities (for hybrid RC H proteins) measured in steady-state absorption spectra of intact LH2\" cells and flash spectroscopy of chromatophores obtained from L H 2 + cells 54 Table 3.2 Summary of RC levels and catalytic activities (for site-specific periplasmic mutations) measured in steady-state absorption spectra of intact LH2\" cells and flash spectroscopy of chromatophores obtained from L H 2 + cells. 66 viii LIST OF FIGURES Figure 1.1 Phylogenetic relationships among life forms Figure 1.2 Membrane system of purple bacterium R. sphaeroides Figure 1.3 Schematic representations of proposed models for the supermolecular organization o f the photosynthetic apparatus of R. sphaeroides Figure 1.4 Ribbon representation of the R. sphaeroides R C Figure 1.5 The R. sphaeroides photosynthesis gene cluster Figure 1.6 Schematic illustration of photosynthetic A T P synthesis in R. sphaeroides Figure 1.7 The catalytic photocycle bf quinone reduction in the R C 1 6 9 11 14 15 18 Figure 2.1 Genetic and restriction map of intermediate plasmid pTZ18U::puhA and pUC19::vpuhA Figure 2.2 Construction of the expression plasmid pATP19P Figure 2.3 Genetic and restriction maps of plasmid pATSHR 29 30 31 Figure 3.1 Absorption spectra of APUHAAPUC (LH2\") intact cells containing relevant plasmids (expressing hybrid RC H mutants) 45 Figure 3.2 Amino acid alignment of R. sphaeroides and B. viridis RC H N-terminal sequences, and a representation of hybrid RC H proteins used 47 Figure 3.3 Photosynthetic growth of APUHAAPUC cells containing relevant plasmids (expressing hybrid RC H mutants) 49 Figure 3.4 Absorption spectra of APUHAAPUC intact cells containing relevant plasmids (expressing hybrid RC H mutants) 52 Figure 3.5 Photosynthetic growth of APUHA cells containing relevant plasmids 55 ix Figure 3.6 Western blot involving hybrid RC H mutants probed with RC H , RC M and RC L antisera 58 Figure 3.7 Kinetics of RC formation 60 Figure 3.8 Photosynthetic growth of APUHAAPUC cells containing relevant plasmids (expressing site-specific mutants of RC H periplasmic domain) 64 Figure 3.9 Absorption spectra of APUHAAPUC intact cells containing relevant plasmids (expressing site-specific mutants of RC H periplasmic domain) 65 Figure 3.10 Western blot involving site-specific mutations of the RC H periplasmic domain probed with RC H , RC M and RC L antisera 68 Figure 3.11 Photosynthetic growth and absorption spectra of APUHAAPUC cells containing relevant plasmids (expressing site-specific mutants of RC H cytoplasmic domain) 70 Figure 3.12 Proton uptake and electron transfer (kAB1) measurements 72 Figure 3.13 Chemical representation of the amino acid histidine 73 Figure 3.14 Proton uptake and electron transfer (kAB2) measurements 74 Figure 3.15 Chemical rescue using imidazole 75 Figure 3.16 Photosynthetic growth of A R C L H cells containing relevant plasmids 78 Figure 3.17 Absorption spectra of A R C L H cells containing relevant plasmids 80 Figure 3.18 Western blot of A R C L H cells containing relevant plasmids 83 Figure 4.1 Proton path through RC H residues His-126 and His-128 91 Figure 4.2 Chemical structure of TRIS 93 Figure 4.3 Model for R. sphaeroides assembly 97 X ABBREVIATIONS Ap(Ap r ) ampicillin (or ampicillin resistant) ATP adenosine 5'-triphosphate ATPase adenosine 5'-triphosphatase bp base pair BChl bacteriochlorophyll cyt bei cytochrome be; oxidoreductase complex HRP horse radish peroxidase I C M intracytoplasmic membrane kb kilobases kDa kilodaltons Kn(Kn r ) kanamycin (or kanamycin resistant) K u Klett unit L H 1 light-harvesting 1 complex L H 2 light-harvesting 2 complex SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis puc operon encoding the of L H 2 complex proteins puf operon encoding RC M / L and L H 1 complex proteins puhA gene encoding the RC H protein RC reaction centre TBS Tris-buffered saline solution Tc (Tcr) tetracycline (or tetracycline resistant) Q A RC quinone A Q B RC quinone B ACKNOWLEDGEMENTS I am indebted to my supervisor, Tom Beatty, for allowing me to mature as a scientist. He is a teacher and a mentor as they should be. Without his support and patience (including \"the handshake\" after the accomplishment of a task), completing this thesis project and the resulting publications would have been most difficult. I would also like to thank the members of my committee: Drs. Hancock, Fernandez, and Mohn, for their guidance and input into this project. I am grateful to Drs. S. Kaplan, and E. Abresch for providing me with antisera, M . Paddock for purified R. sphaeroides RC, and R.C. Prince for flash spectroscopy experiments. Furthermore, I would like to express my gratitude to M . Paddock, R.C. Prince, and Pia Adelroth for all collaborative experiments. I thank Dr. John Nomellini and Ms. Shelley Small for all the help and support they have provided me over the years. Without a doubt my graduate experience would have not been complete i f it wasn't for my colleagues in the Beatty lab (both past and present). I would like to acknowledge and thank Yves Leduc and Steve Seredick for their friendship and many thought provoking discussions in our times at Wesbrook. Finally, I thank my parents (especially my mother), and Lisa for believing in me and staying patient, as I learn my ways. This thesis is dedicated to the memory of my grandmother whom forever will be with me. 1 1. INTRODUCTION Arguably one of the most important biological processes on earth, photosynthesis involves the capture and conversion of light energy into chemical energy (ATP). Within evolutionary schemes there are five distinct bacterial groups (Fig. 1.1), and chloroplasts in eukaryotes that are capable of synthesizing ATP through photosynthesis (40). Figure 1.1 Phylogenetic relationships among life forms based on large subunit rRNA sequences. Adapted from Woese, CR. et al. (71). Photosynthesis is restricted to the Eucarya (plants and microbes), and Bacteria. Photosynthetic bacteria are found within the cyanobacteria, purple bacteria, green sulfur bacteria, Gram-positive bacteria, and green non-sulfur bacteria. However, the generation of ATP alone does not fulfill all of the cellular energy requirements, as N A D P H is also needed. Cyanobacteria and chloroplasts use light energy for the transfer of electrons from H2O to N A D P + to obtain N A D P H . This oxidation of water molecules results in the production and release of molecular oxygen, 2 which categorizes these organisms as oxygenic. Anoxygenic phototrophic organisms produce N A D P H through the oxidation of organic or inorganic compounds. Nevertheless, the general principles of energy transduction are the same for anoxygenic and oxygenic phototrophs (25). In 1941 C B . van Niel described the overall chemistry of photosynthesis with the formulation below, wherein anoxygenic species H2A can for example be H2S, or an organic compound, whereas in oxygenic species H2A is water. C 0 2 + 2 H 2 A + light -» ( C H 2 0 ) + 2A Phototrophs that oxidize organic compounds to obtain N A D P H typically do not reduce very much C 0 2 , making this formulation not directly applicable (25, 40). In purple bacterial phototrophic growth, organic compounds can take the place of C 0 2 and H2A in the above formula. In this case, a revised two-part formula could be written as (the details of light energy conversion to ATP can be found in section 1.5): RC, cyt be 1 light > A T P ATPase organic compound > N A D P H + oxidized metabolic intermediates > amino acids, nucleotides, etc. Oxygenic phototrophs contain a variety of light-harvesting antennae that capture and transfer light energy to two types of pigment-protein complexes, both of which are called a reaction centre: 1) PSII designates the reaction centre where water is oxidized; 2) PSI designates the reaction centre where N A D P + is reduced (9). Anoxygenic phototrophs also contain a variety of light-harvesting systems, but each type of anoxygenic organism 3 1.1 Anoxygenic phototrophs The anoxygenic phototrophic group of bacteria include green sulfur bacteria, green non-sulfur bacteria, heliobacteria, and purple bacteria (Fig. 1.1). Despite many commonalities, these bacteria exhibit a number of differences in the organization of their photosynthetic apparatus, and metabolic characteristics. I provide below a brief summary of three of the four types of anoxygenic phototrophs, followed by a more in depth description of purple phototrophic bacteria, with a focus on Rhodobacter sphaeroides, the subject of this thesis. Green sulfur bacteria. These bacteria are anaerobes that obligately use reduced sulfur compounds as electron donors, while fixing CO2 as a carbon source (27). They primarily capture light though chlorosomes, which are specialized antenna structures found at the cytoplasmic side of the cell membrane. Green sulfur bacteria contain a reaction centre that is thought to be related to the PSI of oxygenic phototrophs (9, 27). Green non-sulfur bacteria. These bacteria are capable of growth under a variety of conditions that include photoautotrophic, photoheterotrophic, and aerobic respiration (58). The fhermophillic bacterium Chloroflexus is the most widely studied member of this group, and has been shown to contain a very similar antenna structure (chlorosomes) to green sulfur bacteria (58). In surprising contrast to the chlorosome similarities, green non-sulfur bacteria contain a reaction centre that is very similar to the purple bacterial reaction centre, and thought to be related to the PSII of oxygenic phototrophs (9). Heliobacteria. The only Gram-positive phototroph, which is also capable of forming spores, these bacteria are strict anaerobes that rapidly die in the presence of oxygen (41). Their reaction centre is thought to be related to the PSI of oxygenic 4 phototrophs (9). Given that they are the most recently discovered photosynthetic organism, not much is yet known about their mechanism of carbon assimilation or metabolic characteristics, other than they require an organic carbon source for growth (9). Purple bacteria. As the metabolically most versatile and easily cultivated of the anoxygenic phototrophs, these bacteria have been the subject of extensive structural and genomic studies (9). Purple phototrophic bacteria contain a reaction centre that is thought to be related to the oxygenic PSII reaction centre, and very similar to the reaction centre of green non-sulfur bacteria (20). On the basis of the ability to oxidize reduced sulfur compounds for CO2 fixation, purple bacteria are subdivided into sulfur and non-sulfur groups. Under highly aerobic conditions, when growth occurs by chemotrophic respiration, the cell membrane of most purple bacteria is morphologically similar to other Gram-negative species such as Escherichia coli (20, 40). However, a lowering of the oxygen partial pressure (typically to <1 kPa), results in the formation of invaginations in the cytoplasmic membrane, which is where the photosynthetic apparatus is found (Fig. 1.2) (20). This elaborate intracytoplasmic membrane system (ICM) is continuous with the plasma membrane and (depending on the species) has been shown to consist of either finger-like intrusions, vesicles, tubular structures, or relatively flat sheets known as lamellae (20). Studies on the purple non-sulfur bacterium Rhodobacter sphaeroides have provided a wealth of information on the fundamentals of subjects such as membrane biogenesis, electron and proton transfer reactions in the membrane, and properties of membrane proteins. The ability of this bacterium to grow under a variety of conditions (i.e. chemotrophically or phototrophically), such that mutations that abolish 5 photosynthetic growth are not lethal, makes this organism ideal for the study of bacterial photosynthesis (20, 36, 48, 51, 66, 74). 6 ICM Figure 1.2 Membrane system of purple bacterium R. sphaeroides. Left panel, electron micrograph of R. sphaeroides, depicting intracytoplasmic membrane invaginations (ICM). Right panel, schematic representation of the photosynthetic membrane of R. sphaeroides, with a closer look at an ICM (see Fig 1.6 for components of the ICM). Figures adapted from www.eawag.ch/research e/ and www.nottingham.ac.uk/~pdzres respectively. 7 1.2 The anoxygenic purple non-sulfur phototrophic bacterium Rhodobacter sphaeroides On the basis of 16S rRNA sequence, R. sphaeroides belongs to the a-subgroup of Proteobacteria, and forms almost spherical rod-shaped cells, which frequently occur in pairs or chains. Under anaerobic conditions, its characteristic yellowrbrown pigmentation is due to the photosynthetic pigments bacteriochlorophyll a, and the carotenoid spheroidene (27). Anaerobically and with light, R. sphaeroides grows phototrophically, although it can also carry out anaerobic respiration in the dark with the addition of a terminal electron acceptor such as dimethyl sulfoxide. Aerobically in the dark, R. sphaeroides utilizes a number of organic compounds such as malate as a carbon source for respiration (74). 1.3 The photosynthetic apparatus Embedded within the ICM, pigment protein complexes of the R. sphaeroides photosynthetic apparatus comprise the light-harvesting antennae called LH2 and LH1, and the reaction centre (RC) complex, which operates as a light-driven two-electron/two-proton quinone reductase (20, 51). X-ray crystallography of LH2 complexes revealed symmetrical pigment-protein ring structures (34, 43), which transfer light energy to the LH1 complex (66). The LH1 complex transfers light energy to the RC, and a variety of experiments indicate the existence of a \"core\" LH1/RC supercomplex (39). Currently, the exact organization of this supercomplex is the subject of extensive research and debate. Recently, Vermeglio and colleagues reported that in R. sphaeroides, LH1 partially encircles the RC, forming a crescent structure that may allow movement of 8 quinones between the RC and cytochrome bc\\ (Fig. 1.3A) (69). Contrary to this model, other research suggested that LH1 forms a closed circular structure around the RC, with LH2 complexes surrounding the monomeric RC/LH1 supercomplex (Fig. 1.3B) (14). Figure 1.3 Schematic representations o f proposed models for the supermolecular organization o f the photosynthetic apparatus o f R. sphaeroides. A Mode l depicting a pair o f R C s , each surrounded by crescent L H 1, associating with cytochrome bct. The core is surrounded by peripheral antenna complexes ( L H 2). A l so shown is the PufX protein (solid dot) located between R C and Cyto beh B Mode l depicting an R C surrounded by a closed-circle L H 1 complex. Although not shown here, in this model the PufX protein forms a gate in L H 1 for quinone exchange in the R C ((38). Figure adapted from H u et al. (73) 10 Light harvesting 2 complex. Classified as an accessory antenna complex, variable amounts of LH2 can be found in the membrane depending on growth conditions. Many species of purple bacteria do not contain a LH2 complex, and mutants of R. sphaeroides that lack LH2 are capable of photosynthetic growth (15). The basic structural unit of this complex consists of a heterodimer of two integral membrane polypeptides known as the a and the P subunits (encoded by pucA and pucB genes), which non-covalently bind three bacteriochlorophyll (BChl) a and one carotenoid molecule (14, 43). Following oligomerization, the subunits form a ring-like structure in which BChl a molecules are arranged as a ring of interactive dimers that absorb light at 850 nm, and monomers that absorb light at the 800 nm wavelength. X-ray diffraction studies have revealed that depending on the species, LH2 rings consist of oligomers of eight or nine a/p subunits (34, 43). A n (a/p)g ring has been demonstrated in a projection structure determined from 2-D crystals of the R. sphaeroides LH2 complex (see J. Mol . Biol. 282:833-845, 1998). Light harvesting 1 complex. Although similar to LH2, the LH1 core antenna complex is clearly distinct as it absorbs light at a different wavelength, aggregates into a larger complex, and LH1 is found in a fixed stoichiometric ratio to the RC. Structurally, this complex is made up of approximately 16 pairs of a/p polypeptide subunits, with each pair binding two BChl a molecules and two carotenoid pigments (20). The pufA and pufB genes encode the L H l a and P protein subunits (26). Unlike LH2, the oligomerized BChl a dimers in LH1 absorb light at the 870 nm wavelength, and transfer the resultant excitation energy directly to the RC. It is not clear how the different oligomerization and/or protein environments of the LH1 and LH2 BChl dimers result in their different wavelength absorption properties. In 2000, Hunter and colleagues reported yet another 11 major difference between L H 1 and L H 2 , in that they found the P subunit in L H 1 to be more bent, which would direct its N-terminus toward the enclosed R C (16). Reaction centre. A widely studied integral membrane pigment-protein complex, the R C of R. sphaeroides is composed of three protein subunits named R C H , R C M and R C L , and nine cofactors (Fig. 1.4) (23). Certain purple bacterial species such as Blastochloris viridis, contain a cytochrome c subunit as the fourth protein component of theRC(18) . Figure 1.4 Ribbon representation of the R. sphaeroides RC. The RC complex is comprised of three proteins subunits (RC H, RC M , and RC L ) , and nine cofactor molecules. Figure adapted from: http://physics.ucsd.edu/~raifeher/rc_ribn.html 12 High-resolution crystal structures of the RCs of B. viridis and R. sphaeroides revealed that RC M and RC L each have five a-helical hydrophobic transmembrane segments, while RC H contains only one such transmembrane segment (36). The RC H protein does not interact directly with cofactors, and can be conceptually divided into three domains: 1) an ~11 amino acid N-terminal domain located on the periplasmic side of the inner membrane of these Gram-negative cells; 2) a single transmembrane a-helix; 3) a C-terminal, large, globular cytoplasmic domain (36). The RC M and RC L proteins, which are homologous and have a pseudo-two-fold axis of symmetry, provide a scaffolding within which the RC associated cofactors are arranged in two symmetrical branches commonly referred to as the A and the B branches. These cofactors are: two BChl a which form a strongly interacting dimer known as the \"special pair\" (sometimes called D), two accessory BChl a which are in close proximity 2_|_ to the special pair, two bacteripheophytins, one non-heme iron atom (Fe ), and a pair of quinones ( Q A , Q B ) (4, 11). 1.4 Organization and regulation of the R. sphaeroides photosynthesis genes. In R. sphaeroides and other purple bacteria, the genetic information required for pigment biosynthesis and the production of photosynthetic pigment-binding proteins is contained within a cluster of genes, commonly referred to as the photosynthesis gene cluster (40,46). The arrangement of the R. sphaeroides photosynthesis gene cluster is shown in Figure 1.5. The a and B polypeptides of the LH1 complex, along with the RC L and RC M protein subunits, are encoded within thepuf operon (26). Relative to thepuf operon, the puhA gene, which encodes the RC H protein, is located at the opposite end of the cluster. The LH2 complex is not essential for photosynthesis and the genes encoding the a and (3 polypeptides of this complex are found within the puc operon (40, 46). This operon is not part of the photosynthesis gene cluster, and is located ~20 kb 3 of the puhA gene. The expression of photosynthesis genes in R. sphaeroides is controlled through the function of several transcriptional regulatory circuits, which are affected by changes in cellular oxygen tension, as well as changes in light intensity (7, 49). Generally, under high oxygen conditions the expression of the photosynthetic apparatus is repressed due to the binding of a repressor protein (PpsR) to a conserved palindrome sequence on the chromosome (48, 75). Additionally, the expression of the RC and L H structural proteins is regulated by a two-component signal transduction system (RegBA), which up-regulates the expression of the puf, puc, and puhA genes in response to a drop in oxygen partial pressure (22, 47, 62). Although not fully understood, the expression of the photosynthesis genes is also affected by light intensity through the AppA protein. Recently Bauer et al. proposed a model in which AppA acts as a transcription factor that controls light repression of photosynthesis gene expression, by mediating the binding of D N A by PpsR. That is, under low O2 and low (blue) light conditions, AppA inhibits D N A binding by PpsR, hence allowing the expression of photosynthesis genes. Under low light conditions, the transcription of puc, puf, and the puhA genes is enhanced, while high light has been reported to repress puc operon gene transcription (42). 14 The above brief summary neglects many details of the fervid and sometimes controversial field of purple bacterial photosynthesis gene expression. The interested reader is directed to recent reviews (6, 29). ( rn \\ CTA) codons of plasmid pTZ18U::puhA, to yield the intermediate plasmids pUC19::xbvir and pTZ18U::xbsph, respectively (both plasmids also contain Xba I sites 5' of their respective RC H genes). After an Xba I digest the 845-bp D N A fragment from plasmid pTZ18U::xbsph was ligated to a 2.7-kb fragment from the plasmid pUC19::xbvir, resulting in plasmid pVtip. The hybrid RC H gene in plasmid pVtip was 28 subcloned as a 900-bp BamR I fragment into the expression vector pATP19P to yield pEVtip. 29 EcoRI Figure 2.1 Genetic and restriction map of intermediate plasmid. A Plasmid pTZ18U::puhA. The R. sphaeroides puhA gene is represented as a filled labelled arrow. This puhA gene was subcloned as a 1.34-kb BamR I DNA fragment, with the start codon (ATG) being at the 474 t h base pair. This plasmid is ampicillin resistant (Ap1). B Plasmid pUC19::vpuhA. The B. viridispuhA gene is represented as filled labelled arrow. This puhA gene was subcloned as a ~1.5-kb EcoR I DNA fragment, with the stop codon being at the 820 t h base pair. This plasmid is ampicillin resistant (Ap r). 30 puc operon ^ promoter i r - 1 — I — i — 1 ' 1 Hind III I Pst I | xbal I Hind III p Hinc UJSal I Sma I A ~ 0.7-kb Hind III fragment containing the R. sphaeroides puc operon promoter was sub-cloned from plasmid pUI612 (37), into plasmid pRK415, yielding plasmid pATP19P rMCS Sma I Figure 2.2 Construction of the expression plasmid pATP19P. Relevant restrictions sites are provided. The bent arrows indicate the location of the R. sphaeroides puc operon promoter, which was used to transcribe all hybrid and mutant puhA genes in this thesis (37). Filled arrow in pATP19P represents tetracycline resistance. 31 Smal Stul Figure 2.3 Genetic and restriction maps of plasmid pATSHR. See sections 2.2.1 and 2.2.3 for construction and use. Bent arrows indicate the puc operon promoter, which drives the transcription of the R. sphaeroides puhA gene. 32 Plasmid pEVtrans encodes the R. sphaeroides periplasmic domain from Met-1 to Asp-11, followed by the B. viridis transmembrane domain from Ile-12 to Arg-34 (Fig. 3.2A), and the R. sphaeroides cytoplasmic domain from Glu-34 onward. This plasmid was created by initially introducing Dde I sites at the location corresponding to the junction between the transmembrane and cytoplasmic domains of both the RC H genes in plasmids pUC19::vpuhA and pTZ18U::puhA. This was done by changing the R. sphaeroides Thr-33 codon (ACC -> ACT) yielding plasmid pTZ18U::mspuhA, and the B. viridis Arg-34 codon (CGT CCT) to yield plasmid pUC19::mvpuhA. A 3.6-kb Hind III (filled in with T4 D N A polymerase) to Dde I fragment containing the D N A sequence of the R. sphaeroides cytoplasmic domain was obtained from pTZ18U::mspuhA, and ligated to a 240-bp Pvu II to Dde I fragment (encoding the B. viridis periplasmic and transmembrane domains) from pUC19::mvpuhA to yield plasmid pVPSCA. The D N A sequence of plasmid pVPSCA RC H codon 34 was changed from CCT back to CGT (restoring Arg-34), creating pVPSCB. EcoR V sites were introduced at the B. viridis RC H Asn-11 codon (GAC -> GAT) in pVPSCB, to yield pVPSCB2, and at the R. sphaeroides RC H Leu-12 codon (CTG ATC) in pTZ18U::puhA, yielding pTZl 8U: :puhA2.1. The 800-bp EcoR V to Hind III fragment of pVPSCB2 was ligated to the 3.2-kb fragment of plasmid pTZ18U::puhA2.1, yielding pVtrans. A 1.4-kb BamH I fragment was subcloned from pVtrans into pATP19P to yield pEVtrans. To obtain plasmid pEVPSCB, which contains a RC H gene encoding the B. viridis periplasmic and transmembrane domains (Met-1 to Arg-34; see Fig. 3.2A) followed by R. sphaeroides sequences from Glu-34 onward, the 1.4-kb EcoR I fragment from plasmid pVPSCB was sub-cloned into the expression vector pATP19P. Plasmid pEVCSP encodes the R. sphaeroides RC H periplasmic and transmembrane domains (amino acids Met-1 to Glu-34; see Fig. 3.2A) followed by the B. viridis RC H cytoplasmic domain (from Asp-36 onward). Plasmid pEVCSP was constructed by ligation of a 3.4-kb Kpn I (resected with T4 D N A polymerase) to Dde I fragment from pTZ18U::mspuhA to a 820-bp Pvu II to Dde I D N A fragment from pUC19::mvpuhA, to yield plasmid pVCSP. The 1.4-kb BamR I fragment from plasmid pVCSP was subcloned into pATP19P to yield pEVCSP. 2.2.2 Construction of plasmids for site-directed mutagenesis of the R. sphaeroides RC H protein domains Site-directed sequence changes were introduced by use of the QuikChange kit (Stratagene), and all mutations were D N A sequenced. Plasmid pEAPeri encodes a mutant R. sphaeroides RC H protein in which Val-2 is followed by Asp-11 (codons 3 through 10 of the RC H periplasmic domain were removed; see Fig. 3.2A). Plasmid pEAPeri was made by first introducing two Hinc II restriction sites, one at the Gly-3 codon (GGT -> A A C ) and the other at the Phe-10/Asp-11 codons (TTC/GAT to GTC/GAC) in plasmid pTZ18U::puhA, resulting in plasmid pTemp2. After Hinc II digestion, the D N A segment encoding Gly-3 to Phe-10 was removed and the plasmid re-ligated to yield plasmid pAPeri. A 1.3-kb BamR I fragment from pAPeri encoding the mutant RC H gene was subcloned into the expression vector pATP19P (see section 2.2.1 for details on this plasmid) to obtain pEAPeri. Mutant RC H genes in plasmids pStiplMT (Thr-5 [ACT] -> Val [GTT]) and pSt iplMA (Asn-9 [AAC] -> Leu [CTC]) were generated in pTZ18U::puhA (see Fig 34 3.2A). Plasmid pStip2M was similarly generated by simultaneous replacement of Thr-5 and Asn-9 with Val and Leu codons, respectively (Fig. 3.2A). These mutant RC H genes were transferred to pATP19P as BamR I fragments. Mutant RC H genes in plasmids pH126A (His-126 [CAC] -» Ala [GCG]) and pH128A (His-128 [CAC] -» Ala [GCG]) were generated in pTZ18U::puhA. These mutant RC H genes were transferred to pATP19P as BamR I fragments yielding plasmids pEH126A and pEH 128A respectively. 2.2.3 Construction of the plasmids used in RC gene (protein) deletion studies Site-directed sequence changes were introduced by use of the QuikChange kit (Stratagene), and all mutations were D N A sequenced. Plasmids pEPUF and pESHPUF were constructed by sub-cloning of a 4.57-kb EcoR I fragment, encoding the R. sphaeroidespufQBALMX genes (Fig. 2.4A) into plasmids pATP19P and pATSHR (see Materials and Methods section 2.2.1 for details on these plasmids) respectively. Thus pEPUF contains only the puf operon, whereas pESHPUF contains the puf operon and the puhA (RC H) gene. Plasmid pESHPUFL\" was constructed by insertion of a 4.2-kb EcoR I D N A fragment (Fig. 2.4B), which contains a translationally in-frame deletion of the RC L (puf L) gene that removed 353 bp between Kpn I (resected with T4 D N A polymerase) and BsaB I sites, into pATSHR. This plasmid also contains the puhA gene. Plasmid pESHPUFM\" was constructed by insertion of a 3.8-kb EcoR I fragment (Fig. 2.4C), which contains a translationally in-frame deletion of the RC M ipufM) gene that removed 767 bp between Rsr II and BstXl sites that were joined by ligation with the 35 linkers 5 ' -GACCGGCCGACGATATCATCTGG-3 ' and 3 ' -GCCGGCTGCTATAGTA-5', into pATSHR. This plasmid also contains the RC H (puhA) gene. Plasmid pESHPUFLM\" was constructed by insertion of a 3.45-kb EcoR I fragment, which contains the combined deletions of the RC M (pufM) and RC (pu/L) genes (as described above), in pATSHR. This plasmid also contains the RC H (puhA) gene. 36 2.3 Construction of the if. sphaeroides APUHAAPUC strain The R. sphaeroides RC\" L H 1 + LH2\" mutant strain APUHAAPUC was constructed through the chromosomal deletion of the pucBA genes (which express the LH2 a and (3 proteins) of the RC\" APUHA strain which contains a translationally in-frame (non-polar) deletion of the puhA gene (RC H protein) (12), using the suicide plasmid pSUP203 (which encodes resistance to ampicillin, tetracycline and chloramphenicol (63)) containing a 3.2-kb Pst I D N A fragment (705 bp of the pucBA genes were replaced with the neo gene (37)). The suicide plasmid pSUP203::705BA was conjugated into the APUHA strain, and heterogenotes arising from a single homologous recombination event were selected on the basis of kanamycin resistance. Clones were grown aerobically in L B medium containing kanamycin for 48 hrs, after which various dilutions were spread onto L B plates containing kanamycin. Clones arising from a second homologous recombination event were selected on the basis of kanamycin resistance, and tetracycline sensitivity. The chromosomal D N A of the resultant colonies was purified and screened using PCR, and Southern blotting to identify the desired mutant strain. 2.4 Construction of the R. sphaeroides ARCLH strain The R. sphaeroides RC\" LH1\" LH2\" mutant strain A R C L H was constructed by step-wise introduction of chromosomal deletions into the puf (LH1 and RC M and RC L genes) and puc (LH2 genes) operons of the RC H\" APUHA strain which contains a translationally in-frame (non-polar) of the puhA gene (RC H protein) (12), using the suicide plasmid p N G H l (which encodes resistance to kanamycin and contains the sacB gene for sucrose counterselection (50)) containing disrupted cloned D N A sequences for 37 the regions of interest. The chromosomal puf operon deletion utilized a cloned 4.57-kb Pst I fragment (which was engineered to contain EcoR I sites adjacent to the native Pst I sites) to replace the BspE I to Bel I intervening sequences with the linkers 5'-C C G G A G G A T A G C A T G A T C A T T - 3 ' and 3' - T C C T A T C G T A C T A G T A A C T A G - 5 ' , yielding a puf deletion extending from 13 bp 5' of the pufB start codon to 19 bp 3' of the pufX stop codon. The chromosomal puc operon deletion utilized a cloned 3.2-kb Pst I D N A fragment (705 bp of the pucBA genes were replaced with the neo gene (37)) that was digested with Kpn I and self-ligated, resulting in the removal of the neo gene. Suicide plasmids containing the deleted puf and puc genes (encoded in plasmids pNGH 1: :PUFDEL and pNGHl : :DELPUC respectively) were conjugated into the APUHA strain (12), and heterogenotes arising from a single homologous recombination event were selected on the basis of kanamycin resistance on solid R C V media aerobically in darkness. Clones were grown aerobically in R C V minimal medium supplemented with 15 pg/ml of biotin and lOmg/ml niacin for seven days, after which 20 pi were plated on this R C V medium supplemented with 15% (w/v) sucrose, 1% (w/v) Bactotryptone, 0.5% (w/v) yeast extract, 0.5% (w/v) NaCI and 0.5% (v/v) dimethylsulfoxide (50). Plates were incubated under anaerobic dark conditions for 10 days, after which the chromosomal D N A of the resultant colonies was purified and screened using PCR to identify the desired mutant strains. 2.5 In vitro D N A techniques Restriction endonuclease digestion, D N A ligation, agarose gel electrophoresis, transformation of E. coli and other recombinant D N A procedures were carried out as 38 described (60). E. coli was transformed using the CaCb competent cell transformation procedure (60), or through electro-transformation, using a Gene Pulser apparatus according to the manufacturer's instruction manual (Bio-Rad Laboratories, Richmond, CA). Plasmid D N A was routinely isolated from E. coli or R. sphaeroides cultures using the QIAprep® Spin Miniprep Kit (QIAGEN Inc., Chatsworth, CA). D N A was purified from agarose gel slices by using the QIAEX® II Gel extraction kit (QIAGEN Inc., Chatsworth, CA). When necessary, D N A was purified from PCR reactions using the QIAquick® PCR Purification kit. D N A sequencing was performed by either the U B C NAPS facility, or by Sadowsky laboratory (Department of Biochemistry & Molecular Biology, UBC) using purified plasmid templates. 2.6 Bacterial conjugation The conjugation of plasmid D N A into R. sphaeroides strains was accomplished by use of E. coli DH10B donor strain, and the E. coli helper strain HB101(pRK2013) in tri-parental conjugations. Sometimes, E. coli strain SI7-1 was used instead. Recipient and donor cultures were grown to stationary phase, mixed at a 4:1 recipient: donor ratio (typically 400 p i : 100 pi), and pelleted (30 seconds, 15,000 x g in a benchtop microcentrifuge), and resuspended in 50 pi of R C V medium. A 10 p.1 portion of the suspension was spotted onto a R C V plate, and once the spot had dried, the plate was incubated at 30 °C overnight. Following the incubation period, a portion of the spot was resuspended in 1 ml R C V medium, and 100-350 pi were spread onto R C V plates with appropriate antibiotic(s). These plates were incubated at 30 °C, typically for 3-5 39 days, and R. sphaeroides exconjugants were purified from E. coli by spreading a single colony onto an L B plate containing appropriate antibiotic(s). 2.7 Absorption spectroscopy of intact cells Cells of semi-aerobic cultures grown to -150 Klett units (-72 hours) were pelleted, resuspended in 0.25 ml of R C V medium, and mixed with 0.75 ml of a 30% bovine serum albumin solution to reduce light-scattering. Spectral data were obtained on a TIDAS II spectrophotometer (World Precision Instruments) and analyzed using the J & M Spectralys program, version 1.82. Samples were scanned for light absorption over light wavelengths ranging from 350 nm to 1000 nm. Spectra were adjusted to the same 650 nm value (A=0.2), due solely to light-scattering of suspended cells, to compensate for minor differences in concentrations. 2.8 Flash spectroscopy Measurements were performed by my collaborator, Roger Prince at Exxon Mobile Research and Engineering Laboratories, using a double-beam spectrophotometer, as previously described (59). Chromatophore suspensions ( A 8 7 0 = 2.0) were reduced with 5 m M sodium ascorbate and allowed to equilibrate in the dark for at least 30 minutes before use, poising the ambient redox potential around E h = +150 mV. The carotenoid bandshift (28) was measured at 490 minus 475 nm after a single actinic flash (10 ps full width at half height, filtered through a Wratten 88 A filter). The RC concentration was determined by the absorbance change at 605 minus 540 nm after a train of eight flashes separated by 32 ms, in the presence of 4 uM valinomycin and 1 u M antimycin (21). 40 2.9 Kinetics of RC formation Cultures of each strain were grown under high aeration (100 ml in 500 ml flasks shaken at 300 rpm) and then used as inocula for 800 ml cultures in 1 L flasks grown under low aeration. Samples of 1.5 ml were removed at time points following the switch to semi-aerobic growth and analyzed using absorption spectroscopy as outlined above. The amounts of RC complex were determined by calculating the area underneath the corresponding 804 nm peak, and plotted as a function of time. 2.10 Electron transfer and proton uptake measurements Electron transfer and proton uptake measurements were performed by my collaborators, Pia Adelroth and Mark L. Paddock at the University of California San Diego, as previously described in (2, 56). 2.11 Chromatophore and soluble fraction preparations Cells from semi-aerobic R. sphaeroides cultures (or B. viridis grown photosynthetically) were pelleted, resuspended in 50 mM Tris-HCI buffer (pH 8.0), disrupted using a French pressure cell, and centrifuged at 25,800 x g for 10 min to obtain a supernatant liquid containing soluble proteins and suspended membrane vesicles (chromatophores); 3 ml of this supernatant were centrifuged for 14 min at 412,000 g, to pellet chromatophores; the supernatant liquid was collected and the chromatophores were resuspended in 3 ml of 50 mM Tris-HCI buffer, and the two fractions were centrifuged a second time for 14 min at 412,000 g; the chromatophore pellet was resuspended in 3 ml 41 of 50 mM Tris-HCI buffer to obtain the membrane fraction, and the top 0.35 ml of the supernatant sample were collected to obtain the soluble fraction. The B. viridis RC was purified as described (45), and the R. sphaeroides purified RC sample was a generous gift from M . Paddock (University of California, San Diego). Samples were stored on ice for flash spectroscopy or at -80 °C for western blots. 2.12 SDS-Gel electrophoresis and western blotting A glycine-SDS polyacrylamide gel (SDS-PAGE) system was used for electrophoresis of purified chromatophores and soluble fractions (61). Chromatophores (10 pg of protein as determined by a modified Lowry method (57)) were mixed with SDS-PAGE sample buffer (35), and heated at 95 °C for exactly 1 minute. Volumes of the soluble proteins proportional to the chromatophores in unfractionated extracts, and empirically determined amounts of pure RCs, were treated the same way. Samples were run on SDS-PAGE (12% acrylamide) at 100 volts for approximately 75 minutes. Gels were either stained using a Coomassie blue solution, or transferred onto PROTRAN pure nitrocellulose membranes (Schleicher & Schuell, NH) for approximately 90 minutes at 100 volts. The RC H, RC M , or RC L proteins were detected, all as recommended in the E C L western blotting kit (Amersham Biosciences). The rabbit antisera raised against purified R. sphaeroides RC proteins were gifts, with the anti-RC H serum kindly provided by S. Kaplan (University of Texas) and the anti-RC M / L sera provided by E. Abresch (University of California, San Diego). The primary antibody was used at a 1:10,000 dilution, while the secondary antibody was used at 1:3000 dilution. Routinely, the incubation with the primary antibody was carried out overnight at 4 °C, whereas the 42 incubation with the secondary (HRP-labelled; see Abbreviations) was for 1 hour at room temperature. When necessary, the membranes were stripped of primary and secondary antibodies, through multiple (2 to 5) 10 minute washes with a Tris-buffered saline (TBS) solution (pH 2.0), followed by a single 15 minute incubation using a TBS-Tween (0.1% Tween-20™) solution (pH 7.6). 43 3. RESULTS 3.1 Studies of RC H protein domain swaps on photosynthesis and RC assembly in R. sphaeroides 3.1.1 Summary Initial in vivo experiments indicated that the RC H protein of B. viridis does not functionally substitute for the R. sphaeroides RC H protein. In attempts to identify specific regions of RC H that were responsible for this species specificity, the three domains of the R. sphaeroides RC H protein were replaced with the corresponding segments from the B. viridis protein. The effects of the mutations were evaluated in terms of photosynthetic growth, steady-state absorption spectra, light-driven electron and proton transfer reactions, western blot analyses of RC proteins in cellular membrane and soluble fractions, and kinetics of RC formation. 3.1.2 Development of the experimental system Plasmid pATVHR was used to express the B. viridis RC H gene in R. sphaeroides APUHA (LH2 +) and APUHAAPUC (LH2) strains, which contain a translationally in-frame chromosomal deletion of the RC H gene (see (12); and Material and Methods section 2.3). The complemented strains were incapable of photosynthetic growth, although expression of the B. viridis RC H gene was indicated by absorption spectra of intact cells. For example, although the 760 and 804 nm RC peaks were absent from APUHAAPUC (pATVHR) cells (Fig 3.1 A.), the LH1 870 nm peak was greater than that 44 of the strain that contained the expression plasmid pATP19P (which lacks a RC H gene; Fig. 2.2), and the broad region of absorbance centred at 770 nm indicates protein-free BChl and/or degradation products. The production of the B. viridis RC H protein in R. sphaeroides was also evaluated in western blot experiments, using the R. sphaeroides RC H antiserum (Fig. 3.IB). This antiserum reacted weakly with the RC H protein in purified B. viridis RC or chromatophore samples, but a band that was absent from the strain that contained pATP19P was obtained using chromatophores from the APUHAAPUC (pATVHR) strain. I attribute this band to the B. viridis RC H protein, which migrated to a slightly lower M W position than in the B. viridis samples, and so perhaps the B. viridis RC H protein is proteolytically cleaved when present in R. sphaeroides. The expression of the B. viridis RC H gene in R. sphaeroides, coupled with the absence of a functional RC complex, provided the basis of a system that I used to exchange segments of these two structurally well-characterized RC H proteins, to investigate RC H protein domain-specific functions (Fig 3.2B). These experiments on hybrid RC H proteins are described in the following section. 45 B 1 2 3 4 5 6 Figure 3.1 A Absorption spectra of APUHAAPUC (LH2\") intact cells containing plasmids: top, pATSHR (R. sphaeroides RC H gene); middle, p A T V H R (B. viridis RC H gene); bottom, pATP19P (expression vector lacking a RC H gene). Arrows point tothe 804 nm RC, and 870 nm L H 1 peaks. The vertical bar indicates A=0.15. B Western blot probed with R. sphaeroides RC H antiserum. Samples used were: lane 1, purified R. sphaeroides RC (control); lane 2, chromatophores from APUHAAPUC(pATSHR); lane 3, chromatophores from APUHAAPUC(pATP19P); lane 4, purified B. viridis RC; lane 5, chromatophores from B. viridis; lane 6, chromatophores from APUHAAPUC(pATVHR). 46 3.1.3 Overview of the hybrid RC H proteins To evaluate the function of the RC H protein periplasmic domain, a hybrid puhA (RC H) gene was constructed in plasmid pEVtip that encodes the B. viridis periplasmic amino acids in place of the corresponding R. sphaeroides sequence (see shaded residues in Fig. 3.2A). The RC H protein transmembrane domain was studied using a hybrid puhA gene constructed in plasmid pEVtrans, which encodes the B. viridis transmembrane domain (see underlined residues in Fig. 3.2A) flanked by the R. sphaeroides periplasmic and cytoplasmic domains. The resultant hybrid RC H protein contains the N-terminal 11 amino acids of R. sphaeroides followed by 23 B. viridis amino acids (Ile-12 to Arg-34) in place of the 22 R. sphaeroides amino acids (Leu-12 to Thr-33) that include the transmembrane a helices. This substitution is followed by the remaining 227 R. sphaeroides C-terminal amino acids that form the bulk of the R. sphaeroides protein in a globular, cytoplasmic domain (36). The periplasmic and transmembrane domain substitutions described above were combined in plasmid pEVPSCB (Fig 3.2B) to study the effects of replacing these two domains simultaneously, as opposed to individually. Thus, pEVPSCB encodes 34 B. viridis residues from the N-terminal Met to Arg-34 in place of the 33 N-terminal R. sphaeroides residues. The effects of a RC H protein cytoplasmic domain substitution were studied using a hybrid RC H gene constructed in plasmid pEVCSP (Fig 3.2B), which encodes B. viridis cytoplasmic residues (starting at Glu-35) in place of the R. sphaeroides sequence that begins with Glu-34. 47 A R.sphaeroides MVGVTAFGNFD' L A S L A I Y S F W I F L A G L I Y Y LQTENMR 3 7 B.viridis MYHGALAQKLD IAQLVWYAQ W L V I W T W L L YLRREDR 37 B p T \\ \\ [ C 0= j pATSHR wild type pEVtip pEVtrans pEVPSCB pEVCSP Figure 3.2 A Amino acid alignment of R. sphaeroides and B. viridis RC H N -terminal sequences. Shaded area indicates periplasmic amino acids, and underlined residues form an a-helix that spans the membrane. Identical amino acids are in bold, and the number 37 refers to the conserved R residue. Asterisks indicate R. sphaeroides residues changed by site-directed mutagenesis (see section 3.2). B Representation of the RC H proteins (with respective plasmid names provided below each) used in this study. P = periplasmic domain; T = transmembrane domain; C = cytoplasmic domain. Thick lines indicate B. viridis sequences. 48 3.1.4 Photosynthetic growth of LH2\" APUHAAPUC strains expressing hybrid RC H genes The R. sphaeroides strain APUHAAPUC was used in these experiments to allow comparison of growth properties with the RC and LH1 levels measured in steady-state absorption spectra (see below). Cultures grown under semi-aerobic dark conditions (to induce the production of photosynthetic pigment-protein complexes) were used to inoculate media that were incubated under anaerobic illuminated (photosynthetic) conditions, with growth monitored over time. Although host cells containing pEVtip were photosynthetically incompetent, the replacement of the transmembrane domain (pEVtrans) supported photosynthetic growth, although to a lower final cell density than the wild type RC H-complemented strain (Fig. 3.3). The strain APUHAAPUC (pEVPSCB), which encodes a RC H protein consisting of both the periplasmic and transmembrane domains of B. viridis RC H in place of R. sphaeroides sequences, showed weak photosynthetic growth (Fig. 3.3). This growth appeared to be intermediate between the growth obtained when only the transmembrane domain was exchanged and the absence of photosynthetic growth when only the periplasmic domain was swapped. When the R. sphaeroides RC H cytoplasmic domain was replaced with the counterpart from B. viridis in pEVCSP, the resultant strain was incapable of photosynthetic growth (Fig. 3.3) The poor photosynthetic growth of APUHAAPUC strains complemented with mutant RC H genes could have been due to either low levels of the RC formed, or to weak catalytic activity of cells replete in a hybrid RC. 49 1000 n 10 4 , , 0 50 100 Time (hrs) Figure 3.3 Photosynthetic growth of R. sphaeroides APUHAAPUC (LH2\") cells containing plasmids: pATSHR (wild type RC H, open circles); pEVtip (hybrid RC H containing B. viridis periplasmic domain, open triangles); pEVtrans (hybrid RC H containing B. viridis transmembrane domain, filled triangles); pEVPSCB (hybrid RC H containing B. viridis periplasmic and transmembrane domains, crosses); pEVCSP (hybrid RC H containing B. viridis cytoplasmic domain, asterisks). Ku, Klett units. 50 3.1.5 Absorption spectroscopy of LH2\" APUHAAPUC strains expressing hybrid RC H genes To gain a better understanding of the reasons for the different photosynthetic growth properties of the strains described above, the absorption spectra of intact cells were recorded after semi-aerobic growth to induce the synthesis of pigment-protein complexes. The R. sphaeroides (LH2\") APUHAAPUC strain was chosen because it is deficient in the production of LH2, thus revealing the positions and amplitudes of RC and LH1 peaks that are normally masked by the LH2 peaks. The primary data are presented in Figure 3.4, where the cellular content of the RC is indicated by the 804 nm peak amplitude, and are summarized in Table 3.1. These steady-state absorption spectra were normalized using the 650 nm value, which is due solely to light-scattering by suspended cells. Exchange of the periplasmic domain (pEVtip) reduced the amount of RC to 21% of the wild type amount. The substitution of the transmembrane domain (encoded by pEVtrans) reduced the amount of RC to 65% of the wild type, and when the periplasmic and the transmembrane domains were simultaneously replaced (pEVPSCB), there was a reduction in the amount of RC to 66% of the wild type. The substitution of the cytoplasmic domain (pEVCSP) resulted in a very low amount of the RC, too little for quantification using this method. The amplitude of the 870 nm LH1 peak was slightly reduced in the pEVtrans-containing cells, and slightly increased in the strain that contained pEVPSCB (Fig. 3.4). Although the amounts of RC indicated by the 804 nm peak area roughly correlate with the photosynthetic growth properties of strains lacking the LH2 complex, it was conceivable that the photosynthetic growth of LH2 strains that contained these mutant RC H proteins would be different, and that the 804 nm peak area does not necessarily indicate the amount of catalytically active RC. Therefore I turned to an L H 2 + background to evaluate photosynthetic growth, and used flash spectroscopy to measure the amount of RC and light-driven RC catalytic activity. 52 pATSHR pATP19P pEVPSCB pEVCSP 700 800 900 nm Figure 3.4 Absorption spectra of APUHAAPUC (LH2~) intact cells containing plasmids: pATSHR (R. sphaeroides RC H); pATP19P (expression vector); pEVtip (hybrid RC H containing B. viridis periplasmic domain); pEVtrans (hybrid RC H containing B. viridis transmembrane domain); pEVPSCB (hybrid RC H containing B. viridis periplasmic and transmembrane domains); pEVCSP (hybrid RC H containing B. viridis cytoplasmic domain). The vertical bar indicates A=0.15. 53 3.1.6 Photosynthetic growth and flash spectroscopy of A P U H A strains expressing hybrid R C H genes The relative photosynthetic growth of the RC H mutants in the L H 2 + background was similar to the growth of the LH2\" strains, except that the periplasmic domain substitution (pEVtip) supported the growth of L H 2 + cells (Fig. 3.5). For flash spectroscopy, cultures were grown semiaerobically, cells disrupted and membrane vesicles (chromatophores) were isolated. The percentage amount of the RC complex in mutants relative to the wild type as indicated by photobleaching (values were normalized to total BChl content) are given in Table 3.1 . This photobleaching technique measured the oxidation of the RC \"special pair\" (D) (see Materials and Methods). The values obtained are consistent with the photosynthetic growth properties and the steady-state absorption spectra of intact LH2\" cells (see above). The carotenoid bandshift of chromatophores is due to changes in the absorption maxima of carotenoids when an electrochemical potential forms across the chromatophore membrane (28). In our experiments this phenomenon results from light-dependent electron and proton transfer reactions of the RC and cytochrome b/ci complexes (38). The carotenoid bandshifts of RC H mutants (relative to the wild type) were similar to the relative RC contents detected by photobleaching and absorbance spectra (Table 3.1). Taken together, the above data indicate that the impairments of photosynthetic growth in these RC H mutants is due to reductions in RC amounts, as opposed to effects on RC catalytic activity, with cellular photosynthetic energy transduction proportional to the amount of RC. This appears to be the case in either the presence or absence of LH2, 54 although the presence of LH2 can partially compensate for a reduced amount of RC when the reduction is not severe (i.e., >20% of the wild type, as in pEVtip). However, it was not clear whether the reduced RC content was due to a low amount of mutant RC H proteins in cell membranes, resulting in a commensurate reduction in the amount of the RC complex, or i f membranes were replete in mutant RC H proteins that were impaired in the assembly of a stable RC complex. T A B L E 3.1. Summary of RC levels and catalytic activities measured in steady-state absorption spectra of intact LH2\" cells and flash spectroscopy of chromatophores obtained from L H 2 + cells. See figure 3.2B on p. 47 for the phenotype of the various constructions. Plasmid % 804 nm % R C % crt bandshift present in peak area LH2 + LH2 + LH2 + or LH2- LH2\" cells3 chromatophoresb chromatophoresb strains pATSHR (wt) 100(1) 100 100 pEVtip 21(2) 24 24 pEVtrans 65 (3) 66 66 pEVPSCB 66(1) N D C N D pEVCSP U d 18 16 amean of experiments on cells from three independent cultures, with standard deviations in parentheses b results of one experiment, with variation typically -10% c not done d undetectable 1000 n 3 (A C 0) Q Q) i_ 3 3 u 100 50 Time (hrs) 100 Figure 3.5 Photosynthetic growth of R. sphaeroides APUHA (LH2 +) cells containing plasmids: pATSHR (wild type RC H, open circles); pEVtip (hybrid RC H containing B. viridis periplasmic domain, open triangles); pEVCSP (hybrid RC H containing B. viridis cytoplasmic domain, asterisks). Ku, Klett units. 56 3.1.7 RC H, RC M, and RC L protein levels in membrane and soluble cell fractions Cells of the APUHAAPUC strain containing relevant plasmids were purified and probed in western blots with antisera directed against RC H, RC M or RC L proteins. The amounts of membrane and soluble fractions used were proportional to the non-fractionated cell extract, and exposure times were chosen to yield the same intensity of a constant amount of a purified RC control, to allow semi-quantitative comparisons of the amounts of these three proteins in the two fractions. As shown in Figure 3.6, the R. sphaeroides 28 kD band detected by the RC H antiserum was present in membranes of cells containing the wild type RC H gene (pATSHR) whereas the amount of RC H in the soluble fraction was extremely low. This band was absent from membrane and soluble fractions of the RC H\" strain containing pATP19P. The amount of RC M in membranes of the RC H\" strain was about the same as in the wild type RC H-complemented strain although, significantly, the amount of membrane-associated RC L was greatly reduced in the absence of the RC H protein. Strikingly, I discovered substantial amounts of RC M and RC L in the soluble fraction of cells that contained the wild type RC H protein in the membrane, whereas RC M and RC L were not detected in the soluble fraction of RC H\" cells. These data indicate that the membrane-associated amount of RC M is independent of the presence of RC H . In contrast, membrane localization of RC L appears to require RC H in the membrane. Furthermore, the presence of substantial amounts of RC M and L in the cytoplasm correlates with the presence of RC H in the membrane. Membrane preparations of cells expressing the RC H periplasmic domain substitution (pEVtip) contained less RC H than the wild type control, about the same amount of RC M , but less of RC L. The amounts of RC M and RC L in the soluble fraction were lower than in the wild type RC H control (pATSHR). Membrane and soluble fractions of cells containing the RC H transmembrane domain substitution (pEVtrans) contained about the same amounts of RC H , RC M and RC L as the wild type control, as did samples of cells containing the simultaneous exchange of periplasmic and transmembrane domains (pEVPSCB). The membranes of cells that produced the cytoplasmic domain-substituted RC H (pEVCSP) appeared to contain an extremely low amount of this mutant protein in western blots probed with the R. sphaeroides RC H antiserum. However, it is likely that this polyclonal antiserum reacts predominantly with cytoplasmic domain residues, which in pEVSCP would give a weak signal (as in the experiments on the B. viridis preparations; Fig. 3.1). I suggest that the amount of this hybrid RC H protein in membranes of R. sphaeroides, although low, is greater than indicated by the band intensity in Fig. 3.6. The amount of RC M was high whereas the amount of RC L was low in the membrane fraction, and the amounts of RC L and RC M in the soluble fraction were greatly reduced. 58 M S M S ! M S ; M S M S M S : H mm • • mm • mm mm • mm • mm P R C pATSHR pATP19P pEVtip | pEVtrans pEVPSCB pEVCSP F i g u r e 3.6 Western blot probed with R. sphaeroides RC H , RC M or RC L antibodies. Samples probed were the R. sphaeroides purified RC (PRC), and membrane (M) and soluble (S) fractions of APUHAAPUC cells containing plasmids: pATSHR (R. sphaeroides RC H); pATP19P (no RC H gene); pEVtip (hybrid RC H containing B. viridis periplasmic domain); pEVtrans (hybrid RC H containing B. viridis transmembrane domain); pEVPSCB (hybrid RC H containing B. viridis periplasmic and transmembrane domains); pEVCSP (hybrid RC H containing B. viridis cytoplasmic domain). 3.1.8 Kinetics of RC formation in the APUHAAPUC (pEVtrans) strain Absorption spectra of APUHAAPUC (pEVtrans) cells, which express a hybrid puhA gene having the R. sphaeroides transmembrane segment replaced with the corresponding B. viridis sequence, revealed a membrane RC complex content lower than that of the wild type control (Fig 3.4), reduced the photosynthetic growth rate (Fig 3.3) and yielded RC complex catalytic activity that was 66% of the wild type control (Table 3.1). However individual RC protein amounts in the membrane appeared to be unaffected (Fig 3.6). To further investigate this interesting mutant, I determined the rate of formation and amounts of the RC complex in induction experiments. This was done by comparing the RC 804 nm peak area of APUHAAPUC cells complemented with either pATSHR (wild type puhA) or pEVtrans initially grown under high aeration (where synthesis of photosynthetic complexes are repressed), followed by a switch to low aeration to induce the production of photosynthetic complexes. In these experiments, the concentration of oxygen is initially high but, as cell numbers increase, the respiration of dissolved oxygen becomes greater than the diffusion of oxygen from the air. Thus, after about 15 to 20 hours of incubation, cells become starved for oxygen, growth slows, and the synthesis of the photosynthetic apparatus is induced (48). As shown in Figure 3.7A, the APUHAAPUC(pATSHR) and APUHAAPUC(pEVtrans) cultures grew identically under semi-aerobic conditions, but the kinetics of RC formation were quite different (Fig. 3.7B). Each data point in Figure 3.7 represents the calculated area under the corresponding peak at a time point after the initiation of low aeration (see Materials and Methods). Based on the results of several 60 experiments, the replacement of the RC H transmembrane domain reproducibly caused a slower initial rate of formation and lower final levels of the RC. 3 o 20 40 60 Time (hrs) 80 B 0 20 40 60 Time (hrs) 80 Figure 3.7. Kinetics of RC formation. A Semi-aerobic growth curves for R. sphaeroides APUHAAPUC cells complemented with either pATSHR (wild type puhA, filled diamonds) or pEVtrans (hybrid puhA open diamonds). B Area under R C 804 nm peak, normalized to light scattering at 650nm, and plotted as a function of time; wild type puhA (encoded in plasmid pATSHR), filled diamonds; hybrid puhA (encoded in plasmid pEVtrans), open diamonds. See Appendix 2 for absorption spectra at various time points. 61 3.2 Site-specific mutagenesis and deletion of the R. sphaeroides RC H protein domain segments 3.2.1 Summary This section details the results of experiments in which selected residues or short segments within the periplasmic or cytoplasmic domains of the R. sphaeroides RC H protein were replaced or deleted. The mutants of the RC H periplasmic domain were evaluated in terms of photosynthetic growth, steady-state absorption spectra, and western blot analyses of RC proteins in cellular membrane and soluble fractions. The mutants of the RC H cytoplasmic domain were evaluated in terms of photosynthetic growth, steady-state absorption spectra, light-driven proton translocation and electron transfer reactions. 3.2.2 Overview of the mutant RC H proteins The domain swapping experiments described in section 3.1 indicated that the B. viridis periplasmic domain substitution reduced the amount of the RC complex in cells to 21-24% of the wild type control (Table 3.1). This amount of RC was incapable of supporting photosynthetic growth in the absence of LH2 (Fig. 3.3), and resulted in slowed growth in the presence of LH2 (Fig. 3.5). It was thought that certain key residues present in the R. sphaeroides RC H (but differing in the B. viridis RC H) periplasmic domain might be important for RC complex assembly and hence photosynthetic growth. 62 The RC H periplasmic amino acids Thr-5 and Asn-9 were individually or together replaced with Vai and Leu, respectively (encoded in plasmids pEStiplMT, pEStiplMA, and pEStip2M). Residues Thr-5 and Asn-9 were chosen because they are the only amino acids in the periplasmic domain with polar side chains capable of forming hydrogen bonds with amino acids of other proteins. Additionally, codons 3 through 10 of the R. sphaeroides RC H periplasmic domain were removed in plasmid pEAPeri, to yield a mutant RC H protein in which Val-2 is followed by Asp-11 (see Fig. 3.2). I also decided to investigate RC H cytoplasmic surface residues that had been implicated as being near the entry point for one of three speculated proton transfer pathways into the RC from the cytoplasm, because of the inhibitory effects of Z n 2 + / C d 2 + binding to this region (see Introduction section 1.6). The RC H cytoplasmic residues His-126 and His-128 were individually or together substituted with alanines (encoded in plasmids pEH126A, pEH128A, and p2XHis), to investigate their role in the uptake of protons from the cytosol. 3.2.3 Photosynthetic growth of APUHAAPUC strains expressing RC H periplasmic domain mutations The LH2\" strain APUHAAPUC was used in these experiments to allow comparison of growth properties with the RC and LH1 levels indicated by absorption spectroscopy of intact cells. The APUHAAPUC (pEAPeri) strain was incapable of photosynthetic growth, while the single amino acid changes to the R. sphaeroides periplasmic domain (Thr-5->Vai in pEStiplMT, and Asn-9->Leu in pEStiplMA) had no significant effects on 63 photosynthetic growth relative to the wild type (Fig. 3.8). In contrast, the mutant RC H gene encoding these two amino acid changes together in the plasmid pEStip2M resulted in slower growth that reached a plateau at a much lower final cell density than the control (Fig. 3.8; compare open squares to open circles). 3.2.4 Absorption spectroscopy of LH2\" APUHAAPUC strains expressing RC H periplasmic domain mutations The individual changes of the periplasmic amino acids (Thr-5->Val in pEStiplMT, and Asn-9^Leu in pEStiplMA) did not result in significant changes to RC levels, whereas simultaneous changes of both amino acids (pEStip2M) decreased the RC level to 61% of the wild type (Fig. 3.9). No RC was detected in cells that produced a RC H protein from which most of the periplasmic domain had been deleted (pEAPeri) and, interestingly, there was a significant reduction in the 870 nm LH1 peak (Fig 3.9). It is noteworthy that the amplitude of the 870 nm LH1 peak was greater in cells that lacked an RC H protein (see pATP19P in Fig. 3.9) than in cells that produced an H protein from which amino acid residues 3 to 10 of the periplasmic domain had been deleted (encoded in plasmid pEAPeri). Thus, the cellular content of LH1 is affected more by the presence or absence of the RC H protein periplasmic domain, than by changes in periplasmic domain amino acid composition, or by the presence or absence of the wild type RC H protein. 64 1000 -! Figure 3.8 Photosynthetic growth of R. sphaeroides APUHAAPUC (LH2~) cells containing plasmids: pATSHR (wild type RC H, open circles); pEAPeri (mutant RC H lacking the periplasmic domain, filled squares); pEStiplMT (mutant RC H with the periplasmic residue Thr-5 replaced with Vai, filled diamonds); pEStiplMA (mutant RC H with the periplasmic residue Asn-9 replaced with Leu, open diamonds); or pEStip2M (mutant RC H with both periplasmic residues Thr-5 and Asn-9 replaced with Vai and Leu, respectively; open squares). 65 pATSHR pATP19P pEStiplMT pEStiplMA pEStip2M pEAPeri 700 800 900 nm Figure 3.9 Absorption spectra of APUHAAPUC (LH2\") intact cells containing plasmids: pATSHR (R. sphaeroides RC H); pATP19P (expression vector); pEVtip (hybrid RC H containing B. viridis periplasmic domain); pEStiplMT (mutant RC H with the periplasmic residue Thr-5 replaced by Vai); pESt iplMA (mutant RC H with the periplasmic residue Asn-9 replaced by Leu); pEStip2M (mutant RC H with periplasmic residues Thr-5 and Asn-9 replaced by Vai and Leu, respectively); or pEAPeri (mutant RC H lacking the periplasmic domain). 3.2.5 Flash spectroscopy of APUHA strains expressing RC H periplasmic domain mutations The percentage amounts of the RC complex in chromatophores from L H 2 + cells containing these mutant periplasmic domains relative to the wild type, as indicated by photobleaching (due to electron transfer from D to QA) are given in Table 3.2. These values are consistent with the photosynthetic growth properties and the steady-state absorption spectra of LH2\" cells (see above). The carotenoid bandshifts of RC H periplasmic domain mutants (relative to the wild type) were similar to the relative RC contents detected by photobleaching and absorbance spectra (Table 3.2). T A B L E 3.2 Summary of RC levels and catalytic activities measured in steady-state absorption spectra of intact LH2\" cells and flash spectroscopy of chromatophores obtained from L H 2 + cells. Plasmid % 804 nm % RC % crt bandshift present in peak area LH2 + LH2 + LH2 + or LH2 LH2 - cells3 chromatophores15 chromatophores15 strains pATSHR (wt) 100(1) 100 100 pEAPeri U c N D D N D pEStip2M 61(2) 69 89 pESt iplMA 89 (3) 100 118 pEStiplMT 91 (2) 100 116 a mean of experiments on cells from three independent cultures, with standard deviations in parentheses b results of one experiment, with variation typically -10% 0 undetectable d not done 67 3.2.6 RC H, RC M and RC L protein levels in membrane and soluble cell fractions of strains expressing RC H periplasmic domain mutations-Cells of the APUHAAPUC strain containing relevant plasmids were disrupted and fractions probed with antisera directed against RC H , RC M or RC L proteins in western blots. The amounts of membrane and soluble fractions used were proportional to the non-fractionated cell extract, and exposure times were chosen to yield the same intensity of a constant amount of a purified RC control, to allow semi-quantitative comparisons of the amounts of these three proteins in the two fractions. Membranes of cells containing the RC H periplasmic domain deletion (pEAPeri) contained a vanishingly small amount of this mutant RC H protein, the amount of RC M was equivalent to the wild type control (pATSHR), and the amount of RC L was very low. A l l three of the RC proteins were undetectable in the soluble fraction (Fig. 3.10). The RC H protein containing both periplasmic domain site-directed amino acid changes (pEStip2M) was present in membranes at a slightly lower level than that observed for the two single amino acid changes, with a slightly reduced amount of RC L, whereas RC M was present at the same level as in the wild type control. The amounts of RC M and RC L proteins in the soluble fraction were comparable to the wild type (Fig. 3.10). Membranes of cells expressing either of the single amino acid changes in the periplasmic domain of RC H (pEStiplMA and pEStiplMT) contained slightly decreased amounts of RC H , whereas the amounts of RC M and RC L were about the same as in the wild type control. The amounts of RC M and RC L in the soluble fractions did not differ greatly from the wild type control (Fig. 3.10). 68 M s M s M S M S M S M S H mm U H • • H i mmm — M *m*m mmm B l mmm. L B i t H i — i ! mmm P R C p A T S H R p A T P 1 9 P pEAPer i | p E S t i p 2 M p E S t i p l M T p E S t i p l M A Figure 3.10 Western blot probed with R. sphaeroides RC H, RC M or RC L antibodies. Samples probed were the R. sphaeroides purified RC (PRC), and membrane (M) and soluble (S) fractions of APUHAAPUC cells containing plasmids: pATSHR (R. sphaeroides RC H); pATP19P (no RC H gene); pEAPeri (mutant RC H lacking the periplasmic domain); pEStip2M (mutant RC H with periplasmic residues Thr-5 and Asn-9 replaced by Val and Leu, respectively); pEStiplMT (mutant RC H with the periplasmic residue Thr-5 replaced by Val); and pESt iplMA (mutant RC H with the periplasmic residue Asn-9 replaced with Leu). 69 3.2.7 Photosynthetic growth and absorption spectroscopy of APUHAAPUC strains expressing RC H cytoplasmic domain mutations Photosynthetic growth of the APUHAAPUC host cells expressing mutant RC H proteins with either single or double substitutions of His-126 and His-128 residues to alanines (encoded in plasmids pEH126A, pEH128A, and p2XHis respectively), closely paralleled that of the wild type control (pATSHR) (Fig. 3.11 A). As shown in Figure 3.1 IB, the absorption spectra of intact cells of APUHAAPUC strains containing pEH126A, pEH128A, or p2XHis plasmids revealed no significant differences between the RC and LH1 peaks of these strains (at 804 and 870 nm, respectively), and that of the wild type control (pATSHR). This is in agreement with the photosynthetic growth characteristics for these strains. 3.2.8 Electron transfer and proton uptake measurements for APUHAAPUC strains expressing RC H cytoplasmic domain mutations The translocation of protons into the RC for the protonation of the Q B quinone is coupled to two sequential light-induced electron transfer reactions (see Introduction section 1.6). To determine the role of RC H surface residues His-126 and His-128 in the delivery of protons into the RC, proton uptake rates and the coupled electron transfer rates were measured using RCs purified from APUHAAPUC host cells containing mutant RC H proteins with individual or simultaneous replacements of the cytoplasmic residues His-126 and His-128 (encoded in plasmids pEH126A, pEH128A, and p2XHis respectively), to alanines. Cultures of these cells were grown semi-aerobically, cells disrupted and RCs purified from isolated membrane vesicles (chromatophores). 70 B pATSHR p2XHis pEH126A pEH128A 900 nm Figure 3.11 A Photosynthetic growth of R. sphaeroides APUHAAPUC (LH2\") cells containing plasmids: pATSHR (wild type RC H , open circles); p2XHis (His-126 and His 128 -» Ala, open triangles); pEH126A (His-126^ Ala, closed squares); pEH128A (His-128^ Ala, open diamonds). B Absorption spectra of APUHAAPUC (LH2\") intact cells containing select plasmids. 71 Electron transfer and proton uptake reactions in the RC were triggered by laser flashes (see Material and Methods section 2.10). The overall electron and proton transfer reactions catalyzed by the RC are summarized in Fig 1.7A. The first electron transfer reaction from Q A to Q B , and the coupled proton uptake can be represented as: Although Q B is not protonated in this step, it is converted into a quasi-stable anionic semiquinone following the acceptance of an electron from Q A . This process also involves the protonation of the QB\" proximal RC L acid residue Glu-212 (51). Following a single laser flash, the rate of proton uptake by Glu-212 for purified RCs of APUHAAPUC host cells containing plasmids pEH126A, pEH128A, or p2XHis was measured by monitoring proton uptake from solution and the coupled electron transfer rate kAB1 • In comparison with the native RC, while the individual changes of the His residues to Ala only slightly decreased the rate of proton uptake and kAB (data not shown), both were significantly decreased for RCs of APUHAAPUC(p2XHis) cells (Fig 3.12). For the native RC, the proton uptake rate and kAB were calculated to be -4000 s\"1. In contrast, the rates for the double mutant was reduced to -500 s\"1. The Q B \" semiquinone is converted to Q B H 2 after a second flash, resulting in the uptake of a second proton from solution and transfer of a second electron from Q A \" ; this process is represented below. kAB QA\" + Q B + H ++(Glu-L212)\" Q A + QB\" + (G1U-L212)H (eq. I) QA\" + QB\" + H + + (Glu-L212)H Q A + Q B H 2 + (G1U-L212)\" (eq. 2) 72 Proton uptake Q O o < t 4 6 — time [ms] Q O a < First electron transfer &AB r • i • i • i 1 i Native i B. If Double mutant j . i 0 •- 4 ; 6 ~~ time [ms] 10 Figure 3.12 Measurement of proton uptake (A), and electron transfer kAB1 (B) processes at pH 8.5 for purified RCs of native R. sphaeroides, and of double His mutant (His-126 and His-128 -» Ala change). Proton uptake was measured at 580 nm using the pH-sensitive dye m-cresol purple. Electron transfer was measured at 750 nm, which is sensitive to the reduction states of the quinones. These experiments were performed in collaboration with the Okamura laboratory (UCSD, Ca). 73 The transfer of a second electron to Q B \" is coupled to its protonation by RC L residue Ser-223, which is rapidly re-protonated by proton uptake from the solvent, yielding the intermediate Q B H \" . A second proton is internally transferred to Q B H \" from Glu-212, which was protonated during the first electron transfer reaction to yield the quinol Q B H 2 (51). Following a second laser flash, the rates of proton uptake by Q B \" for purified RCs of APUHAAPUC host cells containing plasmids pEH126A, pEH128A, or p2XHis were measured by monitoring proton uptake from solution and the coupled electron transfer 2 1 rate kAB (Fig. 3.14). These rates were calculated to be -350 s\" for the native RC. No significant effects on either rate were observed for RCs of APUHAAPUC cells containing either pEH126A, or pEH128A plasmids (data not shown). However, proton uptake and kAB2 rates for the RC of the double mutant were decreased to -100 s\"1. To investigate whether the changes in RC catalytic activity were due specifically to the loss of the His side chains (Fig. 3.13), as opposed to secondary effects such as protein conformational changes, exogenous proton donors such as imidazole were used to determine if they would rescue the observed effects on the proton uptake kinetics of APUHAAPUC (p2XHis) strain. HjN* CH COO-| Imidazole side chain Fig. 3.13. Chemical formula of the amino acid histidine. 74 Proton uptake i — i — i — i — ( — i — i — i — i — i — 0 5 10 15 20 25 30 35 40 time [ms] Second electron transfer &AB 20 Q 15 O __ o 10 c < 5 t 0 - J — t — I — I — I — I — I — I I I I I I I I . 1_ 0 5 10 15 20 25 30 35 40 —\"time [ms] Figure 3.14 Measurement of proton uptake (A), and electron transfer kAB (B) processes at pH 8.5 for purified RCs of native R. sphaeroides, and of double His mutant (His-126 and His-128 -» Ala change). Proton uptake was measured at 580 nm using the pH-sensitive dye m-cresol purple. Electron transfer was measured at 450 nm, the peak absorption of the semiquinone species. The initial rise is due to the formation of the Q A QB\" state, which decays to (QBH)\" resulting in the observed transient. These experiments were performed in collaboration with the Okamura laboratory (UCSD, Ca). 75 The addition of 50 m M imidazole to the double mutant RC sample increased the rate of the both electron transfer reactions to nearly equal to that observed for the native RCs (Fig. 3.15), with the effects being reversible upon the removal of imidazole (data not shown). Addition of imidazole had no effect on the electron transfer rates for the native, or single mutant RCs (data not shown). First electron transfer AAB 0 2 4 6 8 10 12 time [ms] Second electron transfer &AB T — 1 — i — 1 — i — 1 — i — 1 — i — 1 — i — • — i — • — i — • — r 0 5 10 15 20 25 30 35 40 — • time [ms] Figure 3.15 \"Chemical rescue\" by exogenous imidazole of kAB1 (A) and kAB2 (B), for purified RCs of native R. sphaeroides, and of double His mutant (His-126 and His-128 -» Ala change). These experiments were performed in collaboration with the Okamura laboratory (UCSD, Ca). 76 3.3 RC gene deletion studies 3.3.1 Summary Possible interactions between photosynthetic reaction centre (RC) proteins were evaluated by use of RC gene-specific deletions. The RC H, RC M and RC L proteins were produced, either alone or in concert with one or both of the others, in a strain of R. sphaeroides that contained chromosomal deletions of all three RC genes (ARCLH strain). The properties of these strains were studied in photosynthetic growth and absorption spectroscopy experiments. The amounts of the RC H, RC M , and R C L proteins in cell membrane and soluble fractions were assessed in western blots. 3.3.2 Overview of the R. sphaeroides ARCLH strain and the RC gene deletions M y western blots showing RC H, RC M , and RC L levels in cells expressing mutant RC H genes appeared to be at odds with reigning models of RC complex assembly (see Results sections 3.1.7 and 3.2.6). This is because previous publications had proposed that the RC H protein is a \"foundation\" or \"chaperone-like\" protein that is the first protein to accumulate in the membrane for subsequent efficient assembly of RC M and RC L proteins to form the RC holocomplex. Furthermore, the level of R C M in RC H-deleted cells was reported to be low because of degradation (13, 68), whereas my experiments on RC H mutants indicated that the accumulation of RC M in the membrane was independent of RC H . Moreover, the accumulation of RC L had not been studied prior to my research. Therefore, I created a host strain and several plasmids to further 77 investigate the relationships between the RC H, RC M , and RC L proteins, in terms of their accumulation within the cell. The A R C L H mutant strain contains chromosomal deletions in the puh, puf and puc genes, yielding it incapable of producing the RC, LH1 or LH2 complexes. The plasmid pESHPUF, which encodes the native R. sphaeroides puhA and puf operon genes was constructed to assess the recovery of RC and LH1 production in the A R C L H strain to test the functionality of this strain and this plasmid for RC gene deletion studies. The deletion of the RC L gene was studied using plasmid pESHPUFL\", which expresses the RC H and RC M genes. Similarly, plasmid pESHPUFM\" was used to produce the RC H and RC L proteins in the absence of RC M . The effects of the deletion of the RC M and RC L genes on the RC H membrane content was investigated using plasmid pESHPUFL'M\", with the converse (RC M and RC L produced without RC H) examined using plasmid pEPUF. A l l of the RC gene deletions in the complemented A R C L H strain were translationally in-frame deletions, to minimize the possibility of polar effects on other co-transcribed genes. 3.3.3 Photosynthetic growth of ARCLH strain expressing different combinations of RC genes The strain A R C L H containing the expression plasmid pATP19P was incapable of photosynthetic growth, as it was when containing pESHPUFL\", pESHPUFM\", pESHPUFL'M\" or pEPUF (Fig. 3.16). In contrast, the A R C L H (pESHPUF) strain grew 78 almost as well as the wild type, indicating that the puhA (RC H) gene and the puf genes (LH 1 and RC L/M) on plasmid pESHPUF were expressed at a level sufficient for photosynthetic growth. 1000 n Figure 3.16 Photosynthetic growth of wild type R. sphaeroides strain 2.4.1 (filled triangles), and complemented A R C L H strains. Plasmids used to express RC genes in A R C L H were: pESHPUF (expressing RC H , RC M and RC L genes, open circles); pATP19P (expression vector, open triangles); pESHPUFL\" (expressing RC H and RC M genes, open squares); pESHPUFM\" (expressing RC H and RC L genes, asterisks); pESHPUFL'M\" (expressing RC H gene, crosses); and pEPUF (expressing RC M and RC L genes, filled diamonds). 79 3.3.4 Absorption spectra of the ARCLH strain expressing different combinations of RC genes Absorption spectra of intact ARCLH(pATP19P) cells grown semi-aerobically showed that this strain is devoid of RC, LH1 and LH2 complexes, because of the absence of peaks in the 700 to 950 nm range (Fig. 3.17 ). The RC and LH1 complexes were restored by complementation of A R C L H with the plasmid pESHPUF, as indicated by the 760, 804 and 870 nm peaks (Fig. 3.17). The RC 760 and 804 nm peaks were absent from cells containing plasmids pESHPUFL\", pESHPUFM\", pESHPUFL'M\" or pEPUF (Fig 3.17). There was a small 770 nm peak in A R C L H (pEPUF), indicative of free bacteriochlorophyll (40). There were also differences in the amplitude of the L H 1 870 nm peak in the different strains. I cannot readily account for these changes, except to attribute them to a variety of unknown interactions between LH1 and the RC (see Discussion section 4.3). These observations are consistent with the photosynthetic growth results, and together indicate that: 1) the RC and LH1 genes are expressed in the ARCLH(pESHPUF) strain, to produce a functional RC and LH1; 2) the A R C L H strain provides a system to express different combinations of plasmid-borne RC genes for the study of potential interactions between RC proteins in membrane localization and accumulation. pEPUF pESHPUF pESHPUFL\" pESHPUFM\" pESHPUFL\"M\" pATP19P 700 800 900 nm Figure 3.17 Absorption spectra of A R C L H intact cells containing plasmids: pEPUF (encoding RC M and RC L), pESHPUF (encoding RC H, RC M , and RC L), pESHPUFL\" (encoding RC H and RC M), pESHPUFM\" (encoding RC H and RC L), pESHPUFL\"M\" (encoding RC H only), or pATP19P (empty expression vector). 81 3.3.5 RC H, RC M and RC L protein levels in membrane and soluble cell fractions Cultures of the A R C L H strain containing plasmids that express different combinations of RC H, RC M and RC L genes were grown semi-aerobically, cells disrupted, and the soluble and membrane fractions probed in western blots using antisera directed against J?. sphaeroides RC H , RC M or RC L proteins. The amounts of membrane and soluble fractions used were proportional to the non-fractionated cell extract, and exposure times were chosen to yield the same intensity of a constant amount of a purified RC control, for semi-quantitative comparisons of the amounts of these three proteins in the two fractions. As shown in Figure 3.18, the RC H, RC M and RC L proteins were detected in the membrane fraction of A R C L H cells complemented with the positive control plasmid pESHPUF, which was constructed to express all three RC proteins, whereas cells containing the negative control plasmid pATP19P lacked these bands. The RC M protein was detected in the membrane fraction of cells containing the RC M gene, regardless of the presence of the RC H or RC L genes. The RC H protein was detected in membranes only when both RC M and RC L proteins were present, whereas the membrane-associated RC L protein was detected only when the RC M protein was present (Fig. 3.18). However, the amount of RC L present in membranes containing RC M but not RC H was -27 % of the amount detected when both RC H and RC M were present. These data indicate that accumulation of RC H in the membrane requires the presence of both RC M and RC L. Insertion and amounts of RC L in the membrane is 82 low, but detectable in the absence of RC H, and is the greatest in the presence of both RC H and RC M . Steady-state levels of RC M in the membrane are independent of RC H and RC L. No RC H was found in any of the soluble fractions (Fig. 3.18). In contrast, RC M and RC L were readily detected in the soluble fraction of the positive control (pESHPUF), traces of RC M and RC L were present in the soluble fraction of RC H\" (pEPUF) cells, and neither RC M nor RC L was detected when the other of these two proteins was absent from cells (Fig. 3.18). Thus, the accumulation of RC M and RC L in the soluble fraction is mutually dependent, and the presence of the RC H protein in the membrane appears to result in maximal amounts of soluble forms of RC M and RC L proteins destined for membrane insertion and RC assembly. H M L u 3 0-2 UH D o. X OO UJ H J PH OH 00 W CH —1 UH 5 OH s 00 LU O. UH D OH s 00 UJ a UH D CH U J CH II M\" L H M i: H M\" L\" II M L H\" M L B UH 5 0. X 00 UJ CH H J UH 5 OH X oo U J CH H J CH 00 uu OH U-. CH X oo U J CH UH 5 OH w CH H M L H M\" L H M L\" H M\" L H M L H M L Figure 3.18 Western blots of cell fractions probed with RC H , RC M or RC L antibodies. A: membrane fraction. B : soluble fraction. Samples were of A R C H L H cells containing plasmids as indicated. 84 4. DISCUSSION 4.1 Functional analysis of if. sphaeroides RC H protein domains The structurally well characterized, but functionally poorly understood R. sphaeroides RC H protein can be conceptually divided into three domains: 1) a ~11 amino acid N-terminal domain located on the periplasmic side of the inner membrane 2) a single transmembrane a-helix; 3) a C-terminal, large, globular cytoplasmic domain (36). The function of each of these domains was investigated in domain substitution, deletion, and site-specific mutagenesis experiments. The domain substitution approach arose from genetic complementation experiments on R. sphaeroides RC H gene deletion strains that expressed the B. viridis RC H gene, but were incapable of photosynthetic growth. Although the B. viridis RC H protein could in principle have resulted in an equivalently low amount of RC formed in R. sphaeroides, insufficient to support photosynthetic growth or produce a detectable 804 nm RC peak in the absorption spectrum, the appearance of a broad peak centred at 770 nm (the wavelength of protein-free bacteriochlorophyll absorption (40)) (Fig. 3.1) suggests that the B. viridis RC H protein was incapable of interacting with the R. sphaeroides RC M and RC L proteins in RC complex assembly. As there were several ways to explain these results, I reduced the complexity of analysis by creating and using strains that contained only one or two RC H domain sequence changes, to investigate possible contributions of these individual domains to RC catalytic activity, assembly of the RC complex and membrane localization of RC proteins. 85 A l l of the RC H genes were expressed using the puc promoter (32) of pATP19P, and the RC M and RC L genes were expressed from the wild type chromosomal puf operon. Therefore reductions in the steady-state levels of RC proteins are attributed to proteolysis that occurs when these proteins fail to interact to form the RC complex (68). The periplasmic domain. Photosynthetic growth was impaired in LH2\" R. sphaeroides cells that contained the periplasmic domain replacement, deletion or the double amino acid substitution, whereas the single amino acid changes did not greatly affect photosynthetic growth (Figs. 3.3 and 3.8). The absorption spectra in the LH2\" background (Figs. 3.4 and 3.9) and the photobleaching experiments in the L H 2 + background (Tables 3.1 and 3.2) indicate that, regardless of the presence of LH2, the amount of the RC complex in these strains is congruent with their photosynthetic growth properties. Because the carotenoid bandshift of these strains was approximately proportional to their RC content, changes in the periplasmic domain of RC H do not seem to affect light-driven electron transfer and proton translocation in the reduced amounts of RCs that were assembled. It appears that there is a minimal cellular RC content that supports photosynthetic growth under the conditions in this study (of growth medium, light intensity, etc), which is estimated to be in the range of 60 to 70% of the wild type amount when LH2 is absent, and 20 to 30% of the wild type amount when LH2 is present. The western blot results (Figs. 3.6 and 3.10) show the importance of the periplasmic domain for RC H levels in the membrane fraction, and reveal a correspondence between the amount of membrane-associated RC H and membrane-associated RC L, but not RC M . For example, the deletion of periplasmic residues 3 to 86 10 (pEAPeri) resulted in only a trace of RC H in the membrane fraction, similar to the complete absence of RC H . In both cases the amount of membrane-associated RC L was greatly reduced whereas the amount of RC M was about the same as in the wild type control. When the R. sphaeroides RC H periplasmic domain was replaced with the B. viridis sequence (pEVtip), the membrane-associated RC H and L proteins were detected at levels lower than the wild type control, while the amount of RC M in membranes was not reduced. The single substitutions in the periplasmic domain (Thr-5 -> Val in plasmid pEStiplMT, and Asn-9 Leu in pEStiplMA) had little effect on the amounts of RC H and RC L in membranes, whereas the double substitution (pEStip2M) caused a greater reduction in the RC H and L levels. None of these mutations reduced the amount of membrane-associated RC M . Thus it appears that there are several amino acids in the periplasmic domain that together affect RC H (and hence RC L) amounts in the membrane, whereas the amount of membrane-associated RC M is independent of RC H and RC L . It was surprising to discover that in strains which produced the wild type RC H protein, and hence were replete in all three RC proteins, significant amounts of RC M and RC L (but little RC H) were detected in the soluble fraction of disrupted cells (Fig. 3.6). This result is counter-intuitive, because RC M and RC L are much more hydrophobic than RC H (i.e. RC M and RC L each contain five transmembrane oc-helices and relatively short extra-membranous segments, whereas RC H contains a single transmembrane a-helix and a larger cytoplasmic domain). Cells containing RC H proteins with periplasmic domain mutations that impaired membrane localization contained lower amounts of both RC M and RC L proteins in the soluble fraction. 87 Therefore, the presence of the RC H protein in the membrane, which is in part dependent on the periplasmic domain sequence, appears to be needed to sustain wild type levels of RC M and RC L proteins in the cytoplasm. I at first thought that upon membrane insertion, RC H might interact with one or more RC putative assembly factors (3, 12, 72), and that this interaction stabilizes newly translated RC M and RC L proteins in a soluble form for membrane insertion dedicated to RC assembly. However, the subsequent RC gene deletion experiments (see Results section 3.3.5; and Discussion section 4.2) point to membrane accumulation of RC L for accumulation of RC M and L in the cytoplasm. The transmembrane domain. In contrast to the periplasmic domain, substitution of the R. sphaeroides RC H transmembrane domain with the radically different and one amino acid longer B. viridis sequence (Fig. 3.2) encoded by pEVtrans did not appear to affect membrane localization of RC H or RC L, and RC M was again present at the wild type level (Fig. 3.6). The amounts of these proteins in the soluble fraction were also similar to the wild type control. Thus, in spite of a greatly different amino acid sequence, a stretch of 20 to 21 amino acids in the ot-helical transmembrane domain suffices for RC H and, consequently, RC L membrane insertion. However, there appears to be a secondary role of the RC H transmembrane domain in RC assembly. This is because although the substitution of the transmembrane domain did not affect the amounts of individual RC proteins, the initial rate of formation (Fig. 3.7) and steady-state level (Fig. 3.4 and Table 3.1) of the hybrid RC complex were lowered. I attribute these reductions to an inability of the B. viridis transmembrane sequence to properly interact with the R. sphaeroides RC M protein, thus resulting in impaired RC assembly and photosynthetic growth (Fig. 3.3). 88 Analysis of the R. sphaeroides RC crystal structure (Brookhaven PDB accession 1 ALT) indicates that only amino acids of the RC M protein could interact with RC H transmembrane amino acid side chains. RC H transmembrane residues with atoms located <3.5 A of RC M residue atoms are (RC M amino acids given in parentheses): Ser-14 (Trp-297, His-301); Phe-20 (Leu-204); Tyr-30 (Arg-267); Gln-32 (Phe-258); Glu-34 (Arg-267); Asn-35 (Gly-264, Ile-265, Trp-268). Several of these interactions appear to be hydrogen bonds (e.g.: RC H residues Ser-14, Tyr-30, Glu-34 and Asn-35) and so are candidates for determinants of quaternary structure in the R. sphaeroides native RC. A l l of these residues are changed in the transmembrane substitution (Fig. 3.2 A). The carotenoid bandshift of chromatophores containing the transmembrane -substituted RC H was within experimental error proportional to the RC content (Table 3.1). Therefore, the significance of the RC H / M amino acid interactions discussed above appears to be solely in the assembly of the RC, with little or no effect on RC catalytic activity. Simultaneous substitution of periplasmic and transmembrane domains. As noted above, substitution of the R. sphaeroides with the B. viridis periplasmic domain reduced the amount of RC H in the membrane, whereas the analogous transmembrane substitution did not appear to affect membrane localization. These two domain changes were combined (in plasmid pEVPSCB), resulting in a RC H protein consisting of B. viridis sequences from the N-terminus to amino acid Arg-34, followed by the R. sphaeroides cytoplasmic domain (Fig. 3.2). The simultaneous substitution of periplasmic and transmembrane domains yielded cells with photosynthetic growth properties (Fig. 3.3), RC complex absorption and 89 activity (Fig. 3.4 and Table 3.1), and amounts of membrane-associated and soluble forms of RC H , RC M and RC L proteins (Fig. 3.6) that were more like those resulting from the transmembrane than the periplasmic individual substitutions. One way to view these results is that the inhibitory effect of the periplasmic substitution on the level of RC H in the membrane is ameliorated by the addition of the transmembrane substitution. I speculate that species-specific interactions between periplasmic and transmembrane residues enhance RC H insertion into the membrane. This speculation led to the creation of periplasmic domain deletion and site-directed mutants (see below). The cytoplasmic domain substitution. Photosynthetic growth was not observed for LH2\" or L H 2 + host cells that contained a substitution of the R. sphaeroides RC H 226 amino acid cytoplasmic domain with the B. viridis 223 amino acid domain (-40% identity; see Appendix 1) (Fig. 3.3). Absorption spectra (Fig. 3.4), flash spectroscopy (Table 3.1) and western blot experiments (Fig. 3.6) on strains containing plasmid pEVCSP indicate that the significant reductions in steady-state levels of the membrane RC content, and the commensurate loss of photosynthetic growth is the result of impaired RC complex assembly. As with the other domain mutations that reduced the amounts of membrane-associated RC H and RC L, and the cytoplasmic amounts of RC M and RC L, there was not a reduction of the amount of RC M in the membrane. Thus, the RC H cytoplasmic domain also functions to obtain an optimal amount of RC H in the cell membrane, which in turn is needed for accumulation RC L in the membrane, and subsequent accumulation of RC M and RC L proteins in the cytoplasm of for RC assembly. 90 The function of the cytoplasmic domain extends beyond an assembly role, as it was previously implicated in the translocation of protons from the cytoplasm to QB(53). The binding of C d 2 + or Z n 2 + by RC H cytoplasmic residues Asp-124, His-126 and His-128 was shown to result in a lowering of the Q B quinone protonation rate (5). In a collaborative effort with the Okamura laboratory, we determined that individual changes of RC H cytoplasmic surface residues His-126 and His-128 residue to Ala had insignificant effects on steady-state levels of the RC, or its. overall catalytic activity (Fig. • 3.11). However, the simultaneous changes of both His residues to Ala decreased proton uptake rates by 4-10 fold (Figs. 3.12 and 3.13). These reductions are attributed to a slower rate of proton delivery from the entry point to the proton transfer pathway, due to the absence of the His-126 and His-128 imidazole side chains. The addition of exogenous imidazole as a proton donor restored proton uptake rates to almost wild type levels (Fig. 3.15). Therefore, it can be concluded that His-126 and His-128 jointly mediate proton transfer into the RC by acting as proton donors at the entrance of the proton transfer pathway as reported in detail (2, 52), and outlined in Figure 4.1. These conclusions are confirmed by experiments described in (56). 91 Figure 4.1 Part of the structure of the R C in the charge-separated state. Shown are the Q B quinone, and amino acid residues thought to be key for the transfer of protons H + (1), and H + (2) to Q B . The proton path through R C H residues His-126 and His-128 to Q B are represented by a black dashed arrow for H + (1), and a gray dashed arrow for H + (2). Potential hydrogen bonds are indicated by small dashes, and water molecules revealed by x-ray crystallography (1) represented by large dots. This figure was kindly provided by Mark Paddock. Summary. Taken together, the data discussed so far suggest that the RC H mutations that impaired photosynthetic properties did so by reducing the amount of the RC complex in the cell membrane. These effects appear to stem from decreases in the amount of membrane-associated RC H , and override possible effects on the catalytic activity of low amounts of RCs that were assembled. The periplasmic and cytoplasmic domain substitutions greatly reduced the amount of membrane-associated RC H, whereas the transmembrane domain substitution functioned well for membrane targeting. A l l three domains of the RC H protein contain amino acid sequences that are required to obtain wild type levels of the RC complex in the cell membrane. Interestingly, the RC H-His 126/128 mutations did not affect photosynthetic growth of cells, although they clearly impaired purified RC catalysis in vitro. I explain this apparent paradox in the following paragraphs. The in vitro inhibitory effects of the simultaneous changes of RC H residues His-126 and His-128 to Ala, were initially not seen when the catalytic activity of the purified RC was assayed in the standard Tris-HCI buffer. Subsequently, it was realized that the amino group of Tris (Fig 4.2) could act as a proton donor (because it would be partially protonated at the pH = 8.5 of the kAB and A^assays), hence rescuing the effects of the His changes, on RC proton uptake. However follow-up experiments in the absence of Tris, addition of a variety of soluble proton donors (His analogues), and pH titrations revealed the function of His 126/128 as proton donors to a pathway leading to Q B (52). I do not present a detailed physical chemistry argument in support of these collaborative experiments, as frankly they are beyond my microbiological and molecular biological expertise. Instead, the interested reader is directed to the publications (2, 51, 52, 56). 93 Figure 4.2 Chemical structure of TRIS(hydroxymethyl) aminomethane. The N H 2 moiety is in equilibrium between NH2<-» N H 3 + states, which probably provides most of the buffering properties of this molecule (pKa =8.06) The in vitro results on the purified RC indicate that the photosynthetic growth (Fig. 3.11) of the APUHAAPUC (p2xHis) strain may be explained by the vast variety of potential proton donors to the RC in vivo (e.g.; amino acids, and other cytoplasmic molecules), which could rescue the effects of the His-126 / His-128 -> Ala mutations. This could be categorized as an example of how a mutation may not result in a cellular phenotype, yet provide valuable information about a catalytic mechanism by careful in vitro analysis. 4.2 The role of the R C H protein in R. sphaeroides R C complex assembly The pioneering experiments of Kaplan and colleagues on RC complex assembly in R. sphaeroides, which included pulse-chase experiments on RC H and RC M , revealed a dynamic interplay between synthesis, membrane insertion and proteolytic degradation of these two proteins (64, 68). Their data led them to conclude that the RC H protein has a \"chaperone-like\" function in RC assembly, that the amount and stability of RC M in the membrane are dependent on the presence of RC H, and that RC H-stabilized RC M in the 94 membrane leads to \"productive association\" of RC M with RC L, resulting in RC assembly. However, the RC H mutation in the strain used by these investigators (in contrast to the chromosomal RC H mutations used in this thesis) appears to have a polar effect on the expression of genes located 3' of the RC H gene and deletes part of the 5' lhaA gene (3, 12, 64, 68, 72). These genes have been implicated as RC assembly factors, although their exact functions are unknown, which may account for the reduced amount and rapid degradation of RC M in the Varga et al. experiments (68). The RC H protein domain exchanges, and site-directed mutagenesis studies in this thesis indicate that the RC M protein and not RC H is the primary component in the process of RC complex assembly. Combinations of RC H , RC M and RC L gene deletions were constructed to investigate the possibility of stabilizing interactions between these proteins that affect their accumulation within the cell membrane and cytoplasm (see Materials and Methods section 2.2.3 for a description of mutants). As shown in Fig. 3.18A, all three RC proteins were expressed and inserted into the membrane of the control strain ARCLH(pESHPUF), consistent with the photosynthetic growth and the RC complex peaks in absorption spectra (Figs. 3.16 and 3.17). The western blot data (Fig. 3.18) indicate that the absence of either RC M or RC L results in the disappearance of RC H from the membrane fraction, whereas RC M insertion into and steady state levels in the membrane are independent of RC H or RC L. The amount of the RC L protein in the membrane fraction was reduced when RC H was absent, and undetectable when RC M was absent. It appears that RC M and RC L interact in the membrane in the absence of RC H , and so the western blot results are consistent with previous publications that described extremely 95 low amounts of RC catalytic activity in membranes of RC H\" R. sphaeroides and R. rubrum cells (13, 64). Although it was previously shown that RC H and RC M proteins in the membrane are degraded if they are incapable of forming the RC holocomplex (68), the data in this thesis indicate that the rate of RC M synthesis and membrane insertion far exceeds the rate of degradation in the absence of RC H , because the steady-state amount of RC M in the membrane fraction is relatively constant irrespective of the presence of RC H. I also discovered that the amount of RC M in the membrane is independent of RC L (Fig. 3.18A). Thus RC M differs from RC H and RC L , which are absent or present at a low level when either of the other two RC proteins are not produced. These differences are attributed to degradation because the RC H gene transcription unit was identical in the plasmids pESHPUFM', pESHPUFL\" and pESHPUF, and because the deletion of RC M gene sequences in pESHPUFM\" is located 3' (transcriptionally downstream) of the RC L gene. The translationally in-frame deletion of the RC M gene in pESHPUFM\" should not destabilize the 5' RC L mRNA segment because deletions of RC M gene sequences in the closely related species Rhodobacter capsulatus did not decrease the stability of the RC L mRNA segment (33). The data in Fig. 3.18B indicate that there is a maximal accumulation of the extremely hydrophobic RC M and RC L proteins (but not the less hydrophobic RC H) in the cytoplasm when all three RC proteins are present in the membrane, although minute amounts of RC M and RC L proteins were detected in the soluble fraction of the RC H\" strain (compare Figs. 3.18A and 3.18B). These surprising results indicate that cells sense the presence of RC proteins in the membrane and somehow stabilize the hydrophobic RC 96 M and RC L proteins in a soluble form, the amounts of which roughly correlate to the amount of membrane-associated RC L. That is, the soluble forms of RC M and RC L were greatest when all of RC H, RC M and RC L were present, soluble RC L was not detected in the RC M\" strain, but small amounts of RC M and RC L were detected in the RC H\" strain. I use these collective results to propose the following model (Fig. 4.3) of the cellular accumulation of the three RC proteins for assembly of the RC complex in R. sphaeroides. 1) The synthesis of RC proteins increases in response to reductions in oxygen availability (48) with insertion of the three RC proteins into the membrane, followed by rapid degradation of RC H and RC L (but not RC M) i f one or more of the other RC proteins is absent (Fig. 4.3 A). 2) RC M and RC L proteins interact (Fig. 4.3B), although RC L is especially susceptible to degradation in the absence of RC H . 3) The RC H protein interacts with the RC M / L heterodimer, in part through the RC H transmembrane domain and RC M residues, and protects RC L from proteolysis (Fig 4.3C). 4) The formation of a relatively stable RC H / M / L assemblage results in an increase in the levels of soluble RC M and RC L that may increase the rate of RC complex assembly (Fig. 4.3C). Perhaps the presence of RC L in the membrane is sensed by one or more phototrophic bacterial proteins that have been implicated as assembly factors (3, 12, 72) to result in this unusual accumulation of the extremely hydrophobic RC M and RC L proteins in the soluble fraction of cells. 97 A RC L RC M RC H B C Accumulation of RC M and RC L in the cytoplasm Figure 4.3 Model for R. sphaeroides RC assembly in the lipid bilayer of the cellular plasma membrane (ICM). A Membrane insertion and degradation of RC L, RC M and RC H proteins, when synthesized individually under semi-aerobic conditions; upward arrows indicate insertion into the membrane, while downward arrows indicate degradation; dashed arrow indicates a slower rate of degradation for RC M protein. B The RC M and RC L proteins interact to form a heterodimer in the absence of RC H , while the RC L protein is especially susceptible to degradation. C When all three proteins are present the RC H protein interacts with the RC M / L heterodimer, yielding a holocomplex; the formation of this holocomplex results in the accumulation of RC M and RC L in the cytoplasm. 98 4.3 RC-LH1 (core complex) interactions Although the focus of this thesis is on the elucidation of RC H functions in the RC of R. sphaeroides, absorption spectra of a number of mutants revealed potential RC-LH1 (core complex) interactions. For example, APUHAAPUC cells containing a mutant RC H protein with a stretch of seven deleted periplasmic amino acids (encoded in plasmid pEAPeri), exhibited significantly lowered levels of LH1 (Fig. 3.9), whereas the use of a hybrid RC H protein with a substituted cytoplasmic domain (encoded in plasmid pEVCSP) increased LH1 levels (Fig. 3.4). Also note differences in the amplitude of the. LH1 absorption peak in Fig 3.17. As discussed above, these mutations also affected the amounts of individual RC proteins in the membrane, and the RC holocomplex. Thus, the observed effects on LH1 indicate that the presence of the RC complex or mutations in RC proteins affect the amounts of the LH1 complex. Previous work on several purple bacterial species indicated a relationship between the amounts of the RC and LH1 complexes (3, 13, 64, 72). It would be very interesting to use the mutants described in this thesis to carefully evaluate the nature of LH1 and RC interactions. 4.4 Future Directions The data presented in this thesis are generally compatible with the conclusions drawn. Although many minor questions arise, there are several outstanding questions that warrant further investigation. For example: 1) It is not proven whether the steady state amounts of RC H, RC M , and RC L proteins (as indicated by the western blots) really are the result of differential degradation, as opposed to changes in the rates of synthesis and/or membrane insertion. This question could be addressed through the use of RC mutants in pulse-chase experiments, to evaluate the relative rates of protein synthesis, membrane insertion and turnover for the individual RC proteins. A number of RC mutations resulted in changes to the steady-state levels of the LH1 complex. Therefore, it would be interesting to determine the kinetics of LH1 complex assembly for these mutants in absorption spectroscopy experiments on intact cells. If LH1 a and p antisera were obtained, the membrane and soluble fractions could be evaluated in western blots, and in pulse-chase immunoprecipitation experiments, to study the individual LH1 proteins in comparison to the LH1 holocomplex. The speculated interplay between RC and LH1 proteins and/or cofactors, and putative assembly factors begs for additional research. Perhaps a combination of the steady-state measurements that are described in this thesis, as well as pulse-chase experiments, could be used to evaluate the effects of mutations in genes encoding cofactor biosynthesis enzymes and putative assembly factors. It would be interesting to use the RC M and RC L antibodies in immunoprecipitation experiments on the soluble forms of RC M and RC L. Perhaps additional proteins would be co-precipitated and identified by use of mass spectrometry of peptides after trypsin. 100 digestion of SDS-PAGE bands, and searching of the R. sphaeroides genome (http://mmg.uth.tmc.edu/sphaeroides/index2.html) The results of these experiments, combined with the data reported in this thesis, could prove to be instrumental in bringing insight into not only the assembly and catalytic activity of purple bacterial RC and LH1 complexes, but also address fundamental questions about membrane protein complexes in general. 101 5. References 1. Abresch, E. C , M. H. B. Stowell, T. M. McPhillips, D. C. Rees, S. M. Soltis, M. L. Paddock, H. L. A. Axelrod, M. Y. Okamura, and G. Feher. 1998. Improved resolution of X-ray diffraction of crystals of reaction centers from Rb. sphaeroides: Light induced structural changes and elucidation of possible water (proton) channels. Photosyn. Res. 55:119-125. 2. Adelroth, P., M. L. Paddock, A. Tehrani, J. T. Beatty, G. Feher, and M. Y. Okamura. 2001. Identification of the proton pathway in bacterial reaction centers: decrease of proton transfer rate by mutation of surface histidines at HI 26 and H128 and chemical rescue by imidazole identifies the initial proton donors. Biochemistry 40:14538-14546. 3. Aklujkar, M., A. L. Harmer, R. C. Prince, and J. T. Beatty. 2000. The orfl62b sequence of Rhodobacter capsulatus encodes a protein required for optimal levels of photosynthetic pigment-protein complexes. J. Bacteriol. 182:5440-5447. 4. Allen, J. P., G. Feher, T. O. Yeates, H. Komiya, and D. C. Rees. 1988. Structure of the reaction centre from Rhodobacter sphaeroides R-26: protein-cofactor (quinones and Fe2+) interactions. Proc. Natl. Acad. Sci. U S A 85:8487-8491. 5. Axelrod, H. L., E. C. Abresch, M. L. Paddock, M. Y. Okamura, and F. Feher. 2000. Determination of the binding sites of the proton transfer inhibitors C d 2 + and Z n 2 + in bacterial reaction centers. Proc. Natl. Acad. Sci. U S A 97:1542-1547. 6. Bauer, C. E. 1999. Mechanisms for redox control of gene expression. Annu. Rev. Microbiol 53:495-523. 7. Bauer, C. E., and T. H. Bird. 1996. Regulatory circuits controlling photosynthesis gene expression. Cell 85:5-8. 8. Beatty, J. T., and H. Gest. 1981. Biosynthesis and bioenergetic functions of citric acid cycle reactions in Rhodopseudomonas capsulata. J. Bacteriol 148:584-593. 9. Blankenship, R. E. 2001. Molecular Mechanism of Photosynthesis. Blackwell Science, Oxford. 102 10. Brent, R., and M. Ptashne. 1980. The lexA gene product represses its own promoter. Proc. Natl. Acad. Sci. USA 77:1932-1936. 11. Chang, C.-H., O. El-Kabbani, D. Tiede, J . Norris, and M. Schiffer. 1991. Structure of the membrane-bound protein photosynthetic reaction centre from Rhodobacter sphaeroides. Biochemistry 30:5352-5360. 12. Chen, X.-Y., V. Yurkov, M. L. Paddock, M. Y. Okamura, and J . T. Beatty. 1998. A puhA deletion and plasmid complementation system for site directed mutagenesis studies of the reaction centre H protein of Rhodobacter sphaeroides. Photosyn. Res. 55:369-373. 13. Cheng, Y. S., C. A. Brantner, A. Tsapin, and M. L. P. Collins. 2000. Role of the H protein in assembly of the photochemical reaction centre and intracytoplasmic membrane in Rhodospirillum rubrum. J. Bacteriol. 182:1200-1207. 14. Cogdell, R. J . , N. W. Isaacs, T. D. Howard, K. McLuskey, N. J . Fraser, and S. M. Prince. 1999. How photosynthetic bacteria harvest solar energy. J. Bacteriol. 181:3869-3879. 15. Cogdell, R. J . , and J . P. Thornber. 1980. Light-harvesting pigment-protein complexes of purple photosynthetic bacteria. FEBS Lett. 122:1-8. 16. Conroy, M. J . , W. H. J . Westerhuis, P. S. Parkes-Loach, P. A. Loach, C. N. Hunter, and M. P. Williamson. 2000. The solution structure of Rhodobacter sphaeroides LH1(3 reveals two helical domains separated by a more flexible region: Structural consequences for the LH1 complex. J. Mol . Biol. 298:83-94. 17. Debus, R. J . , G. Feher, and M. Y. Okamura. 1985. L M complex of reaction centers from Rhodopseudomonas sphaeroides R-26: characterization and reconstitution with the H subunit. Biochemistry 24:2488-2500. 18. Deisenhofer, J . , K. M. O. Epp, R. Huber, and H. Michel. 1985. Structure of the protein subunits in the photosynthetic reaction centre of Rhodopseudomonas viridis at 3A resolution. Nature (London) 318:618-624. 19. Ditta, G., T. Schmidhauser, E. Yakobsen, P. Lu, X.-W. Liang, D. R. Finlay, D. Guiney, and D. R. Helinski. 1985. Plasmids related to the broad host range vector, pRK290, useful for gene cloning and for monitoring gene expression. Plasmid 13:149-153. 103 20. Drews, G., and J . R. Golecki. 1995. Structure, molecular organization, and biosynthesis of membranes of purple bacteria, p. 231-257. In R. E. Blankenship, M . T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 21. Dutton, P. L., K. M. Petty, H. S. Bonner, and S. D. Morse. 1975. Cytochrome c2 and reaction centre of Rhodopseudomonas sphaeroides Ga. membranes. Extinction coefficients, content, half-reduction potentials, kinetics and electric field alterations. Biochim. Biophys. Acta 387:536-556. 22. Eraso, J . M., and S. Kaplan. 1994. prrA, A putative response regulator involved in oxygen regulation of photosynthesis gene expression in Rhodobacter sphaeroides. J. Bacteriol:32-43. 23. Feher, G., J . P. Allen, M. Y. Okamura, and D. C. Rees. 1989. Structure and function of bacterial photosynthetic reaction centers. Nature 339:111-116. 24. Fejes, A., E. C. Y i , D. R. Goodlett, and J . T. Beatty. 2003. Shotgun proteomic analysis of chromatophores from the purple phototrophic bacterium Rhodopseudomonas palustris. Photosynth Res in press. 25. Gest, H. 1993. Photosynthetic and quasi-photosynthetic bacteria. FEMS Microbiol. Lett. 112:1-6. 26. Gong, L. M., and S. Kaplan. 1996. Translational control of puf operon expression in Rhodobacter sphaeroides 2.4.1. Microbiology 142:2057-2069. 27. Imhoff, J . F. 1995. Taxonomy and physiology of phototrophic purple bacteria and green sulfur bacteria, p. 1-15. In R. E. Blankenship, M . T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 28. Jackson, J . B., and A. R. Crofts. 1969. The high energy state in chromatophores from Rhodopseudomonas sphaeroides. FEBS Lett. 4:185-189. 29. Kaplan, S. 2002. Photosynthesis genes and their expression in Rhodobacter sphaeroides 2.4.1. Photosynth. Res 73:95-108. 30. Keen, N. T., S. Tamaki, D. Kobayashi, and D. Trollinger. 1988. Improved broad-host-range plasmids for D N A cloning in gram-negative bacteria. Gene 70:191-197. 104 31. Keller, S., J . T. Beatty, M. Paddock, J . Breton, and W. Leibl. 2001. Effect of metal binding on electrogenic proton transfer associated with reduction of the secondary electron acceptor (QB) in Rhodobacter sphaeroides chromatophores. Biochemistry 40:429-439. 32. Kiley, P., and S. Kaplan. 1987. Cloning, D N A sequence, and expression of the Rhodobacter sphaeroides light-harvesting B800-850-a and B800-850-C3 genes. J. Bacteriol. 169:3268-3275. 33. Klug, G., and S. N . Cohen. 1990. Combined actions of multiple hairpin loop structures and sites of rate-limiting endonucleolytic cleavage determine differential degradation rates of individual segments within polycistronic puf operon mRNA. J. Bacteriol. 172:5140-5146. 34. Koepke, J . , X. Hu, C. Muenke, K. Schulten, and H. Michel. 1996. The crystal structure of the light-harvesting complex II (B800-850) from Rhodospirillum molischianum. Structure 4:581-597. 35. Laemmli, D. M. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. 36. Lancaster, C. R. D., U. Ermler, and H. Michel. 1995. The structures of photosynthetic reaction centers from purple bacteria as revealed by X-ray crystallography, p. 503-526. In R. E. Blankenship, M . T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 37. Lee, J . K., P. J . Kiley, and S. Kaplan. 1989. Posttranscriptional control of puc operon expression of B800-850 light-harvesting complex formation in Rhodobacter sphaeroides. J. Bacteriol. 171:3391-3405. 38. Lilburn, T. G., R. C. Prince, and J . T. Beatty. 1995. Mutation of the Ser2 codon of the light-harvesting B870 a polypeptide of Rhodobacter capsulatus partially suppresses the pufXphenotype. J. Bacteriol. 177:4593-4600. 39. Loach, P. A. 2000. Supramolecular complexes in photosynthetic bacteria. Proc. Natl. Acad. Sci. USA 97:5016-5018. 40. Madigan, M. T., J . M. Martinko, and J . Parker. 2000. Brock Biology of Microorganisms. Prentice Hall, Upper Saddle River, NJ. 41. Madigan, M. T., and J . G. Ormerod. 1995. Taxonomy, physiology and ecology of heliobacteria, p. 17-30. In R. E. Blankenship, M . T. Madigan, and C. E. Bauer 105 (ed.), Anoxygenic Photosynthetic Bacteria. Kluwer Academic Publishers, Dordrecht. 42. Masuda, S., and C. E. Bauer. 2002. AppA Is a Blue Light Photoreceptor that Antirepresses Photosynthesis Gene Expression in Rhodobacter sphaeroides. Cell 110:613-623. 43. McDermott, G., S. M. Prince, A. A. Freer, A. M. Hawthornthwaite-Lawless, M. Z. Papiz, R. J . Cogdell, and N. W. Isaacs. 1995. Crystal structure of an integral membrane light-harvesting complex from photosynthetic bacteria. Nature (London) 374:517-521. 44. Messing, J . 1983. New M13 vectors for cloning. Methods in Enzymology 101:20-78. 45. Michel, H. 1982. Three-dimensional crystals of a membrane protein complex. The photosynthetic reaction centre from Rhodospeudomonas viridis. J. Mol . Biol. 158:567-572. 46. Naylor, G. W., H. A. Addlesee, L. C. D. Gibson, and C. N. Hunter. 1999. The photosynthesis gene cluster of Rhodobacter sphaeroides. Photosynth. Res. 62:121-139. 47. Oh, J.-I., and S. Kaplan. 1999. The cbb(3) terminal oxidase of Rhodobacter sphaeroides 2.4.1: Structural and functional implications for the regulation of spectral complex formation. Biochemistry 38:2688-2696. 48. Oh, J.-I., and S. Kaplan. 2001. Generalized approach to the regulation and integration of gene expression. Molec. Microbiol. 39:1116-1123. 49. Oh, J.-I., and S. Kaplan. 2000. Interacting Regulatory Circuts Involved in Orderly Control of Photosynthesis Gene Expression in Rhodobacter sphaeroides 2.4.1. J. Bacteriol:3081-3087. 50. Oh, J.-I., and S. Kaplan. 2000. Redox signaling: globalization of gene expression. E M B O J. 19:4237-4247. 51. Okamura, M. Y., M. L. Paddock, M. S. Graige, and G. Feher. 2000. Proton and electron transfer in bacterial reaction centers. Biochim. Biophys. Acta 1458:148-163. 106 52. Paddock, M. L., P. Adelroth, G. Feher, M. Y. Okamura, and J . T. Beatty. 2002. Determination of proton transfer rates by chemical rescue: applications to bacterial reaction centers. Biochemistry 41:14716-14725. 53. Paddock, M. L., M. S. Graige, and M. Y. Okamura. 1999. Identification of the proton pathway in bactyerial reaction centers: inhibition of proton transfer by binding of Z n 2 + or C d 2 + . Proc. Natl. Acad. Sci. USA 96:6183-6188. 54. Paddock, M. L., P. H. McPherson, G. Feher, and M. Y. Okamura. 1990. Pathway of proton transfer in bacterial reaction centers: replacement of serine-L223 by alanine inhibits electron and proton transfers associated with reduction of quinone to dihydroquinone. Proc. Nat. Acad. Sci. U S A 87:6803-6807. 55. Paddock, M. L., S. H. Rongey, G. Feher, and M. Y. Okamura. 1989. Pathway of proton transfer in bacterial reaction centers: replacement of glutamic acid 212 in the L subunit by glutamine inhibits quinone (secondary acceptor) turnover. Proc. Nat. Acad. Sci. U S A 86:6602-6606. 56. Paddock, M. L., L. Sagle, A. Tehrani, J . T. Beatty, G. Feher, and M. Y. Okamura. 2003. Mechanism of proton transfer inhibition by Cd 2+ binding to bacterial reaction centers: Determination of the p K A of functionally important histidine residues. Biochemistry 42:9626-9632. 57. Peterson, G. 1983. Determination of total protein. Meth. Enzymol. 91: 95-119. 58. Pierson, B. K., and R. W. Castenholz. 1995. Taxonomy and physyilolgy of filamentious anoxygenic phototrophs., p. 31-47. In R. E. Blankenship, M . T. Madigan, and C. E. Bauer (ed.), Anoxygenic Photosynthetic Bacteria. Kluwer Academic Publishers, Dordrecht. 59. Prince, R. C. 1986. Light-driven electron and proton transfer in the cytochrome b/c, complex, p. p. 539-546. In L . A . Staehelin and e. C. J. Arntzen (ed.), Encyclopedia of plant physiology, vol. new series, 19. Springer-Verlag, Berlin. 60. Sambrook, J . , E. F. Fritsch, and T. Maniatis. 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N . Y . 61. Schagger, H., and G. von Jagow. 1987. Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Analyt. Biochem. 166:368-379. 107 62. Sganga, M. W., and C. E. Bauer. 1992. Regulatory Factors Controlling Photosynthetic Reaction Centre and Light-Harvesting Gene Expression in Rhodobacter capsulatus. Cell 68:945-954. 63. Simon, R., U. Priefer, and A. Punier. 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Bio/Technology 1:37-45. 64. Sockett, R. E., T. J . Donohue, A. R. Varga, and S. Kaplan. 1989. Control of photosynthetic membrane assembly in Rhodobacter sphaeroides mediated by puhA and flanking sequences. J. Bacteriol. 171:436-446. 65. Stowell, M. H., T. M. McPhillips, D. C. Rees, S. M. Soltis, E. Abresch, and G. Feher. 1997. Light-induced structural changes in photosynthetic reaction centre: implications for mechanism of electron-proton transfer. Science 276:812-816. 66. Sundstrom, V., and R. v. Grondelle. 1995. Kinetics of excitation transfer and trapping in purple bacteria, p. 349-372. In R. E. Blankenship, M . T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 67. Takahashi, E., and C. A. Wraight. 1992. Proton and electron transfer in the acceptor quinone complex of Rhodobacter sphaeroides reaction centers -characterization of site-directed mutants of the two ionizable residues, Glu-L212 and Asp-L213, in the Qs-binding region. Biochemistry 31:855-866. 68. Varga, A. R., and S. Kaplan. 1993. Synthesis and stability of reaction centre polypeptides and implications for reaction centre assembly in Rhodobacter sphaeroides. J. Biol. Chem. 268:19842-19850. 69. Vermeglio, A., and P. Joliot. 2002. Supramolecular organisation of the photosynthetic chain in anoxygenic bacteria. Biochim. Biophys. Acta 1555:60-64. 70. Weaver, P. F., J . D. Wall, and H. Gest. 1975. Characterization of Rhodopseudomonas capsulata. Arch. Microbiol. 105:207-216. 71. Woese, C. R. 1987. Bacterial evolution. Microbiol. Rev. 51:221-271. 72. Wong, D. K.-H., W. J . Collins, A. Harmer, T. G. Lilburn, and J . T. Beatty. 1996. Directed mutagenesis of the Rhodobacter capsulatus puhA gene and pleiotropic effects on photosynthetic reaction centre and light-harvesting I complexes. J. Bacteriol. 178:2334-2342. 108 73. Hu, X., Ritz, T., Damjanovic, A., Autenrieth, F. and Schulten, K. 2002. Photosynthetic apparatus of purple bacteria. Quart. Rev. of Biophysics 35:1-62. 74. Zannoni, D. 1995. Aerobic and anaerobic electron transport chains in anoxygenic phototrophic bacteria, p. 949-971. In R. E. Blankenship, M . T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 75. Zeilstra-Ryalls, J . , M. Gomelsky, J . M. Eraso, A. Yeliseev, J . O'Gara, and S. Kaplan. 1998. Control of photosystem formation in Rhodobacter sphaeroides. J. Bacteriol. 180:2801-2809. 109 APPENDIX 1: Amino acid alignment of R. sphaeroides and B. viridis RC H proteins Cytoplasmic domain sequences • Bv.puh.aa MYHGALAQHL DIAQLVWYAQ WLVIWTWLL YLRREDRREG gPUVEPLgLV 5 0 Rs.puh.aa MVGVTAFGNF DLASLAIYSF W-IFLAGLIY YLQTENMREG JYPIJENEDJG-- 4 7 Bv.puh.aa K L | P E D § Q V Y ELgYPKfglvS JPHG-^VJTVB; RRRP|TlELK S S Q | D G F E G A 99 Rs.puh.aa T P § A N Q ( 3 P - F PfcgKp.KTFjlB raGR|lLfv| G P E S I D | P I A £ 3 5 R | A V S | G F 96 Bv.puh.aa gLQJPTGNgLV BAVGPASYAE gAEWDATVD §KAKWgLRV §TDFjSIAE(3D 14 9 Rs.puh.aa gHAfTCDgMK gGyGPAS.WVA fRDLPELDGH JGHNOKgMKA § A G | H V S A S K 14 6 B v . p u h . a a V D g R G L P J V A AgGVEA^T jV jT g L W V D R S g H Y F § Y j L | L S V A G S A R g j A J D l g L G 1 9 9 R s . p u h . a a - N g i ; G L P V ( R G C g L E I ^ K ^ V g l ^ V E i l p l Q M A S F L E V E L K D GS-gJRJDLgMQ 194 Bv.puh.aa FCD^KKDKIV P ' S I L S E Q F S NVgRLQgRDQ Iffl]REEBKVS AyiYAGGL|LJg| 2 4 9 Rs.puh.aa MYK^QSNRVH j ^ N A L S g D L g s ] GIgTIK§PTE vjfELEEDRIC GMVAGGIiM|Yl 2 4 4 Bv.puh.aa TgEI? AES LO 2 5 8 Rs.puh.aa AgKgKSWAA M S A E Y A 2 6 0 1 APPENDIX 2: Absorption spectra of APUHAAPUC (LH2\") intact cells containing plasmid pATSHR or pEVtrans (see Fig 3.7) Time Ohrs 18 hrs 25 hrs 70 hrs 640 760 880 nm 640 760 880 nm A P U H A A P U C (pATSHR) A P U H A A P U C (pEVtrans) I l l APPENDIX 3 : Publications arising from this thesis research 1. Adelroth, P., M. L. Paddock, A. Tehrani, J . T. Beatty, G. Feher, and M. Y. Okamura. 2001. Identification of the proton pathway in bacterial reaction centers: decrease of proton transfer rate by mutation of surface histidines at HI26 and HI28 and chemical rescue by imidazole identifies the initial proton donors. Biochemistry 40:14538-14546 2. Paddock, M. L., L. Sagle, A. Tehrani, J . T. Beatty, G. Feher, and M. Y. Okamura. 2003. Mechanism of proton transfer inhibition by Cd 2+ binding to bacterial reaction centers: Determination of the p K A of functionally important histidine residues. Biochemistry 42:9626-9632 3. Ali Tehrani, Roger C . Prince, and J . Thomas Beatty. 2003 Effects of Phtosynthetic Reaction Center H Protein Domain Mutations on Photosynthetic Properties and Reaction Center Assembly in Rhodobacter sphaeroides. Biochemistry 42:8919-8928 4. Al i Tehrani and J . Thomas Beatty. Effects of Precise Deletions in Rhodobacter sphaeroides Reaction Center Genes on Steady-State Levels of Reaction Center Proteins: A Revised Model for Reaction Center Assembly. Photosynth. Res. (in-press) "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2003-11"@en ; edm:isShownAt "10.14288/1.0091785"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Microbiology and Immunology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Structure-based functional studies of the rhodobacter sphaeroides reaction centre H protein"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/16097"@en .