Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of the pufX protein in photosynthetic electron transfer Lilburn, Timothy George 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-060012.pdf [ 6.69MB ]
Metadata
JSON: 831-1.0088259.json
JSON-LD: 831-1.0088259-ld.json
RDF/XML (Pretty): 831-1.0088259-rdf.xml
RDF/JSON: 831-1.0088259-rdf.json
Turtle: 831-1.0088259-turtle.txt
N-Triples: 831-1.0088259-rdf-ntriples.txt
Original Record: 831-1.0088259-source.json
Full Text
831-1.0088259-fulltext.txt
Citation
831-1.0088259.ris

Full Text

THE ROLE OF THE PUFX PROTEIN IN PHOTOSYNTHETIC ELECTRON TRANSFER by TIMOTHY GEORGE LILBURN  A THESIS SUBMITTED IN PARTIAL FULFILLMENT 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 May 1995 © Timothy George Lilburn, 1995  In presenting this  thesis  in  partial fulfilment  of  the  requirements  for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of department  or  by  his  or  her  representatives.  It  is  understood  that  copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of  Microbiology  The University of British Columbia Vancouver, Canada  Date June 2 3 , 1995  DE-6 (2/88)  ii  A BSTRA  CT  The role of the puJX gene product of Rhodobacter capsulatus has been unclear, but deletion of the gene renders the organism incapable of photosynthetic growth on a minimal medium. However, suppressor mutants that are able to grow photosynthetically are readily isolated. Two such suppressor mutants were characterized as to their photosynthetic growth properties, their fluorescence at five different incident light intensities, the integrity of their chromatophores and their abilities to generate a trans-membrane potential. I found that the photosynthetic apparatus in the suppressor mutants was less stable than that of the pseudo-wild type and primary mutant strain and that the suppressor mutants used light energy less efficiently than the pseudo-wild type strain. Therefore, the suppressor strains are more correctly designated partial suppressor mutants. The locations and sequences of the suppressor mutations were determined and both were found to change the second codon of the pufA gene. It is hypothesized that the serine residue encoded by this codon is important in interactions between the B870 antenna complex and other membrane-bound polypeptides. Based on these results and those of other investigators, a model for the role of the PufX protein is put forward in which the PufX protein, in conjunction with the B870 ft polypeptide, is proposed to form a gateway in the light harvesting antenna complex surrounding the reaction center. This gateway would allow the passage of quinones and quinols between the reaction center and the cytochrome blc\ complex.  iii  ABSTRACT  ii  LIST OF TABLES  vii  LIST OF FIGURES  viii  ACKNOWLEDGEMENTS  x  1. I N T R O D U C T I O N  1  1.1. Photosynthesis  1  1.2. Rhodobacter capsulatus  2  1.3. The photosynthetic unit  4  1.3.1. Complexes comprising the photosynthetic unit  4  1.3.2. Energy capture by the photosynthetic unit  7  1.3.3. Regulation of expression of photosynthesis genes  12  1.4. The pufKgene  15  1.5. Introduction references  16  2. MATERIALS AND M E T H O D S  25  2.1. Bacterial strains  25  2.2. Construction of plasmids  25  2.3. Mapping and sequencing of suppressor mutations  30  2.4. Growth experiments  30  2.5. Spectroscopy experiments  31  2.5.1. Absorption spectra  31  2.5.2. Flash spectroscopy  31  iv  2.5.3. Reduced minus oxidized difference spectra  32  2.5.4. Chemically-induced and light-induced carotenoid bandshift spectra  32  2.6. Fluorescence measurements  33  2.7. Fractionation of the photosynthetic unit  33  2.7.1. Preparation of chromatophores  33  2.7.2. Isolation of reaction center-B870 and B800-850 complexes  34  2.8. SDS-PAGE  35  2.9. Cytochrome blc\ complex activity assay  36  2.10. Electron microscopy  37  2.11. Polyclonal antibodies to the PufX protein  37  2.12. Over-expression of pufX.  38  2.13. Other assays  39  2.14. Materials and methods references  40  3. RESULTS  44  3.1. The primary mutant 3.1.1. The pufX' phenotype: The effects of the loss of PufX  44 44  3.1.1.1. Growth experiments  44  3.1.1.2. Single-flash spectroscopy  49  3.1.1.3. Quantitation of quinones  50  3.1.1.4. Assay of blc\ complex activity  51  V  3.1.2. Immunological detection of PufX  52  3.1.2.1. Over-expression of PufX  52  3.1.2.2. Polyclonal antibodies to a synthetic peptide based on the PufX sequence  53  3.1.3. Fractionation of the photosynthetic unit  53  3.1.3.1. Electrophoretic fractionation  53  3.1.3.2. Density gradient fractionation  54  3.2. The suppressor mutants  56  3.2.1. Suppressor phenotypes  56  3.2.1.1. Growth experiments  56  3.2.1.2. Fluorescence detection: The efficiency of light energy trapping  65  3.2.1.3. Flash spectroscopy: Monitoring photosynthetic electron transport  66  3.2.1.4. Reduced minus oxidized difference spectroscopy: The relative levels of b- and ctype cytochromes  71  3.2.1.5. Flash spectroscopy: Monitoring the formation of a trans-membrane potential  73  3.2.1.6. Light minus dark difference spectroscopy: The light-induced generation of a membrane potential over a longer time frame  78  3.2.1.7. Transmission electron microscopy of negatively stained chromatophores  78  vi  3.2.1.8. Ion gradient- and continuous light-induced carotenoid bandshifts  81  3.2.1.9. Flash spectroscopy: Monitoring the orientation of the reaction centers in chromatophores 3.2.2. The location of the suppressor mutations 3.3. Results references 4. DISCUSSION  81 88 90 92  4.1. Growth experiments  93  4.2. Fluorescence detection experiments  97  4.3. Single flash spectroscopy  97  4.4. Multiple flash spectroscopy experiments and the integrity of chromatophores  99  4.5. Location and sequences of the plasmid-borne suppressor mutations..101 4.6. A model for the role of the PufX protein  101  4.7. Discussion references  107  VII  LIST  OF  TABLES  TABLE 3.1. Cytochrome blc\ complex specific activity measured in two strains of R. capsulatus.  51  TABLE 3.II. Measurement of photooxidizable c-type cytochromes  70  TABLE 3.III. Measurement of total b- and c-type cytochromes  73  TABLE 3.IV. The generation of a transmembrane potential in four strains of R. capsulatus 75  viii  LIST  OF  FIGURES  Figure 1.1. Light energy transduction in R. capsulatus.  9  Figure 1.2. Electron transfer pathways in R. capsulatus.  11  Figure 1.3. Photosynthesis gene cluster of R. capsulatus  14  Figure 2.1. The construction of plasmids pTB999, pTL2 and pTL30  26  Figure 2.2. The construction of plasmid pTL6  28  Figure 3.1. Light titration experiments on the pseudo-wild type and primary mutant strains  45  Figure 3.2. Photosynthetic growth of the pseudo-wild type and primary mutant strains on minimal, rich and blended media  46  Figure 3.3. Photosynthetic growth of the pseudo-wild type and primary mutant strains on minimal medium and media supplemented with yeast extract  47  Figure 3.4. Photosynthetic growth on DMSO-supplemented medium  48  Figure 3.5. Single flash carotenoid bandshift experiments  50  Figure 3.6. Gel fractionation results  53  Figure 3.7. Density gradient fractionation of detergent solubilized chromatophores  54  Figure 3.8. Monitoring the progress of the purification of the B800-850 lightharvesting and reaction center-B870 light-harvesting complexes with absorption spectroscopy  57  Figure 3.9. SDS-PAGE of samples taken at successive steps in the purification of the B800-850 light-harvesting and reaction center-B870 lightharvesting complexes  59  ix  Figure 3.10. Light titration and photosynthetic growth of five strains of R. capsulatus  60  Figure 3.11. Absorption spectra from photosynthetic cultures grown at two light intensities  61  Figure 3.12 Photosynthetic growth of five strains of/?, capsulatus on minimal medium and media supplemented with yeast extract  63  Figure 3.13. Growth of the crtD strains  64  Figure 3.14. Light-trapping efficiency experiment  65  Figure 3.15. Flash spectroscopy experiments: The bacteriochlorophyll special pair  67  Figure 3.16. Flash spectroscopy experiments: The c-type cytochromes  69  Figure 3.17. Flash spectroscopy experiments: The b-typt cytochromes  72  Figure 3.18. Flash spectroscopy experiments: The carotenoid bandshift  74  Figure 3.19. The steady-state, light-induced carotenoid bandshift  76  Figure 3.20. Transmission electron microscopy of chromatophores  79  Figure 3.21. A comparison of the trans-membrane potentials induced by light and ion gradients Figure 3.22. The orientation of reaction centers  82 85  Figure 3.23. The effect of exogenous cytochrome con the carotenoid bandshift 87 Figure 3.24. Photosynthetic growth of/?, capsulatus ARC6 strains transformed with plasmids from the suppressor mutants  88  Figure 3.25. Mapping and location of the suppressor mutations  89  Figure 4.1 Suppressor mutations in the B870 antenna complex  103  X  ACKNOWLEDGEMENTS This thesis is dedicated to Leslie Fiddler, whose loving support made it all possible. I am deeply grateful to Tom Beatty for the opportunity to do this research and for his ability to give good advice when it is most needed. Thanks also to Roger Prince, who not only taught me the ins and outs of spectroscopy, but also showed a lively sense of hospitality when I visited his lab and home. Finally, I thank my committee members, Julyet Benbasat, Phil Bragg and Tony Warren, for their guidance, and the members of the Beatty lab, especially Mary Forrest, Cheryl Wellington, Farahad Dastoor, Heidi Leblanc and Conan Young, for their help and advice over the years.  1  1.  INTRODUCTION 1.1. Photosynthesis Photosynthesis is the process whereby an organism is able to convert radiant energy, usually from the sun, to chemical energy that can be used for the growth and maintenance of the organism. If we accept this definition of photosynthesis, then true photosynthesis is found only in the Eucarya and the Bacteria (89), in which the conversion of light energy to chemical energy is accomplished by an electron transport pathway. In the Archaea a second type of mechanism is found in which the conversion of light energy to chemical energy is accomplished by a lightdriven trans-membrane proton pump, specifically by bacteriorhodopsin in the halobacteria (84). Since the ability of halobacteria to harvest sufficient energy for growth via this mechanism is a matter of debate, they are perhaps best classified as "quasi-photosynthetic" bacteria (32) and will not be considered further. There are two types of electron-transport mediated photosynthesis: oxygenic and anoxygenic. In both types, electron transport is initiated when a (bacterio)chlorophyll molecule absorbs light energy which creates a strong low-potential reductant. In oxygenic photosynthesis, the electron transport chain is linear; water is indirectly oxidized by the photooxidized chlorophyll P680 of the Photosystem II complex and electrons are transferred to the terminal electron acceptor NADP + via the cytochrome b^f^nd  Photosystem I complex (16).  Oxygenic photosynthesis is found in the Eucarya and the Bacteria, albeit in only one of the ten phylogenetic divisions of the Bacteria proposed by Woese (94), the cyanobacteria and chloroplasts. In anoxygenic photosynthesis, electron transport is cyclic; the pathway of electron transfer from the photo-oxidized bacteriochlorophyll returns to the bacteriochlorophyll molecule via the cytochrome blc\ complex (see below). However, a further distinction can be made amongst the anoxygenic phototrophs. One group, which includes the phototrophs from two divisions of the Bacteria (the Proteobacteria and the Green non-sulfur bacteria and relatives), uses quinones as the primary and secondary electron acceptors in a reaction center that is homologous with the Photosystem II reaction center. The light energy captured is mainly used to make ATP; if reducing power is needed it can be supplied by H2, a reduced sulfur compound or by energy-linked reversed electron transport (72). The second group, whose members are from the Green sulfur bacteria division of the Bacteria, uses an iron-sulfur protein as a secondary electron acceptor (34, 72) in a reaction center that is homologous with  2 the Photosystem I reaction center (34). The mid-point potential of the iron-sulfur acceptor is low enough to allow the direct reduction of NAD + , so there is a branched light-driven electron transport pathway in these organisms in which electrons may return to the reaction center or they may be used to reduce NAD + . Although the existence of anoxygenic photosynthesis was first proposed by Engelmann in 1883, who worked with enriched cultures of purple photosynthetic bacteria, it was first proven by Molisch in 1907 (30, 31). However, anoxygenic photosynthesis was not widely accepted and remained a matter of debate until the definitive experiments were carried out by van Niel in 1932. Using pure cultures of green and purple sulfur bacteria in minimal media, van Niel was able to make stoichiometric measurements of H2S oxidation and CO2 assimilation. The assimilation equation derived from these experiments was remarkably similar to that of green plants and led van Niel to propose a general formula for the lightdependent transfer of hydrogen, from an appropriate donor, H2A, to CO2: C 0 2 + 2H 2 A -» <CH20> + 2A + H 2 0 In van Niel's view, oxygenic photosynthesis was simply a special case of the above general formula, and the unifying feature of all types of photosynthesis was that the process resulted in the fixation of CO2. This hypothesis did not accommodate Molisch's conclusion that the purple photosynthetic bacteria were able to use organic carbon sources for growth (30), and the reduction of CO2 is still often presented as the end result or "purpose" of photosynthesis (6). Gest proposed that the actual unifying aspect of photosynthesis is photophosphorylation, that is, the conversion of light energy to a form that may be used for enzymatic reactions in the cell  (3D. Anoxygenic photosynthesis is found exclusively in the Bacteria, in four of the divisions: the Gram-positives, the green sulfur bacteria, the green non-sulfur bacteria and relatives and the Proteobacteria (82). The wide distribution of the photosynthesis "phenotype" among phylogenetically diverse organisms indicates its ancient origin (71), although it must have arisen after the Bacteria lineage diverged from the lineages that gave rise to the Eucarya and the Archaea. 1.2. Rbodobacter capsulatus Rhodobacter capsulatus is a Gram negative, rod-shaped phototrophic bacterium, from the a subdivision of the Proteobacteria (82, 94). This species of bacterium is among the most  3 metabolically versatile organisms known. Photosynthetically, it can grow both as a heterotroph and as an autotroph. It can grow organotrophically both aerobically and anaerobically, in the latter case by use of anaerobic respiration and substrate level phosphorylation. Finally, this species is capable of lithotrophic growth by respiration of bimolecular hydrogen (59). One consequence of the versatility of R. capsulatus is that it is very widespread in nature. A second consequence of the metabolic versatility of R. capsulatus is that it is a good model organism for the study of a variety of biological problems. Mutations in genes whose products are essential for photosynthetic growth are not lethal to R capsulatus , as they would be for plants, since other modes of growth are available to the bacterium. The mutated photosynthetic apparatus can be induced in chemotrophic cultures of R. capsulatus by lowering the oxygen concentration, and the performance of the mutated apparatus evaluated noninvasively using spectroscopic techniques. Induction of the photosynthetic apparatus is accompanied by the formation of invaginations in the inner membrane that contain the photosynthetic apparatus. Thus, R. capsulatus can serve as a model for the study of membrane biogenesis. As a final example, this bacterium can also serve as an excellent model for the study of basic aspects of electron transport and associated translocation. As mentioned above, the oxidation and reduction of individual electron transport components can be monitored spectrophotometrically. Additionally, the oxidation and reduction turnover of these components can be initiated by shining light on cells, with very brief pulses of light initiating single turnovers of the electron transport pathway. The utility of R. capsulatus as a model organism is enhanced by the relative ease with which genetic analysis and techniques can be applied (24). Not only can the more common plasmid-based techniques for gene deletion and complementation or for gene replacement be used, but also a genetic exchange system unique to R. capsulatus, the gene transfer agent (GTA) (61), is available. The GTA is a virus-like particle that packages uniformly sized, 4.5 kb linear pieces of host DNA and, in a transduction-like process, transfers them to a recipient strain. The genes encoding the GTA proteins have not been discovered. The DNA packaged by the GTA seems to be randomly chosen from the R. capsulatus genome (including plasmids) making the GTA useful for gene deletion or replacement. Genetic studies of R. capsulatus are facilitated by the existence of a high resolution physical and genetic map of its chromosome (27).  4 1.3. The photosynthetic unit A switch from aerobic chemotrophic growth to anaerobic photosynthetic growth in R. capsulatus results in the formation of extensive invaginations of the inner membrane (74). These invaginations contain the components of the photosynthetic apparatus, which are collectively known as the photosynthetic unit, and include a chain of cofactors that mediate the electron transfer pathway of photosynthesis. The photosynthetic unit is often understood to be composed of two types of integral membrane polypeptide complexes: the light-harvesting antenna complexes (B800-850 and B870 complexes), and the reaction center, which catalyzes the reduction of a quinone (ubiquinone 10) on the cytoplasmic side of the membrane and the oxidation of a f-type cytochrome on the periplasmic side of the membrane, using the energy captured by the antenna complexes. In this thesis, the photosynthetic unit will also be considered to include the ubiquinohcytochrome c oxidoreductase {blc\ complex), since this complex is necessary for photosynthetic electron transport and the formation of a proton gradient across the inner membrane of the cell. Electron transfer between the reaction center and the blc\ complex is carried out by quinols and f-type cytochromes (72). A variety of experimental and theoretical data are consistent with the view that in wild type cells of phototrophic Proteobacteria these three types of complexes are organized in the membrane in such a way as to enhance functional interactions with each other (44, 102). Electron microscopy data and detergent fractionation studies indicate that reaction centers and B870 antenna complexes form stable associations (see below). Experiments have been designed to distinguish between a fluid state model of reaction center-^/q complex interactions, in which all the complexes can interact with each other via the free diffusion of quinols and cytochromes through the membrane, and a solid state model, in which the numbers of interacting complexes are limited by the restricted diffusion of the electron carriers. The results have been interpreted to show that so-called supercomplexes, consisting of two reaction centers, a blc\ complex and a cytochrome c, exist in R. sphaeroides (43, 54). There is also biophysical evidence for two separate populations of blc\ complex, one associated with reaction centers and one with the cytochrome oxidase (55). 1.3.1 Complexes comprising the photosynthetic unit The light-harvesting antenna complexes function to gather light energy and funnel it to the reaction center. Amongst the phototrophic Proteobacteria two types of light-harvesting antenna complex are found. Common to all is the so-called "core" complex (13), which is  5 thought to surround the reaction center. In R. capsulatus the core complex has a near infrared absorption maximum at a wavelength of approximately 870 nm and it is consequently known as the B870 light-harvesting complex. The second type of complex is known as the "peripheral" complex. It is not found in Rhodospirillum (Rsp.) rubrum or Rhodopseudomonas (Rps.)viridis (101), but does occur in two forms in the other phototrophic Proteo bacteria (13). In R. capsulatus a. peripheral antenna complex has two near-infrared absorption maxima, at wavelengths of about 800 and 850 nm and is called the B800-850 light-harvesting antenna complex. The basic unit of the B870 complex consists of two polypeptides, the a and /J polypeptides, which bind two bacteriochlorophyll a molecules and a carotenoid molecule (102). The deduced molecular weights of the a and /3 polypeptides are 6.6 and 5.3 kilodaltons (kDa) respectively (97), but their M r values on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels are 12 and 7 kDa respectively. The B870 complex is found in a fixed stoichiometry relative to the reaction center which, as mentioned above, it is thought to surround. There are estimated to be about 16 ccfi dimers around the reaction center (45). The Oc(3 dimers are not the minimal functional subunit of the B870 complex. Through the use of detergents, the purified B870 complex from R. capsulatus has been dissociated into a a/J-2bchl a dimer, which was found to have an absorption maximum at 816 nm (38). Reduction of the detergent concentration by dialysis allowed reformation of a B870 complex that was spectroscopically almost indistinguishable from the native B870 complex (91). Indeed, the dissociation has been taken a step further. Loach and coworkers have been able to use homogenous a and f5 polypeptides from R capsulatus and pigments to assemble B870 complexes in vitro (58). The "B816" dimers were an intermediate in this process. Thus, the reaction center-B870 complex may be envisioned as a reaction center surrounded by sixteen a/3-2bchl a dimers. This combination of complexes (reaction center plus B870 complexes) is quite stable and has been isolated from at least nine species of phototrophic Proteobacteria (20, 38, 73). Evidence for the structure of this B870reaction center complex is provided by electron micrographs of membranes of Rps. viridis (83), Rsp. rubrum (33) and Rsp. molischianum (5). Surrounding and connecting the B870-reaction center complexes is the B800-850 complex. This complex is also composed of bacteriochlorophyll binding polypeptides designated CC and (3, as well as a /polypeptide that does not bind pigments. The complex binds three bacteriochlorophyll a molecules and from one to three carotenoid molecules (37). The deduced molecular weights of the a, y3and /polypeptides are 7,322, 4,579 and 13,493  6 daltons, respectively. Like the B870 polypeptides, the B800-850 polypeptides migrate anomalously on SDS-polyacrylamide gels and have apparent M r 's of 10, 8 and 14 kDa respectively. The B800-850 complex is composed of a/J/-3bchl a subunits that, like the B870 complex, form interacting aggregates. Recently, the B800-850 complex from the (XProteobacterium, Rps. acidophila, has been crystallized and the atomic structure determined to 2.5 A resolution (63). This complex consists of nine a/3-3bchl a dimers (the B800-850 complex of Rps. acidophila does not have a y polypeptide). The dimer of bacteriochlorophyll-binding a and (3 polypeptides forms the fundamental structural subunit of light-harvesting complexes in all phototrophic Proteobacteria. More than twenty such polypeptides have been sequenced and they all have structural features in common, as demonstrated by sequence analysis, biochemical tests and spectroscopic analysis (103). The molecular weight is usually between 4 and 8 kDa, and there are three domains: a hydrophobic trans-membrane domain separates hydrophilic carboxyterminal and amino-terminal domains. The carboxy-terminal domain is found in the periplasmic space and the amino-terminal domain lies on the cytoplasmic side of the membrane. In addition to highly conserved histidine residues (one per a polypeptide and two per p polypeptide), which are involved in bacteriochlorophyll binding, there are conserved residues or groups of residues specific to oc and y3 polypeptides from core and peripheral complexes, which account to some extent for the functional differences seen in these complexes (102). The reaction center cofactors consist of a bchl a dimer (called the special pair), two "voyeur" bchl a molecules (so termed because they were until recently thought not to participate directly in electron transfer, although controversy exists over their function), two bacteriopheophytins, and a pair of quinones (associated with an iron atom) all held in a precise orientation to one another by three polypeptides, the light (L), medium (M), and heavy (H) subunits (21). The M r 's of these proteins are 21, 24, and 28 kDa respectively (69). The L and M subunits each have five trans-membrane CC helices, and may be thought of as the scaffolding for the electron transport-associated cofactors. The H subunit consists largely of a cytoplasmic globular domain with only one trans-membrane a helix, and is thought to be important in maintaining the correct orientation of the reaction center in the membrane (40). The cytochrome blc\ complex in R. capsulatus is made up of three polypeptides. These are a Rieske iron-sulfur (2Fe-2S) protein (M r 22.5 kDa), a b-type cytochrome (M r 42.1 kDa) and a cytochrome c\ (M r 31.2 kDa) (78). The b-type cytochrome apoprotein carries two hemes known as cytochromes bh, a high potential cytochrome, and b\, a low potential  7 cytochrome. Quinone molecules are also associated with the cytochrome blc\ complex at two sites. One site, known as the quinol oxidase, or Qj7, site, is responsible for the reduction of the Rieske iron-sulfur cluster and cytochrome b\. The Q z site is located on the periplasmic side of the blc\ complex. The second site, the quinone reductase, or Q c , site, is responsible for oxidation of cytochrome bh and is found on the cytoplasmic side of the b/c\ complex (29). As mentioned above, electron and proton transfer between the reaction center and b/c\ complex is accomplished by smaller, more mobile cofactors. Electrons are transferred from the cytochrome c\ of the b/c\ complex to the bchl a special pair in the reaction center by two c-type cytochromes, cytochrome ci and cytochrome cy. Cytochrome ci is located in the periplasmic space and cytochrome cy is membrane-associated. Electrons and protons are shuttled from the reaction center to the blc\ complex by quinol molecules, which are found within the lipid bilayer. In R. capsulatus, as in other phototrophic a-Proteobacteria, the genes encoding many of the essential components of the photosynthetic apparatus are clustered in one region of the chromosome. The genes were localized to a 46 kb fragment of the chromosome by Marrs and coworkers (86, 95) and the entire 46 kb fragment has since been sequenced (8). The 46 kb fragment includes the genes for the L, M, and H subunits of the reaction center (puJL and M and puhA), for the fi and CC subunits of the B870 complex (pufB and A) and for carotenoid and bacteriochlorophyll biosynthesis enzymes (see Figure 1.3). The pet opevon, which contains the genes for the apoproteins of the cytochrome blc\ complex (petA, B, and Q (19) is, if we consider puhA gene as zero on the chromosomal map of R. capsulatus , located at 1.5 Mb. The puc operon, which encodes the /?, a, and /subunits of the B800-850 complex (pucB, A, and E) (87, 98) and includes two genes, pucC and pucD, that are required for the formation of the B800-850 complex (87), is found at 1.7 Mb. The gene encoding cytochrome ci (cycA) (18) is located at 3.1 Mb, but the cycYgene, encoding cytochrome cy, has not been located on the chromosomal map. 1.3.2 Energy capture by the photosynthetic unit Figure 1.1 summarizes the capture of light energy by the photosynthetic unit. In photosynthesis, the capture of incident light energy is initially effected by pigment molecules. R. capsulatus has two main types of pigments: bacteriochlorophyll^ and carotenoids. Free bacteriochlorophyll a has an absorption maximum of about 760 nm (11). In vivo, the absorption maximum varies according to the environment in which the molecule is found. Thus, the wavelength of light absorbed by this pigment can be "tuned" through association with  8 different polypeptides, and this accounts for the different absorption maxima of the pigmentbinding complexes. There is more than one type of carotenoid in R. capsulatus, but in anaerobic cultures of R. capsulatus spheroidene is the predominant type whereas in aerobic cultures spheroidenone predominates (36). In addition to their light-harvesting function, carotenoids also protect cells against photo-oxidative damage from the formation of singletstate oxygen in the presence of light and bacteriochlorophyll, essentially by quenching the triplet state of bacteriochlorophyll (12). Both types of pigment are found in the reaction center and in the antenna complexes, with the bulk of the pigments occurring in the antenna complexes. Light energy is usually absorbed by a pigment molecule in an antenna complex, causing an electron to move to a higher energy level. Carotenoids, with their conjugated double bond structures, and bacteriochlorophyll molecules, with their tetrapyrrole structures, are both surrounded by electron clouds and the excitation energy can pass from one pigment to another via a process known as resonance energy transfer (90). Generally, the light energy will migrate from pigment molecules that absorb at shorter (higher energy) wavelengths (e.g., carotenoids that absorb maximally in the range of 400-500 nm) to those that absorb at longer (lower energy) wavelengths (e.g., B870 bacteriochlorophyll that absorbs maximally at 870 nm) until it reaches the reaction center special pair, which absorbs light at 880 nm. The energy transferred to the reaction center by an antenna complex causes the oxidation of the bacteriochlorophyll special pair. The electron lost by the special pair is taken up transiently by a bacteriopheophytin molecule and passes through the "primary electron acceptor", a quinone known as Q A to a second quinone Q B (see Figure 1.2). After a second photo-oxidation event Q B carries two extra electrons. It takes up two protons from the cytoplasm and diffuses laterally through the membrane from the reaction center to the cytochrome blc\ complex (17, 47, 72, 81).  9 The cytochrome blc\ complex helps mediate the formation of the proton gradient through which the captured light energy is used to do work. Formation of the quinol at the Q B  OUT  850  IN Figure 1.1. Light energy transduction in R. capsulatus. The conversion of light energy to a electrochemical potential in R. capsulatus. The heavier lines represent the electrogenic movements of electrons that contribute to the carotenoid bandshift and the wavy lines represent light energy. Abbreviations: R C , reaction center; b / c i , ubiquinolrcytochrome c\ oxidoreductase complex; B870, B870 light-harvesting antenna complex; B800-850, B800-850 light-harvesting antenna complex; Bchl, the special pair of the reaction centre; BPheo, bacteriopheophytin; Q, (ubi)quinone; QH2 , (ubi)quinol; QA and QB, the primary electron acceptors; QZ, quinol oxidase site; Q C , quinone reductase site; FeS, Rieske iron-sulphur complex; c i , cytochrome c\ (membrane bound); C2 , cytochrome Q (diffusible in the periplasm); c y , cytochrome cy (membrane associated); b\, the low potential cytochrome b; bh, the high potential cytochrome b. site of the reaction center can be thought of as an occurrence simultaneous with the oxidation of the iron-sulfur cluster, cytochrome c\ and cytochrome b\, and reduction of cytochrome b\i of the cytochrome blc\ complex. When the quinol from the reaction center binds to the Q z site it is presented with two oxidants: the iron-sulfur cluster and cytochrome b\ (see Figure 1.1). The electrons are thought to follow the Q cycle proposed by P. Mitchell (67) within the  10 cytochrome b/c\ complex. One electron is immediately taken up by the iron-sulfur cluster and the second moves at the same time to cytochrome b\. The electron on the iron-sulfur cluster moves to the cytochrome c\, and then to cytochrome C2 or ty. The two electrons now carried by the b-type cytochrome reduce a quinone at the Q c site, with concomitant uptake of two protons from the cytoplasm. Now it is the turn of this quinol to move to the Q z site, and its oxidation primes the cytochrome blc\ complex for another turn over. The net result is that the electrons removed from the bchl a special pair in the reaction center are returned to the reaction center via the cytochrome blc\ complex and cytochrome Q removed from quinols at the Q z s ' t e  are  or c  y> a n d protons  released to the periplasm. The net stoichiometry is  four protons translocated for two photons absorbed by the reaction center special pair. This forms a gradient of both electrical charge (A\|/) and proton concentration (ApH) (72, 88). Once captured in the form of this proton gradient, energy is used for the phosphorylation of ADP to ATP, for the reduction of NAD + , for the transportation of ionic species across the membrane, and for motility. An interesting and useful side effect of the movement of electrons and protons across the membrane is a shift in the absorption spectrum of the B800-850 antenna complexassociated carotenoid pigments. This electrochromic red shift in absorbency is called the carotenoid bandshift, and the magnitude of the absorption change at selected wavelengths is linear with respect to the membrane potential. Thus, the carotenoid bandshift serves as a means of evaluating the ability of a given bacterial strain to form the trans-membrane potential necessary for growth (35, 41). As shown in Figure 1.2, the photosynthetic electron transport system intersects other electron transport pathways at the level of the quinone/quinol pool. Quinols serve as proton/electron shuttles between the reaction center and the blc\ complex, between dehydrogenases and the blc\ complex, between dehydrogenases and the so-called alternative oxidase, and between dehydrogenases and enzyme(s) mediating electron transfer to alternate electron acceptors used in anaerobic respiration with acceptors such as nitrate and dimethylsulfoxide (DMSO) (26). Photosynthetic electron transport also shares the electron carriers cytochromes Q ar>d ty with the respiratory electron transport chain (26, 39). Under most growth conditions both respiratory and photosynthetic enzymes are expressed. Given this simultaneous expression of components of both pathways and the sharing of proton/electron carriers, it is not surprising that the operations of the various pathways inhibit or enhance one another. For example, respiratory electron transport is inhibited in the presence of light,  Dehydrogenases Hydrogenases  Alternative b/c-i complex V J Figure 1.2. Electron transfer pathways in R. capsulatus. Rectangular boxes represent enzymes and cofactors that are membraneassociated; ovals represent periplasmic enzymes and cofactors. The figure is based on work reviewed by Ferguson et al. (26) and on subsequent work published in (65, 66, 75-77). seemingly because the large light-generated electrochemical potential (Ap) inhibits the protonmotive activity of the respiratory chain dehydrogenases (15, 62) but see also (99).  12 The anaerobic respiratory transport chain has been proposed to play a role in maintaining the redox poise necessary for photosynthetic electron transport (26, 64). The aerobic photosynthetic bacterium Erythrobacter sp. OCh 114 (since renamed Roseobacter denitrificans (80)) has a functional photosynthetic electron transport chain that resembles that of/?, capsulatus, but it does not operate under anaerobic conditions. The cause of this inability is thought to lie in the high midpoint potential (35 raV) of the primary acceptor quinone, Q A (89). The consequence of this higher mid-point potential is that a smaller decrease in the ambient potential (relative to the decrease necessary for the same effect in anaerobic phototrophs, which have Q A midpoint potentials of-25 mV or less) results in the reduction of this quinone. This reduction of the primary acceptor quinone effectively blocks the oxidation of the special pair and, hence, photosynthetic electron transport. By supplying TMAO, an alternative electron acceptor for anaerobic respiration, Takamiya et al. were able to restore anaerobic photosynthetic electron transport (85). It was concluded that oxidation of the anaerobic respiratory chain by TMAO shifted the redox level in the cells towards a more oxidizing level (thus oxidizing Q A ) , and photosynthetic electron transport was able to operate. It was at one time thought that photosynthetic electron transfer in R. capsulatus required only the reaction center and blc\ complexes in combination with cytochrome c% and ubiquinol (72). However, it is now clear that the situation is more complex because a membrane-bound cytochrome, cytochrome Cy, operates in parallel with cytochrome Q (42). Thus, mutants lacking cytochrome Q are still capable of photosynthetic growth and photosynthetic electron transport is entirely membrane-associated. Furthermore, in both R. capsulatus and R. sphaeroides, at least one more protein, the pufiC gene product, is necessary for normal photosynthetic electron transport (25, 57) 1.3.3 Regulation of expression of photosynthesis genes Under aerobic conditions, R. capsulatus grows chemotrophically. When the oxygen concentration falls below about 3 % (70), however, the formation of the photosynthetic apparatus is induced. Pigment synthesis was first shown in R. sphaeroides to be regulated both by oxygen concentration (pC>2) and light intensity by Cohen-Bazire et al. (14). With the cloning (60) and mapping (86) of the genes for the photosynthetic apparatus, it became possible to compare transcription levels from most of the known photosynthesis-related genes under different environmental conditions. Regulation of expression of photosynthesis genes has been shown to occur at the level of transcription (10, 52, 100), and post-transcriptionally (4, 9, 50, 104). Transcriptional regulation is thought to occur via an oxygen sensory transduction  13 cascade (68) modulated by the phosphorylation of the components of the pathway (46). It was proposed that the sensor kinase, encoded by the regB gene, responds to the redox state of the cell rather than to oxygen directly (68). Post-transcriptional regulation is best understood in the pufoperon,  but is also likely to be a factor in the expression of the/»«c operon (56).  mRNA transcripts of the/»«/"operon are differentially degraded (4) such that the ratio of the steady-state amounts of 0.5 kb messages that encode the pufB and A genes to 2.7 kb messages encoding the pufB, A, L, Mand X genes is about ten to one. Oxygen also appears to regulate the rate of />«/"mRNA degradation, but has no effect on the degradation rate of mRNA of other photosynthesis genes (48). Light intensity also regulates the expression of photosynthesis genes (28, 79). Transcription of the puf genes and ofpuhA is increased at lower light intensities (51), and a trans-acting regulatory gene, hvrA, that is believed to be responsible for the differential expression of the puf and puh genes has been identified (7). This gene forms part of a regulatory gene cluster that also includes regA and B (7). Paradoxically, it has been found that the steady-state levels of transcripts from the pucA and B genes are higher at higher light intensities (104). It was proposed that control of B800-850 complex formation is at the level of translation, or post-translational (104). Post-translational regulation of the expression of the genes encoding the light-harvesting and reaction center polypeptides has also been demonstrated. In the absence of bacteriochlorophyll, the genes are translated but the proteins do not accumulate in the membrane and are quickly degraded (22, 23, 53). Synthesis of the pigment and peptide components of the photosynthetic apparatus is coordinated in part at the level of transcription. As can be seen in Figure 1.3 the pufoperon and puhA gene, which encode the structural polypeptides of the reaction center and B870 antenna complex, flank the pigment biosynthesis genes. Young et al. proposed that these genes might be organized into "superoperons" based on genetic evidence for very long transcripts extending from the pigment synthesis enzyme operons through the structural polypeptide operons (96). Wellington et al. demonstrated directly that transcripts originating at the crtEF promoter extended through the bchCXYZ operon and into  14 the/>«/operon (92) (see Figure 1.3). It was shown that mutational disruption of these longer, superoperon transcripts resulted in lower rates of transcription initiation at the internal promoters, in lower levels of constitutive expression of genes in the nested (downstream) operons, and that the net result was a deficiency in the ability of the mutant cells to adapt from respiratory to photosynthetic growth conditions (3, 93). Analysis of the promoter sequences of pigment biosynthesis and pigment-binding protein gene sequences has led to the suggestion that two different RNA sigma factors mediate transcription of the pigment-binding protein versus the pigment biosynthesis genes (1, 49). The pigment biosynthesis gene operons were proposed by Armstrong et al. to contain promoters homologous to E. coli <770 promoters. This is consistent with the observation that these operons are not as strongly O2 regulated as the pigment-binding protein gene operons. The pigment-binding protein gene operons, at the apparent extremes of the photosynthesis gene cluster (see Figure 1.3) seem to have a different promoter sequence from the pigment gene operons (2) and are strongly transcriptionally activated under low oxygen conditions.  Figure 1.3. Photosynthesis gene cluster of R. capsulatus. Genes encoding bacteriochlorophyll biosynthetic enzymes are shown as gray boxes; genes encoding carotenoid biosynthetic enzymes are shown as stippled boxes; genes encoding structural polypeptides are shown as diagonally striped boxes and genes encoding other polypeptides of uncertain or unknown function are shown as cross-hatched boxes. The arrows to the right of the boxes indicate the location of promoters and the demonstrated length of the transcripts originating from the promoters. Dotted lines extending from arrowheads indicate putative transcript lengths.  15 1.4. The puJXgene At the 5' end of the pufoperon  is an open reading frame that was originally designated  C2397 (97) and has since been designated the pufiCgene (4) (see Figures 1.3 and 2.1). No function had been ascribed to the putative protein encoded by this open reading frame at the time I started my M. Sc. thesis work, although it was known that the gene was transcribed as part of a polycistronic mRNA molecule that encodes the other components of the/>«/operon (4). The protein encoded by the pufiigene is 78 amino acids long and has a molecular weight of 8,568 Da (97). Hydropathy plots of the amino acid sequence indicated that it contained a single membrane-spanning segment in the central region of the sequence (57, 97). Sequence alignments between the PufX polypeptide sequence and the sequences of light-harvesting antenna polypeptides from R. capsulatus, Rsp. rubrum, Rps. viridis and Chloroflexus aurianticus, showed that the most similarity (27.6% identity with 5 gaps permitted) existed between PufX and the B870a protein from R. capsulatus (57). In order to assess the role of the pufi( gene product in R. capsulatus, a translationally in-frame deletion of this gene was made. While respiratory growth and anaerobic respiratory growth were unaffected in the mutant strain, deletion of the R. capsulatus pufiCgene impaired the ability of cells to grow photosynthetically on minimal medium at a light intensity of 120 I^E-nv^s" 1 . Loss of pujX also altered the characteristic stoichiometries of the light harvesting antenna and reaction center complexes, and changed the average size of the membrane invaginations that contain the photosynthetic apparatus. Flash spectroscopy experiments, fluorescence measurements and cytochrome blc\ complex assays indicated the components of the photosynthetic unit were functional, but cyclic electron flow was interrupted. Northern blot experiments indicated that the loss of puJX did not alter the level of transcription of the puf bperon or change the relative amounts of the mRNA processing products. The pufiC deletion strain was found to give rise to photosynthetically competent suppressor mutants at a relatively high frequency (10~5). Studies of these suppressor mutants showed that they had different photosynthetic growth rates at a light intensity of 120 |j.E-nr2-s" 1  relative to the pseudo-wild type strain ARC6(pTB999). The absorption spectra and SDS-  PAGE profiles indicated that the stoichiometries of the components of the photosynthetic unit were also altered in the suppressor mutants, but that these stoichiometries were different from those seen in the primary mutant. I concluded that the PufX protein is required for correct assembly of the unit as a structural component and/or as a regulator of its assembly. In work carried out in the laboratory of Dr. Paul Loach, Dr. J. T. Beatty was able to demonstrate that  16 the PufX protein was present in chromatophores prepared from the pseudo-wild type strain of R. capsulatus. However, no specific molecular or biochemical activity of the PufX protein has been obtained. The goal of my Ph. D. thesis work was to improve our understanding of the role of the pujXgene product in the photosynthetic growth of R. capsulatus. By better defining the phenotype of the pufiCmutant and by studying the properties of the pufXsuppressor  mutants, I  hoped to be able to obtain enough information to explain why the PufX protein is necessary for optimal photosynthetic growth of R. capsulatus, and suggest a model for the structure and function of the PufX protein in the photosynthetic unit. 1.5. Introduction references 1. Armstrong, G.A., M. Alberti, F. Leach and J.E. Hearst. 1989. Nucleotide sequence, organization, and nature of the protein products of the carotenoid biosynthesis gene cluster of Rhodobacter capsulatus. Mol. Gen. Genet. 216:254-268. 2. Bauer, C., J. Buggy and C. Mosley. 1993. Control of photosystem genes in Rhodobacter capsulatus. Trends Genet. 9:56-60. 3. Bauer, C.E., J.J. Buggy, Z. Yang and B.L. Marrs. 1991. The superoperonal organization of genes for pigment biosynthesis and reaction center proteins is a conserved feature in Rhodobacter capsulatus: analysis of overlapping bchB and puhA transcripts. Mol. Gen. Genet. 228:438-444. 4. Belasco, J.G., J.T. Beatty, C.W. Adams, A. von Gabain and S.N. Cohen. 1985. Differential expression of photosynthesis genes in R. capsulatus results from segmental differences in stability within the polycistronic rxcA transcript. Cell 40:171-181. 5. Boonstra, A.F., L. Germeroth and E.J. Boekma. 1994. Structure of the light harvesting antenna from Rhodospirillum molischianum studied by electron microscopy. Biochim. Biophys. Acta 1184:227-234. 6. Brock, T.D. and M.T. Madigan. 1991. p. 563. Biology of Microorganisms, 6th ed. Prentice Hall, Englewood Cliffs, NJ. 7. Buggy, J.J., M.W. Sganga and C.E. Bauer. 1994. Characterization of a lightresponding trans-activator responsible for differentially controlling reaction center and lightharvesting-I gene expression in Rhodobacter capsulatus. J. Bacteriol. 176:6936-6943. 8. Burke, D.H., M. Alberti, G.A. Armstrong and J.E. Hearst. 1991. The complete nucleotide sequence of the 46 kb photosynthetic gene cluster of Rhodobacter capsulatus. EMBL Data Library Accession number Z l 1165.  17 9. Chen, C.-YJV.., J.T. Beatty, S.N. Cohen and J.G. Belasco. 1988. An intercistronic stem-loop structure functions as an mRNA decay terminator necessary but insufficient for puf mRNA stability. Cell 52:609-619. 10. Clark, W.G., E. Davidson and B.L. Marrs. 1984. Variation of levels of mRNA coding for antenna and reaction center polypeptides in Rhodopseudomonas capsulata in response to changes in oxygen concentration. J. Bacteriol. 157:945-948. 11. Clayton, R.K. 1966. Spectroscopy of bacteriochlorophyll. Photochem. Photobiol. 5:807-821. 12. Cogdell, R J . and H.A. Frank. 1987. How carotenoids function in photosynthetic bacteria. Biochim. Biophys. Acta 895:63-79. 13. Cogdell, R.J. and A.M. Hawthornthwaite. 1993. Preparation, purification, and crystallization of purple bacteria antenna complexes, pp. 23-42. In]. Diesenhofer and J. R. Norris, The photosynthetic reaction center. Academic Press, Inc., San Diego. 14. Cohen-Bazire, G., W.R. Sistrom and R.Y. Stanier. 1957. Kinetic studies of pigment synthesis by non-sulfur purple bacteria. J. Cell. Comp. Physiol. 49:25-68. 15. Cotton, N.P.J., A.J. Clark and J.B. Jackson. 1983. Interaction between the respiratory and photosynthetic electron transport chains in intact cells of Rhodopseudomonas capsulata. Eur. J. Biochem. 130:581-587. 16. Cramer, W.A. and D.B. Knafif. 1991. Energy transduction in biological membranes: a textbook of bioenergetics. Springer-Verlag, New York. 17. Crofts, A.R. and C.A. Wraight. 1983. The electrochemical domain of photosynthesis. Biochim. Biophys. Acta. 726:149-185. 18. Daldal, F., S. Chen, J. Applebaum, E. Davidson and R.C. Prince. 1986. Cytochrome c is not essential for photosynthetic growth of Rhodopseudomonas capsulata. Proc. Natl. Acad. Sci. USA 83:2012-2016. 19. Daldal, F., E. Davidson and S. Cheng. 1987. Isolation of the structural genes for the Rieske Fe-S protein, cytochrome b and cytochrome c all components of the ubiquinol.-cytochrome c2 oxidoreductase complex of Rhodobacter capsulatus. J. Mol. Biol. 195:1-12. 20. Dawkins, D.J., L.A. Ferguson and R.J. Cogdell. 1988. The structure of the 'core' of the purple bacterial photosynthetic unit, pp. 115 - 127. In H. Scheer and S. Schneider, Photosynthetic light-harvesting systems. Walter de Gruyter and Co., Berlin. 21. Deisenhofer, J. and H. Michel. 1991. Structures of bacterial reaction centers. Ann. Rev. Cell Biol. 7:1-23. 22. Dierstein, R. 1983. Biosynthesis of pigment-protein complex polypeptides in bacteriochlorophyll-less mutant of Rhodopseudomonas capsulata YS. FEBS. Lett. 160:281286.  18 23. Dierstein, R. 1984. Synthesis of pigment-binding protein in toluene-treated Rhodopseudomonas capsulata and in cell-free systems. Eur. J. Biochem. 138:509-518. 24. Donohue, T.J. and S. Kaplan. 1991. Genetic techniques in Rhodospirillaceae. Methods Enzymol. 204:459-485. 25. Farchaus, J.W., H. Gruenberg and D. Oesterhelt. 1990. Complementation of a reaction center-deficient Rhodobacter sphaeroidespufLMX deletion strain in trans with pufBALMdoes not restore the photosynthesis-positive phenotype. J. Bacteriol. 172:977-985. 26. Ferguson, S.J., J.B. Jackson and A.G. McEwan. 1987. Anaerobic respiration in the Rhodospirillaceae. characterization of pathways and evaluation of roles in redox balancing during photosynthesis. FEMS Microbiol. Rev. 46:117-143. 27. Fonstein, M., Zheng, S. and R. Haselkorn. 1992. Physical map of the genome of Rhodobacter capsulatus SB 1003. J. Bacteriol. 174:4070-4077. 28. Garcia, A., G. Venturoli, N . Gad'on, J. Fernandez-Velasco, B.A. Melandri and G. Drews. 1987. The adaptation of the electron transfer chain of Rhodopseudomonas capsulata to different light intensities. Biochim. Biophys. Acta 890:335-345. 29. Gennis, R.B., B. Barquera, B. Hacker, S.R.V. Doren, S. Arnaud, A.R. Crofts, E. Davidson, K.A. Gray and F. Daldal. 1993. The be, complexes of Rhodobacter sphaeroides and Rhodobacter capsulatus. J. Bioenerg. Biomembr. 25:195-209. 30. Gest, H. 1991. The legacy of Hans Molisch (1856-1937), photosynthesis savant. Photosynth. Res. 30:49-59. 31. Gest, H. 1993. History of concepts of the comparative biochemistry of oxygenic and anoxygenic photosynthesis. Photosynth. Res. 35:87-96. 32. Gest, H. 1993. Photosynthetic and quasi-photosynthetic bacteria. FEMS Microbiol. Lett. 112:1-6. 33. Ghosh, R., A. Hoenger, A. Hardmeyer, D. Mihailescu, R. Bachofen, A. Engel and J.P. Rosenbusch. 1993. Two-dimensional crystallization of the light-harvesting complex from Rhodospirillum rubrum. J. Mol. Biol. 231:501-504. 34. Golbeck, J.H. 1993. Shared thematic elements in photochemical reaction centers. Proc. Natl. Acad. Sci. USA 90:1642-1646. 35. Goodwin, M.G. and J.B. Jackson. 1993. Electrochromic responses of carotenoid absorbance bands in purified light-harvesting complexes from Rhodobacter capsulatus reconstituted into liposomes. Biochim. Biophys. Acta 1144:191-198. 36. Goodwin, T.W. 1980. The biochemistry of the carotenoids, Vol. 1: Plants, p. 63. Chapman and Hall, New York. 37. Hawthornthwaite, A.M. and R.J. Cogdell. 1991. Bacteriochlorophyll-binding proteins, pp. 493-528. In H. Scheer, Chlorophylls. CRC Press, Boca Raton, FL.  19 38. Heller, B.A. and P.A. Loach. 1990. Isolation and characterization ofasubunit form of the B875 light-harvesting complex from Rhodobacter capsulatus. Photochem. Photobiol. 51:621-627. 39. Hochkoeppler, A., F.E. Jenney, Jr., S.E. Lang, D. Zannoni and F. Daldal. 1995. Membrane-associated cytochrome Cy of Rhodobacter capsulatus is an electron carrier from the cytochrome bc\ complex to the cytochrome c oxidase during respiration. J. Bacteriol. 177:608-613. 40. Huber, R. 1989. A structural basis of light energy and electron transfer in biology. EMBO J. 8:2125-2147. 41. Jackson, J.B. and P.L. Dutton. 1973. The kinetic and redox potentiometric resolution of the carotenoid shifts in Rhodopseudomonas sphaeroides chromatophores: their relationship to electric field alterations in electron transport and energy coupling. Biochim. Biophys. Acta 325: 102-115. 42. Jenney, F.E. and F. Daldal. 1993. A novel membrane-associated otype cytochrome, cyt cy> can mediate the photosynthetic growth of Rhodobacter capsulatus and Rhodobacter sphaeroides. EMBO J. 12:1283-1292. 43. Joliot, P., A. Vermeglio and A. Joliot. 1989. Evidence for supercomplexes between reaction centers, cytochrome c2 and cytochrome be complex in Rhodobacter sphaeroides whole cells. Biochim. Biophys. Acta 975:336-345. 44. Joliot, P., A. Vermeglio and A. Joliot. 1993. Supramolecular membrane protein assemblies in photosynthesis and respiration. Biochim. Biophys. Acta 1141:151-174. 45. Karrasch, S., P.A. Bullough and R. Ghosh. 1995. The 8.5 A projection map of the light-harvesting complex I from Rhodospirillum rubrum reveals a ring composed of 16 subunits. EMBO J. 14:631-638. 46. Kazuhito, L, J.-l.K. Kouadio, C.S. Mosley and C.E. Bauer. 1995. Isolation and in vitro phosphorylation of sensory transduction components controlling anaerobic induction of light harvesting and reaction center gene expression in Rhodobacter capsulatus. Biochemistry 34:391-396. 47. Kirmaier, C. and D. Holten. 1987. Primary photochemistry of reaction centers from photosynthetic purple bacteria. Photosynth. Res. 13:225-260. 48. Klug, G. 1991. Endonucleolytic degradation of/>«/mRNA in R. capsulatus is influenced by oxygen. Proc. Natl. Acad. Sci. USA 88:1765-1769. 49. Klug, G. 1993. Regulation of expression of photosynthesis genes in anoxygenic photosynthetic bacteria. Arch. Microbiol. 159:397-404. 50. Klug, G., C.W. Adams, J. Belasco, B. Doerge and S.N. Cohen. 1987. Biological effects of segmental alterations in mRNA stability: effects of deletion of the intercistronic hairpin loop region of the Rhodobacter capsulatus puf operon. EMBO J. 6:3515-3520.  20 51. Klug, G., N. Kaufmann and G. Drews. 1985. Gene expression of pigment-binding proteins of the bacterial photosynthetic apparatus: Transcription and assembly in the membrane of Rhodobacter capsulatus. Proc. Natl. Acad. Sci. USA 82:6485-6489. 52. Klug, G., N. Kaufmann and G. Drews. 1985. The expression of genes encoding proteins of B800-850 antenna pigment complex and ribosomal RNA of Rhodopseudomonas capsulata. FEBS Lett. 177: 61-65. 53. Klug, G., R. Liebetanz and G. Drews. 1986. The influence of bacteriochlorophyll biosynthesis on formation of pigment-binding proteins and assembly of pigment protein complexes in Rhodopseudomonas capsulata. Arch. Microbiol. 146: 284-291. 54. Lavergne, J. and P. Joliot. 1991. Restricted diffusion in photosynthetic membranes. Trends in Biochem. Sci. 16:129-134. 55. Lavorel, J., P. Richaud and A. Vermeglio. 1989. Interaction of photosynthesis and respiration in Rhodospirillaceae: evidence for two functionally distinct b-c complex fractions. Biochim. Biophys. Acta 973:290-295. 56. LeBlanc, H.N. and J.T. Beatty. 1993. Rhodobacter capsulatuspuc operon: promoter location, transcript sizes and effects of deletions on photosynthetic growth. J. Gen. Microbiol. 139:101-109. 57. Lilburn, T.G. 1990. The role of the pufiCgene product of Rhodobacter capsulatus. M. Sc. thesis, University of British Columbia. 58. Loach, P.A., P.S. Parkes-Loach, C M . Davis and B.M. Heller. 1994. Probing protein structural requirements for formation of the core light-harvesting complex of photosynthetic bacteria using hybrid reconstitution methodology. Photosynth. Res. 40:231245. 59. Madigan, M.T. and H. Gest. 1979. Growth of the photosynthetic bacterium Rhodopseudomonas capsulata chemoautotrophically in darkness with FL as the energy source. J. Bacteriol. 137:524-530. 60. Marrs, B.L. 1981. Mobilization of the genes for photosynthesis from Rhodopseudomonas capsulata by a promiscuous plasmid. J. Bacteriol. 146:1003-1012. 61. Marrs, B.L. 1983. Genetics and molecular biology, pp. 186-214. In]. G. Ormerod, The Phototrophic Bacteria: Anaerobic Life in the Light. Blackwell Scientific Publications, Oxford. 62. McCarthy, J.E.G. and S.J. Ferguson. 1982. Respiratory control and the basis of light-induced inhibition of respiration in chromatophores from Rhodopseudomonas capsulata. Biochem. Biophys. Res. Commun. 107:1406-1411. 63. 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 lightharvesting complex from photosynthetic bacterium. Nature 374:517-521.  21 64. McEwan, A.G., N.P.J. Cotton, S.J. Ferguson and J.B. Jackson. 1985. The role of auxiliary oxidants in the maintenance of a balanced redox poise for photosynthesis in bacteria. Biochim. Biophys. Acta 810:140-147. 65. McEwan, A.G., D J . Richardson, H. Hudig, S.J. Ferguson and J.B. Jackson. 1989. Identification of cytochromes involved in electron transport to trimethylamine N-oxide / dimethylsulphoxide reductase in Rhodobacter capsulatus. Biochim. Biophys. Acta 973:308314. 66. McEwan, A.G., H.G. Wetzstein, S.J. Ferguson and J.B. Jackson. 1985. Periplasmic location of the terminal reductase in trimethylamine N-oxide and dimethylsulphoxide respiration in the photosynthetic bacterium Rhodopseudomonas capsulata. Biochim. Biophys. Acta 806:410-417. 67. Mitchell, P. 1976. Possible molecular mechanisms of the protonmotive function of cytochrome systems. J. Theor. Biol. 62:327-367. 68. Mosley, C.S., J.Y. Suzuki and C.E. Bauer. 1994. Identification and molecular genetic characterization of a sensor kinase responsible for coordinately regulating light harvesting and reaction center gene expression in response to anaerobiosis. J. Bacteriol. 176:7566-7573. 69. Nieth, K., G. Drews and R. Feick. 1975. Photochemical reaction centers from Rhodopseudomonas capsulata. Arch. Microbiol. 105:43-45. 70. Oelze, J. and G. Drews. 1972. Membranes of photosynthetic bacteria. Biochim. Biophys. Acta 265:209-239. 71. Pierson, B.K. and J.M. Olson. 1989. Evolution of photosynthesis in anoxygenic photosynthetic prokaryotes, pp. 402-407. In Y. Cohen and E. Rosenberg, Microbial Mats, physiological ecology of benthic microbial communities. American Society for Microbiology, Washington, D. C. 72. Prince, R.C. 1990. Bacterial photosynthesis; from photons to Ap, pp. 111-149. In T. Krulwich, The Bacteria. Academic Press, Orlando. 73. Pushkin, A.V., F.A. Eiserling, C.A. Kerfeld and J.P. Thornber. 1993. Isolation and quaternary structure of the photosynthetic core of the marine purple sulfur bacterium Chromatium purpuratum. FEMS Microbiol. Lett. 113:167-174. 74. Remsen, C. 1978. Comparative subcellular architecture of photosynthetic bacteria, pp. 31-60. In R. Clayton and W. Sistrom, The photosynthetic bacteria. Plenum Press, New York. 75. Richardson, D.J., L.C. Bell, A.G. McEwan, J.B. Jackson and S.J. Ferguson. 1991. Cytochrome c~ is essential for electron transfer to nitrous oxide reductase from physiological substrates in Rhodobacter capsulatus and can act as an electron donor to the reductase in vitro. Eur. J. Biochem. 199:677-683.  22 76. Richardson, D.J., A.G. McEwan, J.B. Jackson and S.J. Ferguson. 1989. Electron transport pathways to nitrous oxide in Rhodobacter species. Eur. J. Biochem. 185:659-669. 77. Richardson, D.J., A.G. McEwan, M.D. Page, J.B. Jackson and S.J. Ferguson. 1990. The identification of cytochromes involved in the transfer of electrons to the periplasmic NCL~ reductase of Rhodobacter capsulatus and resolution of a soluble NO.,~-reductasecytochrome-c,-<-2"redox complex. Eur. J. Biochem. 194:263-270. 78. Robertson, D.E., H. Ding, P.R. Chelminski, C. Slaughter, J. Hsu, C. Moonraw, M. Tokito, F. Daldal and P.L. Dutton. 1993. Hydroubiquinone-cytochromeQ oxidoreductase from Rhodobacter capsulatus: Definition of a minimal, functional isolated preparation. Biochemistry 32:1310-1317. 79. Schumacher, A. and G. Drews. 1979. Effects of light intensity on membrane differentiation in Rhodopseudomonas capsulata. Biochim. Biophys. Acta 547:417-428. 80. Shiba, T . 1991. Roseobacter litoralis gen. nov., sp. nov., and Roseobacter denitrificans sp. nov., aerobic pink-pigmented bacteria which contain bacteriochlorophyll a. System. Appl. Microbiol. 14:140-145. 81. Shinkarev, V.P. and C.A. Wraight. 1993. Electron and proton transfer in the acceptor quinone complex of reaction centers of phototrophic bacteria, pp. 194-257. In J. Diesenhofer and J. R. Norris, The photosynthetic reaction center. Academic Press, Inc., San Diego. 82. Stackebrandt, E., R.G.E. Murray and H.G. Truper. 1988. Proteobacteria classis nov., a name for the phylogenetic taxon that includes the "Purple bacteria and their relatives". Int. J. Syst. Bacteriol. 38:321-325. 83. Stark, W., W. Kuhlbrandt, I. Wildhaber, E. Wehrli and K. Muhlethaler. 1984. The structure of the photoreceptor unit of Rhodopseudomonas viridis. EMBO J. 3:777-783. 84. Stoeckenius, W. 1985. The rhodopsin-like pigments of halobacteria. Trends Biochem. Sci. 10:483-485. 85. Takamiya, K.-i., H. Arata, Y. Shioi and M. Doi. 1988. Restoration of the optimal redox state for the photosynthetic electron transfer system by auxiliary oxidants in an aerobic photosynthetic bacterium, Erythrobacter sp. OCh 114. Biochim. Biophys. Acta 935:26-33. 86. Taylor, D.P., S.N. Cohen, W.G. Clark and B.L. Marrs. 1983. Alignment of genetic and restriction maps of the photosynthesis region of the Rhodopseudomonas capsulata chromosome by a conjugation-mediated marker rescue technique. J. Bacteriol. 154:580-590. 87. Tichy, H.V., B. Oberle, H. Stiehle, E. Schiltz and G. Drews. 1989. Genes downstream from pucB and pucA are essential for formation of the B800-850 complex of Rhodobacter capsulatus. J. Bacteriol. 171:4914-4922.  23 88. Trumpower, B.L. and R.B. Gennis. 1994. Energy transduction by cytochrome complexes in mitochondrial and bacterial respiration: The enzymology of coupling electron transfer reactions to transmembrane proton translocation. Ann. Rev. Biochem. 63:675-716. 89. Triiper, H.G. 1987. Phototrophic bacteria (an incoherent group of prokaryotes). A taxonomic versus phylogenetic survey. Microbiologia Sem 3:71-89. 90. van Grondelle, R. 1985. Excitation energy transfer, trapping and annihilation in photosynthetic systems. Biochim. Biophys. Acta 811:147-195. 91. VIsschers, R.W., M.C. Chang, F.v. Mourik, P.S. Parkes-Loach, B j \ . Heller, P.A. Loach and R.v. Grondelle. 1991. Fluorescence polarization and low-temperature absorption spectroscopy of a subunit form of light-harvesting complex I from purple photosynthetic bacteria. Biochemistry 30:5734-5742. 92. Wellington, C.L. and J.T. Beatty. 1989. Promoter mapping and nucleotide sequence of the bchC bacteriochlorophyll biosynthesis gene from Rhodobacter capsulatus. Gene 83:251-261. 93. Wellington, C.L., A.K.P. Taggart and J.T. Beatty. 1991. Functional significance of overlapping transcripts of crtEF, bcbCA, and /^/photosynthesis gene operons in Rhodobacter capsulatus. J. Bacteriol. 173:473-482. 94. Woese, C.R. 1987. Bacterial evolution. Microbiol. Rev. 51:221-271. 95. Yen, H.C. and B.L. Marrs. 1976. Map of genes for carotenoid and bacteriochlorophyll biosynthesis in Rhodopseudomonas capsulata.]. Bacteriol. 126:619-629. 96. Young, D.A., C.E. Bauer, J.C. Williams and B.L. Marrs. 1989. Genetic evidence for superoperonal organization of genes for photosynthetic pigments and pigment-binding proteins in Rhodobacter capsulatus. Mol. Gen. Genet. 218:1-12. 97. Youvan, D.C., E.J. Bylina, M. Alberti, H. Begusch and J.E. Hearst. 1984. Nucleotide and deduced polypeptide sequences of the photosynthetic reaction-center, B870 antenna and flanking polypeptides from R. capsulata. Cell 37:949-957. 98. Youvan, D.C. and S. Ismail. 1985. Light harvesting II (B800-850 complex) structural genes from Rhodopseudomonas capsulata. Proc. Natl. Acad. Sci. USA 82:58-62. 99. Zannoni, D . and F. Daldal. 1993. The role of c-type cytochromes in catalyzing oxidative and photosynthetic electron transport in the dual functional plasma membrane of facultative phototrophs. Arch. Microbiol. 160:413-423. 100. Zhu, Y.S. and J.E. Hearst. 1986. Regulation of expression of genes for lightharvesting antenna proteins LH-I and LH-II; reaction center polypeptides RC-L, RC-M, biosynthesis in Rhodobacter capsulatus by light and oxygen. Proc. Natl. Acad. Sci. USA 83:7613-7617. 101. Zuber, H. 1986. Structure of light-harvesting antenna complexes of photosynthetic bacteria, cyanobacteria and red algae. Trends Biochem. Sci. 10:414-419.  24 102. Zuber, H. 1993. Structural features of photosynthetic light-harvesting systems, pp. 43-100. In J. Deisenhofer and J. R. Norris, The Photosynthetic Reaction Center. Academic Press, Inc., San Diego. 103. Zuber, H. and R.A. Brunisholz. 1991. Structure and function of antenna polypeptides and chlorophyll-protein complexes: Principles and variability, pp. 627-703. In H. Scheer, Chlorophylls. CRC Press, Boca Raton, FL. 104. Zucconi, A.P. and J.T. Beatty. 1988. Posttranscriptional regulation by light of the steady-state levels of mature B800-850 light-harvesting complexes in Rhodobacter capsulatus. J. Bacteriol. 170:877-882.  25  2.  MATERIALS  AND  METHODS  2.1. Bacterial strains The wild type strain BIO of R. capsulatus has been described (26, 39). Strain ARC6 is a puf derivative of BIO that has a 3,316 bp chromosomal DNA deletion. The deletion extends from the Sail site located 40 bp downstream of the start of the puJQ gene to the Xho II site 488 bp downstream of the 5' end of the pufX gene, with the neomycin phosphotransferase structural gene interposed between these two sites (8). R. capsulatus ARC6crtD  is identical to R. capsulatus ARC6 except that it carries a  mutation in the crtD gene encoding a carotenoid biosynthetic enzyme, introduced by gene transfer agent transduction with strain DE442 as donor (22, 41). Since the final product of the carotenoid biosynthetic pathway in the crtD mutant is neurosporene, the 500-600 nm region of the spectrum is simplified making evaluation of the carotenoid bandshift and cytochrome absorptions (see below) much easier. R capsulatus ARC6* is a B800-850", Crt" derivative of ARC6 that carries a pufX suppressor mutation on the chromosome ([23]; and see below). Escherichia coli C600 r m + (a hsdR derivative of C600 [4]) was used as a host strain for cloning and modifying R. capsulatus DNA. E. coliHBtlOl (pRK2013) is a helper strain used for conjugating plasmids from host E. coli strains into R. capsulatus in tri-parental matings (12). E, coli DH50C [(18) and Life Technologies, Inc. Burlington, ON] was used as a host for DNA sequencing vectors. E. coli BL21 (DE3) and HMS174(DE3) (34) are host strains for the over-expression vector pET-3a. 2.2. Construction of plasmids Plasmid pTB999 (Figure 2.1) carries the pu/Q, B, A, L, M, and Xgenes on a DNA fragment that extends from the Xho II site located 348 bp upstream of the puf opevon promoter to the Nru I site 620 bp downstream of the end of the pufXgene (24).  26 Deletion of the DNA between the filled-in Tth\ 11 I and the Fsp I sites in the pufiC gene, followed by insertion of a Sma I linker at the site of the deletion, created a 174 bp translationally in-frame deletion in pufiC. Substitution of this deleted gene for the wild type gene carried on pTB999 created the plasmid pAX. Plasmid pTL2 (Figure 2.1) is identical to pAX except that it has an Omega (Q) fragment (Amersham) inserted at the deletion site, accomplished by insertion of a Hind III linker in the Smal site of pAX followed by insertion of an Q. fragment in this Hind III site  H B  pTL2  Lr HEM  M  Q' H  pTL30  L-EH-  M  -lr1 NhB  tr  Figure 2.1. The construction of plasmids pTB999, pTL2 and pTL30. Plasmid pTB999 and the two derivative pufX deletion plasmids used in the experiments, pTL2 and pTL30. The pufoperon genes pu/Q, B, A, L, M and X and the tetracycline resistance gene (Tc r ) are shown as open boxes. For the methods of plasmid construction, please see the text. The omega insertion cartridge is shown as a heavy line with the appropriate symbol (Q). Restriction enzyme sites are abbreviated as follows: B, Bam HI; Bs, Bss HII; E, Eco RI; F, Fsp I; H, Hind III; N, Nco I; Nh, Nhe I; S, Stu I; Sp, Sph I; T, Tth 1111.  27 (24). The Q. fragment carries transcriptional and translational terminators at both ends and the truncated puJXpolypeptide  carried on pTL2 is only 17 amino acids in length, of which only  the N-terminal methionine is a pufX residue (Figure 2.1). Plasmids pTL2-7 and pTL2-9 were isolated from PS + suppressor mutants of ARC6 (pTL2) and shown to carry the suppressor mutations (23). The suppressor mutations were isolated by plating approximately 107 cells from low oxygen grown cultures of ARC6(pTL2) on RCV minimal medium plates and incubating the plates under photosynthetic growth conditions. Colonies that appeared at about the same time as colonies on plates inoculated with pseudo-wild type cells were tested for the loss of the />«/Vphenotype. About thirty such suppressor mutants were initially isolated, and three phenotypically diverse isolates were selected for further study (22, 23). Plasmid pTL30 is identical to pTL2 except that the Hind III site at the 5' end of the Q. cartridge (see Figure 2.1) has been eliminated by a fill-in reaction. This created an Nhe I site. Cells of R. capsulatus ARC6 reconstituted with pTL2 or pTL30 were phenotypically identical in all ways tested. Plasmid pTL6 (Figure 2.2) was created by digesting pTL2 with Hind III, isolating the 3.3 kb fragment carrying the />«/promoter and pu/Q, B, A, L, and Af genes along with the first codon of the pufiCgene (see Figure 2.1) and ligating this fragment into the Hind III site of p U C l 3 . Plasmids pTL3 and pTL40 were created in the same way except that the 3.3 kb Hind III fragments used to make these plasmids were isolated from the suppressor strain plasmids pTL2-9 and pTL2-7, respectively. Plasmid pA4 contains the pufoperon promoter and the pufQ gene (1), expression of which is necessary for optimal formation of the B800-850 complex in ARC6 (17). Plasmid p U C l 3 (27) was used as a vector for cloning and manipulating R. capsulatus DNA, and plasmids pTZ18 and p T Z l 9 (40) were used as sequencing vectors. Plasmid pET-3a (34) was used as an over-expression vector (see below). This plasmid is a pBR322-based vector. It carries the bacteriophage T7 gene 10 promoter, its Shine-Dalgarno sequence and the first ten codons of this gene followed by a BamH I site. Downstream of the BamH I site there is a T7 transcriptional terminator. There is also a unique Nde I site incorporating the initial ATG codon of gene 10. Plasmid pLysS is a pACYC construct carrying the phage T7 lysozyme gene (34).  28  Figure 2.2. The construction of plasmid pTL6. Vector sequences are shown as thin lines, none-coding R. capsulatus DNA is shown as open lines, the Q cartridge is shown as a heavy black line and the pufoperon genes are shown as open boxes. The abbreviations for the restriction enzymes are as in Figure 2.1.  Hindm  29  EcoRI  HindHI  SphI BssHII  30  Plasmids from E. coli were conjugated into R. capsulatus using tri-parental mating as described in Johnson et al (20). Ten (0,1 volumes from log-phase cultures of the E. coli donor strain, the R. capsulatus recipient strain and E. «>/z(pRK2013), the helper strain, were mixed in a microcentrifuge tube, pelleted and resuspended in 10 u.1 of RCV medium. The resuspended cells were then spotted on RCV medium plates. The auxotrophic donor and helper strains were incapable of growth on this minimal medium. After growth appeared on the initial plate, cells were streaked on an RCV plate containing the antibiotic selective for the plasmid to be transferred. An isolated colony was selected from this second plate and used to inoculate an RCV plate and a YPS plate both of which contained the selective antibiotic. The absence of any E. coli colonies on the YPS plate indicated that a pure isolate of the exconjugant R. capsulatus strain had been obtained, since the auxotrophic E. coli strains used would be capable of growth on this medium had they been carried over. 2.3. Mapping and sequencing of suppressor mutations The suppressor mutations on the plasmids pTL2-7 and pTL2-9 were first localized by mapping experiments. The mapping strategy is outlined in Figure 3.25A. Fragments of puf DNA were subcloned from pTL3 and exchanged for the homologous fragment in pTL6. The Hind III fragment from the hybrid plasmid was then cloned into plasmid pTL30. After being moved into pTL30, the fragments were tested for suppressor mutations by conjugating the plasmid into strain ARC6 and testing this strain for the ability to grow photosynthetically. The strain ARC6(pTB999) was used as a positive control in these experiments. Segments from plasmids pTL2-7 and pTL2-9 that were found to suppress the^w/^phenotype were sequenced using the dideoxy termination method (32). 2.4. Growth experiments All R. capsulatus strains were grown in RCV or YPS medium at 34° C (37, 39). RCV is a minimal medium containing malate as the carbon source and YPS (Yeast, Peptone, Salts) medium is a complex medium. High oxygen cultures were grown in flasks filled to 10% of their nominal volume and shaken at 300 RPM in a rotary shaking water bath. Oxygen-limited cultures were grown in flasks filled to 80% of their nominal volume and shaken at 150 RPM. Photosynthetically grown cultures were grown in screw-cap tubes filled to capacity and held in a glass-sided water bath in front of 60 W Lumiline lamps (General Electric, Ltd.). The light intensity was adjusted when necessary by changing the number of lamps and/or by changing the distance from the cultures to the lamps, and was measured with a model LI-185B photometer  31 equipped with a LI-190SB quantum sensor (Li-Cor; Lincoln, NB). All cultures used in growth experiments were inoculated to an optical density of 20 Klett units (ca. 8 x 10 7 cfu / ml) and growth was followed by measuring the turbidity of the culture using a Klett-Summerson Colorimeter equipped with filter No. 66. Photosynthetic cultures were inoculated from oxygen-limited cultures in stationary phase. Plate cultures were grown on RCV or YPS medium supplemented with 15 g / 1 agar. Media for plasmid-carrying R. capsulatus strains were supplemented with 0.5 fig of tetracycline or 10 fig spectinomycin per ml as appropriate. Photosynthetic plate cultures were grown in anaerobic jars at 34° C. E. coli strains were normally grown in LB medium (32) or, for the preparation of single stranded phage DNA for sequencing, in 2X YT medium (32) . The medium was supplemented with 10 fig tetracycline, 200 fig ampicillin or 50 fig spectinomycin per ml as appropriate. 2.5. Spectroscopy experiments 2.5.1. Absorption spectra A Hitachi U-2000 double-beam spectrophotometer (bandwidth of 2 nm) was used to obtain absorption spectra. About 2 x 109 cells were suspended in 1 ml of a solution of 22.5 % bovine serum albumin in RCV medium and scanned over the wavelength range of 350 to 1000 nm. Chromatophores were diluted to 400 fig of protein per ml in chromatophore buffer ( 20 mM morpholinopropanesulfonate (MOPS), 100 mM potassium chloride, 1 mM magnesium chloride, pH 7.0) and scanned over the same wavelength range. 2.5.2. Flash spectroscopy Flash spectroscopy experiments were carried out on a dual wavelength, double-beam spectrophotometer constructed by the Bio-Instrumentation group at the University of Pennsylvania. Spectra were accumulated using an IBM personal computer interfaced to the spectrophotometer via a Nicolet 3091 oscilloscope. Actinic flashes were supplied by a xenon flash lamp filtered through a Schott RG780 filter. The duration of the flash at half-height was 12 fls and the intensity of the flash under the conditions used was sufficient to photo-oxidize more than 9 5 % of the reaction centers in a sample (30). Typically, chromatophores were suspended in chromatophore buffer at a given concentration, the ambient potential (E/,) was adjusted to approximately 160 mV by adding a  32 few crystals of ascorbate, and any inhibitors used were added to the required concentration. After mixing in the cuvette, the preparations were usually subjected to a train of eight flashes in the apparatus described above. Intact cells were treated in much the same way except the cells were suspended in 20 % sucrose in chromatophore buffer in order to reduce light scattering. 2.5.3. Reduced minus oxidized difference spectra Reduced minus oxidized difference spectra were obtained using a Beckman DU-7 spectrophotometer. A chromatophore suspension was first oxidized by the addition of a few crystals of ferricyanide, a spectrum was recorded over the wavelength range 510 - 580 nm and then the suspension was reduced by the addition of a few crystals of ascorbate plus the redox mediator N-methylphenazonium methosulfate (PMS). A second spectrum was then recorded and the first, oxidized, spectrum subtracted from the result. This difference spectrum was generated three times in succession to ensure that complete reduction with ascorbate had occurred. The ascorbate minus ferricyanide spectrum revealed the relatively high redox midpoint potential cytochromes, mainly c-type cytochromes. Dithionite was then added to the chromatophore suspension, a spectrum recorded and a second difference spectrum was generated. This dithionite minus ferricyanide difference spectrum revealed the lower potential cytochromes, mainly £-type cytochromes. 2.5.4. Chemically-induced and light-induced carotenoid bandshift spectra Chemically-induced and light-induced carotenoid bandshift spectra were obtained with a Hitachi 557 spectrophotometer operating in the dual beam scanning mode. In both cases pairs of identical cuvettes were made up and light reaching the photomultiplier tube was filtered through a blue cutoff filter (which transmitted wavelengths < 600 nm). In the lightinduced carotenoid bandshift experiments a baseline was first recorded and then light filtered through a Wratten 87 filter (which transmitted wavelegths > 800 nm) was used to induce the bandshift in the sample cuvette while the reference cuvette was kept dark. The difference spectrum was recorded over the wavelength range of 550 to 400 nm at 50 nm per minute. In the chemically-induced carotenoid bandshift experiments valinomycin was added to both cuvettes to a final concentration of 2 |lM and a baseline was recorded. Immediately following the simultaneous addition of a pulse of saturated KCl to the sample cuvette and of the same volume of buffer to the reference cuvette, the difference spectrum was recorded over the 550 to 400 nm wavelength range at 50 nm per minute.  33 Light-induced membrane potential formation was also monitored by measuring the bandshift of the potential-sensitive dye Oxonol VI (Molecular Probes Inc., Eugene, OR) (2). In these experiments, the dye was added to chromatophore suspensions to a final concentration of 1.5 JiM and, after recording a baseline, the light minus dark bandshift was measured over the wavelength range of 400 to 650 nm at 50 nm per minute. These measurements were also carried out in the presence of 20 U.M antimycin A. 2.6. Fluorescence measurements Infrared fluorescence was evaluated at different incident light intensities using a technique based on that of Youvan et al. (42). Cultures were grown to stationary phase under low aeration, the cells pelleted and then resuspended to a density of 1.2 x 10 1 0 cfu/ml.  Serial  two-fold dilutions were made from this suspension and 100 |ll of each dilution were placed in a microtiter plate with flat-bottomed wells. The plate was placed over a 1 cm thick 1 M cupric sulfate solution illuminated from below with fluorescent lamps, and photographed through a Wratten 87C infrared filter (Eastman Kodak) using infrared film. The intensity of the excitation light was varied using neutral density filters. 2.7. Fractionation of the photosynthetic unit 2.7.1. Preparation of chromatophores Chromatophores were prepared from cells grown under reduced aeration, which were harvested by centrifugation. Cell densities were typically 4.5 x 10 8 to 5.5 x 10 8 cfu/ml (120 to 150 Klett units). After two washes in chromatophore buffer (see section 2.5.1), cells were concentrated approximately forty fold followed by resuspension in the same buffer, and disrupted by passage through a French pressure cell at 15,000 psi. Following a fifteen minute, room temperature incubation with 5 (ig / ml DNase I and 5 Jig / ml RNase A, the disrupted cells were cleared of unbroken cells and other undesirable debris by centrifugation at 30,000 x g for 20 min. The resultant supernatant fluid was centrifuged at 149,000 x g for 75 min to pellet the chromatophores. The chromatophore pellets were resuspended, washed and finally suspended in a small volume of chromatophore buffer at a concentration of 15 to 50 mg of protein per ml. These preparations were further purified, when necessary, using sucrose gradient centrifugation. Step gradients were made with sucrose solutions prepared using 50 mM Tris - HCl (pH 8.0) buffer. Crude chromatophores were layered on a 10 - 20 - 30 - 40% (wt / wt) sucrose step gradient over a 60% sucrose cushion, and centrifuged at 97,000 x g for  34 230 min. Chromatophores were collected from the 20/30% sucrose interface, diluted with buffer, pelleted (149,000 x g for 75 min) and resuspended in the chromatophore buffer. 2.7.2. Isolation of reaction center-B870 and B800-850 complexes Three approaches were taken to the separation and purification of reaction centers and the B870 light-harvesting antenna complexes associated with them (the so-called reaction center-B870 core complexes) and B800-850 antenna complexes. Initially, two different electrophoretic techniques were used. The first system, developed for fractionation of higher plant chloroplasts (25, 28) and first applied to purple bacteria by Ferguson et al. (15), involves the solubilization of chromatophores with a non-ionic detergent (or mixture of detergents) followed by non-denaturing polyacrylamide gel electrophoresis. This gel system had relatively low ionic strength buffers, no detergent in the gel buffer and 0.2% Deriphat in the cathode buffer. Deriphat 160 (disodium N-lauryl-fi-iminodipropionate; Henkel Corp., Mississauga, ON) is a zwitterionic detergent and substitution of this detergent for SDS in fractionation experiments with chloroplasts resulted in much less disruption of non-covalent protein-protein and pigment-protein interactions (25). Four non-ionic detergents were tested for their ability to solubilize chromatophores. These were dodecylmaltoside [1-O-ndodecyl-P-D-glucopyranosil (1—»4) a-D-glucopyranoside; Boerhinger-Mannheim], LDAO (lauryl dimethylamine oxide), octylglucoside (l-0-n-octyl-(3-D-glucopyranoside; Boerhinger-Mannheim), and MEGA-8 [N-(D-gIuco-2,3,4,5,6-pentahydroxyhexyl)-Nmethyloctanamide; Boerhinger-Mannheim]. Chromatophores were solubilized using these detergents at detergent to bacteriochlorophyll ratios of 20, 25, 30, 35, 40, 45 and 50 to 1 (w/w). The bacteriochlorophyll a concentration was determined as outlined in section 2.13. Normally, samples were loaded directly on to the gel after centrifuging briefly (13,000 x g for two minutes) to remove insoluble material, but incubations of up to thirty minutes at 4° C were done in attempts to improve solubilization. After the samples were loaded, 0.75 mm gels were typically run at 20 mA constant current for thirty minutes at 4° C (15). The second electrophoretic fractionation system used was a lithium dodecyl sulfate (LiDS) gel procedure introduced for the fractionation of chlorophyll ^-protein complexes of Chlamydomonas reinhardtii (11) and adapted for pigment protein complexes of R. sphaeroides by Broglie et al. and Hunter et al. (6, 19). This system essentially involves the substitution of LiDS for SDS in the standard SDS-PAGE system (21), but detergent is eliminated from the gel and EDTA is added to the cathode buffer. The use of LiDS allows gels to be run at 4° C; at this temperature pigment-protein interactions are stabilized. The  35 LiDS to bacteriochlorophyll ratios tested in the sample buffer were 0.1, 0.25, 1, 2, 5, 10, 15, 20, 25, 30, 35 and 40 to 1 (w/w). Two non-ionic detergents, dodecylmaltoside and octylglucoside, were also tested in this system at detergent to bacteriochlorophyll ratios of 10, 20 and 40 to 1 (w/w). The sample buffer was treated with 1 mM phenylmethylsulfonyl fluoride to inhibit protease activity. Chromatophores were mixed with sample buffer and immediately loaded on the gel. Gels were run at 3 W constant power in the cold room for 9 to 12 hours. After gels had finished running in either of the above systems, pigmented bands were cut out of the gel, placed in a cuvette with chromatophore buffer and scanned over the wavelength range 350 to 1000 nm. The slices were then transferred to a microcentrifuge tube, 50 JU.1 of 2x SDS-PAGE buffer were added and the gel was crushed. This mixture was separated by spinning it through silanized glass wool into a second tube. The polypeptides in the band could then be resolved using SDS-PAGE (see below). A third method of fractionation involved solubilization of chromatophores followed by differential centrifugation, using a protocol based on that of Firsow and Drews (16) and Dawkins et al. (10). The purified chromatophores were first diluted with buffer to give an A860 of 50 cm"1. One volume of diluted chromatophores was then mixed with two volumes of 1.5% (wt/wt, prepared in buffer) dodecylmaltoside and stirred for 30 min at 4° C. Insoluble material was removed by centrifugation (13,000 x g for two minutes) and the solubilized chromatophores were layered onto a 10-20-40% sucrose (w/w, made up in 50 mM Tris buffer, pH 8.0) step gradient and centrifuged for 16.5 hours at 200,000 x g. Approximately eighty 0.5 ml fractions were collected and each fraction was scanned over the 700 to 1000 nm region. Fractions showing spectra of pure complexes were pooled, diluted with buffer, pelleted and resuspended to a protein concentration of approximately 20 mg per ml. The polypeptide composition of the fractions was then analyzed using SDS-PAGE (see below). 2.8. SDS-PAGE Separation of proteins by electrophoresis was carried out using the Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis system developed by Schagger and von Jagow (33). This system is based on that of Laemmli (21), but it uses Tricine rather than glycine as the trailing ion. This results in superior resolution of proteins in the 1 to 20 kDa molecular weight range. The separating gels (16.5% T, 3 % C) were made up in 1.0 M TrisHC1 (pH 8.45), 10% glycerol, and 0.03% SDS; the stacking gel (4% T, 3 % C) and a  36 separating or spacer gel (10% T, 3 % C) that was formed between the stacking and separating gel were made up in the same buffer, but with no glycerol. The gels were cast and run using a Bio-Rad Protean II apparatus (Bio-Rad Inc., Richmond, CA). The anode buffer was 200 mM Tris-HCl (pH 8.9), the cathode buffer was 100 mM Tris, 100 mM Tricine and 0 . 1 % SDS (pH 8.25), and the sample loading buffer was 125 mM Tris-HCl, 50 mM dithiothreitol, 50% glycerol 0.1% bromophenol blue (pH 6.8). Samples (20 to 50 JXg protein) were mixed 1:1 with loading buffer, heated to 37° C for 10 min and loaded on the gel alongside pre-stained low molecular weight protein (3 to 43 kDa) standards (Life Sciences Ltd., Mississaugua, ON). After electrophoresis at 10 mA for about one half hour, the current was increased to 18 mA (constant current) and the gels run until the dye front, which ran just ahead of the 3 kDa molecular weight marker, reached the bottom of the gel (typically about 18 hours at room temperature). Gels were fixed in 40% methanol, 10% acetic acid for one half hour and then stained in 40% methanol, 10% acetic acid, 0.025% Coomassie Blue R (Sigma Chemicals) for 1 to 2 hours. Gels were destained in the fixative solution and dried or photographed. 2.9. Cytochrome b/cj complex activity assay The cytochrome b/cj complex activity was assayed as described by Berry and Trumpower (3). Chromatophores to be assayed were purified on sucrose gradients and solubilized in dodecyl-pVD-maltoside (1.5 mg detergent per mg protein, stirred 30 min at 4° C). Any insoluble material was removed by centrifugation (30 min at 138,000 x g). The b/cj complex activity was assayed at each step of purification. In the assay, the b/cj complex was supplied with 60 |lM 2,3-dimethoxy-5-methyl-6-ndecyl-l,4-benzoquinol (DBH2), a ubiquinol analog, as a substrate. Transfer of electrons from DBH2 to oxidized horse heart cytochrome c was followed by monitoring the absorption change at 550 nm on a Hitachi U2000 spectrophotometer. The amount of cytochrome c reduced per minute per milligram of protein was calculated using an extinction coefficient (reduced minus oxidized) of 18.5 mM" 1 cm"1. The assay was carried out in 40 mM phosphate buffer (pH 6.8) with EDTA added to a final concentration of 0.5 mM. Potassium cyanide was added to a final concentration of 250 |lM to inhibit oxidation of reduced cytochrome c. DBH2 was synthesized according to the protocol published by Wan et al. (38) except that flash column chromatography was substituted for preparative TLC. The synthesis was carried out in the laboratory of E. Piers. The substrate was reduced prior to use using a modification (E. Berry, personal communication) of the method published by Trumpower and Edwards  37 (36) in which the initial reductant was sodium thiosulfate and the reduced substrate was finally dissolved in DMSO. 2.10. Electron microscopy Electron microscopy was carried out on sucrose gradient-purified chromatophores from low-oxygen grown cultures. The chromatophores were dialyzed overnight against 2% ammonium acetate, the suspensions mixed with 3 % ammonium molybdate, and placed on copper grids coated with collodion and carbon (31). The excess suspension was gently blotted from the grid and observations were made immediately using a Zeiss EM CIO electron microscope. 2.11. Polyclonal antibodies to the PufX protein To obtain polyclonal antibodies to the PufX protein, rabbits were immunized using a synthetic polypeptide homologous to the first twenty amino terminal amino acids of the putative PufX protein. The polypeptide was obtained from HSC / Pharmacia Biotechnology Service Centre in two forms. In one form, it was conjugated to Ultrasyn D (Kieselguhr) beads; the second form of the antigen was a free peptide with a cysteine residue at the carboxy terminus to facilitate conjugation to chromatographic supports. Two weeks after obtaining pre-immune serum, three rabbits were immunized with about 600 |lg of the polypeptide conjugated to Ultrasyn D using the Ribi adjuvant system (RIBI ImmunoChem Research Inc.) according to the manufacturer's protocol. After two weeks, the rabbits were boosted with a second injection of the antigen plus adjuvant. Two weeks later the rabbits were bled and serum obtained by differential centrifugation. This serum was evaluated by trying Western immunoblotting of chromatophores and by doing ELISA's using both forms of the polypeptide as the antigen. The rabbits were subsequently boosted twice and evaluated for anti-PufX antibodies by doing immunodetection of serial dilutions (1 ng to 10 pig) of both forms of antigen blotted on nitrocellulose paper. In order to reduce the background seen in Western blots, the IgG fractions from the rabbit sera were purified by column chromatography. Serum was first desalted by size exclusion chromatography on a column packed with Bio-Gel P-6 DG gel. The desalted serum was then passed over a DEAE Affi-Gel Blue gel column. This chromatography matrix is a bifunctional affinity gel which removes most of the protein from the serum except IgG and transferrin. Finally, an anti-PufX antibody affinity column was prepared by conjugating the cysteine-derivatized form of the synthetic PufX polypeptides to Affi-Gel 15 activated  38 affinity supports. The purified IgG sera were mixed with the affinity matrix in a batch process overnight at 4" C. Bound antibodies were separated from the matrix by pouring the matrix into a column and using acidic elution conditions. The elution buffer was glycineHC1, pH 2.5; the eluate was collected in tubes containing a neutralizing buffer. The eluate was evaluated for the presence of anti-PufX antibodies using dot blots of the synthetic polypeptides and Western blots of SDS-polyacrylamide gels of sucrose gradient-purified chromatophores. Primary antibody probing of dot blots and Western blots was carried out according to the protocol supplied with the ImmunoSelect immunoblotting system (Life Sciences). This system uses a streptavidin-alkaline phosphatase conjugate to detect biotinylated goat antirabbit antibodies that have bound to the primary antibodies. When the probed blot is incubated with the chromogenic substrates (5-bromo-4-chloro-3-indolylphosphate p-toluidine salt and nitroblue tetrazolium chloride) an insoluble purple precipitate is formed where the alkaline phosphatase has been bound. The procedure used for Western blotting was essentially that given by Sambrook et al. (32). Three blotting buffers were evaluated for their ability to transfer relatively small proteins to the blots. They were Towbin buffer (35), the buffer system of Dunn (13) and the buffer system of Bjerrum and Nielsen-Schafer (5). The latter two buffer systems, at pH 9.9 and 9.2 respectively, are more alkaline than Towbin's buffer, which is a pH 8.3 buffer. Towbin's buffer with SDS added to a final concentration of 0.1% was also evaluated, since SDS can increase the transfer of some proteins (7). Various blotting media were also tested to ensure that the absence of the PufX polypeptide was not due to selective retention properties of the initial blotting medium (BA85 nitrocellulose paper pore size 0.45 U.M; Schleicher and Schuell, Keene.NH). The other media tested were BA83 nitrocellulose paper (pore size 0.22 (iM; Schleicher and Schuell), Nylon 88 membrane and polyvinyldiflouride (PVDF) membrane (Westran; Schleicher and Schuell). Blotting was carried out in a BioRad TransBlot electrophoretic transfer cell (Bio-Rad Laboratories, Richmond, CA). Gels were blotted overnight at 30 V, and the voltage was increased to 60 V the following morning for about two hours. 2.12. Over-expression of pufiC Over-expression of the protein encoded by pufX was attempted using the pET vector system developed by Studier et al. (34). The entire puJX sequence plus approximately 620 bp of downstream R. capsulatus DNA were cloned into the Nde I to BamH I sites downstream  39 of the phage T7 <J>10 promoter in the expression vector pET-3a. Consequently, the initial ATG codon of pupC replaced the corresponding codon of the T7 gene 10, and expression of the cloned gene was under the control of the T 7 promoter and translational initiation signals. This plasmid was used to transform the host strains E. coli BL21 (DE3) and E. coli BL21(DE3)(pLysS). These strains have been lysogenized with a X derivative, called DE3, that has had the gene encoding the T7 RNA polymerase inserted into the X int gene. Expression of the T7 RNA polymerase gene in this construct is controlled by the lac UV5 promoter which is inducible by isopropyl-f3-D-thiogalactopyranoside (IPTG). Since the ^>10 promoter is recognized only by T7 RNA polymerase, there should be little or no expression of  pufXuntil  IPTG is added to the growth medium. At this point, because T 7 RNA polymerase is such an active enzyme, the cloned gene should be transcribed and translated at very high levels. The strain E. coli BL21(DE3)(pLysS) carries a plasmid encoding the T7 lysozyme protein, a protein known to degrade the T7 RNA polymerase. Supplying this protein in the same cell as the over-expression vector ensures that any background expression of the polymerase from the lysogenic phage and, hence, any low-level expression of the gene targeted for over-expression will be suppressed. This is important if the target gene encodes a product that is toxic to the cell. The lysozyme is unable to degrade the amount of polymerase produced when expression of the polymerase gene has been induced. 2.13. Other assays Protein measurements were done using the Lowry assay as modified by Peterson (29) with bovine serum albumin as the standard. The concentrations of bacteriochlorophyll a and reaction center for the spectroscopy experiments were determined by the method of Dutton et al. (14). The extinction coefficients for bacteriochlorophyll (£865 = 9 5 mM*1 cm -1 ) and the reaction center bacteriochlorophyll dimer (£605-540 = 29.8 mM" 1 cm"1) as determined for Rhodobacter sphaeroides Ga (14) were used. Total bacteriochlorophyll a was extracted from cells or chromatophores with acetone:methanol (7:2) and the concentration determined using the extinction coefficient (£770 = 76 mM" 1 cm"1; [9]). 2.14. Materials and methods references 1. Adams, C.W., M.E. Forrest, S.N. Cohen and J.T. Beatty. 1989. Structural and functional analysis of the Rhodobacter capsulatus puf operon. J. Bacteriol. 171:473-482.  40 2. Bashford, C.L., B. Chance and R.C. Prince. 1979. Oxonol dyes as monitors of membrane potential their behavior in photosynthetic bacteria. Biochim. Biophys. Acta 545:4657. 3. Berry, E. and B. Trumpower. 1985. Isolation of ubiquinol oxidase from Paracoccus denitrificans and resolution into cytochrome blc and cytochrome aa^ complexes. J. Biol. Chem. 260:2458-2467. 4. Bibb, M.J. and S.N. Cohen. 1982. Gene expression in Streptomyces: construction and application of promoter-probe plasmid vectors in Streptomyces lividans. Mol. Gen. Genet. 187:265-277. 5. Bjerrum, O.J. and C. Schafer-Nielsen. 1983. pp. 315-. In M. J. Dunn, Analytical Electrophoresis. VCH, Weinheim. 6. Broglie, R.M., C.N. Hunter, P. Delepelaire, R.A. Niederman, N.-H. Chua and R.K. Clayton. 1980. Isolation and characterization of the pigment-protein complexes of Rhodopseudomonas sphaeroides by lithium dodecyl sulfate/polyacrylamide gel electrophoresis. Proc. Natl. Acad. Sci. USA 77:87-91. 7. Burnette, W.N. 1981. "Western blotting" Electrophoretic transfer of proteins from sodium dodecyl sulfate polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem. 112:195-203. 8. Chen, C.-Y.A., J.T. Beatty, S.N. Cohen and J.G. Belasco. 1988. An intercistronic stem-loop structure functions as an mRNA decay terminator necessary but insufficient for puf mRNA stability. Cell 52:609-619. 9. Clayton, R.K. 1966. Spectroscopy of bacteriochlorophyll. Photochem. Photobiol. 5:807-821. 10. Dawkins, D.J., L.A. Ferguson and R.J. Cogdell. 1988. The structure of the 'core' of the purple bacterial photosynthetic unit, pp. 115 - 127. In H. Scheer and S. Schneider, Photosynthetic light-harvesting systems. Walter de Gruyter and Co., Berlin. 11. Delepelaire, P. and N.-H. Chua. 1979. Lithium dodecyl sulfate/polyacrylamide gel electrophoresis of thylakoid membranes at 4° C: Characterizations of two additional chlorophyll a-protein complexes. Proc. Natl. Acad. Sci. USA 76:1 11-115. 12. 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. 13. Dunn, S.D. 1986. Effects of the modification of transfer buffer composition and the renaturation of proteins on the recognition of proteins on Western blots by monoclonal antibodies. Anal. Biochem. 157:144-153. 14. Dutton, P.L., K.M. Petty, H.S. Bonner and S.D. Morse. 1975. Cytochrome c^and reaction center of Rhodopseudomonas spheroides Ga. membranes. Extinction coefficients,  41 content, half-reduction potentials, kinetics and electric field alterations. Biochim. Biophys. Acta 387:536-556. 15. Ferguson, L., E. Halloran, A.M. Hawthornwaite, R. Cogdell, C. Kerfeld, G.F. Peter and J.P. Thornber. 1991. The use of non-denaturing Deriphat-polyacrylamide gel electrophoresis to fractionate pigment-protein complexes of purple bacteria. Photosynth. Res. 30:139-143. 16. Firsow, N.N. and G. Drews. 1977. Differentiation of the intracytoplasmic membrane of Rhodopseudomonas palustris induced by variations of oxygen partial pressure or light intensity. Arch. Microbiol. 115:299-306. 17. Forrest, M.E., A.P. Zucconi and J.T. Beatty. 1989. The pu/Qgene product of Rhodobacter capsulatus is essential for formation of B800-850 light-harvesting complexes. Curr. Microbiol. 19:123-127. 18. Hanahan, D. 1983. Studies on transformation of Escheria coli with plasmids. Mol. Biol. 166:557-580. 19. Hunter, C.N., J.D. Pennoyer, J.N. Sturgis, D. Farrelly and R.A. Niederman. 1988. Oligomerization states and associations of light-harvesting pigment-protein complexes of Rhodobacter sphaeroides as analyzed by lithium dodecyl sulfate-polyacrylamide gel electrophoresis. Biochemistry 27:3459-3467. 20. Johnson, J.A., W.K.R. Wong and J.T. Beatty. 1986. Expression of cellulase genes in Rhodobacter capsulatus by use of plasmid expression vectors. J. Bacteriol. 167:604-610. 21. Laemmli, D.M. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. 22. Lilburn, T.G. 1990. The role of the pufi(gene product of Rhodobacter capsulatus. M. Sc. thesis, University of British Columbia. 23. Lilburn, T.G. and J.T. Beatty. 1992. Suppressor mutants of the photosynthetically incompetent puflCdeletion mutant Rhodobacter capsulatus ARC6(pTL2). FEMS Microbiol. Lett. 100:155-160. 24. Lilburn, T.G., C.E. Haith, R.C. Prince and J.T. Beatty. 1992. Pleiotropic effects of pufXgene deletion on the structure and function of the photosynthetic apparatus of Rhodobacter capsulatus. Biochim. Biophys. Acta 1100:160-170. 25. Markwell, J.P., J.P. Thornber and R.T. Boggs. 1979. Higher plant chloroplasts: Evidence that all the chlorophyll exists as chlorophyll-protein complexes. Proc. Natl. Acad. Sci. USA 76:1233-1235. 26. Marrs, B.L. 1974. Genetic recombination in R. capsulata. Proc. Natl. Acad. Sci. USA 71:971-973. 27. Messing, J. 1983. New M13 vectors for cloning. Methods. Enzymol. 101:20-78.  42 28. Peter, G.F. and J.P. Thornber. 1991. Electrophoretic procedures for fractionation of photosystems I and II pigment-proteins of higher plants and for determination of their subunit composition, pp. 195-210. In L. J. Rogers, Methods in Plant Biochemistry. Academic Press, San Diego. 29. Peterson, G. 1983. Determination of total protein. Meth. Enzymol. 9 1 : 95-119. 30. Prince, R.C., E. Davidson, C.E. Haith and F. Daldal. 1986. Photosynthetic electron transfer in the absence of cytochrome c'm Rhodopseudomonas capsulata: cytochrome c2 is not essential for electron flow from the cytochrome be complex to the photochemical reaction center. Biochemistry 25:5208-5214. 31. Robards, A.W. and A.J. Wilson. 1993. Negative staining of subcellular components, bacteria, viruses and macromolecules, pp. 5:8.1-2. In Procedures in Electron Microscopy. John Wiley and Sons, Chichester. 32. Sambrook, J., E.F. Fritsch and T. Maniatis. 1989. Molecular Cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Plainview, New York. 33. Schagger, H. and G. von Jagow. 1987. Tricine-sodium dodecyl sulfatepolyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal. Biochem. 166:368-379. 34. Studier, F.W., A.H. Rosenberg, J.J. Dunn and J.W. Dubendorff. 1990. Use of T7 RNA polymerase to direct expression of cloned genes. Methods Enzymol. 185:60-89. 35. Towbin, H., T. Staehelin and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4353. 36. Trumpower, B. and C. Edwards. 1979. Purification of a reconstitutively active iron-sulfur protein (oxidation factor) from succinate-cytochrome c reductase complex of bovine heart mitochondria. J. Biol. Chem. 254:8697-8706. 37. Wall, J.D., P.F. Weaver and H. Gest. 1975. Gene transfer agents, bacteriophages, and bacteriocins of Rhodopseudomonas capsulata. Arch. Microbiol. 105:217-224. 38. Wan, Y.-P., R.H. Williams, K. Folkers, K.H. Leung and E. Racker. 1975. Low molecular weight analogs of coenzyme Q as hydrogen acceptors and donors in systems of the respiratory chain. Biochem. Biophys. Res. Commun. 63:1-15. 39. Weaver, P.F., J.D. Wall and H. Gest. 1975. Characterization of Rhodopseudomonas capsulata. Arch. Microbiol. 105:207-216. 40. Yanisch-Perron, C , J. Vieira and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of M13 and pUC vectors. Gene 33:103-119. 41. Yen, H.C., N.T. Hu and B.L. Marrs. 1979. Characterization of the gene transfer agent made by an overproducer mutant of Rhodopseudomonas capsulata. J. Mol. Biol. 131:157-168.  43 42. Youvan, D.C., J.E. Hearst and B.L. Marrs. 1983. Isolation and characterization of enhanced fluorescence mutants of Rhodopseudomonas capsulata. J. Bacteriol. 154:748-755.  44  3.  RES U L T S 3.1. The primary mutant 3.1.1. The puflC phenotype: The effects of the loss of PufX 3.1.1.1. Growth experiments As mentioned in the Introduction, it was determined that the pufXgene product was essential for photosynthetic growth of R capsulatus when the cells were grown in minimal medium at a light intensity of 120 tlE-nr 2 ^" 1 (13); cultures of the primary mutant strain inoculated and placed under photosynthetic growth conditions did not grow appreciably until a suppressor mutation occurred. At this light intensity, the length of time that passed before the growth of a suppressor mutant was apparent varied considerably, as did the growth rates of the suppressor mutants. Some evidence indicated that the length of time before growth occurred in cultures of the primary mutant strain was shorter at high light intensities. The effects of the light intensity and of the composition of the growth medium on the ability of the primary mutant strain to grow photosynthetically were further investigated. The effects of light intensity on the photosynthetic growth of cultures of the pseudowild type strain ARC6(pTB999) and the primary mutant strain ARC6(pTL30) are shown in Figure 3.1. Six light intensities were used: 1400, 250, 120, 60, 30 and 15 (lE-m"2^"1. As shown in Figure 3.1 A, the pseudo-wild type strain showed little or no lag time before initiation of exponential growth under any light condition; the primary mutant strain (Figure 3.IB) appeared to be able to grow at 1400 uE-nr 2 ^" 1 , since there was little or no lag time before photosynthetic growth started (see inset). At light intensities of less than 1400 |lE-m~ 2  -s" 1 the time passing before suppressor mutations appeared in cultures of the primary mutant  strain increased with decreasing light intensity. Interestingly, although the pseudo-wild type strain grew more slowly at lower light intensities (inset, Figure 3.1A), the suppressor mutants grew with about the same kinetics at all light intensities, once growth began. The effect of providing a richer growth medium for the cells on their ability to grow photosynthetically was assessed in two experiments. In the first experiment photosynthetic growth of the primary mutant strain in minimal, rich and blended media was compared. In  45  500  100  D ts  0  10  20  30  40  50  60  10  </2  § Q  500  3  u 100  10 100  200  300  400  500  600  Time (Hours) Figure 3.1. Light titration experiments on the pseudo-wild type and primary mutant strains. Phototrophic growth of the R. capsulatus ARC6 strains (pTB999) (A) and (pTL30) (B), at six light intensities. A , 1400 jlE-nr 2 ^- 1 ; A, 250 u E - n r V 1 ; • , 120 uE-nr 2 ^- 1 ; • , 60 15 M-E-nr 2 ^" 1 . The insets show the phototrophic growth of the |lE-m -2.C-1. l,30\iE-m-2-s-hO, same cultures during the first sixty hours.  46  the second, increasing amounts of yeast extract were added to minimal RCV medium and the effect of this addition on the photosynthetic growth of the primary mutant strain was assessed. Figure 3.2 compares the 1000  photosynthetic growth at 120 JiE-nr 2 ^" 1 of the pseudo-wild type strain and the primary mutant strain in minimal (RCV)  100  medium, rich (YPS) medium and a 1:1 blend of these two media. It was evident that the  a  pseudo-wild type strain had a  10  I  40  80  120  200  160  shorter doubling time when the minimal medium was blended 1:1  4>  with the rich medium or when it  o  1  1000  B  3  was supplanted by the rich medium. With the mutant strain, enriching the growth medium had two effects: it shortened the lag  100  time before photosynthetic growth could occur and it shortened the doubling time once growth began. The growth curve of the mutant  10 40  80  120  160  200  strain in the 1:1 blend of rich and minimal media was diauxic. This  Time (Hours)  implies that some component or Figure 3.2. Photosynthetic growth of the pseudo- components in the rich medium wild type and primary mutant strains on minimal, rich  allowed the primary mutant to  and blended media. Photosynthetic growth of the  overcome the loss of pufi(without  R. capsuktus ARC6 strains (pTB999) (A) and (pTL30)  undergoing a secondary mutation.  1  (B) at a light intensity of 120 fjE-nr^s" in: O , minimal  In the blended medium, this  RCV medium; • , YPS medium; and • , a 1:1 blend of  component was exhausted before  these media.  the culture entered stationary phase and growth stalled until a  47 suppressor mutation arose. This hypothesis is supported by the two different growth rates {i.e., T j = 3.8 hr vs. Td = 9.7 hr) seen in the two separate log growth phases. In Figure 3.3, the results of supplementing minimal medium with three different concentrations of yeast extract are shown. Once again, cultures were 1000  A  grown at 120 plE-m"2^"1, and the results were similar to those seen in Figure 3.2. As the growth medium is enriched, the growth  100  rate of the pseudo-wild type strain increases as does that of the primary mutant. However, at a  c D  yeast extract concentration of 10  20  40  60  80  0.4% growth of the primary mutant was initially inhibited. The lag time seen in the primary  c o Q  e 3  mutant decreased in richer media and a diauxic growth curve was  1000  observed when the yeast extract  •4-1  "3 u  concentration was 0.1%, although in a different experiment a yeast extract concentration of 0.1% was  100  sufficient to carry the culture into stationary phase. The observations that  10 20  40  60  80  Time (Hours) Figure 3.3. Photosynthetic growth of the pseudowild type and primary mutant strains on minimal medium and media supplemented with yeast extract. Photosynthetic growth of the R. capsulatus ARC6 strains (pTB999) (A) and (pTL30) (B), at a light intensity of 120 ,uEinsteins • m~2 . s*1 in: O , minimal RCV medium; • , RCV plus 0.1 % yeast extract; # , RCV plus 0.2 % yeast extract; and • , RCV plus 0.4 % yeast extract.  cultures of the primary mutant were able to grow under photosynthetic growth conditions when supplied with a richer medium, and that this growth seemingly stopped when some component of the richer medium was exhausted, led me to suspect that an alternative electron  48 transport pathway was being used by the cells in these cultures. One candidate pathway was the branch of the respiratory pathway that uses dimethylsulfoxide (DMSO) or tetramethylammonium N-oxide (TMAO) as a terminal electron acceptor (see Introduction). This pathway allows growth by sugar fermentation under anaerobic conditions. When cultures of the pseudo-wild type and primary mutant strains were placed in RCV minimal medium supplemented with DMSO they were unable to grow in the dark. 500  This was expected, since R. capsulatus cells require a fermentable carbon source for generation of ATP during dark anaerobic growth, and RCV medium contains the non-fermentable carbon source malate. When identical cultures were placed in front of a light source, not only was the pseudo-wild 10  20  30  40  50  Time (Hours)  60  type strain able to grow as expected, but also the primary mutant was now able to grow as well as the pseudo-  Figure 3.4. Photosynthetic growth on DMSO-supplemented medium. Photosynthetic growth of the R. capsulatus ARC6 strains (pTB999) and (pTL30) at a light intensity of 120 \\E-m'2-srl. A, (pTB999), in RCV minimal medium supplemented with dimethylsulfoxide at a concentration of 20 mM; O , (pTL30), in unsupplemented RCV minimal medium; and • , (pTL30), in RCV minimal medium supplemented with dimethylsulfoxide at a concentration of 20 mM. The expression of the enzyme dimethylsulfoxide reductase was pre-induced in all cultures by the addition of dimethylsulfoxide to the low oxygen grown cultures used to inoculate these cultures.  wild type (see Figure 3.4). When the cultures used as inocula for the growth experiments were induced for DMSO reductase, the lag time prior to the start of photosynthetic growth was uniformly short. When "naive" cultures were used as the inocula, the lag times in the primary mutant cultures varied widely (data not shown). Subcultures from all DMSO-supplemented, photosynthetically grown primary mutant cultures were unable to grow photosynthetically in RCV medium  that was not supplemented with DMSO, so the growth seen in the original cultures was not due to the occurrence of suppressor mutations. Four concentrations of DMSO were tested (10, 20, 40 and 80 mM; data not shown). At the lowest concentration (10 mM) growth was seen to stop  49 at about 250 Klett units; this was analogous to the stall in growth seen in primary mutant cultures supplemented with 0 . 1 % or 0.2% yeast extract. 3.1.1.2. Single-flash spectroscopy The inability of the primary mutant to grow photosynthetically did not correlate with the surprising observations that the individual components of the photosynthetic unit were fully functional (13). An experiment was designed to test the interaction between the blc\ complexes and the reaction centers. Electron movement through the reaction center and the blc\ complex can be monitored by observing the carotenoid bandshift, which is divided into three phases. Phase I is associated with electron movement from the reaction center special pair to the primary acceptor Q A (i-e-< photooxidation) and Phase II is associated with electron movement from cytochrome Q  or c  y to the special pair (i.e., the rereduction of the special  pair) (16). Phases I and II, the reaction center's contribution to the carotenoid bandshift, occur within 300 (Is after a flash. Phase III arises from electron movement through the blc\  complex  from cytochrome b\ to cytochrome bh (60%) and from cytochrome b^ to the Q c site (40%), and reaches its maximum amplitude within about 30 ms after a flash (16), although electron movement from the Q z to Q c only takes about 8 ms (2). We observed the carotenoid bandshift after a single saturating flash of light which stimulated one turnover of the cyclic photosynthetic electron transport pathway. Two sets of conditions were used. In the first set of experiments, the ambient potential (E/) of the chromatophore preparations was poised at ~ 90 mV. At this potential, which matches the midpoint potentials (Em7) of the quinone pool and the Q z site, oxidation of the quinol at the Q z site is not thought to be limited by diffusion of quinols to this site (6). In the second set of experiments, the Ef, was poised at = 225 mV. Under these conditions, the quinone pool and the Q z s ' t e  are  more than 90%  oxidized, and quinol oxidation at the Q z s ' t e dependsupon the arrival of quinols from the reaction center. One might expect that the reaction center Qg quinone, which has a midpoint potential of 0 mV, would be oxidized and that formation of a quinol after a single flash would therefore be precluded under these conditions. However, it is known that the Qg semiquinone is very stable and when the special pair is reduced, as it is under these conditions, the semiquinone has been shown to persist for up to twenty minutes under oxidizing conditions (2, 18). A control experiment was carried out in which preparations, which had been poised oxidized, were flashed in the presence of antimycin A. This inhibitor blocks electron movement through the blc\ complex.  50 Thus, the results of flashing chromatophores in the presence of antimycin A, shown in Figure 3.5, confirm that reaction centers in the primary mutant are largely functional, since the change in absorption, which is proportional to the sum of Phases I and II of the carotenoid bandshift (not resolved on this time scale), was 80% of that seen in the pseudo-wild type. In Figure \lil\^^^t^^'u^^^l^AMr^^|^f^J^l<fl'  3.5A, which shows the results for the pseudo-wild type strain, we see NWv»^ A A<Jk* , '' w W\*^^  A= 0.011  10 ms  Qzox. + Ant  a small but definite reduction in  Qzox. Qz red.  the oxidizing potential (trace  the extent and rate of Phase III at marked Q z ox.). The results for  B  the primary mutant strain, shown in Figure 3.5B, are quite different. When chromatophores were poised oxidized, Phase III was almost abolished (trace marked Q z  ox  -)  and the rate sharply reduced. Even Figure 3.5. Single flash carotenoid bandshift experiments. Single flash carotenoid bandshift kinetics in chromatophores from R. capsulatus crtDAKC6 strains (pTB999) (A) and (pTL2) (B). Chromatophores were suspended in chromatophore buffer to a final bacteriochlorophyll concentration of 25 |J-M. 5 (J.M diaminodurene-,2,3,5,6-tetramethylphenylenediamine, N-methylphenazonium methosulfate and N-ethylphenazonium ethosulfate were added as redox mediators. The redox midpoint potential was poised at Eh » 90 mV (Qz red.) or Eh » 225 mV (Qz ox.), and antimycin added (+ Ant.) to 2 |lM where indicated. A single flash was delivered and the kinetics of the optical change at the wavelength pair 490 - 475 nm were monitored.  when the Q z was reduced prior to flashing the extent of Phase III was less than half the extent seen in the pseudo-wild type (trace marked Q z red.), perhaps because of an impairment in the rate of rereduction of Q z by quinols from the reaction center. 3.1.1.3. Quantitation of quinones One possible explanation  of the impaired kinetics of quinol oxidation by the blc\  complex and poor photosynthetic  growth of the primary mutant was that the increased membrane volume (membrane invaginations) induced by low pC>2 was diluting the quinone pool disproportionately relative to the pseudo-wild type. This would prevent the cells from growing when placed under photosynthetic growth conditions until enough quinones had been synthesized to allow normal  51 electron flow. However, comparison of relative concentrations of quinones in chromatophores from the primary mutant strain and the pseudo-wild type strain showed that, at membrane protein concentrations of=10 mg/ ml, chromatophores from the pseudo-wild type strain contained quinones at a concentration of 1.8 |J.M whereas the concentration in the primary mutant strain was 1.7 |LlM. Additionally, the demonstration that low oxygen grown mutant strain cultures showed no lag when switched to high oxygen growth conditions (as they did when switched to photosynthetic growth conditions) spoke against this possibility since the quinone pool is also a necessary part of the respiratory electron transport chain. 3.1.1.4. Assay of blc\ complex activity Chromatophore preparations from the crtD* background strains were previously used to assay electron flow through the solubilized cytochrome blc\  complex using an excess of DBH2  as electron donor. These experiments were repeated with the ARC6crtD background strains to ensure that the crtD mutation did not affect this aspect of the pufXdeletion  mutant phenotype.  The rates obtained (see Table 3.1) indicated that there was no great difference in the specific activities of solubilized cytochrome blc\  complexes from the two strains, as was observed in  the ARC6 background.  TABLE 3.1. Cytochrome b/cl complex specific activitya measured in two strains of R. capsulatus.  ARC6  Crude Chromatophores  Solubilized Crude Chromatophores  Solubilized Chromatophores (Pellet)'3  Solubilized Chromatophores (Supernatant Liquid) b  76  107  20  192  107  121  26  173  crtD(pTL2) ARC6 m£>(pTB999) a  Umoles cytochrome c reduced • min "' • mg protein "'  b  solubilized chromatophores were spun at 100,000 x g for 90 minutes  52 An attempt was also made to titrate the levels of DBH2  to test the affinity of the  cytochrome blc\ complex from each strain for this substrate. Both strains reacted in a similar way as the concentration of DBH2was reduced to 4 flM from 60 |lM (data not shown) 3.1.2. Immunological detection of PufX It was thought that if there was an in vitro assay for the PufX protein, then we might learn something about its function by fractionating pseudo-wild type cells, assaying the fractions for the presence of PufX and determining with which proteins PufX was associated. Therefore, we attempted to obtain anti-PufX antibodies. 3.1.2.1. Over-expression of PufX We first tried to obtain sufficient PufX protein for the immunization of rabbits by producing high levels of PufX in E. coli. One of the pET vectors constructed in W. Studier's lab was used in attempts to over-express pufi(. In this system, the puJXgene was placed on the plasmid pET-3a downstream of the phage T7 promoter and two strains of £ coli were transformed: BL21 DE3 and a strain that controls expression from the T7 promoter more stringently, BL21 DE3 (pLysS). Expression was induced by addition of IPTG to the growth medium. Transcription of the gene of interest normally occurs at such a high level that the cell's translational machinery produces virtually no other protein. Five controls were done to ensure over-expression was occurring or could occur. Plasmid DNA isolated from these strains was shown to carry the pufi(gene by restriction enzyme digest. Control platings of cultures of £ coli BL2\ DE3 (pET-3x.:pufX) and E. coli BL2\ DE3 (pLysS) (pET-3x.:pufiO were done to show that IPTG was indeed inducing expression of theT7 RNA polymerase (i.e., there was no growth on plates containing the inducer). These platings also indicated that the expression vector was not being lost prior to induction, since 70 to 100% of the cells taken from cultures just prior to induction were resistant to ampicillin, which was the antibiotic resistance marker carried by the expression vector. However, evidence that the pufi(gene was being transcribed, as assessed by agarose gel electrophoresis of total cellular RNA, or translated, as assessed by SDS-polyacrylamide gel electrophoresis of total cell protein, was not obtained. Therefore, this approach was abandoned.  53 3.1.2.2. Polyclonal antibodies to a synthetic peptide based on the PufX sequence Since we were unable to obtain PufX protein, we turned to immunization of rabbits with a synthetic polypeptide having the same sequence as the first twenty-one amino acids of PufX Three attempts to raise polyclonal antibodies to PufX were made as outlined above (Section 2.12). The purified IgG obtained detected the synthetic polypeptide used as the antigen in dot blot experiments, but it did not detect any unique bands on Western blots of chromatophores from the pseudo-wild type strain when compared to the primary mutant. 3.1.3. Fractionation of the photosynthetic unit Simultaneous with the first attempts to raise antibodies to PufX, the development of protocols for the fractionation of the photosynthetic unit was started. The original intent was to probe the fractions with anti-PufX antibodies, but it was also hoped that the simplified banding patterns of the purified fractions on SDS-polyacrylamide gels would allow us to identify a PufX band by comparing the gel 0.5  I  profiles of the pseudo-wild type and primary mutant strains. 3.1.3.1. Electrophoretic fractionation The first protocols tried were based on existing electrophoretic fractionation protocols i  1  700  800  1  1  850 900  Figure 3.6. Gel fractionation results. Absorption spectra of eight gel slices from a lithium dodecyl sulfate gel of sucrose gradient purified chromatophores from R. capsulatus (pTB999). The spectra are arranged in the same relative position as the bands excised from the gel, with slice (a) located nearest the origin and slice (h) located just above the free pigments.  developed for other phototrophic Proteobacteria. None of the electrophoretic fractionation protocols used led to reproducible isolation of B800-850 and B870-reaction center complexes. Figure 3.6 shows scans of the bands from a lithium dodecyl sulfate gel. Apparently pure B800-850 light harvesting complexes were obtained using this technique (e.g., slices c and d), as well as with the Deriphat gel protocol (data not shown), but no B870-reaction center complex-specific band was ever seen using  54 either system. The different bands seen on the lithium dodecyl sulfate gel may correspond to B800-850 complexes in different oligomerization states (8). 3.1.3.2. Density gradient 2.5  fractionation In further attempts to isolate B870-reaction center  1.5  and B800-850 complexes, I next tried detergent solubilization followed by  0.5 B o  ^s&  «aa  0  I  differential centrifugation as  V  10  20  30  40  50  60  70  80  described by Fidai et al. (4). Initially, promising results were obtained using n-octyl-GD-glucopyranoside  B  (octylglucoside) as the solubilizing detergent. The optimal detergent  1.5  concentration was found to be  ^  3.5% (wt/wt). Optimization  is  experiments were done on relatively small volumes (80 fll) of chromatophores from  0.5 L  « miilllH—  10  20  30  *  strains AKC6crtD (pTB999)  ***  40  and (pTL30), but I was unable 50  60  70  80  to successfully apply the procedure to larger volumes of  Fraction Number Figure 3.7. Density gradient fractionation of detergent solubilized chromatophores. Absorption at 855 nm ( • ) and 875 nm (A) of eighty fractions from a sucrose density gradient separation of dodecylmaltosidesolubilized chromatophores from the R. capsulatus ARC6 strains (pTB999) (A) and (pTL30) (B). Fractions were taken starting at the 40%/60% sucrose interface.  chromatophores from the crtD* versions of these strains. Returning to the fractionation protocol this technique was based upon (3, 5) and trying different detergents led to a successful result using dodecyl  55 maltoside. Chromatophores from strains ARC6(pTB999) and (pTL30) were solubilized and fractionated on a sucrose step gradient. The gradient was divided into eighty fractions and the absorption of each fraction was measured at 855 and 875 nm, which are the approximate wavelengths of the long-wavelength absorption maxima for B800-850 and B870 lightharvesting antenna complexes respectively. These data are shown in Figure 3.7. When the absorption profiles of the two strains are compared it can be seen that both preparations had similar levels of the B800-850 light-harvesting complex. This was because the initial amount of chromatophores from each strain was normalized to absorption at 850 nm. It was also apparent that the primary mutant strain (Figure 3.7B) had higher levels of B870 complex, as would be predicted from absorption spectra of intact cells. Absorption spectra were taken at each step of the purification process and are shown in Figure 3.8. As mentioned above, the higher amount of the B870 complex in the primary mutant is visible as a pronounced shoulder on the red-most 850 nm peak in spectra of intact cells and chromatophores (Figure 3.8[i], [ii] and [iv]). Spectra of partially purified complexes from each strain are shown in Figure 3.8 [v] and [vi]. Comparison of the spectra of the fraction enriched in the B870-reaction center complexes from the two strains (Figure 3.8[v]) indicates that the complexes from the pseudo-wild type strain are not free of B800-850 complex, as evidenced by the shoulder on the left-hand side of the B870 peak and the greater relative absorption at around 800 nm. It was also seen that the B870-reaction center enriched fraction from the pseudo-wild type strain had more absorption in the carotenoid region (around 500 nm) than did the comparable fraction from the primary mutant. After the two types of complex had been enriched from each strain, equal amounts (of total protein) were run on a denaturing polyacrylamide gel. Figure 3.9 shows a photograph of such a gel. The reaction center-B870 complex lanes showed what appeared, at first glance, to be a significant number of differences between the two strains. The differences, however, were interpreted as due to the greater amount of the B870 complex per unit protein in the primary mutant strain, and the better resolution of the B870-reaction center complexes from the B800850 complexes in the sucrose gradient with this strain. None of the bands that were present in the pseudo-wild type lanes were absent from the primary mutant lanes.  56 3.2. The suppressor mutants 3.2.1. Suppressor phenotypes As mentioned above, it was hoped that, by learning how the suppressor mutants got around the impairments caused by the lack of PufX, we might learn more about the function of PufX. Therefore, we subjected the suppressor strains chosen for further study, ARC6*(pTL2), ARC6(pTL2-7) and ARC6(pTL2-9), to many of the same experiments carried out with the pseudo-wild type and primary mutant strains. 3.2.1.1. Growth experiments The suppressor strains, were compared to the pseudo-wild type strain ARC6(pTB999) and the primary mutant strain ARC6(pTL30) in the same type of growth experiments described in section 3.1. That is, the effects of different light intensities and of different compositions of the growth medium were examined independently. Cultures of the five strains were grown at light intensities of 15, 30, 60, 120 and 250 (lE-m^-s"1 (Figure 3.10). Reductions of the light intensity resulted in decreases in the growth rates of the pseudo-wild type (pTB999) and the suppressor strains. The decreases in growth rates of the suppressor strains were larger than the decreases seen in the pseudo-wild type and at light intensities <30 |lE-m~2-s-1 the growth of the suppressor strains was very slow or virtually non-existent until after long lags. It was assumed that the growth that was eventually obtained was due to a second suppressor mutation. These mutations seemed to allow these strains to use light energy more efficiently since the slopes of the growth curves changed positively. In the strain ARC6*(pTL2) (Figure 3.10C) these mutations invariably resulted in the appearance of the B800-850 complex, as seen when the spectra in Figure 3.1 IB are compared. After growth at 120 l l E - n r 2 ^ 1 the spectrum of the cells was typical of this strain: there is virtually no carotenoid absorption (in the 485 nm region) and, in the near infra-red, the only significant absorption is due to the reaction center at 800 nm and the B870 light-harvesting complex peak. The obvious change in absorption in the remaining strains was an increase in absorption at 860 nm after growth at 15 uE-nv-^s"1. This is due to increased production of the B800-850 complex, a change that is normal when R. capsulatus strains are grown at low light intensities. However, since the suppressor strains were believed to have accumulated cells with an additional mutation in response to the selective pressure of the low light intensities, it was  57  Figure 3.8. Monitoring the progress of the purification of the B800-850 lightharvesting and reaction center-B870 light-harvesting complexes with absorption spectroscopy. Purification of B800-850 light-harvesting and reaction center-B870 light-harvesting complexes from R. capsulatus ARC6 strains (pTB999) (A) and (pTL30) (B) evaluated using absorption spectroscopy. Spectra shown are of: (i), intact cells grown under low oxygen conditions; (ii), crude chromatophores prepared from these cultures; (iii), the supernatant liquid above the crude chromatophore pellet; (iv), the chromatophores after sucrose gradient purification; (v), fraction 49 from sucrose density gradient separations of solubilized chromatophores (see Figure 3.7); and (vi), fraction 60 from the same sucrose density gradients.  (i)  (ii)  (iii)  (iv)  (v)  (vi) I 1000  —1 400  Nanometers  1600  -T 800  1 1000  B870 , ^ ON ON ON  M H a  B800-850 ^^  ,*—^  o CO J H a.  ON ON ON  M HQ.  ^ • ^  o CO J H o.  43.1 —  29.2 —  H M —L  18.8 — 16.5  — B800-850y  — B870a — B800-850a — B870(3 N 6800-850(3 Figure 3.9. SDS-PAGE of samples taken at successive steps in the purification of the B800-850 light-harvesting and reaction center-B870 light-harvesting complexes. Photograph of a SDS-polyacrylamide gel of enriched B870-reaction center complex (lanes 1 and 2) and enriched B800-850 complex (lanes 3 and 4) from the R. capsulatus ARC6 strains (pTB999) (lanes 1 and 3) and (pTL30) (lanes 2 and 4). Approximately 30 |ig of protein was loaded per lane. The molecular weights and positions of protein standards are shown on the left. The positions of the light (L), medium (M) and heavy (H) subunits of the reaction center, of the (X and ^polypeptides of the B870 light-harvesting complex and of the a, /?and /polypeptides of the B800-850 light-harvesting complex are shown on the right.  60 500  0  A  100 200 300 400 500 600  0  100 200 300 400 500 600  100 200 300 400 500 600  0  100 200 300 400 500 600  C0 • H  500  2 Q 0 =3  Time (Hours)  0  100 200 300 400 500 600  Time (Hours)  Figure 3.10. Light titration and photosynthetic growth of five strains of R capsulatus . Photosynthetic growth of the R. capsulatus strains ARC6(pTB999) (A), ARC6(pTL30) (B), ARC6*(pTL2) (C), ARC6(pTL2-7) (D) and ARC6(pTL2-9) (E) at five light intensities. A, . e - l . >, 30 uE-m-2-s-1; O , 15 u E - n v V 1 250 u E - n r V 1 ; • , 120 u E - n r V 1 - I, 60 uE-nr 22-s  61  120 |iEinsteins • nr 2 • s  802 880 1000 350  15 fxEinsteins • nr 2 • s-1  T  r  475  590  IT 802 860  1 1000  Wavelength (Nanometers)  Figure 3.11. Absorption spectra from photosynthetic cultures grown at two light intensities. Absorption spectra of samples from cultures of R. capsulatus strains ARC6(pTB999) (A), ARC6*(pTL2) (B), ARC6(pTL2-7) (C) and ARC6(pTL2-9) (D) after photosynthetic growth at 120 \±E-m~2-s~] and 15 uE-nr 2 ^" 1 .  62  concluded that the suppressor strains used light energy less efficiently than the pseudo-wild type strain. In the case of ARC6*(pTL2) this is a not unexpected result, since this strain lacks the B800-850 light-harvesting complex, which is thought to enhance the ability of/?, capsulatus to grow at low light intensities. It was concluded that there are more than one successive suppressor mutations possible, arranged in a hierarchy, and it seems the three suppressor mutants studied have suppressor mutations that effectively suppress the/>«^f phenotype only at relatively high light intensities. The additional mutations seem to result in significant increases in the amounts of the B870 complex. When RCV minimal medium was supplemented with yeast extract, the suppressor mutants typically had faster growth rates than in the minimal medium (Figure 3.12). It can be seen in Figure 3.12 that the three suppressors have lag times comparable to the primary mutant, when the latter strain is grown in a medium supplemented with 0.1% or 0.2% yeast extract, whereas the pseudo-wild type strain shows no lag. Unlike the primary mutant strain, however, the suppressors are able to grow photosynthetically with no yeast extract supplement at all, as was observed in previous experiments (14). Three of the five strains tested (ARC6[pTL30], ARC6[pTL2-7] and ARC6[pTL2-9]) appeared to be inhibited by the highest concentration of yeast extract (0.4%). The cause for this sensitivity is not known. The phenotype of the pufX deletion mutant was first characterized in the strain ARC6, but important flash spectroscopy experiments and reduced minus oxidized quantitations of cytochromes were carried out in the ARC6<r£D strain (see below). In order to determine whether or not the /»«^deletion mutation had the same phenotype in the crtD strain, some of the growth experiments were repeated with this strain (Figure 3.13). When cultures were grown aerobically, under either high (data not shown) or low oxygen conditions, no difference was seen between ARC6cr/£>(pTB999) and ARC6m£>(pTL30), and their two crtD*  analogs. This  was not the case when cultures were grown photosynthetically. Both carotenoid background strains complemented with (pTB999) grew with similar kinetics at 120 laE-nr 2 ^" 1 or at 30 jj-E-m"2^"1 (compare Figures 3.1A and 3.13B and C). However, at 120 (lE-nr^s" 1 ARC6m£>(pTL30) showed a reduced lag when compared with ARC6(pTL30) (compare Figure 3.13B with Figure 3.IB). Even when the light intensity was reduced to 30 (J-E-m"2^"1 the lag time for ARC6mZ)(pTL30) was shorter, but growth did seem to pause at about 50 Klett units and then start again about 75 hr later (Figure 3.13C). Possibly the second phase of  63  D  A  J? 20  40  60  0  20  40  60  80  D 4-1  •*—>  2 60  C  Q 0  20  40  60  =5  Time (Hours)  20  40  60  Time (Hours)  Figure 3.12 Photosynthetic growth of five strains of R. capsulatus on minimal medium and media supplemented with yeast extract. Photosynthetic growth of the R. capsulatus strains ARC6(pTB999) (A), ARC6(pTL30) (B), ARC6*(pTL2) (C), ARC6(pTL2-7) (D) and ARC6(pTL2-9) (E) at a light intensity of 120 uE-nr^s- 1 in: O , minimal RCV medium; • , RCV plus 0.1 % yeast extract; # , RCV plus 0.2 % yeast extract; and • , RCV plus 0.4 % yeast extract.  64  500  500  B  100  10  50  100  Time (Hours)  150  0  50  100  150  150  o  50 100 Time (Hours)  150  Figure 3.13. Growth of the crtD strains. Growth of the R. capsulatus ARC6crtD strains (pTB999) (A), (pTL30) (O), (pTL2-7) (•) and (pTL2-9) (•). (A), aerobic growth at low oxygen concentration; (B), photosynthetic growth at 120 flE-nr2^"1; (C), photosynthetic growth at 35 flE-m"2^"1; and (D), photosynthetic growth at 35 IJ-E-nr2^"1 in a 1:1 mix of minimal RCV medium and rich YPS medium.  65  logarithmic growth was due to the growth of suppressor mutant cells. The "diauxic" growth curve for strain ARC6cr£D(pTL30) was even more pronounced when cultures were grown in a medium that was a blend of rich and minimal media (Figure 3.13D). The growth of this strain under these conditions was similar to that of its atD* analog under the same conditions (see Figure 3.2B). 3.2.1.2. Fluorescence detection: The efficiency of light energy trapping As noted above, the suppressor strains seemed to use light energy less efficiently than the pseudo-wild type strain. The efficiency of light energy trapping can be assessed by observing the infrared fluorescence from intact cells  Light Intensity  at low light intensities. Cells which cannot use the harvested  (jiEinsteins • n r 2 • s"1)  light lose the "excess" energy in  0.5  4.1 7.5  13  24  the form of fluorescence. This fluorescence can be measured by the use of infra-red film and the appropriate filter to record the light emitted from intact cells in response to excitation light. Using this technique, the light-trapping efficiency of the suppressor strains relative to the pseudowild type and primary mutant strains was assessed. As can be seen in Figure 3.14E and F, the two suppressor strains ARC6(pTL2-7) and ARC6(pTL2-9) showed  Figure 3.14. Light-trapping efficiency experiment. Fluorescence from R. capsulatus strains ARC6(pTB999) (A), ARC6(pTL30) (B), ARC6(pA4) (C), ARC6*(pTL2) (D), ARC6(pTL2-7) (E) and ARC6(pTL2-9) (F), when irradiated with light of five different intensities.  observable fluorescence at a lower light intensity than the pseudo-wild type, indicating that the suppressor strains were  66 not able to use light efficiently. At light intensities above 4.1 JlE-nr 2 ^" 1 fluorescence from ARC6(pTL2-9) becomes visible and fluorescence from ARC6(pTL2-7) becomes visible at 7.5 llE-m" 2 ^" 1 , the same point at which fluorescence from the primary mutant is first seen. The apparent inefficiency of these strains is reconciled with their ability to grow photosynthetically when we recall that (a) the stoichiometry of the light-harvesting complexes has been perturbed in these strains and so the transfer of light energy may be impaired, (b) both these strains have a mutation in the a polypeptide of the B870 complex (see below), and (c) neither strain grows as well as the pseudo-wild type under photosynthetic growth conditions. The strain ARC6(pA4) was a positive control; it contains the B800-850 complex, but no B870 and reaction center complexes and therefore all the light energy harvested by the B800-850 antenna complexes was released as fluorescence. 3.2.1.3. Flash spectroscopy: Monitoring photosynthetic electron transport The extent and kinetics of oxidation and reduction of components of the photosynthetic electron transport pathway can be monitored by measuring the absorption changes at suitable wavelength pairs in response to flashes of light. For the bacteriochlorophyll special pair of the reaction center, absorption changes are monitored at 605-540 nm; for ctype cytochromes absorption changes at 550-540 nm are monitored and for b-type cytochromes absorption changes at 572-560 nm are monitored. In each case oxidation of the monitored component causes a decrease in absorption whereas reduction causes an increase in absorption. Additionally, measuring the total absorption change after a train of eight saturating flashes in the presence of an electron transport inhibitor such as antimycin A allows estimation of the total relative amount of the component. Antimycin A blocks electron transfer from the £-type cytochrome pathway to the quinol reductase (Qc) site. The results of such flashing experiments carried out on different chromatophore preparations from the three suppressor strains as well as the pseudo-wild type and primary mutant strains are shown in Figures 3.15, 3.16 and 3.17. The data show the oxidation (as a downward movement of the trace) and reduction (as an upward movement of the trace) of the reaction center special pair (Figure 3.15), c-type cytochromes (Figure 3.16) and £-type cytochromes (Figure 3.17) when subjected to a sequence of eight saturating flashes. As mentioned in the Introduction, a brief flash of light can cause a single turnover of the light-driven electron transport pathway. The flash causes the almost instantaneous oxidation of the bchl a special pair, and, when monitoring at 605-540 nm, the photo-oxidation  67  1  (i)  (ii)  M •*r^~~  +  m (iii)  WMty^' (iv)  (v) AA = 0.0022  AA = 0.0022 100 ms  100 ms  Figure 3.15. Flash spectroscopy experiments: The bacteriochlorophyll special pair. Oxidation and reduction kinetics of the bacteriochlorophyll special pair monitored at 605 minus 540 nm in two chromatophore preparations (1 and 2) from R. capsulatus strains ARC6m£>(pTB999) (i), ARC6mD(pTL30) (ii), ARC6*(pTL2) (iii), ARC6o-/£>(pTL2-7) (iv) and ARC6m£>(pTL2-9) (v) in response to eight saturating flashes. Chromatophores were suspended in chromatophore buffer to a concentration of 25 |lM bacteriochlorophyll a (1) and a reaction center concentration of=220 nM (2). Traces were recorded separately in the presence (+) and absence of 2 |J.M antimycin A, and overlaid for ease of comparison.  68  is seen in the spectroscopy traces as a vertical, downward movement of the trace. Subsequent reduction of the special pair is seen as a curved, upward movement of the trace. The trace is curved because re-reduction occurs on a time scale of milliseconds (rather than picoseconds as is the case for photo-oxidation). For a train of eight saturating flashes, the results are as seen in Figure 3.150). In the absence of obstacles to electron transport, reaction centers from the pseudo-wild type strain were re-reduced subsequent to each flash, and, after eight flashes cyclic electron flow was still occurring. When an electron transport inhibitor, in this case antimycin A, is added to the chromatophore suspension (such traces are marked "+" in Figure 3.15) the cyclic flow of electrons is interrupted. Re-reduction of the special pair decreases with each flash as the pool of c-type cytochromes is oxidized; hence the reaction centers are almost all oxidized after eight flashes in the presence of the inhibitor. When such experiments were carried out on the primary mutant strain ARC6m£>(pTL30) (Figure 3.15[ii]) a difference between this strain and the pseudo-wild type strain was apparent. In the primary mutant strain, even in the absence of antimycin, reaction centers were almost totally oxidized after two flashes. This indicates that cyclic electron flow in the primary mutant was impaired; reaction centers were readily photo-oxidized, i.e., electrons were being transferred to the acceptor quinones, but the special pairs were not being reduced because they were "starved" for electrons. Figure 3.15 shows the results of flashing experiments on two preparations of chromatophores from the suppressor strains as well as the pseudo-wild type and primary mutant strains. Surprisingly, the suppressor strain ARC6*(pTL2) appeared, like the primary mutant, to be starved for electrons (Figure 3.15 [iii]); the other two suppressor strains seemed to have improved cyclic electron flow compared to the primary mutant strain, although re-reduction of reaction centers in uninhibited preparations was not as rapid as was seen in the pseudo-wild type. In preparation 2, the reaction centers of the suppressor strains (pTL2-7) and (pTL2-9) were re-reduced to a greater extent in the presence of antimycin A than the pseudo-wild type (Figure 3.15 [iv] and [v]). This phenomenon was probably related to the higher levels of photo-oxidizable c-type cytochromes in these preparations (see below). Light-initiated oxidation and reduction of otype cytochromes in two chromatophore preparations are shown in Figure 3.16. Here again, oxidation (due to electron transfer to the special pair) occurs rapidly upon flashing and reduction (due to electron transfer through the blc\ complex) occurs between flashes. It was apparent that re-reduction of c-type cytochromes was impaired in the primary mutant relative to the pseudo-wild type strain, as had been seen  69  (i) •VwrV^V^twfc ^ L i N t * ^ ^ * yW<»<M»MWr _J_  ^^wW*****"  (ii)  •**^i ^.^Vd^^viwvow  +  (iii)  (iv) "v^-^vV-j-  (v) • * ^ * r - m » . •Wt^Mr-V ~ i _  AA = 0.0022  AA = 0.0029 100 ms  100 ms  Figure 3.16. Flash spectroscopy experiments: The otype cytochromes. Oxidation and reduction kinetics of the c-type cytochromes monitored at 550 minus 540 nm in two chromatophore preparations (1 and 2) from R. capsulatus strains ARC6m£>(pTB999) (i), ARC6m£>(pTL30) (ii), ARC6*(pTL2) (iii), ARC6mZ>(pTL2-7) (iv) and ARC6mZ)(pTL2-9) (v), in response to eight saturating flashes. Chromatophores were suspended in chromatophore buffer to a concentration of 25 \xM bacteriochlorophyll a (1) and a reaction center concentration of =220 nM (2). Traces were recorded separately in the presence (+) and absence of 2 JJ.M antimycin A, and overlaid for ease of comparison.  70  previously (13). In preparation 1, the extents and kinetics of oxidation and reduction in suppressor mutants ARC6cnZXpTL2-7) ar>d ARC6atD(pTL2-9) were comparable to those seen in the pseudo-wild type strain (Figure 3.16 l[i], [iv] and [v]), but in preparation 2, although the kinetics of oxidation and reduction were comparable to the pseudo-wild type, the amounts of photo-oxidizable c-type cytochromes in these strains were twice the level seen in the pseudo-wild type. When the results of flashing experiments were normalized and tabulated (Table 3.II), the variability in c-type cytochrome levels in these two strains is readily apparent. The pseudo-wild type and primary mutant strains show much less variability. In the case of ARC6*(pTL2), it was not clear that re-reduction of the c-type cytochromes was occurring (Figure 3.16 2[iii]) but the level of photo-oxidizable c-type cytochromes in this strain was similar to the level seen in pseudo-wild type. Overall, the average level of photo-oxidizable c-type cytochromes as determined by flashing experiments agreed quite closely with the amounts as determined by reduced minus oxidized spectra (see section 3.2.1.4, below).  TABLE 3.II. Measurement of photooxidizable otype cytochromes. The amount of photooxidizable c-type cytochromes measured as AA 550 - 540 (x 10^ ) after eight saturating flashes in chromatophores from five strains of R. capsulatus: (pTB999) = the pseudowild type ARC6m£>(pTB999); (pTL30) = the primary mutant ARC6tnZXpTL30); (pTL2) = the suppressor mutant ARC6*(pTL2); (pTL2-7) = the suppressor mutant ARC6tr£D(pTL2-7) and (pTL2-9) = the suppressor mutant ARC6m£>(pTL2-9). The values are adjusted to a reaction center concentration of 150 nM.  Mean ±SD Relative  (PTB999)  (pTL30)  (pTL2)  (pTL2-7)  (pTL2-9)  (n=4)  (n = 3)  (n=l)  (n = 3)  (n = 3)  3.15 ±0.28  2.76  4.12  4.12  -  (±9%)  2.06 ±0.08 (±4%)  ±1.95 (±47%)  ±1.35 (±33%)  1  0.65  0.88  1.31  1.31  to 999 Figure 3.17 shows the reduction of the high potential £-type cytochromes in response to eight flashes, monitored as absorption changes at 572-560 nm. These traces were acquired in the presence of antimycin A, which prevents the oxidation of the high-potential &-type cytochrome by the quinone at the Q c site of the reaction center (see Figure 1.1). Thus, if electrons are not able to reach the £-type cytochrome the extent of the absorption change will be smaller than that seen in the pseudo-wild type, as it indeed is for the primary mutant  71 (Figure 3.17[ii]). Turning to the suppressor strains, all three suppressor strains in preparation 1 appeared to have regained the ability to transfer electrons to the high potential b-type cytochrome; ARC6*(pTL2) actually surpasses the pseudo-wild type. This apparent recovery may be due simply to the higher levels of b-typt cytochromes in these strains relative to the pseudo-wild type and primary mutant strains. In chromatophores in preparation 2 the spectra for strains ARC6o"*Z)(pTL2-7) and ARC6m£>(pTL2-9) (Figure 3.17[iv] and[v]) are distorted because of the overlap in absorption spectra between c- and b-type cytochromes and the very high levels of the former in these preparations as visualized in the reduced minus oxidized spectra discussed below (see preparation 2 in Table 3.III). 3.2.1.4. Reduced minus oxidized difference spectroscopy: The relative levels of band c- type cytochromes Flash spectroscopy indicated that there were different levels of photooxidizable cytochromes in different chromatophore preparations from the same strains. The relative total amounts of b- and c-type cytochromes can be measured by examining reduced minus oxidized difference spectra. The use of ascorbate to reduce oxidized cytochromes causes primarily the c-type cytochromes to be reduced, since they generally have higher midpoint potentials than the b-type cytochromes. The latter are reduced when dithionite is used as a reducing agent. Table 3.Ill shows that the relative amount of these cytochromes in the primary mutant strain was about 75% to 80% of that found in the pseudo-wild type strain regardless of the reductant used. When we look at the data for the suppressor mutants, the strain ARC6*(pTL2) has slightly higher levels of cytochromes than the pseudo-wild type, but only one measurement was made on freshly prepared chromatophores. Two separate preparations of the other two suppressors, ARC6crtD{pTL2-7) and ARC6at£>(pTL2-9), were measured and in both cases levels of cytochromes appeared to be about 30% to 40% higher than in the pseudo-wild type when ascorbate was used as the reductant. When dithionite was used as the reductant, the relative levels of cytochromes in these two strains were only 5% to 28% higher. This implies that the ratios of b- to r-type cytochromes in the suppressor strain may have changed. However, the spectra can at best be regarded as semi-quantitative measurements of intra-strain amounts of b- and c-type cytochromes and are most useful for inter-strain comparisons of cytochrome levels.  72  (i)  (iii)  (iv)  yy^^^  AA = 0.0003 100 ms 100 ms Figure 3.17. Flash spectroscopy experiments: The £-type cytochromes. Oxidation and reduction kinetics of the £-type cytochromes monitored at 572 minus 560 nm in two chromatophore preparations from R. capsulatus strains A R C 6 m Z ) ( p T B 9 9 9 ) (i), ARC6crfZ)(pTL30) (ii), ARC6*(pTL2) (iii), ARC6cr;£>(pTL2-7) (iv) and ARC6c7tZ)(pTL2-9) (v), in response to eight saturating flashes. Chromatophores were suspended in chromatophore buffer to (1) a reaction center concentration of = 100 nM and (2) a reaction center concentration of =220 nM. Traces were recorded in the presence of 2 \lM antimycin A.  73  TABLE 3.III. Measurement of total b- and c-type cytochromes. The chemically-induced absorption changes in cytochromes in two preparations (1 and 2) of chromatophores from R. capsulatus ARC6*(pTL2) and four strains of R. capsulatus ARC6crtD measured at 550 minus 540 nm. Normalized to a reaction center concentration of 100 nM. Ascorbate/PMS - Ferricyanide  Dithionite - Ferricyanide  AA  AA Relative  AA  AA Relative  (x 10-* absorption units)  (relative to 999)  (x 10-* absorption units)  (relative to 999)  1  2  1  2  1  2  1  2  (PTB999)  6.52  9.26  1.00  1.00  8.92  13.22  1.00  1.00  (pTL30)  4.88  6.84  0.75  0.74  7.27  10.03  0.82  0.76  ARC6*(pTL2)  -  10.34  -  1.12  -  16.11  -  1.22  (PTL2-7)  8.58  11.73  1.32  1.27  9.47  15.23  1.06  1.15  (pTL2-9)  8.96  12.89  1.37  1.39  10.53  16.88  1.18  1.28  3.2.1.5. Flash spectroscopy. Monitoring the formation of a trans-membrane potential Flash spectrophotometry can also be used to measure the formation of a membrane potential in the pseudo-wild type and suppressor strains. A consistent, linear correlation has been observed between changes in membrane potential and changes in the carotenoid absorption spectrum in R. capsulatus and R. sphaeroides (9, 10). Since the change in absorption is instantaneous, this so-called carotenoid bandshift can be used to monitor changes in membrane potential induced by light on any timescale. As noted above, there was variability in the light-driven electron transport capabilities of the suppressor strains as monitored by flash spectroscopy from one chromatophore preparation to the next. Although this variability was also seen in the bandshift experiments,  74  AA = 0.008  100 ms  B  AA = 0.002  100 ms Figure 3.18. Flash spectroscopy experiments: The carotenoid bandshift. The carotenoid bandshift (monitored at 490-475 nm) in chromatophores (A) and intact cells (B) of R. capsulatus ARC6crtD strains (pTB999) (i), (pTL30) (ii), (pTL2-7) (iii) and (pTL2-9) (iv), in response to eight saturating flashes.  75  one result is very clear: the chromatophores from the suppressor strains that contained carotenoids were unable to generate a membrane potential significantly larger than the primary mutant, as seen in the experimental results shown in Figure 3.18A. The experiment shown in Figure 3.18A represents a typical result. The normalized results of six to seven experiments carried out with three chromatophore preparations are summarized in Table 3.IV, and show that these strains are greatly deficient in the generation of a membrane potential. Although it seems reasonable to expect that if the suppressor strains are able to grow as well as the pseudowild type strain ARC6triD(pTB999) under photosynthetic growth conditions they should demonstrate a carotenoid bandshift comparable to that seen in the pseudo-wild type strain, or at least bigger than that seen in the primary mutant, it clearly was not happening. However, when these experiments were repeated using intact cells, rather than chromatophores, it was evident that the suppressor strains were capable of generating a trans-membrane potential comparable to that observed in the pseudo-wild type (Figure 3.18B). These results show that the process of chromatophore preparation from cells of the suppressor mutants ARC6m£>(pTL2-7) and ARC6mZ)(pTL2-9) caused a disruption of the in vivo properties of the photosynthetic unit.  TABLE 3.IV. The generation of a transmembrane potential in four strains of R. capsulatus. The carotenoid bandshift in chromatophores from four strains of R. capsulatus ARC6crtD measured as AA 490 minus 475 nra (x 10 3 ) after one and eight flashes. Adjusted to a reaction center concentration of 150 nM. ( P TB999)  (pTL30)  (n = 6)  (n = 7)  ( P TL2-7)  (pTL2-9)  (n = 6)  (n = 6)  1st Flash  8th Flash  1st Flash  8th Flash  1st Flash  8th Flash  1st Flash  8th Flash  Mean  14.82  52.51  6.60  16.43  4.06  14.24  0.77  3.66  ±SD  ±3.29  ±12.81  ±1.87  ±5.23  ±2.15  ±8.80  ±0.40  ±1.92  (±22%)  (±24%)  (±28%)  (±32%)  (±53%)  (±62%)  (±52%)  (±52%)  1  1  0.44  0.31  0.27  0.27  0.05  0.07  AA Relative to  76  Figure 3.19. The steady-state, light-induced carotenoid bandshift. Light minus dark difference spectra measured using chromatophores from R. capsulatus strains ARC6m£>(pTB999) (A), ARC6m£>(pTL30) (B), ARC6*(pTL2) ( Q , ARC6m£>(pTL2-7) (D) and ARC6a*Z)(pTL2-9) (E), in the presence of the membrane potential-sensitive dye oxonol VI. Spectra were recorded in the presence (gray trace) and absence (black trace) of antimycin A.  77  0.1 A  r 400 nm  T  500 nm  ~1  650 nm  78 3.2.1.6. Light minus dark difference spectroscopy: The light-induced generation of a membrane potential over a longer time frame One possible explanation for the lack of a carotenoid bandshift in the chromatophore preparations from the suppressor strains is that for some reason, perhaps because the position of the carotenoid molecules in the membrane has been altered, the absorption spectrum of the carotenoids was no longer changing in response to the membrane potential. Therefore, the suppressor strains would be able to generate a membrane potential that was sufficient for photosynthetic growth, but not detectable using carotenoid bandshift measurements. To test this hypothesis the membrane potential was measured using an oxonol dye, OX-VI, which responds to changes in membrane potential by shifting its absorption spectrum (1). The results, shown in Figure 3.19, indicate that the carotenoid bandshifts of the suppressor mutants obtained with chromatophores were accurate indicators of their ability to generate a membrane potential in response to light. The results in Figure 3.19C also show that ARC6*(pTL2) (the suppressor strain whose carotenoid bandshift could not be measured because it lacks B800-850 light-harvesting antenna complexes and their associated carotenoids) is not able to generate a membrane potential of a magnitude comparable to the pseudo-wild type strain. 3.2.1.7. Transmission electron microscopy of negatively stained chromatophores The physical integrity of the chromatophores was evaluated by transmission electron microscopy. As can be seen in Figure 3.20, there were variations in the sizes of the chromatophores. The significance of the size differences, if any, is not clear, but it is interesting to note that, in R. sphaeroides, it has been reported that chromatophores formed in the absence of the B800-850 complex can be smaller than chromatophores formed in the presence of the B800-850 complex (17). This is evidently not the case in R. capsulatus, since the chromatophores from strain ARC6*(pTL2), which lacks the B800-850 complex, were larger than the chromatophores from the other strains (Figure 3.20C). Other defects reported as resulting from the absence of B800-850 complexes include the formation of linear or spiral tubules (7, 11, 12); such defects would probably prevent the formation of normal spherical chromatophores such as those seen in Figure 3.20. Chromatophores from the primary mutant and the suppressor mutants, although they have a reduced ability to form a trans-membrane potential, appear intact and similar in gross structure to wild type chromatophores.  Figure 3.20. Transmission electron microscopy of chromatophores. Transmission electron micrographs of negatively stained, sucrose gradient-purified chromatophores from R. capsulams strains ARC6cr*D(pTB999) (A), ARC6cr/£>(pTL30) (B), ARC6*(pTL2) (C), ARC6m£>(pTL2-7) (D) and ARC6m£>(pTL2-9) (E). The bar represents 0.24 Urn.  j  : ^  '  -fir  v '  --  81 3.2.1.8. Ion gradient- and continuous light-induced carotenoid bandshifts It was possible that, despite the apparent integrity of the chromatophores from the suppressor strains, as judged by their gross morphology, these chromatophores had small holes that allowed the passage of ions (such as protons). The permeability of chromatophores to ions was assessed by measuring the membrane potential induced by a chloride ion gradient in the presence of the uncoupler valinomycin (10). As shown in Figure 3.21A, upon the addition of KC1, a stable ion gradient is formed, as indicated by the induced carotenoid bandshift. This indicates that the chromatophores are not permeable to small ions. The time scale of the ion gradient experiments is identical to that of the continuous light-induced carotenoid bandshift experiments, shown in Figure 3.2 IB. However, the results of the continuous light-induced experiments show a much greater degree of strain to strain variability. Thus, all strains had a comparable ability to form a chemically induced membrane potential, but their ability to form a membrane potential in response to continuous illumination was impaired in chromatophores from the primary mutant and suppressor mutants. The absorption difference at 503 - 490 nm for light-induced bandshift experiments divided by the 503 - 490 difference seen in the chemically-induced bandshift experiments yields the following values: the pseudowild type strain = 4.2; the primary mutant strain = 2.0; ARC6(pTL2-7) = 2.7 and ARC6(pTL2-9) = 0.6. Thus, the long time-scale variations in the membrane potentials in response to continuous illumination between these chromatophore preparations matched the differences of membrane potentials in response to the short time-scale flash spectroscopy experiments (see Figure 3.18A). 3.2.1.9. Flash spectroscopy: Monitoring the orientation of the reaction centers in chromatophores If the reaction centers in the suppressor strain chromatophores were reversed in the membrane or turned sideways, this would certainly interfere with electron transport and would also decrease the carotenoid bandshift, since electrons moving within the reaction center would no longer be crossing the membrane. In order to test the orientation of the reaction center within the chromatophores, the oxidation and reduction kinetics of the reaction center special pair were monitored at 605-540 nm in the presence of the electron transport inhibitor antimycin A, and the uncoupler valinomycin. After a set of control measurements had been made, reduced horse heart cytochrome cwas added as an exogenous reductant of the reaction center and a second set of measurements made. Normally, when chromatophores are formed, most of the reaction centers are oriented with the special pair on the inside of the  82  Figure 3.21. A comparison of the trans-membrane potentials induced by light and ion gradients. Ion-gradient-induced (A) and light-induced (B) carotenoid bandshift spectra in chromatophores from R. capsulatus strains ARC6mZ)(pTB999) (i), ARC6<T*D(pTL30) (ii), ARC6m£>(pTL2-7) (iii) and ARC6o-£D(pTL2-9) (iv).  83  B (i)  (ii)  (iv)  400  490  550  400  Wavelength (nm)  490  550  84 chromatophore membrane (15). That is, the cytoplasmic side of the membrane constitutes the external face of the chromatophore, and the cytochrome c docking site on the reaction center is on the inside of the chromatophore. This means that the addition of exogenous cytochrome c would affect the oxidation and reduction kinetics only to the extent that the chromatophores have adopted a topologically reversed or open conformation, such as a sheet. Usually about 10 - 18% of chromatophores in a normal preparation are in this form (16). As can be seen in Figure 3.22, in the presence of exogenous cytochrome c the pseudo-wild type (Figure 3.22A) and the primary mutant (Figure 3.22B) both showed only a small increase in reaction center special pair reduction after the first flash relative to the control, indicating that these chromatophore preparations were in the normal orientation and that the reaction centers were inserted in the membrane correctly. On the other hand, the suppressor strains ARC6crtD(pTL2-7)  and ARC6mZ)(pTL2-9) both showed greater levels of reduction of the  special pair after the first flash in the presence of exogenous cytochrome c (Figure 3.22D and E). This means that the exogenous cytochrome rwas able to transfer electrons to the reaction center special pair, which in turn implies that these chromatophores contained reaction centers oriented such that the cytochrome c docking site was on the outside, since the electron microscopy data showed that the chromatophores were not present as topologically open sheets (see figure 3.20). Addition of exogenous cytochrome chad no significant effect on the extent or kinetics of the carotenoid bandshift in these chromatophores (Figure 3.23), indicating that the reaction centers which were able to oxidize external cytochrome cdid not generate a transmembrane potential.  85  Figure 3.22. The orientation of reaction centers. Oxidation and reduction kinetics of the bacteriochlorophyll special pair in chromatophores from R. capsulars strains ARC6crtD(pTB999) (A), ARC6mD(pTL30) (B), ARC6*(pTL2) (C), ARC6crtD(pTL2-7) (D) and ARC6<rfZ)(pTL2-9) (E), in response to eight saturating flashes. The experiments were carried out in the presence (+) and absence of exogenous cytochrome c.  86 WWW  J ^ ^ ^ ^ c  H  l^Vl^^  B  w 1 AA = 0.0016  i^*Vv<*w^*i** sw **' t  hW  v\-\s*«*»t*^^  80 ms  B  >f  D * W W ^ " ' 'M^WM#*#feft|ll|tM<W|MI«M' +  87  AA = 0.023  100 ms  c  A  ^zczzz  +  xc<? Figure 3.23. The effect of exogenous cytochrome con the carotenoid bandshift. The carotenoid bandshift in chromatophores from R. capsulatus strains ARC6m£>(pTB999) (A), ARC6mZ)(pTL30) (B), ARC6m£>(pTL2-7) (C) and ARC6mZ)(pTL2-9) (D), in response to eight saturating flashes. The experiments were carried out in the presence (+) and absence of exogenous cytochrome c.  88  3.2.2. The location of the suppressor mutations Preliminary to high-resolution mapping of the locations of the suppressor mutations, experiments were done to determine whether these mutations were pTL2 plasm id-derived or were on the chromosome. Figure 3.24 shows the result of an experiment in which plasmids from the suppressor mutants were conjugated into strain ARC6, which had not been subjected to photosynthetic growth conditions. The ex-conjugants were then tested for the ability to grow photosynthetically. It was concluded that ARC6(pTL2-7) and ARC6(pTL2-9) carried their suppressor mutations on the pufoperon segment of the plasmid, since the ex-conjugants carrying the plasmid from these strains were able to grow without the long lag seen with the plasmid pTL2 (pTL30). The strain ARC6(pTL2-l) carried its suppressor mutation on the chromosome since transfer of the plasmid to ARC6 did not transfer the ability to grow photosynthetically (Figure 3.24). Since the mutation in ARC6(pTL2-l) must be located on the chromosome of the host strain, this new strain was designated ARC6*(pTL2) to indicate this fact. In order to map the  500  plasmid-borne mutations, defined fragments of DNA  a D  s  from the "suppressor plasmids" were excised and used to  100  a  replace the corresponding  Q  fragments in pTL2 DNA. These higher resolution  3  complementation analyses  u  with puf operon segments 10  localized the suppressor 20  40  60  80  100  Time (Hours)  mutations on the plasmids pTL2-7 and pTL2-9 to a region within the pufBA genes.  Figure 3.24. Photosynthetic growth of R. capsulatus ARC6 strains transformed with plasmids from the suppressor mutants. Photosynthetic growth at 120 U-E-nr-^s"1 of the R. capsulatus ARC6 strains (pTB999) (A) and (pTL30) (O) and of R. capsulatus ARC6 after transformation with the plasmids from the R. capsulatus A R C 6 / ^ s u p p r e s s o r strains (pTL2-l) ( • ) , (pTL2-7) ( • ) and (pTL2-9) ( • ) .  As shown in Figure 3.25A, four puf operon fragments from the suppressor plasmids were tested for the ability to suppress the/>#/Vphenotype.  89 Two of the fragments, the Xho I to Bsp EI and the Stu I to Bss HII fragments did not suppress. The _ .a JZ  X X  remaining two fragments, the Xho  UJ  ou ".5g  o PQ  co  "oi-ipTAi^r  CQ  Z  EC  M  I to Stu I and Nco I fragments, which did suppress, overlapped in the Bsp EI to Stu I region. This region, which encodes the pufBA genes, was sequenced. Both plasmid-borne suppressor mutations were found to be in the  I  second codon of the pufA gene; in  1 300 bp  plasmid pTL2-7 this mutation changed the normally encoded  B ARC6(pTB999)  serine to a proline codon, whereas CTGAAA ATG TCC AAG Met S e r Lys  ARC6(pTL2-7)  CTGAAA ATG CCC AAG Met P r o Lys  ARC6(pTL2-9)  CTGAAA ATG TTC AAG Met P h e Lys  Figure 3.25. Mapping and location of the suppressor mutations. (A): mapping suppressor mutations in R. capsulatus ARC6(pTL2-9). The fragments shown beneath the map of the/>w/operon were used in complementation experiments; (+) indicates that a PS + phenotype resulted from complementation and (-) indicates that it did not. (B): the sequence of the 5prime end of the pufA gene in the plasmids (pTB999), (pTL2-7) and (pTL2-9) showing the deduced amino acid sequence.  in pTL2-9 this serine codon has been changed to a phenylalanine (see Figure 3.25B).  90  3.3. Results references 1. Bashford, C.L., B. Chance and R.C. Prince. 1979. Oxonol dyes as monitors of membrane potential their behavior in photosynthetic bacteria. Biochim. Biophys. Acta 545:4657. 2. Crofts, A.R. and C.A. Wraight. 1983. The electrochemical domain of photosynthesis. Biochim. Biophys. Acta. 726:149-185. 3. Dawkins, D.J., L.A. Ferguson and RJ. Cogdell. 1988. The structure of the 'core' of the purple bacterial photosynthetic unit, pp. 115 - 127. In H. Scheer and S. Schneider, Photosynthetic light-harvesting systems. Walter de Gruyter and Co., Berlin. 4. Fidai, S., S.B. Hinchigeri, T.J. Borgford and W.W. Richards. 1994. Identification of the PufQ protein in membranes of Rhodobacter capsulatus. J. Bacteriol. 176:7244-7251. 5. Firsow, N.N. and G. Drews. 1977. Differentiation of the intracytoplasmic membrane of Rhodopseudomonas palustris induced by variations of oxygen partial pressure or light intensity. Arch. Microbiol. 115:299-306. 6. Gennis, R.B., B. Barquera, B. Hacker, S.R.V. Doren, S. Arnaud, A.R. Crofts, E. Davidson, K.A. Gray and F. Daldal. 1993. The be, complexes of Rhodobacter sphaeroides and Rhodobacter capsulatus. J. Bioenerg. Biomembr. 25:195-209. 7. Golecki, J.R., S. Ventura and J. Oelze. 1991. The architecture of unusual membrane tubes in the B800-850 light-harvesting bacteriochlorophyll-deficient mutant 19 of Rhodobacter sphaeroides. FEMS Microbiol. Lett. 77:335-340. 8. Hunter, C.N., J.D. Pennoyer, J.N. Sturgis, D. Farrelly and RA. Niederman. 1988. Oligomerization states and associations of light-harvesting pigment-protein complexes of Rhodobacter sphaeroides as analyzed by lithium dodecyl sulfate-polyacrylamide gel electrophoresis. Biochemistry 27:3459-3467. 9. Jackson, J.B. 1988. Bacterial photosynthesis, pp. 317-376. InC. Anthony, Bacterial Energy Transduction. Academic Press, London. 10. Jackson, J.B. and A.R. Crofts. 1969. The high energy state in chromatophores from Rhodopseudomonas spheroides. FEBS Lett. 4:185-189. 11. Kiley, P.J., A. Varga and S. Kaplan. 1988. Physiological and structural analysis of light-harvesting mutants of Rhodobacter sphaeroides. J. Bacteriol. 170:1103-1115. 12. Lang, H.P. and C.N. Hunter. 1994. The relationship between carotenoid biosynthesis and the assembly of the light-harvesting LH2 complex in Rhodobacter sphaeroides. Biochem. J. 298:197-205. 13. Lilburn, T.G. 1990. The role of the pufX gene product of Rhodobacter capsulatus. M. Sc. thesis, University of British Columbia.  91 14. Lilburn, T.G. and J.T. Beatty. 1992. Suppressor mutants of the photosynthetically incompetent pufXdeletion mutant Rhodobacter capsulatus ARC6(pTL2). FEMS Microbiol. Lett. 100:155-160. 15. Prince, R.C., A. Baccarini-Melandri, G.A. Hauska, B A Melandri and A.R. Crofts. 1975. Asymmetry of an energy transducing membrane: the location of cytochrome cj in Rhodobacter sphaeroides and Rhodobacter capsulatus. Biochim. Biophys. Acta 387:212227. 16. Robertson, D.E. and P.L. Dutton. 1988. The nature and magnitude of the chargeseparation reactions of ubiquinol cytochrome c oxidoreductase. Biochim. Biophys. Acta 935:273-291. 17. Sturgis, J.N., C.N. Hunter and R.A. Niederman. 1990. Assembly of intracytoplasmic membranes in Rhodobacter sphaeroides mutants lacking light-harvesting and reaction center complexes, pp. 219-226. In G. Drews and E. A. Dawes, Molecular biology of membrane-bound complexes in phototrophic bacteria. Plenum Press, New York. 18. Vermeglio, A. 1977. Secondary electron transfer in reaction centers of Rhodopseudomonas sphaeroides. Out-of-phase periodicity of two for the formation of ubisemiquinone and fully reduced ubiquinones. Biochim. Biophys. Acta 459:516-524.  92 4.  DISCUSSION The goal of my thesis project was to elucidate the role of the pufXgene product (the PufX protein) in phototrophic growth of R. capsulatus. Previous work (15) had established that the PufX protein was essential for phototrophic growth at a light intensity of 120 (J-E-m"2^"1 on a minimal medium, and that electron transport and the consequent formation of a transmembrane potential was impaired in strains of R. capsulatus [ARC6(pTL2) and ARC6(pTL30)] that had the pufiC gene deleted. It had also been demonstrated that these pujX deletion mutants readily gave rise to suppressor mutants when placed under phototrophic growth conditions. The frequency with which these suppressor mutations appeared (10"5) (15) led me to speculate that a mutation at more than one locus could suppress the/Jw/X'mutation (15, 16). The approach taken to the characterization of the role of the PufX protein in this project has essentially been to broaden our understanding of the phenotypes of the pujX, or primary, mutant, and of three suppressor mutants. By learning what functions were affected by the loss of the PufX protein, on the one hand, and by learning what functions were restored by the suppressor mutations, on the other hand, I hoped to be able to infer what role the PufX protein had in R. capsulatus. This is not an ideal approach, and I had hoped to be able to raise antibodies to the PufX protein which would have allowed direct identification of the PufX protein on gels etc., but this did not prove to be possible. Farchaus et al. (9) were able to generate antibodies to a segment of the R. sphaeroides PufX protein and localize the PufX protein to a B870-reaction center fraction from chromatophores from R. sphaeroides. (Note: the core light-harvesting complex from R. sphaeroides has been called the B875 complex. However, to avoid confusion I will refer to all core light-harvesting complexes as B870 complexes.) While visiting the laboratory of Dr. Paul Loach, Dr. J. T. Beatty was able to demonstrate the presence of the PufX protein in chromatophores from R. capsulatus. His approach was to prepare chromatophores from pseudo-wild type and primary mutant strain cells grown at low O2, lyophilize and extract the chromatophores with a solution of chIoroform:methanol (1:1) plus 0.1 M ammonium acetate, and purify the proteins using size fractionation gel chromotography and reverse phase HPLC (17). Comparison of the HPLC elution profile from the two strains showed a peak in the pseudo-wild type eluate which was absent from the primary mutant. This peak fraction was lyophilized and subjected to N-terminal sequencing. The sequence matched the deduced  93 amino acid sequence of the PufX protein, but the N-terminal methionine was missing (Beatty, unpublished results). Although this demonstrated the existence of the pujXgene product, the position of the PufX protein within the photosynthetic unit remains unknown. Thus, a broad understanding of the phenotypes of the primary and suppressor mutants relative to the pseudowild type strain seemed to be the best route to defining the function of the PufX protein. To this end, the growth characteristics of these strains, their photosynthetic electron transport capabilities and the structure of their photosynthetic units were studied using the techniques outlined in the Methods section. Before proposing a model for the role of the PufX protein, I will discuss the implications of the experimental results obtained with the primary and suppressor mutants. 4.1. Growth experiments The aim of the growth experiments done with the primary mutant was to determine if the/>«/Xrphenotype could be suppressed by supplying cells with more light energy, or more nutrients. At first glance, the results of the light titration experiment, shown in Figure 3.1, seem to indicate that it is possible to completely overcome the loss of the PufX protein by increasing the light intensity to 1400 jlE-nr 2 ^" 1 . However, when we examine the insets on the plots in Figure 3.1, which show the first sixty hours of growth, it appears that there is still a lag before phototrophic growth occurs in the primary mutant, and that the rate of growth is less than that seen in the pseudo-wild type. With decreasing light intensity, the lag time in the primary mutant cultures increases. This implies that the length of time that passes before a suppressor mutation occurs, or is manifested by growth, is dependent on the amount of light energy available to the cells in the culture, which in turn implies that the mutation rate is dependent on the activity of cell metabolism. While the rates of phototrophic growth of the pseudo-wild type strain decrease with decreasing light intensity (Figure 3.1A), the growth rates of the suppressors, once they arise, are remarkably similar, despite the different light intensities. This growth rate phenomenon probably occurs because more than one type of suppressor mutation can arise within a culture, and the one best adapted to growth at a given light intensity comes to predominate. In any case, the conclusion to be drawn from the light titration experiments is that supplying more light to pufl( cultures hastens the appearance of suppressor mutants, but it does not enable the primary mutant cultures to grow. Providing cultures of the primary mutant with richer media has rather different results from providing them with more light energy. As shown in Figures 3.2 and 3.3, the lag times of phototrophically grown cultures of the primary mutant that are provided with a richer  94 medium or with a yeast extract supplement are shortened in a uniform way to approximately fifteen hours. The growth rates of both pseudo-wild type and primary mutant strains seem to increase on richer media, but growth of the primary mutant appears to stop when some component of the richer media is exhausted, as is seen in the growth on a 1:1 blend of RCV and YPS shown in Figure 3.2B, and in the growth on 0.1 % yeast extract shown in Figure 3.3B. I propose that the primary mutant strain is able to grow photosynthetically when supplied with a rich medium, at least until some component of the medium is exhausted, whereupon growth stalls until a suppressor mutation arises in the culture. The exception to this trend was growth on RCV plus 0.4 % yeast extract. Although this supplement enhanced the growth rate of the primary mutant, the initial growth rate was lower than that of the cultures supplemented with 0.5% or 1% yeast extract (Figure 3.3). It appears that this concentration of yeast extract was too high for optimal enhancement of photosynthetic growth. Although this was not observed with the pseudo-wild type strain, the high concentration effect was seen with the suppressor strains (see Figure 3.10). The mechanism by which the primary mutant was able to gain sufficient energy for photosynthetic growth on the richer media probably involves an alternative electron transport pathway. As mentioned in the Introduction, DMSO is a terminal electron acceptor for an anaerobic respiratory transport chain in R. capsulatus. When this compound was supplied in the growth medium of cultures of the primary mutant, the cultures were able to grow at a rate that was essentially equivalent to the growth rate of the pseudo-wild type (see Figure 3.4). The operation of an anaerobic respiratory electron transport pathway in these cultures may allow the phototrophic growth of the primary mutant by supplementing the proton-translocating activity of the photosynthetic electron transport pathway, which appears to be sufficient to maintain cell viability but not sufficient to enable growth. Flash spectroscopy experiments showed that the special pair of the reaction center in the primary mutant photooxidizes normally (see Figure 3.15), so electrons are transferred to the quinones at the Q A and Q B s i t e s ( s e e Figure 1.1). Since the quinol from the Qg site does not seem to reach the quinol oxidase (Qz) site of the cytochrome blc\ complex (where release of protons to the periplasm occurs) at a rate high enough to allow phototrophic growth (see below), the quinone pool may be overly reduced. Thus, the rate-limiting step in the formation of a proton gradient would be the oxidation of quinols. Quinols are oxidized in the anaerobic respiratory pathway, however, and the additional proton translocation catalyzed by this pathway could be enough to raise the transmembrane electrochemical potential over a threshold value required for growth (5, 26). Richardson et al. have shown that it is possible for R. capsulatus MTCBCl (8), a mutant strain that lacks a cytochrome blc^ complex, to grow photosynthetically, albeit very slowly, when  95 RCV plus DMSO is used as the growth medium (24). These workers also showed that electron transport to DMSO in MTCBCl generated a trans-membrane potential. Thus, an alternative anaerobic, energy-yielding pathway for oxidizing quinol exists in R. capsulatus. Normally, electron transport along this pathway is inhibited by the trans-membrane potential generated by light-driven electron transport (21), but the pufi(primary  mutant strain probably  does not generate a high enough potential to inhibit electron transfer to DMSO. Growth experiments carried out on the suppressor strains allowed an estimation of the relative efficiency of the photosynthetic apparatus in these strains by comparing the kinetics of growth at lower light intensities. The suppressor strains were first characterized using a light intensity of 120 jlE-nr 2 ^" 1 . At this light intensity, two of the suppressors, ARC6(pTL2-7) and ARC6(pTL2-9), grew at a slightly slower rate than the pseudo-wild type and a third suppressor ARC6*(pTL2) grew at about the same rate as the pseudo-wild type strain (see Figure 3.10). As shown in Figure 3. IOC, D and E, the growth rates of all three suppressors dropped off sharply at light intensities of 30 jj.E-m'2-s"1 or less, whereas the pseudo-wild type strain showed a more moderate decline in growth rate with a decrease in light intensity. This is taken to indicate that the suppressor strains were not able to use light energy as efficiently as the pseudo-wild type strain. It is interesting to see that all three suppressor strains could improve their efficiency of light use, apparently by undergoing further mutations, as evidenced by the positive change in the slopes of their growth curves after 150 to 300 hours. It would seem that the initial suppressor mutations, which occur after 35 to 120 hours at 120 llE-m'^-s"1, might never occur at low light intensities if the primary mutant were maintained at a low light intensity: not only would there be an increased lag before the initial suppressor mutation occurred, but a second mutation would be required that would allow growth at low light intensities. However, as Figure 3.10B shows, this is not the case. Despite the low light intensity, a suppressor mutation capable of growth at the lowest light intensity appeared within 300 hours. Could this mean that the "second" suppressor mutations appearing in the cultures shown in Figure 3.10C, D and E would be capable of suppressing the lack of photosynthetic growth of the pup( mutant in the absence of the suppressor mutations that were isolated at 120 flE-m"2^"1? Aerobic subculture from these cultures, followed by photosynthetic subculture at low light, resulted in photosynthetic growth without a lag (data not shown). This indicates the "second" suppressor mutations are real {i.e., genetic changes) and not physiologically induced, reversible adaptations. The "second" suppressor mutation in cultures of ARC6*(pTL2) is interesting in that it restores expression of the B800-850 antenna complex (Figure 3.11). This implies that  96 the loss of the complex was probably not essential to the suppression of pufX. No attempt was made to further characterize the "second" mutations. The uncharacterized "second" suppressor mutations were not the only mutations that allowed the suppressor mutants to grow at lower light intensities. The effect on growth of inserting a known mutation is shown in Figure 3.13. This mutation, in the carotenoid biosynthesis gene crtD, was originally inserted to simplify the absorption spectrum of the host cells in the 400 to 500 nm region of the spectrum for spectroscopy experiments. The enzyme encoded by this gene is a dehydrogenase (18), and mutation of the gene blocks carotenoid biosynthesis at the level of hydroxyneurosporene or methoxyneurosporene. Consequentially, five downstream carotenoids, including the two most commonly found in R. capsulatus, spheroidene and spheroidenone, are not synthesized (1), and cultures appear green rather than brown-orange in color. For some reason this mutation relieves to an extent the effects of the deletion of the puJXgene. As can be seen in Figure 3.13B, the primary mutant in this background shows almost no lag, although it still does not grow as well as the pseudo-wild type. The PufX protein is still required for photosynthetic growth at 35 U-E-nr2^"1 in this background, but in the experiment shown in Figure 3.13C, two of the suppressors with the crtD background reached stationary phase in less than seventy-five hours, almost without a lag. This is a marked improvement over their performance in the crtD+ background. It is conceivable that, given the structural role of carotenoids in the light-harvesting antenna complexes (30), the differences in the structures of neurosporene and the spheroidene derivatives alter the structure of these complexes in such a way as to facilitate light-driven electron transfer. The crtD strain growth experiments shown in Figure 3.13 were intended to determine whether or not the crtD mutation altered the properties of the pujX~ strains, such that the relevance of the flash spectroscopy experiments to the phenotype of the original primary mutant strain was questionable. Although the crtD mutation does change the photosynthetic growth abilities of the mutant strains, I argue that this does not detract from the significance of the flash spectroscopy experiments. This is because obvious differences are still seen between the pufiC strains and the pseudo-wild type strain in the flashing experiments (see below). Even though the crtD mutation alleviates the effects of the loss of pufli on photosynthetic growth at high light intensities, it does not eliminate the effects of the loss of pu/Xon  light-driven electron transport.  97 4.2. Fluorescence detection experiments A second way of measuring the relative efficiencies of light trapping is to measure the fluorescence from suspensions of intact cells under varying light conditions. If the light energy is not trapped at the reaction center, the light energy is lost as fluorescence. The results are shown in Figure 3.14. The positive control (row C) only begins to show fluorescence when the light intensity is 4.1 jlE-m" 2 ^" 1 . This represents the limit of detection in this experiment since this strain, ARC6(pA4), has no B870 or reaction center complexes and hence cannot trap any light energy. The suppressor strains ARC6(pTL2-7) and (pTL2-9), and the primary mutant strain, all show fluorescence at a light intensity of 7.5 (lE-nr 2 ^" 1 , but the pseudo-wild type strain only begins to fluoresce at 24 llE-nr 2 ^' 1 . I interpret this to mean that the suppressor strains and the primary mutant strain are not trapping light energy as efficiently as the pseudowild type strain. The inefficiency in the primary mutant may arise from a "back-up" in electron transport arising from the restricted movement of quinones and quinols between the reaction center and the blc\ complex. The inefficiencies of the suppressor strains may be due to the altered stoichiometrics of their photosynthetic units [(15) and see Figure 3.11] and/or a disruption in the structure of the photosynthetic apparatus arising from the nature of the suppressor mutations (see below, section 4.6). Despite the apparent inefficiency of light energy trapping on the part of the suppressor strains, they are still able to grow photosynthetically, albeit not as well as the pseudo-wild type strain. 4.3. Single flash spectroscopy Since the primary mutant strain could not grow photosynthetically, its inability to form a trans-membrane potential via light-driven electron transport was not a surprise. However, it was somewhat puzzling in that the individual components of the photosynthetic unit all seemed to be fully functional ([15] and Results section 3.1.1.4 above). The ensemble performance of the photosynthetic unit requires successful interaction in three places: firstly, energy harvested by the antenna complexes must be efficiently transferred to the reaction center; secondly, protons and electrons from the reaction center must move (via quinols) to the blc\ complex; and thirdly, quinones and electrons must return from the blc\ complex to the reaction center (see Figure 1.1). Energy transfer from the antenna complexes to the reaction center seemed irrelevant because the flashes used in the flash spectroscopy experiment were saturating (i.e., the special pair of the reaction center could be directly excited), yet the fault in cyclic electron transport still existed. It seemed possible to rule out a defect in the transfer of electrons from otype cytochromes to the reaction center as well, because these cytochromes  98 were oxidized with the same kinetics as the reaction center ([15], and see Figures 3.15 and 3.16). This means that a block or restriction in electron transport lay upstream of the c-type cytochromes; that is, between the reaction center Q B site and the FeS center of the blc\ complex. The single flash spectroscopy experiments were designed to look at quinol transfer between the reaction center Q B site and the blc\ complex Q z site. In these experiments, the use of the carotenoid bandshift to measure changes in trans-membrane potential (as outlined in sections 3.1.1.2 and 3.2.1.5) allowed monitoring of electron movement through the blc\ complex, since Phase III of the carotenoid bandshift is mediated by the blc\ complex. By poising the ambient potential appropriately I created optimal conditions for electron movement through the blc\ complex (Qz red. =90 mV), or conditions such that this electron movement depends on the arrival of a quinol from the Q B site of the reaction center (Qz ox. =225 mV). As the results in Figure 3.5 show, when the quinone pool (and hence the quinone at the Q z site) was poised oxidized, electron movement through the cytochrome blc\ complex in the primary mutant strain was severely impaired, whereas chemical reduction of the Q z site largely overcame this impairment. The extent of the bandshift increased about two fold when the Q z site was poised reduced, and the time to reach half the maximum absorption change was reduced by over 50%. By comparison, when Q z site was poised reduced in the pseudowild type strain the extent of the absorption change increased by only 15% over that seen when the Q z site was poised oxidized, and the time to half height was reduced by only 10%. There was a marked difference in the magnitude of the carotenoid bandshift in the two strains. The sum of Phases I and II in the pseudo-wild type strain was 20% greater than in the primary mutant strain, and phase III was almost 70% greater. The difference in the magnitude of the carotenoid bandshift could arise from a second electron flow bottleneck downstream of the Qg to Q z impairment. This bottleneck would not have been detected in the multi-flash experiments; if the upstream bottleneck were circumvented then the second impairment would be manifest. The results of the single flash experiment imply a directed, or facilitated, movement of electrons from the reaction center to the blc\ complex in the pseudo-wild type strain, and that in the primary mutant strain the electrons associated with the Q B quinol molecule do not efficiently reach the Q z site of the cytochrome blc\ complex. If this disruption in movement of quinols from the reaction center to the blc\ complex were caused by a disruption in the physical proximity of the two complexes, I would think that electron transfer via the c-type cytochromes might also be impaired, if only because the slower rate of quinol oxidation would mean a slower rate of c-type cytochrome oxidation as well.  99 The possibility existed that the impairment in electron/proton flow to the blc\ complex from the reaction center was due a difference in the size of the quinone pool in the two strains. Quantitation of quinones ruled this possibility out when it was shown that the quinone concentration was almost identical in the two strains (see Results section 3.1.1.3). 4.4. Multiple flash spectroscopy experiments and the integrity of chromatophores A summary of experimental results from the first multiple flash spectroscopy experiments (Figures 3.15, 3.16, 3.17 and 3.18, Table 3.II), the reduced minus oxidized difference spectra (Table 3.Ill), and the light minus dark spectra carried out in the presence of the membrane potential-sensitive dye oxonol-VI (Figure 3.19), all done with the suppressor strains alongside the pseudo-wild type and primary mutant strains, left me initially rather baffled. Although the kinetics and extents of oxidation and reduction of the bacteriochlorophyll special pair and of the otype cytochromes appeared to be better in the suppressor strains ARC6(pTL2-7) and ARC6(pTL2-9) than in the primary mutant strain, there was a large degree of variability in the response of these suppressor strains from one chromatophore preparation to another. The suppressor strain ARC6*(pTL2) showed kinetics and extents of oxidation and reduction of these same components that were very similar to those of the primary mutant. What was most puzzling, though, was that none of the suppressor strains were able, in any of the chromatophore preparations tested, to generate a transmembrane potential that was significantly larger than the primary mutant strain, whether the potential was monitored using the carotenoid bandshift or oxonol-VI absorption changes. Yet, these data were contradicted by the ability of the suppressor strains to grow photosynthetically, which is good evidence to indicate that light-driven electron transport is successfully carried out in these strains. This contradiction was resolved when flash-induced carotenoid bandshift experiments were carried out on intact cells of all strains (see Figure 3.18). It was evident that the suppressor strains ARC6(pTL2-7) and ARC6(pTL2-9) were able to generate a membrane potential using light energy. Unfortunately, it was not possible to confirm this in the suppressor strain ARC6*(pTL2) because this strain lacks the B800-850 complex, which is the complex that contains the membrane potential-sensitive population of carotenoid molecules (7, 10). For this reason, most of the following discussion will focus on the two suppressor strains ARC6(pTL2-7) and ARC6(pTL2-9). The results from the carotenoid bandshift experiments carried out on intact cells suggested that for some reason the process of making chromatophores destroyed their ability to carry out photosynthetic electron transport. (It should be noted that chromatophore preparations from the wild type and pseudo-wild type strains reproducibly yielded a significant carotenoid bandshift.) I thought  100 that an understanding of why the release of chromatophores from cells disrupts light energy transduction in these suppressor strains would shed light on the nature of the suppressor mutations and, by extension, on the role of the PufX protein. Two possible reasons for the disruption of successful light energy transduction in chromatophores from ARC6(pTL2-7) and ARC6(pTL2-9) presented themselves: (i) the suppressor strains made leaky chromatophores, so that there was no long-lived membrane potential; or (ii) chromatophore isolation from cells somehow disrupted the photosynthetic electron transfer machinery, so that it was not capable of generating a trans-membrane potential. The gross structure of the chromatophores was evaluated in electron micrographs (Figure 3.20). These showed that the chromatophores were spherical and appeared to be topologically closed. Differences in size between the five strains were seen, but these could not be correlated with differences in light energy transduction. The possibility of membrane defects not visible in the micrographs, but which could still cause the membrane to be permeable to ions, was evaluated by the chemically induced carotenoid bandshift experiments (see Results section 3.2.1.8). Membrane potentials generated by a valinomycin/KCl pulse (11) show that a significant membrane potential can be maintained over a relatively long time frame (Figure 3.21), so membrane permeability to ions does not account for the reduced carotenoid bandshifts seen in the flashing experiments. Work carried out by Jackson et al. showed that the reaction center of/?, capsulatus strain PBS 108 is not inserted correctly into the membrane, resulting in the loss of phototrophic growth, because of a point mutation in the gene encoding the a subunit of the B870 complex (12, 13). This mutation changed a highly conserved threonine codon at position 37 (29) to a proline codon. By analogy with strain PBS 108, it is possible that the reaction center of strains ARC6(pTL2-7) and ARC6(pTL2-9) was displaced during disruption of the cells to make chromatophores. I therefore assessed the ability of chromatophores from these strains to oxidize external cytochrome c. Figure 3.22 shows the results from these experiments. It is apparent that the suppressor strains ARC6(pTL2-7) and ARC6(pTL2-9) possessed a greater number of reaction centers that were accessible to external cytochrome c, indicative of abnormal reaction center orientation. The failure of the addition of external cytochrome clo change the carotenoid bandshift (Figure 3.23) probably means two things: (1) the chromatophores are not a mixture of inside-out and right side-out vesicles, since otherwise adding exogenous cytochrome c would result in a reduction in the magnitude of the bandshift, because the potential of the inverted chromatophores would be opposite in polarity to that of the normal chromatophores, and the resulting electrochromic blue shift would cancel out the  101 normal electrochromic red shift; and (2) the mis-oriented reaction centers are not completely inverted in the membrane, since if this were the case, movement of electrons through these reaction centers would also diminish the bandshift. 4.5. Location and sequences of the plasmid-borne suppressor mutations Transfer of the plasmids from the suppressor strains to the photosynthetically "naive" cells of strain of/?, capsulatus ARC6 showed that two of the suppressor strains, ARC6(pTL27) and ARC6(pTL2-9), carried the suppressor mutation on the ^>w/operon-bearing plasmid (Figure 3.24). Subsequent high resolution mapping and DNA sequencing of the suppressor mutations in strains ARC6(pTL2-7) and ARC6(pTL2-9) showed that they are both located in the second codon of the pufA gene (Figure 3.25), which normally encodes a serine in the <X polypeptide of the B870 antenna complex. This polypeptide forms dimers with the B870 ft polypeptide and a ring of these dimers is believed to surround the reaction center as the B870 antenna complex (6, 14). The N terminus of the B870 a polypeptide has been shown to reside on the cytoplasmic side of the inner membrane (25). It is possible that the presence of serine at the second residue of the B870 complex a polypeptide enhances interaction between this complex and a soluble cytoplasmic factor. If so, it might explain why the ability to form a trans-membrane potential was lost when chromatophores were prepared from the suppressor strains ARC6(pTL2-7) and ARC6(pTL29). That is, chromatophores enclose components from the periplasm, whereas cytoplasmic components, unless bound to the outer face of the chromatophore, would be lost when the chromatophores are subjected to differential centrifugation. Alternatively, the presence of this serine residue could enhance interaction between the B870 complex and cytoplasmic segments of other integral membrane proteins of the photosynthetic unit (such as of the reaction center, the PufX protein or other B870 complexes). 4.6. A model for the role of the PufX protein The recent publications of the atomic structure of the B800-850 complex from Rps. acidophila (20), and an 8.5 A resolution projection map of B870 complexes from Rsp. rubrum (14), have greatly increased our understanding of the structures of light-harvesting antenna complexes in bacteria. There are numerous inter-species similarities in sequences, motifs and domains of the polypeptides that make up these complexes (28, 29), and preliminary crystallographic data indicate that there are also structural similarities between species (4, 23).  102 Consequently, this information is used to provide a foundation for a model of the role of the PufX protein. As outlined in the Introduction, the B870 antenna complex is thought to surround the reaction center, with a stoichiometry of about sixteen a(3 dimers per reaction center (14). This antenna structure appears to be closed (14, 20) in that there is no obvious pathway for quinols, which are membrane soluble, and cytochrome Cy, which is membrane associated, to move between the reaction center and the blc\ complex. Data from both Karrasch et al. and McDermott et al. indicate that the two transmembrane a-helices in the ccj3 dimer are not in contact with each other. This was an unexpected result; apparently the two types of apoprotein form the inner (a) and outer ((J) walls of a transmembrane cylinder, with the bacteriochlorophyll and carotenoid pigments between these two "walls" (14, 20). Although the distances between the a and (3 proteins are perhaps greater than was anticipated, the overall structure would probably form a barrier to the lateral diffusion of quinols within the membrane, since the non-covalently bound carotenoid and bacteriochlorophyll molecules fill the space between the a and (5 proteins (at least in the B800-850 structure). It should be noted that this proposed barrier would not be an absolute barrier, that is, some movement of quinones and quinols between the reaction center and the blc\ complex can occur, as evidenced by the primary mutant's ability to generate a limited membrane potential. The serine residue encoded by the second codon of pufA that is mutated in the suppressor mutants is not widely conserved within the B870 a proteins from the phototrophic (X-Proteobacteria that have been sequenced; only R. sphaeroides also has a serine at this position (29). Of the 48 independent suppressor mutations of a R. sphaeroidespuJX deletion strain that were analyzed, 90% were found to occur in the genes encoding the B870 complex (2). However, none of the B870-associated suppressor mutations occurred in the Ser2 codon of the OC polypeptide; instead, 91.2% were found to be grouped in sequences encoding residues on the periplasmic side of the trans-membrane segment of the a or (3 polypeptides. Because of the cytoplasmic and periplasmic locations of suppressor mutations in R. capsulatus and R. sphaeroides, the effects of these mutations are not easily explained by changes in interactions with soluble proteins. Instead, they are best explained by changes in interactions between B870 peptides and membrane-proximal segments of other membrane proteins. It is worth noting that in the B800-850 complex from Rps. acidophila the only direct contacts between the two polypeptides in the a(3 dimer are found in the cytoplasm or in the periplasm (20).  103 Cytoplasm The suppressor S F  M-A-D-K-*J-D-L- - '  nT'  mutations from R. capsulatus and R. sphaeroides that are found in the genes encoding the B870 complex are shown in Figure 4.1, and can be summarized as follows: the plasmid-borne suppressor mutations identified in this study both occur in the aSer2 residue, and the B870 complexes in the suppressor  T J A Tip - Stop | |Ser-Phe  strains have altered spectral properties (see Figure 3.11). The work done on  Periplasm Figure 4.1 Suppressor mutations in the B870 antenna complex. The schematic representation shows the location of mutations in the B870 complexes of R. capsulatus and R. sphaeroides that suppressed the effects of the deletion of the pufiC gene. Mutated codons are circled and the nature of the mutation shown in the linked box. The solid line represents the approximate limits of the hydrophobic r r r . , , , , , ,. i r , portion or the membrane; the dashed lines represent the approximate outer edges of the polar head groups of the phospholipids. After Jackson and Prince (12).  R. sphaeroides pufiC suppressors was carried out with strains that were merodiploid; there were co  P i e s o f t h e PufA a n d Pufi genes on the chromosome j L i • and on the complementing r ° plasmid (2). (Interestingly, >90% of the suppressor mutations found in these  experiments mapped to the the plasmid-borne copy of the genes.) Fifty-five per cent of the suppressor mutations characterized by Barz et al. converted either the crTrp43 or the /TTrp44 to a stop codon and it was found that, in order to suppress the effect of the puJXmutation  on  photosynthetic growth, these mutations required the presence of a wild-type copy of the pufA and pufB genes. A further 35% of the suppressor mutations converted the /TTrp47 codon to an Arg codon and these mutants did not assemble B870 complexes in the absence of a wild-type copy of the pufA and pufB genes, although they were able to grow photosynthetically. Finally, 4% of the R. sphaeroides suppressor mutations resulted in codon changes that allowed the suppressors to grow photosynthetically in the absence of a wild-type copy of the pufA and pufB genes, but which had spectrally altered forms of the B870 complex. In addition to a fiH'islO to Arg mutation (the only mutation located in sequences on the cytoplasmic side of  104 the membrane found by Barz etaL), which changes a residue thought to lie just at the cytoplasmic surface of the membrane (29), there was also an CtSer47 to Phe mutation. I distinguish at least two types of suppressor mutation in this summary. The first type of suppressor mutation causes a structural change in the B870 complex, and occurs when there is only a single copy of the pufA and pufB genes in the cell. This category includes «Ser2 to Pro and «Ser2 to Phe mutations in R. capsulatus and the /?His20 to Arg and aSer47 to Phe changes in R. sphaeroides. I argue that this type of mutation disrupts or destabilizes the structure of the B870 antenna complex, allowing exchange of quinones and quinols between the reaction center and the blc\ complex. For example, the change of the serine residue to phenylalanine or proline might allow the B870 complexes to be more easily displaced from the reaction center, or change the packing density of this complex around the reaction center, thus facilitating quinone and quinol exchange at the reaction center Qg site. This decreased stability also explains why the reaction centers of the suppressor strains were more liable to displacement during cell breakage, and why there was a large degree of variability in the response of these strains in the multiple-flash spectroscopy experiments. Unfortunately, no attempt was made to isolate B870-reaction center complexes from the suppressor strains; the model would predict that the detergent solubilization might result in the disassociation of B870 and reaction center complexes in these strains. The second type of suppressor mutation only occurs in a pufBA merodiploid, and it results in changes to the highly conserved Trp residues in the periplasmic domains of the Ctfi dimer. The residue PTrp44, for example, is conserved in the B870 (5 apoproteins of eleven different strains of bacteria, and has been proposed to interact with the bacteriochlorophyll molecule that is liganded to the His residue six positions N-terminal (29). Changing this residue to a stop codon also results in the loss of the /347Trp residue, also highly conserved in the j3 apoprotein from both B870 and B800-850 complexes. In Rps. acidophila this residue is known to be in contact with the B850 bacteriochlorophyll pigment (20). None of the Type 2 suppressor mutants were able to form the B870 complex, and those mutants which had changed the Trp codon to a stop codon were unable to grow photosynthetically in the absence of wildtype copies of the pufB and A genes. Also, the Trp to stop mutants had reduced levels of absorption due to the B870 complex when they were in the pufBA+ background. The {FTrp47 to Arg suppressors were, as mentioned, photosynthetically competent, but the B870 complex in this strain was spectroscopically undetectable and expression of the a. and /J apoproteins (as measured by Western blots with anti-B870 antibodies) were low (2). I speculate that the second type of suppressor mutation allows the mutated polypeptide to, in effect, replace the  105 PufX protein, in that the mutation supplies a polypeptide that does not bind pigment but that is able to take a place in the antenna ring structure. The rather minimal requirements for a PufX protein-like protein may in part explain the low degree of amino acid sequence similarities in alignments between the PufX proteins from R. capsulatus and R. sphaeroides, and between the PufX and the B870 a proteins. As long as the polypeptide maintains a three domain structure, with the periplasmic and cytoplasmic sequences available for interaction with the antenna polypeptides, and does not bind pigment, the sequence is somewhat free to change. It is possible that the reason Barz et al. (2) saw lower B870 absorption in their Trp mutation merodiploids is that the level of the non pigment-binding polypeptides was higher than it would have been in a wild-type situation, where the levels of pufXtranscription  are  regulated by differential decay of the principal puf operon transcript such that the number of pujX transcripts is the same as the number of transcripts encoding the L and M subunits of the reaction center (3). As might be inferred from the mechanisms of suppression put forward above, I propose a structural role for the PufX protein in the photosynthetic unit. Thus, the PufX protein would not be directly involved in the movement of electrons and protons between the reaction center and the b/c\ complex; rather it would facilitate this movement, possibly by forming a quinone/quinol gateway in the ring of B870 complexes surrounding the reaction center. In this model, the PufX protein would replace the a polypeptide in one of the B870 Ct/?dimers. As mentioned above, the PufX protein co-purifies with the B870-reaction center complex in R. sphaeroides (9) and with B870 polypeptides from R capsulatus (J. T. Beatty, personal communication). It has the characteristic three domain structure found in the apoproteins of light-harvesting complexes from phototrophic Proteobacteria as evaluated by computer analysis (15), but the PufX protein has none of the conserved residues associated with pigment binding (20, 29). The absence of pigments in the region of a PufX protein insertion into the B870 complex could conceivably allow the passage of quinone/quinols through the antenna ring. This proposed role for the PufX protein is consistent with the phenotype of the primary mutant strain. The growth experiments and single-flash spectroscopy results indicated that quinols did not reach the blc\ complex at a rate sufficient to support photosynthetic growth. This deficiency could be overcome by supplying a second quinol oxidase as part of the inducible DMSO reductase pathway, but not by increasing the turnover of the reaction center with increases in light intensity.  106 In the absence of the PufX protein, the B870 antenna ring structure would contain an extra set of bacteriochlorophyll molecules, and this could account for the increased absorption at 870 nm seen in the mutant strains. It has been shown that, in R. sphaeroides, the loss of the PufX protein results in changes in the organization of the B870 complexes around the reaction center (27). These authors presented fluorescence polarization data showing that in the pufX mutant the number of interacting pigment molecules in the B870 antenna complex was higher than in the pseudo-wild type strain. More evidence pointing to a role for the PufX protein in the structure of the B870-reaction center complex of/?, sphaeroides was presented by McGlynn et al. (22). A 16% increase in the size of the B870 complex in pufi(mutants  was  reported by these authors, and they also showed that the PufX protein was not required in the simultaneous absence of the B870 and B800-850 complexes. These results are consistent with my model, since in the simultaneous absence of the PufX protein and light-harvesting complexes, quinones and quinols would be capable of unobstructed diffusion from the reaction center Qg site to the blc\ complex Qz site. Finally, direct evidence that physical restriction of the diffusion of quinones and quinols in the membrane can stop electron transfer has been presented (19). These authors restricted the void space in chloroplast and mitochondrial membranes by decreasing the osmolality of the medium surrounding vesicles prepared in moderately high osmolality buffers. As the membrane was stretched, diffusion of quinones and quinols was restricted to the point of immobility, and electron transport ceased (19). This demonstrates the importance of intramembrane voids, such as might be formed between a PufX:B870 (3 heterodimer, to the movement of quinones and quinols in the membrane. It should be noted that the proposed model relies heavily on the crystallographic structure of the B800-850 antenna complex from Rps. acidophila and the fairly low resolution electron density map of the isolated B870 antenna complex from Rsp. rubrum. These data are extrapolated to R. capsulatus on the basis of sequence similarities between the constituent polypeptides of the antenna complexes. Other oligomeric structures for light-harvesting antenna complexes have been proposed (see the review by Zuber and Brunisholz [30], for examples) which fit the observed spectrophotometric and biochemical data. The external thesis examiner proposed that a combination of the reaction center - blc\ complex supercomplex proposed by Joliot et al. (13a) with two B870 complexes would fit the bulk of the data from R. capsulatus. In such a model, the PufX protein could serve as a structural element tha ensured an optimal alignment between the two reaction centers and the blc\ complex or between the two B870 complexes and the reaction centers. In the latter case, it  107 could still be incorporated as part of the B870 complex. This would accord with the nature of the suppressor mutations, but would not explain the inhibition of quinol movement between the reaction center and the blc\ complex. The accuracy of my proposed model for the PufX protein might soon be tested by the generation of higher resolution projection maps of the B870 complex from Rsp. rubrum. If insufficient resolution is attainable, less direct methods could be employed. For example, the number of Oc(3 bchltf dimers in the B870 complex of puJXstrains of R. capsulatus that lack the B800-850 complex could be measured using fluorescence polarization spectra. Such measurements have been made in R. sphaeroides, but they were complicated by the presence of the B800-850 antenna complex and by the formation of B870 oligomers that were not associated with reaction centers (27). One could also try domain switching between B870 a and PufX proteins in order to localize the sequences necessary for the function of the PufX protein. Another approach would be to substitute a mutant copy of the pufA gene for the pufiC gene. By eliminating the bacteriochlorophyll-binding His residue from the B870 a polypeptide, one could test the hypothesis that the PufX protein can be replaced by a nonpigment-binding polypeptide with the required three domain structure. However, until further data become available the most suitable proposed role for the PufX protein would seem to be as an integral part of the B870-reaction center core complex, facilitating the passage of quinones and quinols to and from the reaction center through the B870 ring structure. 4.7. Discussion references 1. Armstrong, GA., A. Schmidt, G. Sandmann and J.E. Hearst. 1990. Genetic and biochemical characterization of carotenoid biosynthesis mutants of Rhodobacter capsulatus.}. Biol. Chem. 265:8329-8338. 2. Barz, W.P. and D. Oesterhelt. 1994. Photosynthetic deficiency of a pufXdeletion mutant of Rhodobacter sphaeroides is suppressed by point mutations in the light-harvesting complex genes pufB or pufA. Biochemistry 33:9741-9752. 3. Belasco, J.G., J.T. Beatty, C.W. Adams, A. von Gabain and S.N. Cohen. 1985. Differential expression of photosynthesis genes in R. capsulatus results from segmental differences in stability within the polycistronic rxcA transcript. Cell 40:171-181. 4. Boonstra, A.F., L. Germeroth and E.J. Boekma. 1994. Structure of the light harvesting antenna from Rhodospirillum molischianum studied by electron microscopy. Biochim. Biophys. Acta 1184:227-234. 5. Clark, A.J., N.P.J. Cotton and J.B. Jackson. 1983. The relation between membrane ionic current and ATP synthesis in chromatophores from Rhodopseudomonas capsulata. Biochim. Biophys. Acta 723:440-453.  108 6. Cogdell, R.J. and A.M. Hawthornthwaite. 1993. Preparation, purification, and crystallization of purple bacteria antenna complexes, pp. 23-42. In J. Diesenhofer and J. R. Norris, The photosynthetic reaction center. Academic Press, Inc., San Diego. 7. Crielaard, W., F.v. Mourik, R.v. Grondelle, W.N. Konings and K.J. Hellingwerf. 1992. Spectral identification of the electrochromically active carotenoids of Rhodobacter sphaeroides in chromatophores and reconstituted liposomes. Biochim. Biophys. Acta 1100:914. 8. Daldal, F., E. Davidson and S. Cheng. 1987. Isolation of the structural genes for the Rieske Fe-S protein, cytochrome b and cytochrome c. all components of the ubiquinohcytochrome c2 oxidoreductase complex of Rhodobacter capsulatus. J. Mol. Biol. 195:1-12. 9. Farchaus, J.W., W.P. Barz, H. Grunberg and D. Oesterhelt. 1992. Studies on the expression of the pufX polypeptide and its requirement for photoheterotrophic growth in Rhodobacter sphaeroides. EMBO J. 11:2779-2788. 10. Goodwin, M.G. and J.B. Jackson. 1993. Electrochromic responses of carotenoid absorbance bands in purified light-harvesting complexes from Rhodobacter capsulatus reconstituted into liposomes. Biochim. Biophys. Acta 1144:191-198. 11. Jackson, J.B. and A.R. Crofts. 1969. The high energy state in chromatophores from Rhodopseudomonas spheroides. FEBS Lett. 4:185-189. 12. Jackson, W.J. and R.C. Prince. 1987. Genetic and DNA sequence analysis of a Rhodobacter capsulatus mutant unable to properly insert photochemical reaction centers into the membrane, pp. 725-728. In}. Biggins, Progress in Photosynthesis Research. Martinus NijhofT Publishers, Dordrecht. 13. Jackson, W.J., R.C. Prince, G.J. Stewart and B.L. Marrs. 1986. Energetic and topographic properties of a Rhodopseudomonas capsulata mutant deficient in the B870 complex. Biochem. 25:8440-8446. 13a. Joliot, P., A. Vermeglio and A. Joliot. 1989. Evidence for supercomplexes between reaction centers, cytochrome ci and cytochrome bc\ complex in Rhodobacter sphaeroides whole cells. Biochim. Biophys. Acta 975:336-34514. Karrasch, S., P.A. Bullough and R. Ghosh. 1995. The 8.5 A projection map of the light-harvesting complex I from Rhodospirillum rubrum reveals a ring composed of 16 subunits. EMBO J. 14:631-638. 15. Lilburn, T.G. 1990. The role of the pufX gene product of Rhodobacter capsulatus. M. Sc. thesis, University of British Columbia. 16. Lilburn, T.G. and J.T. Beatty. 1992. Suppressor mutants of the photosynthetically incompetent pufi(deletion mutant Rhodobacter capsulatus ARC6(pTL2). FEMS Microbiol. Lett. 100:155-160.  109 17. Loach, P.A., P.S. Parkes-Loach, C M . Davis and B.M. Heller. 1994. Probing protein structural requirements for formation of the core light-harvesting complex of photosynthetic bacteria using hybrid reconstitution methodology. Photosynth. Res. 40:231245. 18. Marrs, B.L. 1982. Genetic analysis of carotenogenesis in Rhodopseudomonas capsulata, pp. 273-277. In G. Britton and T. W. Goodwin, IUPAC Carotenoid Chemistry and Biochemistry. Pergamon Press, New York. 19. Mathai, J.C., Z.E. Sauna, O. John and V. Sitaramam. 1993. Rate-limiting step in electron transport. J. Biol. Chem. 268:15442-15454. 20. 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 lightharvesting complex from photosynthetic bacterium. Nature 374:517-521. 21. McEwan, A.G., N.P.J. Cotton, S.J. Ferguson and J.B. Jackson. 1984. The inhibition of nitrate reduction by light in Rhodopseudomonas capsulata is mediated by the membrane potential, but the inhibition by oxygen is not. Adv. Photosynth. Res. 2:449-452. 22. McGlynn, P., C.N. Hunter and M.R. Jones. 1994. The Rhodobarter sphaeroides PufX protein is not required for photosynthetic competence in the absence of a light harvesting system. FEBS Lett. 349:349-353. 23. Meckenstock, R.M., K. Krusche, R.A. Brunisholz and H. Zuber. 1992. The lightharvesting core-complex and the B820-subunit from Rhodopseudomonas marina. Part II. Electron microscopic characterisation. FEBS Lett. 311:135-138. 24. Richardson, D.J., A.G. McEwan, J.B. Jackson and S.J. Ferguson. 1989. Electron transport pathways to nitrous oxide in Rhodobacter species. Eur. J. Biochem. 185:659-669. 25. Tadros, M.H., R. Frank and G. Drews. 1986. Localization of the N-terminal regions of the B870CC, (3 and reaction center L polypeptides on the cytoplasmic surface of the chromatophores of Rhodopseudomonas capsulata. FEBS Lett. 196:233-236. 26. Taylor, M.A. and J.B. Jackson. 1985. Threshold dependence of bacterial growth on the protonmotive force. FEBS Lett. 192:199-203. 27. Westerhuis, W.H.J., J.W. Farchaus and RA. Neiderman. 1993. Altered spectral properties of the B875 light-harvesting pigment-protein complex in a Rhodobacter sphaeroides mutant lacking pufiC. Photochem. Photobiol. 58:460-463. 28. Zuber, H. 1993. Structural features of photosynthetic light-harvesting systems, pp. 43-100. In). Deisenhofer and J. R. Norris, The Photosynthetic Reaction Center. Academic Press, Inc., San Diego. 29. Zuber, H. and R.A. Brunisholz. 1991. Structure and function of antenna polypeptides and chlorophyll-protein complexes: Principles and variability, pp. 627-703. In H. Scheer, Chlorophylls. CRC Press, Boca Raton, FL.  110 30. Zurdo, J., C. Fernandez-Cabrera and J.M. Raminez. 1993. A structural role of the carotenoid in the light-harvesting II protein of Rhodobacter capsulatus. Biochem. J. 290:531537.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088259/manifest

Comment

Related Items