Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Structural and functional relationships of the bacteriopheophytin environment in a photosynthetic reaction… Saer, Rafael Gonzalo 2014

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

Item Metadata

Download

Media
24-ubc_2014_november_saer_rafael.pdf [ 4.63MB ]
Metadata
JSON: 24-1.0167255.json
JSON-LD: 24-1.0167255-ld.json
RDF/XML (Pretty): 24-1.0167255-rdf.xml
RDF/JSON: 24-1.0167255-rdf.json
Turtle: 24-1.0167255-turtle.txt
N-Triples: 24-1.0167255-rdf-ntriples.txt
Original Record: 24-1.0167255-source.json
Full Text
24-1.0167255-fulltext.txt
Citation
24-1.0167255.ris

Full Text

STRUCTURAL AND FUNCTIONAL RELATIONSHIPS OF THE BACTERIOPHEOPHYTIN ENVIRONMENT IN A PHOTOSYNTHETIC REACTION CENTER PROTEIN  by RAFAEL GONZALO SAER B.Sc., The University of Miami, 2008 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF    DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Microbiology and Immunology) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) AUGUST 2014 © Rafael Gonzalo Saer, 2014  ii  ABSTRACT   The photosynthetic reaction center (RC) of the bacterium Rhodobacter sphaeroides is a pigment-protein complex in which electron transfer reactions efficiently capture solar energy in the form of charge separation. Embedded in this protein are two molecules of bacteriopheophytin (BPhe), one of which mediates electron transfer (HA). Efficient electron transfer through HA, as well as other cofactors of the RC, relies on both the spatial arrangement of the cofactors in the protein scaffold, as well as favorable interactions between the cofactors and their surrounding protein components. In this work, the interplay between the HA cofactor and its protein environment are explored. Specifically, a leucine residue whose side chain projects orthogonally to the plane of the HA macrocycle was changed by site-directed mutagenesis, creating a series of RC mutants with different side chains at this position (termed (M)214). The results derived from this work reveal that: (i) (M)214 plays a role in the pigment selectivity of the HA binding pocket; (ii) the volume of the (M)214 side chain is important for fast forward electron transfer; and (iii) the (M)214 residue affects the configuration of a nearby bacteriochlorophyll pigment. In a subsequent chapter of this work, I describe the effects of HA coordination state in RCs that assemble exclusively with zinc-bacteriochlorophyll in place of bacteriochlorophyll and BPhe. My results reveal that the coordination state of the HA Zn2+ metal plays a role in determining the yield of charge separation within the protein. In addition, it was found that these so-called Zn-RCs do not assemble with a full occupancy of a cofactor. Collectively, the results from this work serve to increase our basic understanding of the protein-cofactor interplay within photosynthetic systems.  iii  PREFACE  Much of the work presented in this thesis is derived from publications of which I am either the first or second author. The text from Sections 2.9, 2.10, 2.11, 3.1, 3.3, 3.4, 3.5, 4.1, 4.2, and 4.3, including Table 3.1 and Figures 3.4, 3.5, 3.6, 3,7, 3.8, 3.9, 3.10, 4.1, and 4.2, have all been published in Saer, R.G., Hardjasa, A., Rosell, F.I., Mauk, A.G., Murphy, M.E., and Beatty, J.T., (2013) Role of Rhodobacter sphaeroides photosynthetic reaction center residue M214 in the composition, absorbance properties, and conformations of the H(A) and B(A) cofactors. Biochemistry, 52(13), 2206-17. F. I. Rosell obtained the low temperature spectra. A. Hardjasa crystallized the protein samples, produced Figures 3.9, 3.10, Appendix 1, and the text corresponding to sections 2.10 and 3.5. M. E. Murphy refined the crystallography dataset and assisted with sections 2.10 and 3.5. The low temperature spectroscopy equipment was the property of A. G. Mauk. I performed the remaining experiments, and conceived the design of the experimental workflow. All other figures and text in the above sections were written by me under the supervision of J. T. Beatty.  The time-resolved spectroscopy was performed in collaboration with the group of Dr. Neal Woodbury at Arizona State University. Sections 3.6, and 4.5, including Table 4.2 and Figures 3.11, 3.12, 3.13, 3.14, 4.3, and 4.4 were published in Pan, J., Saer, R. G., Lin, S., Guo, Z., Beatty, J.T., and Woodbury, N.W., (2013) The protein environment of the bacteriopheophytin anion modulates charge separation and charge recombination in bacterial reaction centers. J. Phys. Chem. B. 117(24), 7179-89. The text for sections 3.6 and 4.5 is my own, which is written based upon the results and discussion presented in Pan et al. The table and figures in the above publication are the work of J. Pan, and reproduced with iv  permission in this thesis. J. Pan assembled the time-resolved spectrometer acquired the data. During this period, I visited the group of Dr. Woodbury to learn this spectroscopic technique, including the analysis of the obtained datasets, which were also performed by J. Pan. Experimental design was the work of Drs. J. Pan, S. Lin, N. W. Woodbury, J.T. Beatty, and I. All the protein mutants were created by me.   The text from Sections 2.12, 3.9, 4.6, 4.7, and 4.8, including Tables 3.3 and 3.4, as well as Figures 3.18, 3.19, 3.20, 3.21, and 3.22, are published in Saer, R.G., Pan, J., Hardjasa, A., Lin, S., Rosell, F. I., Mauk, A. G., Woodbury, N. W., Murphy, M. E., and Beatty, J. T. (2013) Structural and kinetic properties of Rhodobacter sphaeroides photosynthetic reaction centers containing exclusively Zn-coordinated bacteriochlorophyll as chlorin cofactors. BBA: Bioenergetics. (Dec 5). In this work, J. Pan performed the time-resolved spectroscopy, A. Hardjasa, crystallized the protein, M. E. Murphy performed the crystallographic data refinement, created Table 3.3 and Figure 3.9.3, as well as Appendix 2. F.I. Rosell obtained the low temperature spectroscopy data. The low temperature spectrometer was the property of A. G. Mauk. The experimental design was conceived by Drs. S. Lin, N.W. Woodbury, J. T. Beatty, and I. The protein material for this work was created by me. All other figures and tables, as well as the manuscript text, were produced by me.   The recipe of the RLB culture media (Section 2.1) is published in Jun, D., Saer, R.G., Madden, J.D., and Beatty, J.T. (2013) Use of new strains of Rhodobacter sphaeroides and a modified simple culture medium to increase yield and facilitate purification of the reaction centre. Photosynth. Res. (Jun 14). None of the experiments presented in Jun et al. are included in this thesis.  v   A description of the plasmid pRS1 (Section 2.6) and its uses as an expression system for the R. sphaeroides photosynthetic reaction center has been published in Jaschke, P.R., Saer, R.G., Noll, S., and Beatty, J.T. (2011) Modification of the genome of Rhodobacter sphaeroides and construction of synthetic operons. Methods Enzymol. 497, 519-38. No text or figures in this thesis are derived from the above publication.    vi  TABLE OF CONTENTS  ABSTRACT ............................................................................................................................. ii PREFACE ............................................................................................................................... iii TABLE OF CONTENTS ...................................................................................................... vi LIST OF TABLES ................................................................................................................. ix LIST OF FIGURES ................................................................................................................ x LIST OF ABBREVIATIONS ............................................................................................. xiv ACKNOWLEDGEMENTS ............................................................................................... xvii 1. INTRODUCTION............................................................................................................... 1 1.1 Definitions and types of photosynthesis.......................................................................... 1 1.2 Pigments of photosynthetic bacteria. .............................................................................. 4 1.3 The photosynthetic apparatus of R. sphaeroides. ............................................................ 8 1.4 Reaction center types, structure and function. .............................................................. 12 1.5 Interactions between the chlorins and the protein component of the RC. ..................... 18 1.6 Analysis of RCs (RC pigments) by optical absorption spectroscopy. .......................... 19 1.7 The protein environment around HA, and the origin of BPhe. ...................................... 28 1.8 Thesis aims and approach. ............................................................................................ 32 2. MATERIALS AND METHODS ..................................................................................... 34 2.1 Bacterial strains and plasmids. ...................................................................................... 34 2.2 Cloning. ......................................................................................................................... 40 2.3 Modification of the puhA gene to facilitate the creation of RC expression systems and purification of the RC. ......................................................................................................... 40 2.4 Removal of excess DNA from pAli2::puf. ................................................................... 53 2.5 Mutagenesis of the reaction center M gene. .................................................................. 53 2.6. Assembly of the RC genes onto a broad host range expression vector for in trans expression of mutant RCs. .................................................................................................. 56 2.7. Purification of the RC. ................................................................................................. 58 2.8. Quantification of pigments from RCs. ......................................................................... 59 2.9. Low temperature absorption spectroscopy. .................................................................. 59 2.10. X-ray crystallography. ................................................................................................ 60 2.11. Photoheterotrophic growth experiments. ................................................................... 62 vii  2.12 Time-resolved spectroscopy. ....................................................................................... 62 2.13. Partial purification of chlorins. .................................................................................. 63 2.14. Semi-preparative HPLC purification and quantification of chlorins. ........................ 65 2.15. Pheophytinization of chlorins by chemical means. .................................................... 66 2.16. Pigment substitution of RCs under mild conditions................................................... 66 3. RESULTS .......................................................................................................................... 67 3.1 Optimization of chemotrophic growth conditions for R. sphaeroides. ......................... 67 3.2. Pigment composition of RCs with amino acid changes at position 214 on the M subunit. ................................................................................................................................ 75 3.3 Ground-state analysis of mutant RCs by low temperature absorbance spectroscopy. .. 78 3.3.1 Effect of replacement of (M)L214 with nonpolar or S-containing amino acid residues on the Qy and Qx transitions of RC BPhes. ....................................................... 78 3.3.2 Effect of replacement of (M)L214 with nonpolar or S-containing amino acid residues on the Qy and Qx transitions of RC BChls. ........................................................ 81 3.3.3 The Qy and Qx regions in the absorbance spectra of (M)214 polar mutants. ......... 84 3.4 Functional analysis of (M)214 mutant RCs in photoheterotrophic growth experiments. ............................................................................................................................................. 88 3.5 X-ray crystallography of (M)L214G, (M)L214A, and (M)L214N mutant RCs. .......... 91 3.6 Kinetic analysis of wild type and mutant RCs with nonpolar or sulfur-containing side chains at (M)214.................................................................................................................. 96 3.7 Pigment substitutions of wild type and (M)L214G RCs. ............................................ 107 3.8. Bacteriopheophytin incorporation into the RC in the presence of adventitious ligands. ........................................................................................................................................... 113 3.9 Structural, spectroscopic, and functional studies of RCs containing zinc-bacteriochlorophyll (Zn-BChl) in place of both BChl and BPhe. ..................................... 121 3.9.1 Comparison of the wild type RC, the (M)L214H RC, the Zn-RC, and the Zn-β-RC in low temperature absorption spectroscopy. ................................................................ 122 3.9.2 X-ray crystallography and anomalous scattering of Zn- and Zn-β-RCs. ............. 125 3.9.3 Time-resolved spectroscopy of wild type, (M)L214H, Zn-, and Zn-β-RCs. ....... 130 4. DISCUSSION .................................................................................................................. 140 4.1 Pigment discrimination at HA. ..................................................................................... 140 4.2 Spectroscopic and structural consequences of nonpolar (M)L214 mutations............. 144 4.3 Spectroscopic and structural consequences of polar (M)L214 mutations................... 150 4.4 Effects of changing the volume of (M)214 on the structure of RCs in vivo. .............. 152 viii  4.5 In vitro performance of RCs with nonpolar moieties at HA. ....................................... 154 4.6 Functional analysis of Zn- and Zn-β RCs. .................................................................. 168 4.7 Consequences of the absence of Mg-BChl in RCs. .................................................... 172 4.8 Biological implications of unnatural cofactor incorporation. ..................................... 175 4.9 Why do natural type II RCs contain a BPhe in the HA or analogous pocket, regardless of the presence of Mg2+ versus Zn2+ in their (B)Chls?...................................................... 176 5. SUMMARY AND FUTURE DIRECTIONS ................................................................ 179 REFERENCES .................................................................................................................... 181 APPENDIX 1. Tabulated photoheterotrophic growth rates of wild type and (M)214 mutants at 5 µE.m-2.s-1 light intensity. ................................................................................................ 192 APPENDIX 2. X-ray data collection and refinement statistics for (M)L214G, (M)L214A, and (M)L214N RCs. ............................................................................................................. 193 APPENDIX 3. X-ray data collection and refinement statistics for the Zn- and Zn-β-RCs. 194 APPENDIX 4: Multiple sequence alignment of the PufM protein sequence ...................... 195 APPENDIX 5: Light minus dark difference spectra of wild type and (M)L214G RCs before and after incubation with ubiquinone-10. ............................................................................. 199     ix  LIST OF TABLES  Table 2.1 Bacterial Strains ..................................................................................................... 36 Table 2.2. List of plasmids used in this study ........................................................................ 37 Table 2.3. Primers used for the deletion of the hairpin 3’ of puhA ........................................ 48 Table 2.4. Primers utilized for the addition of a C-terminal 6x his tag .................................. 50 Table 2.6. Primers used to change the (M)L214 residue on the RC M gene. ........................ 55 Table 3.1. Pigment composition of (M)214 mutant RCs. ...................................................... 77 Table 3.2. Calculated BChl:BPhe ratios of wild type and (M)L214G RCs purified in the presence and absence of 2 mM Imidazole ............................................................................ 120 Table 3.3. Anomalous map statistics for metals in the Zn-RC and Zn-β-RC from x-ray diffraction datasets. ............................................................................................................... 127 Table 3.4. Calculated transient state lifetimes, reaction rates and final yields of charge separation for RCs in this study based on global analysis. ................................................... 131 Table 4.1 Organisms that do not encode a Leu at the analogous (M)214 position from a search of 100 genome sequences that contain a pufM homologue ....................................... 153 Table 4.2. Populations (in %) of charge-separated states in (M)L214 mutant RCs derived from linear combinations of the wild type hundreds of picoseconds and non-decaying EADS................................................................................................................................................ 162    x  LIST OF FIGURES  Figure 1.1. Chemical structures of the pigments from the photosynthetic apparatus of R. sphaeroides. .............................................................................................................................. 7 Figure 1.2. A schematic of photosynthesis in purple bacteria. ................................................ 9 Figure 1.3. The three dimensional structure of the RC from PDB 2J8C. .............................. 16 Figure 1.4. Electron transfer time constants within the RC of R. sphaeroides. ..................... 17 Figure 1.5. Bonding interactions of the BChl dimer with the surrounding protein. .............. 21 Figure 1.6. Bonding interactions of the BA and HA cofactors with the surrounding protein. 22 Figure 1.7. The two molecular axes of BChl a. ..................................................................... 25 Figure 1.8. Ground state absorption spectrum of the wild type R. sphaeroides RC taken at 10 K. ............................................................................................................................................. 27 Figure 1.9. Two stages of integral membrane protein development. ..................................... 31 Figure 2.1. General schematic for the design of a synthetic operon with the aim of expressing RCs in trans. .......................................................................................................... 42 Figure 2.2. Predicted secondary structure of the region 3’ of the puhA stop codon. ............. 44 Figure 2.3. A schematic of the SLIM protocol for site-directed mutagenesis. ...................... 46 Figure 2.4. A genetic and restriction map of pRS1. ............................................................... 57 Figure 3.1. Whole cell absorption spectra of chemotrophically grown R. sphaeroides cultures incubated in the presence of various RCV media components............................................... 69 Figure 3.2. Whole cell absorption spectra of chemotrophically grown R. sphaeroides cultures grown in the presence of various Super Salts components. .................................................... 71 Figure 3.3. Comparison of LH1 output values from cells grown in flasks filled to different volumes. .................................................................................................................................. 74 xi  Figure 3.4. Low temperature (~11 K) ground-state absorption spectra of (A) the Qy and (B) the Qx transitions of the nonpolar (M)214 mutant series. ...................................................... 80 Figure 3.5. Overlay of the BA,B Qy absorbance band of wild type and (M)L214G/A/C/V mutant RCs from the 11 K absorbance spectra. ...................................................................... 83 Figure 3.6. Low temperature (~11 K) steady-state absorption spectra of (A) the Qy and (B) the Qx transitions of the polar (M)214 mutant series. ............................................................ 85 Figure 3.7. Overlay of the Qy transitions corresponding to the BA,B and HA,B regions in the ~11 K absorbance spectrum of the wild type and (M)L214H/N/Q RCs. The peaks denoted “β” correspond to the HA BChl absorbance in the (M)L214H/N/Q RCs. .............................. 86 Figure 3.8. Photoheterotrophic growth of R. sphaeroides strains containing mutant RCs cultured with 5 µE.m-2.s-1 illumination. .................................................................................. 90 Figure 3.9. Omit difference electron density maps of: (A), the (M)L214G RC; (B), the wild type RC (PDB:2J8C); and (C), the (M)L214A RC................................................................. 93 Figure 3.10. Omit difference electron density map of the (M)L214N RC ............................ 95 Figure 3.11. Normalized kinetics of stimulated emission decay in wild type and (M)L214 mutant RCs measured at 926 nm with an excitation pulse at 865 nm. ................................... 99 Figure 3.12. Normalized kinetic traces of the electrochromic shift of the BA absorbance, measured at 785 nm, for wild type and (M)L214 mutant RCs. ............................................ 101 Figure 3.13. Normalized kinetic traces measured at 675 nm for wild type and (M)L214G/A/C mutant RCs. ................................................................................................ 103 Figure 3.14. Normalized kinetic traces (circles) and their respective fits (solid lines) measured at 837 nm for wild type and (M)L214G/A/C mutant RCs. .................................. 105 xii  Figure 3.15. Absorption spectra of wild type (left) and (M)L214G (right) RCs after incubation at 37o C with a molar excess of heterologous pigment. ...................................... 109 Figure 3.16. Optical density of chemotrophically grown R. sphaeroides strain ΔRCLH containing plasmid pRS1 under varying concentrations of imidazole. ................................ 116 Figure 3.17. Absorbance spectra of wild type and (M)L214G RCs from cultures grown photoheterotrophically in the presence or absence of 2 mM imidazole ............................... 118 Figure 3.18. Ground state absorption spectra obtained at 10 K for (A) the wild type (WT, black), and (M)L214H mutant RCs (green), and (B) the Zn-RC (red) and Zn-β RCs (blue)................................................................................................................................................ 124 Figure 3.19. Stick models and electron density (mesh) of the HA cofactor and axial (M)214 residues from the crystal structures of the (A) Zn-RC and (B) Zn-β-RC. ............................ 129 Figure 3.20. Transient absorption difference spectra of the Qx transition of wild type (WT) (M)L214H, Zn-, and Zn-β-RCs 25 ps after excitation .......................................................... 133 Figure 3.21. Kinetics of P+IA- state decay at 675 nm for wild type (WT), (M)L214H, Zn-, and Zn-β-RCs measured up to 5500 ps ................................................................................. 136 Figure 3.22. Evolution associated difference spectra (EADS) of the Qx and Qy transitions of wild type (WT), (M)L214H, Zn-, and Zn-β-RCs. ................................................................ 139 Figure 4.1. Organization of the pigment cofactors in crystal structures of the RC ............. 148 Figure 4.2. Orientation of the BA phytyl tails in RCs for which high-resolution data are obtainable. ............................................................................................................................. 149 Figure 4.3. Evolution associated difference spectra (EADS) and kinetic models of the wild type (WT) and (M)L214G/A/C RCs. .................................................................................... 156 xiii  Figure 4.4. Overlay of the hundreds of picoseconds EADS for the (M)L214G, (M)L214A, and (M)L214C RCs............................................................................................................... 160    xiv  LIST OF ABBREVIATIONS  β mutant A mutant reaction center that assembles with BChl in place of a BPhe  μE Microeinstein ADP  Adenosine 5’ diphosphate ApR Amplicillin resistance ATP Adenosine 5’ triphosphate bchD Gene encoding the delta subunit of the magnesium chelatase enzyme BChl Bacteriochlorophyll BLAST Basic local alignment search tool Bp  Base pair BPhe Bacteriopheophytin Bx High energy absorption band of chlorins. Its absorption overlaps with the By band and collectively, these bands are called the Soret band By High energy absorption band of chlorins. Its absorption overlaps with the Bx band and collectively, these bands are called the Soret band Chl Chlorophyll DEAE Diethyl amino ethyl DNA Deoxyribonucleic acid DNase A Deoxyribonuclease A DOXP 1-deoxy-D-xylulose-5-phosphate E Escherichia EADS Evolution-associated difference spectrum EDTA Ethylenediaminetetraacetc acid FFT Fast Fourier transform xv  FPLC Fast protein liquid chromatography HPLC High pressure liquid chromatography ICM Intracytoplasmic membrane IR Infrared Kb kilobase LDAO Lauryldimethylamine-n-oxide LH Light Harvesting LH1 Light harvesting 1 LH2 Light harvesting 2 mRNA Messenger ribonucleic acid NADP Nicotinamide adenine dinucleotide phosphate NTA Nitrilotriacetic acid PCR Polymerase chain reaction PDB  Protein data bank Phe Pheophytin Puc Operon encoding the LH2 complex proteins Puf Operon encoding the RC M/L and LH 1 complex proteins puhA Gene encoding the RC H protein Qx The intermediate energy absorption band for chlorins in the visible region of the electromagnetic spectrum Qy The low energy absorption band for chlorins in the visible to near-IR region of the electromagnetic spectrum R. Rhodobacter RC Reaction center RPM Revolutions per minute SLIM Site-directed ligation independent mutagenesis xvi  TCR Tetracycline resistance Zn-BChl Zinc-bacteriochlorophyll Zn-RC A RC containing Zn-BChl in place of BChl and BPhe Zn-β-RC The Zn-RC with the (M)L214H mutation    xvii  ACKNOWLEDGEMENTS  I would like to thank my supervisor Dr. J. Thomas Beatty for allowing me to undertake graduate studies at his laboratory. Without his patience and guidance, I would have never been able to make it as far as I have. I would also like to thank him for teaching me that science is a team sport. Furthermore, I would like to thank Jeanette Beatty for being a top notch laboratory manager. Her organizational skills and attention to detail are unsurpassed, and I’m ever grateful for her effort to make the laboratory run as smoothly as possible (especially during times when us graduate students were acting like 10 year olds with ethanol squirt bottles).  My committee members, Dr. Lindsay Eltis, Dr. Michael Murphy, and Dr. Rachel Fernandez, were guiding forces throughout my entire thesis project. I thank them for their scientific insight and helpful suggestions inside and outside of my committee meetings.  I thank the group of Dr. Neal Woodbury at Arizona State University for their immense assistance with the time-resolved spectroscopy. Visiting them in Arizona was always a great learning experience. I am especially thankful to Drs. Su Lin and Jie Pan for undertaking the challenge of teaching biophysical data analysis to someone with a background in microbiology, which could not have possibly been easy for them.  I would like to thank the Murphy lab, especially Amelia Hardjasa, for providing the laboratory resources and expertise to conduct the X-ray experiments. Several publications arising from this project would not have been possible without her and Michael’s hard work.  Dr. Fred Rosell has been incredibly helpful in aiding with the low temperature spectroscopy experiments, editing my manuscripts, as well as providing insightful discussion on absorption spectroscopy.  I am grateful to the laboratory of Dr. Julian Davies for providing me with coffee at 10 in the morning every day. Some days, that 10 AM coffee was all I looked forward to. My lab mates, especially Cedric Brimacombe, Daniel Jun, Alexander Westbye, and Hao Ding, have been a great source of insightful and not-so-insightful discussions on life, the universe, and everything. I also thank them for taking me on numerous hiking experiences in the past years. I thank Drs. Sheldon Greer and Paul Schiller at the University of Miami for first exposing me to research at the undergraduate level and inspiring me to pursue a higher level of education in the field.  I would also like to thank my parents Rafael and Rosa Saer, as well as my brother Diego, for stealing me to Miami twice a year and giving me encouragement on my research project.  1  1. INTRODUCTION  1.1 Definitions and types of photosynthesis. Photosynthesis is a fundamental biological process which has shaped the atmospheric composition of Earth, as well as provided the vast amounts of food and energy resources that our modern civilization relies upon. The word “photosynthesis” has its roots in 1893, when Charles Barnes sought a term to distinguish the assimilation of carbon dioxide in plants from the metabolic processes of animals (such as eating food). Before this, both the processes were united under the term “assimilation” (Gest 2005). Barnes proposed two alternative words for the “assimilation” of carbon, as performed by higher plants: photosyntax and photosynthesis. Although Barnes preferred the former term, the latter was more widely adopted by the rest of the scientific community, and thus became established in the literature. While the word photosynthesis has its historical roots in carbon assimilation, in the 1950s it was discovered that the true nature of a photosynthetic organism is the ability to conduct photophosphorylation: the production of ATP from light energy. More recently, Howard Gest proposed a revised definition of what constitutes a photosynthetic organism by suggesting that such an organism derives a “major fraction” of their energy for cellular syntheses from light (Gest 1993). The reason for this was to exclude several genera of so-called quasi-photosynthetic bacteria, which can produce photosynthetic pigments, yet require oxygen for growth using oxidative phosphorylation (respiration). Furthermore, this revised definition excludes bacteria which can generate ATP via light-driven proton pumps called bacteriorhodopsins, as these bacteria are unable to support their own growth using these proton pumps as the sole source of ATP. 2  Nature was quick to exploit the immense and virtually inexhaustible source of solar energy (currently, ~3 x 1024 J/year (Bolton and Hall 1979)) that bombards Earth on a daily basis, as evidenced by the proximity of the evolution of photosynthesis to the origin of life on our planet. Photosynthesis is thought to have originated between 3.7 and 2.4 billion years ago. Our planet is estimated to be 4.54 billion years old, with the earliest biomarkers of life appearing in banded iron formations in rocks dated to be 3.8 billion years old (Dalrymple 2001; Bjorn and Govindjee 2009; Dauphas et al. 2004). When coupled to the consumption of inorganic carbon from the oceans and atmosphere, photosynthesis allows our planet to sustain the 600-1000 gigatons of living biomass that exists today (Falkowski et al. 2000). Photosynthesis has branched over evolutionary timescales into two major forms. Oxygenic photosynthesis is the form that is performed by plants, algae and cyanobacteria. These organisms are commonly grouped as photoautotrophs, because they can produce organic carbon compounds from inorganic carbon (i.e., carbon dioxide). This form of photosynthesis, which is thought to be responsible for the oxygenation of Earth’s atmosphere as early as 3 billion years ago (Crowe et al. 2013), utilizes two separate photosystems to catalyze the light-driven coupling of electron transfer to proton translocation across a membrane. The terminal electron acceptor is NADP+, a molecule whose reduced form, NADPH, is used in a variety of biochemical reactions in the cell as a source of reducing power.  As its name states, this form of photosynthesis evolves oxygen from the splitting of water as a source of electrons. The produced NADPH is subsequently used in the Calvin–Benson-Bassham cycle to produce fixed (organic) carbon compounds, such as sugars, and potentially all major components of a cell, and storage compounds such as starch (Blankenship 2002). 3  In contrast to oxygenic photosynthesis, anoxygenic photosynthesis does not result in the production of oxygen. This form of photosynthesis is thought to have preceded the oxygenic form due to the relative simplicity of the photosynthetic machinery, as well as the presence of the pigment bacteriochlorophyll (BChl) a in these systems. BChl a was proposed to have been ancestral to chlorophyll (Chl) a (Xiong et al. 2000; Hohmann-Marriott and Blankenship 2011). Organisms that perform anoxygenic photosynthesis may be photoautotrophic, but many are photoheterotrophs and are capable of using organic carbon compounds as the source of cell carbon. In anoxygenic photosynthesis, the captured light energy is used to drive electron transfer coupled to proton translocation across the inner membrane of photosynthetic bacteria. The resultant proton gradient gives rise to a proton motive force that is used to generate ATP. Due to its relative simplicity, the anoxygenic photosynthetic machinery of photosynthetic bacteria has been widely studied in photosynthesis research to date, ranging from a purely fundamental understanding of the underlying mechanisms to applied research in light-driven energy generation for photovoltaic systems (Moser et al. 1992; Mahmoudzadeh et al. 2011). Furthermore, the close association of photosynthesis with the beginning of life on Earth makes studies of photosynthetic organisms useful for understanding how life began on this planet.  Bacterial photosynthesis is best understood in the purple bacteria, most of which are members of the α proteobacteria. These organisms have been the subject of a wealth of photosynthesis research in the decades since they were first characterized by Hans Molisch (Molisch 1907), and later in an extensive report by C. B. van Niel (van Niel 1944). Of these organisms, Rhodobacter sphaeroides is the most well-characterized in terms of the structure and function of the photosystem. In addition to being amenable to study using almost all 4  forms of spectroscopy, this organism features a sequenced (and recently revised with new annotations some errors corrected) genome (Mackenzie et al. 2001; Kontur et al. 2012), high resolution crystal structures of its photosynthetic apparatus (Allen et al. 1987b, a; Yeates et al. 1987), as well as a repertoire of genetic tools for both genomic and in trans manipulations. R. sphaeroides is also easily cultured, being capable of growing aerobically in darkness (a trait that some anoxygenic photosynthetic bacteria lack) and anaerobically in the absence illumination, as well as by use of photosynthesis. 1.2 Pigments of photosynthetic bacteria. The pigments utilized by photosynthetic bacteria to harvest light energy are of two main types, chlorins and carotenoids, and vary between different species. Representative structures of these pigments are shown in Figure 1.1.   In R. sphaeroides, BChl a is the most abundant pigment. This molecule is similar to plant Chl a in that it consists of a large tetrapyrrole macrocycle containing an additional "isocyclic" 5th ring, as well as a hydrophobic phytyl tail that makes this pigment soluble in the cell membrane and a variety of organic solvents. Two modifications, a reduction of a double bond on ring B and the presence of a 3’ acetyl group instead of a vinyl group on ring A, distinguish this pigment from the Chl a of higher plants (Scheer 2006). Such modifications result in a shifting of the pigment’s absorption spectrum from the visible region (where Chl a absorbs) to the far red/near infrared region. This makes BChl a a useful pigment for photoheterotrophic bacteria, as it allows them to avoid competition (with plants and cyanobacteria) for photons by using wavelengths of light that are not absorbed by Chl-containing species such as plants and cyanobacteria. BChl a is functional as both a light-harvesting antenna pigment, as well as a redox-active cofactor in the capture of light energy. 5  The magnesium (Mg2+) at the center of the all BChl macrocycles is bound to the four pyrrole nitrogen atoms from each surrounding ring. In addition, an exogenous fifth ligand must be present to stabilize this complex, as the Mg2+ of BChl is never found in a 4-coordinate state. In the photosynthetic pigment-protein complexes of R. sphaeroides, this ligand is almost always an axial histidine residue, although in other systems, other ligands such as asparagine, glutamine, water, certain organic solvents, and even other BChl molecules can provide a fifth coordinate (Heimdal et al. 2007; Evans and Katz 1975). Furthermore, mutagenic studies on BChl in protein complexes, in addition to studies of free BChl in organic solvents, have shown that it is possible for BChl to exist in a 6-coordinate state, with a second axial ligand coordinating the Mg2+ from the face of the macrocycle opposite to the fifth coordinate (Frolov et al. 2010; Evans and Katz 1975).   Bacteriopheophytin a (BPhe a) differs from BChl a by the absence of the central Mg2+. The loss of this ion makes BPhe a more hydrophobic and electronegative molecule compared to BChl a (Noy et al. 1998). This gives the pigment properties relevant to electron transfer reactions in the capture of energy from sunlight. BPhe can be easily prepared by acid treatment of BChl, however the in vivo biosynthetic pathway to BPhe is not completely known. Enzymatic studies on the late stages of BChl biosynthesis indicate that BPhe is synthesized from BChl, because Mg2+-lacking analogues of BChl synthesis intermediates are not good substrates for their respective enzymes in the BChl biosynthetic pathway (Oster et al. 1997).  Carotenoids are a class of pigments characterized by a long, conjugated series of carbon-carbon bonds with a variety of different functional groups located on the ends of the molecule. These pigments are all synthesized from the 5-carbon precursor molecule isoprene, 6  which can be formed via the mevalonate pathway, or the more recently discovered 1-deoxy-D-xylulose-5-phosphate, or DOXP pathway. Both pathways seem to be employed by purple bacteria, albeit without any sort of obvious taxonomic pattern (Rohmer 1999). In bacterial photosynthesis, carotenoids play an important role in light-harvesting, as well as protection from photo-oxidative stress (Griffiths et al. 1955; Cogdell and Frank 1987). Most of the carotenoids produced by anoxygenic photoheterotrophic organisms differ from those of higher plants, algae, fungi, animals, and non-photosynthetic bacteria (Takaichi 2008). R. sphaeroides primarily synthesizes two carotenoids, sphaeroidene and sphaeroidenone, with sphaeroidene synthesized under anaerobic conditions. In the presence of oxygen, the enzyme sphaeroidene monooxygenase adds a keto group to the C2 carbon of sphaeroidene, generating sphaeroidenone (Takaichi 2004).   7    Figure 1.1. Chemical structures of the pigments from the photosynthetic apparatus of R. sphaeroides. Note the phytyl tails of BChl and BPhe are displayed as a separate molecule in the free alcohol form. The rings of the BChl and BPhe macrocycles are labeled according to the IUPAC convention.     8  1.3 The photosynthetic apparatus of R. sphaeroides. Despite the inherent photoactivity of photosynthetic pigments, their diverse functionality is ultimately achieved in the context of pigment molecules in an ordered structure. Nature accomplishes this by arranging the pigments in a protein scaffold. In R. sphaeroides, the photosynthetic machinery is composed of several proteins and pigment-protein complexes that work in concert to efficiently capture light energy in the visible and near infrared region, ultimately generating ATP. These proteins consist of two light-harvesting (LH) antenna complexes, a reaction center (RC) protein, a cytochrome bc1 protein complex, and an F0F1 ATP synthase protein. All of these proteins are present as integral membrane proteins in the inner membrane, or water-soluble proteins in the periplasm. A schematic of these proteins and their functions is illustrated in Figure 1.2.    9    Figure 1.2. A schematic of photosynthesis in purple bacteria. Note: the RC/LH1 core complex is shown as a monomer for simplicity. Abbreviations: ADP = adenosine diphosphate; ATP = adenosine triphosphate; cyt = cytochrome; ox = oxidized; red = reduced; hv = light energy.   10  The LH complexes are pigment-protein complexes that capture light energy and funnel this excitation energy to the RC. In R. sphaeroides, two types of LH are present. Light-harvesting 2, or LH2, is the most abundant antenna structure during photosynthetic growth under low light intensity. LH2 consists of nine heterodimeric α/β protein subunits arranged in a membrane-spanning ring structure, with each subunit containing 3 molecules of BChl, and one carotenoid, for a total of 36 components (Sturgis and Niederman 2008; Strumpfer and Schulten 2009; Scheuring et al. 2003). This structure is not associated with the RC and exists as its own entity in the photosynthetic membrane. In fact, the regulation of LH2 expression at the genetic level differs from the RC and the light-harvesting 1 complex (LH1) and, the ratio of LH2 to LH1/RC in R. sphaeroides and other photosynthetic bacteria grown under different conditions is highly variable. Previous studies have shown that this variability may be attributed to both changes in light intensity, as well as the stage of acclimation from aerobic to photosynthetic growth. Specifically, the LH2 complex is maximally expressed (relative to LHI) under conditions of low light intensity, as well as during the later stages of acclimation to photosynthetic growth and/or O2-limited chemotrophic growth in the dark (Woronowicz et al. 2011; Woronowicz et al. 2012).The BChls associated with LH2 absorb light maximally at wavelengths of 800 and 850 nm. The LH1 complex is a larger antenna complex than LH2, although both LH complexes are multimeric ring structures composed of membrane-spanning α/β helical subunits. Unlike the LH2 complex, however, LH1 is genetically regulated in an identical manner to the RC because the genes coding for both of LH1 and the RC are part of the same puf operon (Choudhary and Kaplan 2000). The LH1 and RC complexes come together to form a “core complex”, which in R. sphaeroides is dimeric in structure, with two RCs 11  partially encircled by 28 LH1 α/β polypeptide subunits and a small polypeptide called PufX (Qian et al. 2005; Qian et al. 2013). LH1 contains two molecules of BChl in every α/β subunit, which absorb infrared light maximally at 875 nm. The role of this complex is to capture light energy, as well as relay excitation energy captured by LH2 to the RC.  The core complex, in combination with LH2 constitute the vast majority of the photosynthetic machinery in R. sphaeroides. These complexes are located in spherical vesicles within the cell called intracytoplasmic membranes (ICMs), which are formed by invagination of the cytoplasmic membrane, perhaps due to electrostatic interactions of the LH2 and LH1 polypeptides (Chandler et al. 2009). These ICM vesicles were proposed to be either attached to the cytoplasmic membrane as buds, or “free floating” cytoplasmic entities released from the cytoplasmic membrane, functioning analogously to a eukaryotic organelle (Tucker et al. 2010).  Atomic force microscopy on ICMs revealed that these structures do not consist of randomly interspersed LH2 and core complexes, but rather clusters of core complexes appear to be arranged in stacks, surrounded by “lakes” of LH2 (Bahatyrova et al. 2004). To date, only the core complexes and LH2 have been resolved by imaging techniques, and thus the localization of the other proteins of the photosynthetic apparatus (the ATPase and the cytochrome bc1 complex) remains unknown.  The process of photosynthesis may be thought of as initiating with the absorption of light energy, usually by a photoactive pigment in an LH complex. Absorption of light energy promotes the pigment systems in the LH complex to an excited state. This excitation energy is then transferred to the RC, which can store this energy in the form of charge separation. 12  This is an extremely important reaction as it marks the point at which the light energy is trapped by conversion to chemical energy. The R. sphaeroides RC converts energy conversion via a network of pigments that catalyze a series of redox reactions, embedded in the protein, ultimately leading to the reduction of a diffusible quinone that functions as a proton/electron shuttle. This quinone diffuses laterally in the ICM, and docks with the cytochrome b/c1 complex. The role of this complex is twofold. First, it oxidizes the quinol molecule back to a quinone, translocating the released protons from the cytoplasm into the periplasmic space (or the interior of the ICM in the case of a free floating vesicle). This accumulation of protons in the periplasm allows the ATPase enzyme to generate ATP from ADP, by utilizing the potential energy of the generated ion gradient. The second function of the cytochrome b/c1 complex is to reduce a periplasmic c-type cytochrome (cytochrome c2 in R. sphaeroides). This cytochrome serves to reduce the RC, which has become oxidized from the conversion of light energy to chemical energy. In addition, the cytochrome b/c1 complex, in a process called the Q cycle, reduces a quinone to additionally contribute to uptake of protons from the cytoplasm (Fig. 1.2). 1.4 Reaction center types, structure and function. RCs can be grouped into two main types: type I and type II. Type I RCs are easily identified by the presence of three redox-active iron-sulfur clusters of the [4Fe-4S] type, which act as electron transfer intermediates. The protein component of these RCs may be either heterodimeric, as in photosystem I from plants and cyanobacteria, or homodimeric, as in the RCs from the Heliobacteriaceae, Chlorobiaceae, and candidatus Chloroacidobacterium thermophilum (Liebl et al. 1993; Buttner et al. 1992; Bryant et al. 2007). The role of these RCs is to generate a strong 13  reductant capable of reducing NADP+ to NADPH. Because of this, they operate at very negative redox potentials (~ -1200 to -540 mV for photosystem I (Brettel 1997)).   Type II RCs are heterodimeric in structure and lack an iron-sulfur center. These RCs are found in cyanobacteria and higher plants (photosystem II is a type II RC), as well as in all purple photosynthetic bacteria. Instead of Fe-S clusters, type II RCs contain quinones as electron transfer intermediates. Since quinones are two-electron shuttles, a type II RC undergoes two turnovers before the quinone is fully reduced into a quinol state, at which point this molecule diffuses laterally through the membrane, away from the RC. Type II RCs generally operate at higher redox potentials than type I RCs (~ -580 to -80 mV for photosystem II, and ~ -1000 to ~ -80 mV for the R. sphaeroides RC), which, in the case of photosystem II, also create a powerful oxidant capable of splitting a water molecule (in concert with the oxygen-evolving complex (Grabolle and Dau 2005; Blankenship 2002)). Also present in these RCs are (B)Phe molecules, which are electron transfer intermediates that apparently have been selected by nature to facilitate charge transfer between (B)Chls and quinones.  The first high-resolution 3-D structure of a membrane protein was of the type II RC from Blastochloris viridis (then Rhodopseudomonas viridis), and consisted of 4 proteins -- 3 RC proteins and a bound, tetraheme cytochrome (Deisenhofer et al. 1985). Subsequently the RC of R. sphaeroides was crystallized and the structure solved, showing that this complex differs from the B. viridis RC by being composed of only 3 protein subunits, homologues of the B. viridis RC proteins and termed L, M, and H (Allen et al. 1987b). Apart from the tetraheme cytochrome the two structures are similar, and predicted amino acid sequences of RC proteins from a wide variety of purple photosynthetic bacteria indicate little variation 14  from the B. viridis/R. sphaeroides paradigm. This view was confirmed by the crystal structure of the RC from the moderate thermophile Thermochromatium tepidum (Nogi et al. 2000). The L and M proteins are structurally symmetrical, sharing an axis of symmetry orthogonal to the plane of the membrane bilayer, and contain 5 transmembrane helices that house the cofactors buried inside. The cofactors consist of two excitonically coupled BChls (PA and PB; together, often called P), two accessory BChls (BA and BB), two molecules of BPhe (HA and HB), a non-heme Fe2+ ion (Boso et al. 1981), and two molecules of ubiquinone 10 (QA and QB). The overall structural arrangement of the RC is shown in Figure 1.3. The pigments of the RC are organized into two structurally symmetrical, but functionally asymmetrical branches, termed the A (active) and B (inactive) branches. The P BChls are located on the periplasmic side of the protein, and the quinones are located on the cytoplasmic side. Between P and QA, BA and HA serve as electron transfer mediators that also assist in extending the longevity of the charge-separated state, by creating a large physical distance between oxidized P and reduced QA. These cofactors are arranged such that when P is promoted to an excited state, electron transfer to an accessory pigment is the most rapid decay mechanism. The electron transfer cascade, as well as the approximate electron transfer rates, are shown in Figure 1.4. Ultrafast visible and IR spectroscopic studies of the RC indicate that the primary electron acceptor is BA, which is reduced on a time scale of ~3 ps (Khatypov et al. 2012; Pawlowicz et al. 2008). This P+BA- radical then reduces HA within 1 ps to generate the P+HA- state. The HA- radical anion reduces QA in ~200 ps followed by electron transfer to the QB quinone in a reaction (~200-300 µs) that is coupled to protonation of QB to produce an electronically neutral ubisemiquinone radical (Paddock et al. 2003). The 15  non-heme Fe2+ ion is located between the two quinones (not shown in Figure 1.3), and is presumed to play a role in mediating electron transfer from HA to QA by tuning the electronic properties of the quinone.(Ishikita and Knapp 2006; Kirmaier et al. 1986). A mobile cytochrome c reduces the oxidized P+, and the RC is set for a second turnover in which the QB ubisemiquinone is fully reduced to a quinol. This quinol is released from the RC and is oxidized at the cytochrome b/c1 complex.    16    Figure 1.3. The three dimensional structure of the RC from PDB 2J8C. Left: The protein component surrounding the photoactive pigment cofactors. The L subunit is colored red, the M subunit is colored gray, and the H subunit is colored orange. Right: The embedded cofactors shown with the protein component removed. The two QB molecules present in the figure represent the different binding states of QB in the RC. The black circle denotes the non-heme iron atom. The shorter QB tail (relative to QA) is a result of this specific RC being reconstituted at the QB site with ubiquinone-2.   17     Figure 1.4. Electron transfer time constants within the RC of R. sphaeroides. The grey arrows and numbers indicate the direction and time constants of the back reaction. The cofactors that are not directly involved in electron transfer are not shown. Also missing from this diagram is the non-heme iron.       18  In experiments intended to elucidate the structure-function relationship of the RC, this pigment-protein complex has been subject to a vast degree of mutagenic and spectroscopic analysis, leading to the understanding that this protein exhibits a remarkable tolerance to changes in both the protein and pigment composition (Jones 2008). A key discovery in RC research was the ability to alter the pigment composition of the RC by the use of site-directed mutagenesis. Examples include the swapping of photoactive BPhe with BChl, in the HA binding pocket, as well as the reverse exchange in both the P and BA pockets (Bylina and Youvan 1988; Kirmaier et al. 1988; Nabedryk et al. 2000; Katilius et al. 2004; Kirmaier et al. 1991; Heller et al. 1995). RCs containing a molecule of BChl in place of BPhe at HA are called “beta” mutants, and the HA BChl molecule is commonly referred to as the “β” pigment. In addition to mutagenesis, the pigments of the RC can also be exchanged through biochemical means, allowing for the incorporation of synthetic or chemically modified chromophores into the RC (Scheer and Hartwich 1995). Such experiments have provided insights into the nature of biological electron transfer, as well as an understanding of the requirements of pigment binding to the protein scaffold.  1.5 Interactions between the chlorins and the protein component of the RC. The crystal structure of the R. sphaeroides RC coupled with site-directed mutagenesis experiments have greatly contributed to our understanding of the interactions present between the photoactive cofactors and the protein component. Several pigment-protein interactions assist in fixing the pigments to the protein matrix, as well as tuning the electronic properties of the photoactive pigments in the RC. Figure 1.5 shows the known binding interactions of the BChl dimer (P) with the surrounding protein, to give an example of the different kinds of protein-cofactor interactions available in the RC. The P BChls are oriented 19  nearly perpendicular to the plane of the membrane, with their A rings overlapping. The magnesium ions of the PA and PB BChls are coordinated by histidine residues (L)173 and (M)202, respectively. An additional hydrogen bond between the ring A acetyl group of PA and His (L)168 serves to raise the midpoint potential of the dimer for optimal electron transfer (Mattioli et al. 1994; Lin et al. 1994). The BA monomer BChl contains no such protein-macrocycle hydrogen bonding, however this BChl Mg2+ ion is pentacoordinated by His (L)153 (Yeates et al. 1988). The interaction between BA and His (L)153 is shown in Figure 1.6A. The HA cofactor, unlike the previously mentioned BChls, contains no axial histidine ligand, and no Mg2+ is present in the center of the macrocycle. A leucine residue at position (M)214 resides in an analogous axial position, however (Figure 1.6B). A nearby glutamate residue, Glu (L)104, forms a hydrogen bond with the keto carbonyl of ring E (Figure 1.6B), giving rise to a change in the absorbance properties of this cofactor relative to HB (Yeates et al. 1988). The presence of a BPhe molecule at HA, in addition to the hydrogen bond by Glu (L)104, ensures a high yield of the P+QA- state during electron transfer. Yields as low as 27% have been reported in a mutant RC in which Leu (M)214 is changed to histidine (resulting in a molecule of BChl at HA) and Glu (L)104 is changed to valine (removing the hydrogen bond from the ring E carbonyl group of HA (Kirmaier et al. 1995b)). 1.6 Analysis of RCs (RC pigments) by optical absorption spectroscopy. Optical absorption spectroscopy has been an invaluable tool in the study of all photosynthetic systems. The photoactive nature of photosynthetic pigments gives them characteristic spectra which reveal a wealth of information about the pigment’s properties in a non-destructive manner. The use of absorption spectroscopy in the study of bacterial photosynthesis dates 20  back to the middle of the twentieth century, when Duysens began to characterize purple bacteria using the techniques of absorption and fluorescence spectroscopy (Duysens 1951). Interactions between pigments and other molecules (such as solvents, other pigments, polypeptides, etc.) give rise to changes, or shifts, in their absorption spectrum. This is especially evident in the light-harvesting complexes of R. sphaeroides, where the absorption bands of BChl are shifted over 100 nm to the red as a result of their spatial arrangement.   21     Figure 1.5. Bonding interactions of the BChl dimer with the surrounding protein. Histidines (M)202 and (L)173 act as fifth coordinates to the PA and PB, respectively, while his (L)168 donates a hydrogen bond to the ring A acetyl group of PA. Ligating nitrogen atoms on histidines are shown in blue. Histidine residues are otherwise colored red. Carbon and oxygen atoms on the BChls are colored green and red, respectively. The yellow dashed lines denote ligation between the BChl nitrogen atoms and the central Mg2+ (green pluses). The red dashed lines denote ligation between histidine nitrogen atoms and the BChl Mg2+ or carbonyl oxygen atoms. RC structural data was obtained from PDB 2J8C (Koepke et al. 2007).   22   Figure 1.6. Bonding interactions of the BA and HA cofactors with the surrounding protein. (A) Histidine (L)153 (red) acts as the fifth ligand to the Mg2+ of BA through one of its nitrogen atoms (blue). (B) At HA, Glu (L)104 (blue) donates a hydrogen bond to the carbonyl group on ring E of the HA macrocycle. Leu (M)214 (yellow) projects towards the center of the HA macrocycle. The dashed lines denote ligation or hydrogen bonding interactions. RC structural data was obtained from PDB file 2J8C (Koepke et al. 2007).   23  All Chls and BChls contain four principle absorption components which represent the physically allowed π -> π* transitions that give rise to absorption bands in the visible and near IR region. The transitions are grouped along two nearly perpendicular axes that run along the plane of the chlorin macrocycle. These axes are denoted X (running from ring D to B), and Y (running from ring A to C), and are displayed in Figure 1.7. The two transitions for each axis are denoted B and Q, which carry the axis label as a subscript. The stronger-absorbing Bx and By bands normally overlap in chlorin type of pigments, and thus are collectively referred to as the Soret band. For BChl a, this is a large heterogeneous absorption band with a maximum around 360 nm. The other two transitions, Qx and Qy, give rise to longer wavelength absorption bands at 590 nm and 770 nm, respectively. The Qx and Qy transitions have been especially useful in characterizing the environment of BChls because of their sensitivities to the surrounding environment. The RC is a case in point, as the Qx and Qy bands from the absorption spectrum of the RC can (almost) be resolved for each individual chlorin, at cryogenic temperatures where the absorption bands become sharper.  The absorption spectrum of the R. sphaeroides WT RC at 10 K is shown in Figure 1.8. This spectrum contains seven characteristic peaks (as well as shoulders) in the 250 to 950 nm range. Like all proteins, the RC has a strong absorbance band at 280 nm, which mainly represents contributions from aromatic amino acids, especially tryptophan, which has the strongest absorbance. The Soret band of all chlorins present (BChl and BPhe) is located ~80 nm to the red of the protein band. Most spectroscopic analysis of RCs and the corresponding literature, however, focus on the Qx and Qy transitions. The Qx transitions of the R. sphaeroides RC at 10 K are represented by three resolved peaks. The two bands at 24  ~535 and ~545 nm represent contributions from the QX transition of HB and HA, respectively. The overlapping peaks at ~590 to 605 nm represent contributions from all BChls, with the 590 nm peak assigned to the BA,B BChls, whereas the 605 nm shoulder is assigned to a combination of BB and PA,B (Shochat et al. 1994; Kirmaier et al. 1985). The carotenoid sphaeroidenone (the predominant carotenoid when cultures grow in the presence of O2) also absorbs in this region of the spectrum. The absorbance of this pigment is broadband and ill-defined in the region from approximately 500 nm to 650 nm.   25    Figure 1.7. The two molecular axes of BChl a. The dashed grey arrows depict the X and Y axes of the BChl molecule, respectively (Scheer 2006). Note: the positioning of electronic transition dipoles do not run exactly along the molecular axes because the angle of the transition dipole relative to the molecular axis is sensitive to the surrounding environment.    26   The Qy transition region contains three characteristic peaks, all in the far red to near IR region of the spectrum. The 760 nm peak is composed of contributions from HA,B, with the HB peak on the blue side and the HA peak on the red side. The BA,B accessory BChls contribute to the large, overlapping ~ 802 nm peaks, with the red shoulder representative of BB. Finally, the peak at 890 nm is assigned to PA,B. The position of this peak is temperature-sensitive, and absorbs nominally at 865 nm when an absorption spectrum is recorded at room temperature. For this reason, PA,B is frequently referred to as P865.    27     Figure 1.8. Ground state absorption spectrum of the wild type R. sphaeroides RC taken at 10 K. The labels in bold indicate the transition dipole of the photoactive pigments that give rise to the absorbance peaks. Terms that are not in bold indicate the specific cofactors that contribute to the amplitude of the peak. Assignments of peaks and shoulders were obtained from Jones and Kirmaier et al (Jones 2009; Kirmaier et al. 1985).     28  1.7 The protein environment around HA, and the origin of BPhe. As mentioned in section 1.5, the protein backbone of the RC is responsible for tuning the electronic properties of the embedded pigments, as well as providing a fifth coordinate to the Mg2+ central ion of the BChls. In addition, the protein, in combination with the long hydrophobic tails of the pigments, serve to create a well-insulated environment in which electrons tunnel from pigment to pigment with little risk of a side reaction that would result in a decrease in the efficiency of energy transduction. Studies aimed at understanding this structure-function relationship between the pigments and the protein component are useful in developing a knowledge base on biological electron transfer reactions. Such reactions could prove useful in future engineering applications that capitalize on electron transfer reactions using biological, as opposed to synthetic, systems. In addition, an understanding of the mechanisms governing biological electron transfer within, and to and from different types of RCs provides insight into the evolution of such systems. Because an attempt to obtain a comprehensive understanding of all electron transfer reactions in the RC would be beyond the scope of a single PhD thesis, I focus on the electron transfer events which involve the BPhe cofactor.  The use of (B)Phe in native type II RCs is critical to ensure a successful electron transfer event to a quinone molecule of much lower free energy. The role of this molecule in the electron transfer cascade is twofold. First, it creates an anion radical intermediate that is physically further away from the oxidized donor molecule (P865), which greatly decreases the rate of the back reaction. Second, the (B)Phe is energetically distant from the (B)Chl monomer (BA in R. sphaeroides), such that the (B)Chl and (B)Phe anionic states do not exist in equilibrium, ensuring that the large distance between the oxidized donor and its electron is 29  maintained. In the R. sphaeroides RC, the lack of an axial ligand in the HA,B pockets is crucial to the incorporation of this pigment into the RC. Instead of a ligand, a Leu residue is positioned axial to the BPhes in both the A and B branches at the (M)214 and (L)185 positions, respectively. The fact that BChl is found in the HA pocket when (M)L214 is changed to His suggests that BPhe incorporates into the RC by one of two mechanisms: (i) the axial Leu residue contributes to a BPhe binding pocket selective for an already synthesized molecule of BPhe; or (ii) the axial Leu residue physically assists in the de-chelation of Mg2+ from a BChl molecule, perhaps during RC assembly, resulting in BPhe. It is difficult to distinguish between these two possibilities for several reasons. The first is that, apart from the RC itself, BPhe is not detectable in the chromatophore membrane. Second, no Mg-dechelatase enzyme has been discovered in purple bacteria to date, although a heat-stable Mg-dechelating substance has been purified in senescent plant leaves, in addition to a Mg-dechelation activity by the magnesium chelatase enzyme itself during the greening of plant leaves (Shioi et al. 1996; Ignatov and Litvin 1994).  It was thought that RC-mediated dechelation could occur in one of two stages of RC complex assembly, depicted in Figure 1.9. The first is during the assembly of the apo-complex, where (B)Chls are thought to initially bind the protein and, in studies conducted on plant photosystems I and II, assist in the proper folding of the protein (Mullet et al. 1990; Eichacker et al. 1996; Muller and Eichacker 1999). In purple bacteria, it is not known how the photosynthetic machinery is translocated into the bacterial inner membrane, however evidence from cyanobacterial photosystems reveals the requirement for the SecYEG translocation complex, in addition to a protein called YidC (Sobotka 2013). Assuming binding of (B)Chl to the protein at this stage of protein folding, prior to complex assembly, 30  the Mg2+ ion may depart from the (B)Chl macrocycle unless an axial ligand is present. In the second stage of complex assembly, the independently stable transmembrane domains of the RC are formed, and the final structure of the protein begins to take shape via interactions between these helices in combination with the exclusion of lipids between domains (Popot and Engelman 2000). At this point in assembly, the bulky Leu side chains found orthogonal to the macrocycle plane of (B)Phes in the RC may act as a steric obstruction to the Mg2+ ion, resulting in its removal from the (B)Chl molecule.    31     Figure 1.9. Two stages of integral membrane protein development. In stage I, the nascent polypeptide forms stable, but disperse transmembrane domains in the lipid bilayer. In stage II, the domains begin to interact, forming the final protein structure with exclusion of the lipid component between helical domains. Figure reprinted with permission from Popot and Engelman (2000).    32  1.8 Thesis aims and approach. The goal of this work is multifaceted, in part because certain results of early experiments shed light on and led to other experiments. In one instance, I aimed to provide further insight into the origin of BPhe as it appears in the RC of R. sphaeroides, using a combination of in vivo, genetic, and biochemical techniques. The RC-mediated dechelation hypothesis (described above in Section 1.7) was tested. The work in this section focuses on the use of a series of nonpolar “cavity” mutants in which the Leu at position 214 on the M subunit of the RC was substituted with smaller, nonpolar moieties. The rationale behind such changes was that these moieties cannot act as a fifth coordinate to a BChl molecule, nor can they act as a steric impediment to the Mg2+ center of a BChl molecule during RC assembly. The binding criteria of HA was further investigated using a series of mutant RCs in which polar amino acid residues (Gln, His, Asn) were introduced at this site. It was found that RCs with amide-containing side chains at position (M)214, were also capable of assembling with BChl at HA. Because of this, I also examined the spectroscopic differences between these RCs, as the amide-containing mutants used an oxygen atom for ligation (compared to a nitrogen in and Leu -> His mutant), and the BChl-ligand distances differed between the amide-containing mutants. The RCs described above were studied in the context of whole cells during periods of photosynthetic anaerobic growth, as well as studies of isolated protein using absorption spectroscopy, biochemical pigment substitution methods, and X-ray crystallography.   The results derived from spectroscopic analysis of RCs containing small, nonpolar moieties at (M)214 led me to pursue a functional investigation of these proteins. In this arm of my research, I collaborated to investigate the performance of these RCs through ultrafast spectroscopic techniques. The results obtained allowed me to propose explanations of 33  changes in electron transfer rates due to structural perturbations that resulted from engineered mutations at (M)214. These data bring new insight into the means by which RCs generate charge separation with extremely high efficiency.   In a subsequent section of my research, I employed the same techniques used to study the small, nonpolar mutant RC series on a different RC that assembles entirely with Zn-BChl in place of both BChl and BPhe. These RCs were originally characterized in a Mg-chelatase (bchD) mutant of R. sphaeroides (Jaschke and Beatty 2007). Our ultrafast spectroscopic studies on these proteins have enabled the discrimination of the effects on electron transfer that can be ascribed to the type of central metal ion, versus the presence of a fifth coordinate at HA. The crystal structures of these RCs are also reported, revealing the structural consequences of assembling RCs of wild type amino acid sequence but unnatural pigment composition.   Collectively, these data further the understanding of the structure-function relationship of photosynthetic RCs. The molecular mechanisms relating RC structure and electron transfer rates and pathways are extremely complex, and experiments such as the ones presented in this work may aid in the creation of more refined and accurate models to describe these phenomena. Finally, in the process of optimizing the preparation of sufficient material for these experiments, several new strains, plasmids, and protein isolation techniques were developed, which may aid the scientific community in research on this remarkable organism and its iconic photosystem.     34  2. MATERIALS AND METHODS  2.1 Bacterial strains and plasmids. Table 2.1 lists the different strains of Escherichia coli and R. sphaeroides used in this thesis. E. coli Strains DH5α and DH10B were used for cloning. E. coli strain HB101 (pRK2013) was used as a helper strain for mobilization of plasmids from E. coli into R. sphaeroides via tri-parental conjugation (Ditta et al. 1985).  R. sphaeroides strain ΔRCLH was used as a background strain for the in trans expression of photosynthetic RC-LH1 core complexes (Tehrani and Beatty 2004). This strain lacks the genes required for the synthesis of RC-LH1 core complexes, in addition to LH2. R. sphaeroides strain ΔpuhA was used as a negative control for photoheterotrophic growth experiments (Chen et al. 1998).   For the growth of E. coli, cultures were grown aerobically on LB medium in glass test tubes at 37 oC. LB contains 1% tryptone, 0.5% yeast extract, and 1% NaCl per liter of media (Sambrook et al. 1989). For aerobic growth of R. sphaeroides, cultures were grown in sterile glass test tubes filled with RCV medium at 30 oC (Beatty and Gest 1981). RCV medium contains per liter: D,L-malic acid, 4 g; (NH4)2SO4, 1 g; potassium phosphate buffer (pH 6.8), 10 mM; MgSO4 . 7 H2O, 200 mg; CaCl2 . 2 H2O, 75 mg; FeSO4 . 7 H2O, 12 mg; Na2 EDTA, 20 mg; trace element solutions (containing, per 250 mL of dH2O: H3BO3, 0.7 g; MnSO4 . H2O, 398 mg; Na2MoO4 . 2 H2O, 188 mg; ZnSO4 . 7 H2O, 60 mg; Cu(NO3)2 . 3 H2O, 10 mg), 1 mL; thiamine hydrochloride, 1 mg; biotin, 15 μg; niacin, 1 mg. For semi-aerobic growth, cultures were grown in the dark in Erlenmeyer flasks filled to 70% nominal capacity with RLB medium. RLB medium contains per liter: 10% tryptone, 5% yeast extract, and 0.23% NaCl, in addition to MgCl2 and CaCl2 at final concentrations of 810 μM and 510 μM, 35  respectively (Jun et al. 2013). Semi-aerobically grown cells were shaken at 150 RPM, instead of the 200 RPM used for aerobic growth conditions. For anaerobic, photoheterotrophic growth of R. sphaeroides, cultures were grown in 16.5 mL screw cap tubes filled completely with RCV media. These tubes were incubated at 30 oC in a temperature-controlled aquarium and illuminated with an array of incandescent light bulbs.   For plasmid selection in E. coli, media was supplemented with antibiotics at the following final concentrations: tetracycline-HCl at 10 μg/mL, kanamycin sulfate at 50 μg/mL, ampicillin at 100 μg/mL. For plasmid selection in R. sphaeroides, tetracycline was used at 2 μg/mL final concentration.    Liquid culture densities of R. sphaeroides cells were measured with a Klett-Summerson photometric colorimeter equipped with a red filter (100 Klett units ~ 3 x 108 CFU mL-1).   36  Table 2.1 Bacterial Strains Strains Relevant characteristics Source or reference E. coli   DH5α F-, Δ(lacZY A-argF) deoR, recA1, endA1 Invitrogen DH10B F-, mcrA, recA1, endA1, araΔ 139, ΔlacX74  Invitrogen HB101 pRK2013 contains RK2 transfer genes (Ditta et al. 1985) R. sphaeroides   ΔRCLH RC-, LH1-, LH2- (Tehrani and Beatty 2004) ΔpuhA RC-, LH1+, LH2+ (Chen et al. 1998) ΔpuhAΔpucΔbchD RC-, LHI+, LH2-, Mg Chelatase- (Jaschke and Beatty 2007) ΔRCLHΔbchD RC-, LHI-, LH2-, Mg Chelatase- (Neupane et al. 2012) R. capsulatus   ΔgtaIΔ1081 C16 homoserine lactone-, capsular polysaccharide- (Brimacombe et al. 2013)    37  The plasmids used in this work are listed in table 2.2. Plasmids created in this work are described in detail following this table.  Table 2.2. List of plasmids used in this study Plasmid Relevant characteristics Source or reference pATP19P Broad host range expression vector containing the hypoxia-inducible puc promoter, Tcr (Tehrani and Beatty 2004) pTZ18U Cloning vector, Apr Pharmacia pTZ18U::puhA pTZ18U containing the RC-H gene from R. sphaeroides as a 1.34 Kb BamH I fragment (Chen et al. 1998) pTZ18U::puhA T- pTZ18U::puhA with a putative Rho-independent transcriptional terminator removed between the puhA stop codon and the BamH I restriction site This study pZT18U::6 his puhA T-  pTZ18U::puhA T- containing a c-terminal 6x histidine tag on the puhA gene, in addition to the changing of the GCC codon immediately 5’ of the tag to a GTG codon, thereby introducing an ApaL I site This study pZT18U::6 his puhA T- 3’sac pZT18U::6 his puhA T- with the BamH I site downstream of the puhA gene changed to a SacI site This study p6His-C pATP19P containing a c-terminal 6x his-tagged RC-H gene as a 1.3 kb BamH I fragment    (Jaschke and Beatty 2007) 38  Plasmid Relevant characteristics Source or reference pATSHR:: 6 his T- 3’ sac pATP19P containing the 1.3 kb BamH I – Sac I fragment from pTZ18U::puhA T- 3’ sac This study pAli2::puf pAli2 containing a 4.57 kb EcoR I fragment, encoding the R. sphaeroides pufQBALMX genes (Tehrani and Beatty 2004) pAli2::puf sph- PAli2::puf with a 936 bp Sph I fragment removed from the pufQBALMX insert This study pAli2::puf sph- (M)L214G pAli2::puf sph- containing an L -> G mutation at position 214 on the pufM gene This study pAli2::puf sph- (M)L214A pAli2::puf sph- containing an L -> A mutation at position 214 on the pufM gene This study pAli2::puf sph- (M)L214C pAli2::puf sph- containing an L -> C mutation at position 214 on the pufM gene This study pAli2::puf sph- (M)L214V pAli2::puf sph- containing an L -> V mutation at position 214 on the pufM gene This study pAli2::puf sph- (M)L214I pAli2::puf sph- containing an L -> I mutation at position 214 on the pufM gene This study pAli2::puf sph- (M)L214M pAli2::puf sph- containing an L -> M mutation at position 214 on the pufM gene This study pAli2::puf sph- (M)L214N pAli2::puf sph- containing an L -> N mutation at position 214 on the pufM gene This study pAli2::puf sph- (M)L214Q pAli2::puf sph- containing an L -> Q mutation at position 214 on the pufM gene This study pAli2::puf sph- (M)L214H pAli2::puf sph- containing an L -> H mutation at position 214 on the pufM gene This study pRS1 pATSHR:: 6 his T- 3’ sac containing a 3.3 kb Sac I – EcoR I fragment from pAli2::puf sph- containing the pufQBALMX genes  This study 39  Plasmid Relevant characteristics Source or reference pRS1:: (M)L214G pRS1 containing the an L -> G mutation at position 214 on the pufM gene This study pRS1:: (M)L214A pRS1 containing the an L -> A mutation at position 214 on the pufM gene This study pRS1:: (M)L214C pRS1 containing the an L -> C mutation at position 214 on the pufM gene This study pRS1:: (M)L214V pRS1 containing the an L -> V mutation at position 214 on the pufM gene This study pRS1:: (M)L214I pRS1 containing the an L -> I   mutation at position 214 on the pufM gene This study pRS1:: (M)L214M pRS1 containing the an L -> M mutation at position 214 on the pufM gene This study pRS1:: (M)L214N pRS1 containing the an L -> N mutation at position 214 on the pufM gene This study pRS1:: (M)L214Q pRS1 containing the an L -> Q mutation at position 214 on the pufM gene This study pRS1:: (M)L214H pRS1 containing the an L -> H mutation at position 214 on the pufM gene This study       40  2.2 Cloning. Plasmid DNA smaller than 10 kb was isolated using mini spin columns (Epoch Biolabs). For plasmids greater than 10 kb (e.g. pATP19P), the spin columns did not appear to work at all. Thus, these plasmids were purified using the alkaline lysis procedure outlined in Sambrook et al., however the final cesium chloride purification step was not performed (Sambrook et al. 1989). All restriction digests were performed using restriction enzymes purchased from NEB and used according to the manufacturer’s instructions. T4 DNA ligase was also purchased from NEB and used according to the manufacturer’s instructions with a few exceptions. Room temperature ligation reactions were performed for a full hour instead of 10 minutes. For overnight reactions, ligations were carried out in an ice bucket stored in a room maintained at 4 oC. Because cloning reactions involving large fragments of DNA (> 10 kb) were prone to much failure and troubleshooting, The following procedure was used to make such constructs: 1. The vector and insert were combined into a single microcentrifuge tube at a molar ratio of ~1:5 vector to insert.  2. The combined vector and insert DNA were digested with restriction enzymes in a 20 μL reaction. 3. Following digestion, 2.5 μL of 10x ligase buffer, 1.5 μL of dH2O, and 1 μL of T4 DNA ligase were added directly to the reaction tubes. 4. 10 μL of the completed ligase reactions were used to transform CaCl2 competent E. coli. It is highly recommended that the competency of the cells be higher than 108 CFU/μg pUC19 plasmid DNA. 2.3 Modification of the puhA gene to facilitate the creation of RC expression systems and purification of the RC. Nickel-NTA affinity chromatography of the R. sphaeroides RC has been previously shown to be an effective tool in the rapid and relatively simple purification the protein compared to previous methods (Goldsmith and Boxer 1996). 41  Because the work presented in this thesis relies on the creation of many point mutations in the RC, measures were taken to ensure that the creation of RC expression systems was a streamlined process. The overall synthetic operon for RC expression is derived from a previous system used in this laboratory, and is illustrated in Figure 2.1.    42        Figure 2.1. General schematic for the design of a synthetic operon with the aim of expressing RCs in trans. The hypoxia-inducible puc promoter is denoted by the forward arrow, and the boxes represent the restriction fragments which include the puhA and puf QBALMX genes, respectively. Asterisks represent restriction sites for mobilization of genetic elements into the expression vector.     43   The first roadblock in creating this expression system was a putative rho-independent transcriptional terminator present immediately 3’ of the puhA stop codon, between the codon itself and the natural BamH I restriction site. Secondary structure predictions using the RNAfold web server (http://rna.tbi.univie.ac.at/cgi-bin/RNAfold.cgi, (Gruber et al. 2008)) of the 84 bp that constitute this region indicated the presence of a large hairpin structure present within the first 40 bp of this 3’ region, shown in Figure 2.2 (Hofacker et al. 1994; Zuker and Stiegler 1981; McCaskill 1990). Transcription termination at this point would result in a large decrease or lack of pufQBALMX transcription, thus the terminator sequence needed to be removed.    44    Figure 2.2. Predicted secondary structure of the region 3’ of the puhA stop codon. This stretch of sequence lies between the end of the coding sequence and the BamH I site. The 5’ and 3’ ends of the sequence are denoted by the arrows. The large hairpin loop occurring within the first 40 bp is shaded in red.    45   The removal of this putative hairpin terminator was accomplished via site-directed mutagenesis. The so-called SLIM method (Chiu et al. 2004)for site-directed mutagenesis was employed for two reasons: (i) the amplification reaction did not rely on a relatively low frequency event in which the flanking sequence of the region to be deleted must bind to the primer throughout each amplification cycle, and (ii) the SLIM technique is capable of exponentially amplifying the mutagenic product, resulting in higher transformation efficiencies than other site-directed mutagenesis methods which employ linear amplification. This technique relies on the use of four primers: a forward and reverse “tailed” primer pair that incorporate the desired change in addition to ~20 bp of sequence complementary to the template, and two “short” primers which only contain the complementary sequence of the “tailed” primers. A schematic of this procedure is shown in Figure 2.3 A.    46    Figure 2.3. A schematic of the SLIM protocol for site-directed mutagenesis. (A) The forward and reverse tailed primer pair (FT and RT) incorporating the desired changes (the checkered portion of the primer) are used in a PCR reaction with short primers (FS and RS) that lack these regions. The reaction produces two productive and two non-productive PCR products which are Dpn I digested, and subsequently heated and cooled to yield self-annealing, transformable nicked plasmids. (B) Application of SLIM to produce a deletion of any size. The checkered region of the tailed primers is simply a stretch of sequence flanking the region to be deleted (dotted line).  47  The mutagenesis reaction, described in detail in Chiu et al. (Chiu et al. 2004), creates four amplification products. Two of the amplification products (those created from the binding of one “tailed” and one “short” primer to the DNA template) are linear copies of the DNA template containing the desired change on either end. After Dpn I digestion of the template DNA, these two products can be repeatedly heated and cooled to give products of DNA that are capable of self-annealing, yielding a nicked, transformable plasmid which contains the desired change. In the case of a deletion (Figure 2.3 B), flanking sequence from the region to be deleted is incorporated into the tailed primers. The result is a PCR reaction in which the region to be deleted is simply excluded from the amplification. Deletion of the hairpin sequence 3’ of the puhA gene was accomplished with the primers listed in Table 2.3.    48   Table 2.3. Primers used for the deletion of the hairpin 3’ of puhA Primer name Sequence (5’ -> 3’) puhA T- SLIM FT GGATCCGATCCGCATCCGATCCGCCGCAGACGCACCCG puhA T- SLIM FS ATCCGCCGCAGACGCACCCG puhA T- SLIM RT CGGATGCGGATCGGATCCGATCAGGCGTATTCGGCCAG puhA T- SLIM RS GATCAGGCGTATTCGGCCAG    49  PCR amplification of the GC-rich genes from the R. sphaeroides requires the use of certain additives to the reaction mixture in order to ensure a successful reaction. For this reason, all mutagenic PCRs in this study were performed with the addition of 5% DMSO (final concentration), as well as 2 μl of 100 mM MgSO4 (NEB). Aside from these modifications, the mutagenic PCR reaction for the deletion of the hairpin region 3’ of the puhA gene was performed under the same conditions described in Chiu et al (Chiu et al. 2004). Removal of this terminator sequence resulted in the creation of the plasmid pTZ18U::puhA T-. Following the removal of the hairpin 3’ of the puhA gene, a 6x his tag was added to the C-terminus of the puhA gene on pTZ18U::puhA T- to allow for purification of the RC via metal affinity chromatography. The same mutagenic technique was employed again in order to add the histidine tag. This tag was added immediately 5’ of the stop codon for the puhA gene. In addition to the tag, the penultimate GCC codon of the puhA gene was changed to GTG, creating an ApaL I restriction site (GTGCAC) that allowed for a simple means of screening for clones. The primers utilized for the addition of this tag are listed in Table 2.4. The PCR amplification of 6x his tagged puhA gene was performed under the same conditions as those employed for the removal of the hairpin terminator. Successful cloning of the mutagenic PCR product resulted in the creation of the plasmid pTZ18U::6his puhA T-.   50   Table 2.4. Primers utilized for the addition of a C-terminal 6x his tag Primer name Sequence (5’ -> 3’) 6his PuhA T- SLIM FT GTGCACCACCACCACCACCACTGATCGGATCCTCTAGAGTC 6his PuhA T- SLIM FS TGATCGGATCCTCTAGAGTC 6his PuhA T- SLIM RT GTGGTGGTGGTGGTGGTGCACGTATTCGGCCAGCATCGC 6his PuhA T- SLIM RS GTATTCGGCCAGCATCGC    51  Attempts to clone the puhA gene from pTZ18U::6 his puhA T-  as a BamH I fragment into plasmid pATP19P repeatedly resulted in either a complete cloning failure or a preference for the fragment to insert in the wrong direction. For this reason, the BamHI site downstream of the gene was changed to a SacI site using the SLIM method once more. The primers used for this reaction are listed in Table 2.5. The PCR reaction conditions are the same as above. A successful mutagenesis reaction resulted in the creation of the plasmid pTZ18U::6 his puhA T- 3’ Sac. This plasmid allowed for the insertion of the puhA gene into pATP19P as a BamH I to SacI fragment, generating the plasmid pATSHR:: 6 his T- 3’ sac.   52   Table 2.5 Primers utilized to change the BamH I site 3’ of the puhA gene to a Sac I site Primer name Sequence (5’ -> 3’) Bam2Sac PuhA FT CACTGATCGAGCTCTCTAGAGTCGACCTGCAGGCATGCAAGCT Bam2Sac PuhA FS TCGACCTGCAGGCATGCAAGCT Bam2Sac PuhA RT CTCTAGAGAGCTCGATCAGTGGTGGTGGTGGTGGTGCACGTATTC Bam2Sac PuhA RS GTGGTGGTGGTGGTGCACGTATTC    53  2.4 Removal of excess DNA from pAli2::puf. The plasmid pAli2::puf contained the pufQBALMX operon present as a large 4.57 kb EcoR I fragment. Because this plasmid was going to be utilized for PCR based mutagenesis, it was helpful to remove some of the excess DNA, which would facilitate any PCR-based methods applied on the plasmid. While there did not seem to be any obvious methods of DNA removal with restriction enzymes for the region upstream of the pufQBALMX operon, the sequence downstream of the operon contained 936 bp of sequence flanked by Sph I sites. Thus, pAli2::puf was simply digested with Sph I (NEB) and re-ligated with T4 DNA ligase (NEB), generating the plasmid pAli2::puf sph-.  2.5 Mutagenesis of the reaction center M gene. In order to change the residue at position 214 on the M subunit of the RC from R. sphaeroides, the plasmid pAli2::puf sph-, which contains the pufQBALMX operon was utilized for site-directed mutagenesis. Because this series of mutations consisted of single codon changes, the more simple Quickchange protocol from Stratagene was used. In this method, two complementary oligonucleotides containing the point mutation of interest at the center are utilized such that they can bind a DNA template through their flanking sequence. Amplification of a template with such primers results in the production of single-stranded products which are capable of self-annealing, generating a nicked plasmid product. Since the amplification reaction is linear (i.e. the primers can only generate useful product when bound to the original starting template), more starting material is required for the reaction. In addition, fewer amplification cycles are employed since it has previously been shown that the proofreading polymerase that I used for these mutagenic PCR reactions (Vent from NEB) begins to degrade the amplified product when 20 or more cycles are used in the PCR reaction (Byrappa et al. 1995). Mutagenic PCR 54  reactions for point mutations in the RC M gene were set up as follows: 50-100 ng of starting template DNA, 1 μL of each mutagenic primer (10 μM stock concentration), 1 μL of dNTPs (5 mM stock), 5 μL of 10x Thermopol buffer (NEB), 2.5 μL of DMSO (Acros Organics), 2 μL of 100 mM MgSO4 (NEB), 1 unit of Vent polymerase (NEB), and dH2O to 50 μL. A table of the primers used in this study to change the (M)L214 residue to different amino acids is listed in Table 2.6.    55   Table 2.6. Primers used to change the (M)L214 residue on the RC M gene. Primer name Desired residue Forward primer * (M)L214G Glycine 5’ CTCTACGGGTCGGCCGGCCTCTTCGCGATGCAC 3’ (M)L214A Alanine 5’ CTCTACGGGTCGGCCGCCCTCTTCGCGATGCAC 3’ (M)L214C Cysteine 5’ CTCTACGGGTCGGCCTGCCTCTTCGCGATGCAC 3’ (M)L214V Valine 5’ CTCTACGGGTCGGCCGTCCTCTTCGCGATGCAC 3’ (M)L214I Isoleucine 5’ CTCTACGGGTCGGCCATCCTCTTCGCGATGCAC 3’ (M)L214M Methionine 5’ CTCTACGGGTCGGCCATGCTCTTCGCGATGCAC 3’ (M)L214N Asparagine 5’ CTCTACGGGTCGGCCAACCTCTTCGCGATGCAC 3’ (M)L214Q Glutamine 5’ CTCTACGGGTCGGCCCAGCTCTTCGCGATGCAC  3’ (M)L214H Histidine 5’ CTCTACGGGTCGGCCCACCTCTTCGCGATGCAC 3’  * Reverse primers were the reverse complement of the forward primer   56   The reaction conditions for the (M)L214 mutagenic PCR reactions proceeded as follows. An initial denaturation step was performed at 95 oC for 3 minutes. This was followed by 15 cycles of: 95 oC for 20 seconds, 50 oC for 20 seconds, and 68 oC for 5 minutes and 50 seconds. After the amplification step, the reaction was held at 68 oC for 7 minutes and then held at 10 oC indefinitely. Following the PCR, 0.5 μL of Dpn I enzyme (NEB) was added to each reaction mixture without any further purification in order to remove any remaining template DNA. Due to the increased amount of template required for the reactions, the restriction digests were incubated for two hours or overnight in order to ensure complete digestion of the template DNA. Ten microliters of each of these restriction digests were subsequently used to transform CaCl2 competent E. coli.  2.6. Assembly of the RC genes onto a broad host range expression vector for in trans expression of mutant RCs. The cloning of the pufQBALMX genes into a broad host range vector was previously achieved via mobilization of the operon from pAli2::puf as an EcoR I fragment. This method, however, failed to succeed in my hands because of a tendency for this fragment to always insert in the wrong orientation. Because of this, the pufQBALMX genes on pAli2::puf sph- (M)L214X (where X denotes the single letter notation of the amino acid replacing leucine at position 214 on the RC M gene) were subcloned into pATSHR:: 6 his T- 3’ sac as Sac I to EcoR I fragments. The cloning method described in section 2.3 was used, as it gave good success in cloning the puhA gene into pATP19P. The resultant plasmid was called pRS1::(M)L214X. A map of this plasmid is shown in Figure 2.4.   57       Figure 2.4. A genetic and restriction map of pRS1. The bent arrow indicates the puc operon promoter. Curved arrows represent transcribed genes from the plasmid. Tc = tetracycline resistance.   58  2.7. Purification of the RC. Reaction centers were purified using a modified version of the protocol of Goldsmith and Boxer (Goldsmith and Boxer 1996). The cells from 21 L of R. sphaeroides semi-aerobic culture were collected by centrifugation and the cell paste was resuspended in 10 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2 mM MgCl2 to a volume of 175 mL. For the purification of Zn-BChl-containing RCs, the culture volume was up-scaled to 72 liters in order to obtain amounts of protein for spectroscopic analyses. A few crystals of DNase I were added to the suspension before lysing the cells in a French press at 18,000 PSI. Lysed cells were centrifuged to pellet debris and unbroken cells. The supernatant was then centrifuged at 66,226 × g at 4o C overnight to pellet chromatophores. Following ultracentrifugation, the supernatant was discarded and the remaining pellet was resuspended in 10 mM Tris, pH 8.0, 150 mM NaCl, to a volume of 100 mL. Samples (1 mL) of this suspension were placed in 1.7 mL microcentrifuge tubes, brought to room temperature, and lauryldimethylamine oxide (LDAO; Fluka) was added to each tube at concentrations ranging from 0.5% to 2.75% to determine the optimal concentration of detergent for RC solubilization. These samples were rocked in the dark for 30 minutes, and 900 µL were centrifuged at 107,400 × g for 30 min at 4o C.  The 875 nm absorbance maximum of the supernatants was used as a measure of the degree of photosynthetic complex solubilization.  In our experiments, the optimal concentration of LDAO for chromatophore solubilization varied between 1.25% and 1.75% LDAO. Once the optimal LDAO concentration was found, the remaining chromatophore suspension was brought to this detergent concentration and stirred in the dark for 30 min at room temperature, and centrifuged at 117,734 × g at 4 oC for 15 min.  The supernatant solution was collected and imidazole (10 mM) and NaCl (200 mM) added to the concentrations indicated before loading onto a Ni NTA column (Qiagen).  The 59  column was washed with 10 mM Tris, pH 8.0, 150 mM NaCl, 0.1% LDAO until the maximum absorbance over the range of 500 – 950 nm was less than 0.01. The RC was eluted from this column with 10 mM Tris, pH 8.0, 150 mM NaCl, 300 mM imidazole, 0.1% LDAO, and dialyzed against TL buffer (10 mM Tris, pH 8.0, 0.1% LDAO). For crystallization, RCs were purified further by anion exchange chromatography with an ÄKTA Explorer FPLC system (GE Healthcare) equipped with a SourceQ column and elution with a NaCl gradient (0 – 200 mM), with the majority of RCs eluting at 120 mM NaCl. The eluted RCs were desalted by passage through a 10DG column (Bio Rad) and concentrated with an Amicon 30 kDa centrifugal ultrafilter (Millipore). RCs used for crystallization experiments were not frozen and used immediately following purification.  2.8. Quantification of pigments from RCs. Pigments of RCs were extracted with a solution of acetone:methanol (7:2) and quantified according to Van der Rest and Gingras (Van der Rest and Gingras 1974). Following the extraction, the samples were briefly centrifuged to remove a precipitate that formed upon addition of the organic solvent to the RC samples. Absorbance spectra of the isolated pigments were obtained on a Hitachi U-3010 spectrophotometer (Hitachi). 2.9. Low temperature absorption spectroscopy. Cryogenic absorbance spectra (0.6 nm bandwidth) were obtained at ~10-13 K using a Cary 6000 spectrophotometer (Agilent) equipped with a closed-cycle helium cryostat (Omniplex OM-8; ARS Inc.) as described by Lin et al. (Lin et al. 2009). RC samples in 10 mM Tris buffer, pH 8.0, 0.1% LDAO were concentrated to an A802 of ~40 by centrifugation using a 30 kDa cut-off Centricon ultrafilter (Millipore). Samples were supplemented with sodium ascorbate (final concentration of 1 60  mM) and diluted 1:1 with spectroscopic grade glycerol prior to freezing between two quartz plates separated by a 25 m polycarbonate spacer. 2.10. X-ray crystallography. Fresh solutions of RC purified as described above were used to set up hanging drop crystallization trials. Instead of mixing a reservoir solution with the protein, a separate precipitant solution (1 M potassium phosphate pH 7.4, 3.5% 1,2,3-heptanetriol, and 0.1% LDAO) was prepared and mixed with the protein solution on cover slips before sealing over the reservoir. Various ratios of precipitant and protein solution were found to result in crystals of widely varying quality. Diffraction quality crystals of (M)L214G (~0.3 × 0.3 × 0.3 mm) grew at a ratio of 2 µL RC solution (7 mg/mL, or 20.6 OD802/mL):1 µL precipitant solution at 298 K, protected from light, over a reservoir of 1.52 M potassium phosphate buffer, pH 7.4. Crystals appeared after approximately 72 hours. Repeated exposure of the crystals to light appeared to diminish diffraction quality, so immediately after initial exposure to light, crystals were transferred to a solution of 1 M potassium phosphate buffer (pH 7.4) and 30% glycerol before flash-freezing in liquid nitrogen. The (M)L214A and (M)L214N crystals were produced and frozen in a similar fashion, except that the former grew from a 5 mg/mL protein solution at a 3:1 protein:precipitant ratio, whereas the (M)L214N crystals formed from a 3 mg/mL protein solution at a 2:1 drop ratio, over a reservoir of 1.5 M potassium phosphate buffer, pH 7.4. Diffraction data from a single (M)L214G crystal were collected at the Canadian Light Source on beamline ID-1 and processed using AUTOXDS, Pointless, and Scala to 2.2 Å resolution (Kabsch 2010; Winn et al. 2011). Data sets for (M)L214A and (M)L214N crystals were collected at the Stanford Synchrotron Radiation Lightsource on beamline 7-1 and processed using HKL2000 (Otwinowski and Minor 1997) for (M)L214A, or using Mosflm, 61  Pointless and Scala for (M)L214N (Winn et al. 2011; Leslie and Powell 2007). All three mutant RCs crystallized in the space group P3121, with one molecule in the asymmetric unit. Solvent content was approximately 75%, which is typical for a membrane protein crystal. All data sets were isomorphous with the wild type (WT) RC (PDB entry 2J8C, chosen for comparison because it is the highest resolution (1.85 Å) RC structure in the PDB). The (M)214 mutant structures initially underwent five cycles of rigid body refinement with Refmac5 (Murshudov et al. 1997); the input model used was an edited version of PDB entry 2J8C, with residue (M)L214 changed to the appropriate side chain. Subsequently, models were improved first by alternating between inspection of 2Fo-Fc and Fo-Fc electron density maps and manual editing in Coot (Emsley and Cowtan 2004), and followed by restrained refinement with Refmac5. Manual editing consisted primarily of adding and removing solvent molecules (mainly water, but also glycerol, 1,2,3-heptanetriol, and LDAO). Water molecules were added using peaks > 3σ in the Fo–Fc difference map with plausible hydrogen bonding geometry. Omit difference density maps, in which the phytyl tail of BA and the nearby LDAO molecule were omitted from the refinement model, were generated to remove model bias from the refinement procedure. To produce these maps, the occupancies of the phytyl tail atoms from CGA to C20, as well as LDAO and/or glycerol, were set to 0, and the model was refined by 8 cycles of restrained refinement in Refmac5. Unweighted omit maps were generated using the FFT program in the CCP4 suite (Winn et al. 2011).  For the Zn-BChl-containing RCs, diffraction data were collected at Stanford Synchrotron Light Source on beamline 7.1 using a wavelength of 1.12709 Å and processed with Mosflm and Aimless to 2.85 Å resolution (Leslie and Powell 2007; Winn et al. 2011). Anomalous differences were detected to 6 Å resolution. All data sets were isomorphous with 62  wild type RC (PDB entry 2J8C), which was used as the starting point for refinement with removal of the Mg2+ ions and for Zn-β-RC the side chain atoms of (M)214. Initial phases were obtained by limited refinement with Refmac5 (Vagin et al. 2004) and Fo-Fc and anomalous difference maps were computed with Coot (Emsley and Cowtan 2004) and FFT, respectively. Based on the peaks observed in both maps, Zn-BChl was modeled at each of the P, B, and H cofactor sites in both the A and B branches. Density for a histidine side chain was observed at residue (M)214 in the Zn-β-RC and modeled accordingly using the program Coot. Refinement of the structures was continued with Refmac5 and the removal of solvent atoms with unrealistic B-factors. 2.11. Photoheterotrophic growth experiments. For photoheterotrophic growth of R. sphaeroides, inoculum cultures were grown in RCV medium aerobically on a rotary shaker at 200 RPM. Two hundred µL of this culture were used to seed 16.5 mL photoheterotrophic starter cultures in screw cap tubes at 30 oC. These tubes were placed under saturating light intensity for 48 hours to adapt cells to photoheterotrophic conditions prior to low light growth. Adapted cultures were transferred to fresh 16.5 mL screw cap tubes and diluted with RCV medium to a density of approximately 30 Klett units. Cultures were grown at a light intensity of 5 µEm-2s-1 (measured with a LI-185B photometer; Li-Cor) and culture density measurements were taken with a Klett photoelectric colorimeter (100 Klett units = ~3 X 108 CFU/mL). 2.12 Time-resolved spectroscopy. Transient absorption measurements were performed at room temperature using a home-built spectrometer based on a kilohertz Ti:sapphire laser, according to the methods of Pan et al. (Pan et al. 2011; Pan et al. 2013). Briefly, Pulses of 1 mJ, at a repetition rate of 1 kHz (150 fs pulse duration at 800 nm), were generated from a 63  regenerative amplifier system (Tsunami and Spitfire, Spectra-Physics). Part of this pulse energy was used to pump an optical parametric amplifier to generate 865 nm excitation pulses. The broadband probe pulse was generated by focusing a weak 800-nm beam into a 3-mm sapphire plate, and sent to an optical compressor composed of a pair of prisms, before it was focused onto the sample. The white-light probe pulses were then dispersed by a spectrograph and detected using a charge-coupled device camera (DU420, Andor Technology). The polarization of the pump pulse was set to the magic angle (54.7o) with respect to the probe pulse. Samples were loaded onto a spinning wheel with a path length of of 1.2 mm. Global analysis of the time-resolved spectra was performed using ASUFIT, a MATLAB program provided by Dr. Su Lin at Arizona State University. The quinone-lacking RC control for transient spectroscopy measurements was prepared by treatment of native RCs with 4% LDAO and 10 mM o-phenanthroline (Sigma) for three hours in the dark, according to the methods of Okamura et al (Okamura et al. 1975).  2.13. Partial purification of chlorins. A novel method of partial pigment purification was devised for the extraction of BChl a from photosynthetic bacteria. Existing methods rely on the use of large amounts of acetone:methanol mixtures, followed by a tedious drying step which is conducive to chlorin allomerization from exposure to light and oxygen (Gauthier-Jaques et al. 2001), or via a lengthy precipitation procedure involving the carcinogenic chemical dioxane, which also involves extended periods of time in which the chlorins are exposed to air and light (Iriyama et al. 1974). In this new protocol, a combination of solvent and detergent removes the chlorin from the broken cell mixture, yielding a high concentration of pigments in a small volume. The solvent extraction and drying steps are 64  very short, and do not allow for much chlorin allomerization, even when working under aerobic conditions in the light.  BChl a was purified from photosynthetically grown R. capsulatus strain ΔgtaIΔ1081, because this strain achieves unusually high cell densities, thus yielding more BChl per liter of culture compared to wild type R. capsulatus or R. sphaeroides cells. In addition, this strain does not suffer from centrifugation issues stemming from capsular polysaccharide production (Brimacombe et al. 2013). Cells were centrifuged at 7000 x g in 1 L bottles, in a JLA 8.1 rotor (Beckman Coulter). The supernatant was discarded, and cells were resuspended in ~ 30 mL of 10 mM Tris, pH 8.0, 150 mM NaCl, 2 mM MgCl2. A few crystals of lyophilized DNAse was added to the cell suspension before lysis via a single passage through a French press. Cell debris was removed by discarding the pellet after centrifugation at 10,000 x g for 10 minutes in a JA-20 rotor (Beckman Coulter). To the remaining supernatant, LDAO was added to a final concentration of 1.5-2%, and 1/5th to 1/10th volume of ether was added to the mixture. The resultant mixture was poured into a 50 mL screw cap graduated polypropylene tube, inverted several times, and allowed to settle. The upper ether layer containing the pigments was removed, dried under nitrogen and resuspended in a desired volume of methanol. For the partial purification of Chl a, the detergent/ether method was not tested, and thus an acetone:methanol extraction method was used. 100 g of dandelion leaves were harvested from plants growing rampant around campus. The leaves were cut into small pieces with a pair of scissors, placed in a 1 L beaker, and dried as much as possible with paper towels. It is important to note that water was not completely removed in this step, and the remaining water assisted in the precipitation of the pigments. To the partially dried leaf 65  pieces, 500 mL of ice cold 7:2 acetone:methanol was added. The plant cell suspension was then sonicated to assist in cell rupturing. The mixture was stored at -20 oC in the dark for one hour. Leaf pieces were removed from the solvent by passing the suspension through a piece of filter paper placed inside a glass powder funnel. The filtrate was then dried under a stream of nitrogen until the pigments precipitated out of solution from the gradually increasing water concentration in the drying mixture. These pigments were rapidly collected by laying a piece of filter paper on a porcelain Buchner funnel and slowly pouring the pigment solution onto the paper. The filter paper was placed in a Petri dish, and ~ 10 mL of acetone was added to the dish, removing the majority of the pigments from the filter paper. The pigments in acetone were transferred to a glass test tube, dried under nitrogen, and resuspended in methanol. 2.14. Semi-preparative HPLC purification and quantification of chlorins. For the purification of Chl a and BChl a, 1.5 mL of partially purified pigments in methanol were filtered through a 0.45 μm nylon membrane. The filtrate was placed in a glass HPLC vial, and injected into a Waters 2695 HPLC equipped with a photodiode array (Waters) and a semi-preparative C18 column (Luna C18, 250 x 10, 5 μm bead size), using 100% methanol as the mobile phase (Shioi et al. 1983). For the purification Chl a, the detection wavelength was set to 666 nm. For BChl a, the detection wavelength was set to 771 nm. Concentrations of collected Chls and BChls were determined by absorption spectroscopy using their extinction coefficients in methanol (79.95 and 60 l g-1 cm-1 for Chl a and BChl a, respectively (Namsaraev 2009)). The concentration was calculated as C = A/a, where C is the concentration of the solute in mg/sample, A is the absorbance value at 665 or 771 nm for Chl a and BChl a, respectively, and a is the extinction coefficient.  66  2.15. Pheophytinization of chlorins by chemical means. Chl a and BChl a were pheophytinized by the addition of 3 μL of concentrated HCl per 100 μL of pigment solution. This mixture was a allowed to sit for 5 minutes on ice in the dark before the solvent was completely dried under a stream of nitrogen gas. The dried pigments were resuspended in the desired volume of solvent. 2.16. Pigment substitution of RCs under mild conditions. The protocol for the substitution of pigments was modified from the methods of Franken et al. since the (M)L214G mutant RC denatured under these conditions (Franken et al. 1997). Purified pigments were concentrated by drying and subsequent resuspension in acetone, such that 50 μL of this concentrate contained a 20-fold molar excess of pigments compared to RC. The pigment concentrate was combined with a buffer consisting of 10 mM Tris, pH 8.0, 0.1% LDAO. To this mixture, 0.5 mg of RC was added. The final volume of the reaction was 500 μl. Once set up, the reactions were stored in the dark at room temperature for 2 hours. Following this incubation, the reactions were transferred into multiple PCR tubes and incubated at 37 oC in a PCR machine for 90 minutes. To remove excess pigments and denatured protein, the reactions were purified by anion exchange chromatography using DEAE 650M (Tosoh). The resin was equilibrated in 10 mM Tris, pH 8.0, 0.1% LDAO, after which the reaction mixtures were bound to the column. RCs bound to the column were washed with 10 column volumes of 10 mM Tris, pH 8.0, 0.1% LDAO, 25 mM NaCl. The protein was eluted with the same buffer containing 120 mM NaCl.    67  3. RESULTS  3.1 Optimization of chemotrophic growth conditions for R. sphaeroides. Despite having a functional expression system for the purification of the RC, initial attempts at purifying the pigment-protein complex produced yields that were not sufficient to provide enough starting material for the spectroscopic and crystallographic analyses used in this study. The existing protocol used in the laboratory for the chemotrophic (semi-aerobic in the dark) growth of R. sphaeroides cultures employed 2 L Erlenmeyer flasks filled to 80% nominal capacity with LB medium (1% tryptone, 0.5% yeast extract, 1% NaCl), shaken at 150 RPM. This technique produced cultures that were approximately 220 – 300 Klett units in cell density after 96 hours. In this study, the growth conditions were optimized by changing the culture medium and volume such that higher cell densities would be attained upon harvesting.  Optimization of the culture growth medium began with the observation that R. sphaeroides cultures appeared to contain stronger LH1 absorbance intensity (measured at 875 nm) when grown in the synthetic medium RCV than in the complex medium LB, after 96 hours. Because the production of > 20 L of a synthetic medium like RCV is both labor intensive and costly, I wanted to see if one or more of the ingredients of RCV medium was responsible for the observed increase in LH1 peak intensity per mL of cell culture. Thus, flasks were prepared in which selected RCV components were added to LB medium (in the same concentration as in RCV medium). The cultures were allowed to grow for 96 hours before a whole cell absorption spectrum was taken. In addition, a sample was included in which the culture was grown in LB medium in complete darkness by wrapping the entire flask in aluminum foil. Because light negatively regulates RC-LH1 complex production, this 68  sample was included to test whether greater quantities of RCs can be expressed from samples grown in complete darkness (Shimada et al. 1992). The results are compiled in Figure 3.1.   69    Figure 3.1. Whole cell absorption spectra of chemotrophically grown R. sphaeroides cultures incubated in the presence of various RCV media components. The spectra were normalized to A650 = 0.2. LB = LB medium. RCV = RCV medium. LB Dark = LB culture grown in complete darkness. Each variable culture had one component of RCV added to LB in its respective final concentration in the synthetic medium (See Section 2.1 for the exact concentrations of each solute). Super salts contains: MgSO4, CaCl2, FeSO4, EDTA, trace elements, and thiamine hydrochloride. Absorption spectra were taken after 96 hours of growth.  70   The results in Figure 3.1 show that one of the RCV medium components, Super Salts, had a relatively large stimulatory effect on the LH1 peak. Because the amplitude of this peak is proportional to the amount of the RCs in a sample, a larger LH1 peak intensity is indicative of a greater amount of the RC. Also, because Super Salts consists of a mixture of magnesium, calcium, iron, thiamine, and several trace elements, a subsequent experiment of the same fashion was performed in which each of the Super Salts components was added separately to individual flasks of LB medium. The results of this experiment are compiled in Figure 3.2.    71     Figure 3.2. Whole cell absorption spectra of chemotrophically grown R. sphaeroides cultures grown in the presence of various Super Salts components. The spectra were normalized to A650 = 0.2. Each variable culture had one component of Super Salts added to LB in its respective final concentration in the synthetic medium RCV (See Section 2.1 for the exact concentrations of each solute). Absorption spectra were taken after 96 hours of growth.   72   The results in Figure 3.2 show that three constituents of the Super Salts mixture had stimulatory effects on R. sphaeroides LH1 peak intensity after 96 hours of growth: magnesium, calcium, and thiamine. Although this growth enhancement was reproducible with magnesium and calcium, the thiamine effect was not reproducible. Thus, thiamine was ruled out as having a stimulatory effect on chemotrophic LH1 production in R. sphaeroides cultures, and only magnesium and calcium were used as enhancers in the growth medium of preparative-scale R. sphaeroides cultures. As a final modification to the medium recipe, the amount of sodium chloride was reduced from ~170 mM to 4 mM because it was discovered over half a century ago that R. sphaeroides grows optimally at much lower salt concentrations (Sistrom 1960). The final composition of the so-called RLB medium consists of: 10% tryptone, 5% yeast extract, 4 mM NaCl, 810 µM MgCl2, and 510 μM CaCl2.(Jun et al. 2013)   Following the development of a better medium for the growth of R. sphaeroides, I sought to investigate whether a greater amount of the RC could be produced from RLB cultures when the volume of medium in the flask was changed. The rationale behind this idea was that chemotrophic growth of R. sphaeroides cultures in flasks filled to 80% nominal capacity did not produce culture densities as high as the same cultures grown aerobically or photoheterotrophically (anaerobic under illumination). Because filling an Erlenmeyer flask beyond 80% nominal capacity would result in little to no aeration at 150 rpm, the flask volume was titrated from 50% to 80% nominal capacity. After 48 hours of growth, the cells in all the flasks began to turn red, becoming a deeper red after 72 hours of growth. Therefore, measurements were taken after 72 hours of growth, with a subsequent measurement taken after 96 hours. The results are displayed in terms of a factor called “LH1 output”. This term 73  takes into account culture density as well as LH1 absorbance amplitude at 875 nm to compare the relative amounts of LH1 in a given volume of culture. The absorbance of LH1 at 875 nm was used as an estimate for RC concentrations because of the stoichiometric relationship between LH1 and the RC in core complexes of R. sphaeroides. Because cell light-scattering also contributes to the absorbance value at 875 nm, this scattering component was subtracted. This was done by estimating an 875 nm scatter value, which was calculated as the mean of the absorbance at 825 nm and 925 nm (assuming that the amount of light-scattering is linear along this portion of the near IR spectrum), and subtracting the 875 nm scatter value from the 875 nm absorbance to yield an estimated 875 nm absorbance value attributed to LH1 light absorbance alone. This LH1 contribution was multiplied by the absorbance of the sample at 700 nm (a region of the spectrum where absorbance is affected by light scattering, and thus cell density, alone), yielding the “LH1 output” value. This was done in order to account for increases in RC production due to the presence of more cells per ml of culture. The equation for LH1 output is as follows:  LH1 output = A700 * (A875 – ((A825 + A925) / 2))   [1] The results of applying equation 1 to R. sphaeroides cultures grown in flasks filled with RLB media at 50% – 80% nominal capacity are presented in Figure 3.3.    74     Figure 3.3. Comparison of LH1 output values from cells grown in flasks filled to different volumes. The flasks were filled to 50%, 60%, 70%, or 80% nominal capacity with RLB medium. The first four bars represent samples taken at 72 h after inoculation, and the last four bars represent measurements after 96 h. Bars are averages of three independent measurements. Error bars represent standard deviations about the mean.   00.050.10.150.20.250.30.3550% 72h 60% 72h 70% 72h 80% 72h 50% 96h 60% 96h 70% 96h 80% 96hRELATIVE LHI OUTPUT75  The results in Figure 3.3 demonstrate that much higher levels of LH1, and therefore RC, are achievable from cultures grown in RLB medium when a particular volume is used. Specifically, the flasks filled to 50%, 60%, and 70% nominal capacity produced two to three times as much LH1 per unit culture volume as the flasks filled to 80% nominal capacity. Between the 50%, 60%, and 70% samples, however, there did not seem to be a significant difference. Because of this, I chose to grow all future preparative scale R. sphaeroides cultures in Erlenmeyer flasks filled to 70% nominal capacity. This allowed me to use the most amount of media per flask, which in turn allowed me to grow the highest amount of cells in the limited shaker space provided in a shaker system that is shared among several laboratories.   The aim of the work presented above was not directly geared towards the answering of my research question. However, it enabled me to proceed faster through the subsequent experiments presented in Section 3 of this thesis. In addition, I have devised a simple and inexpensive means to grow preparative scale cultures of R. sphaeroides. Despite the fact that these growth conditions were only analyzed in this strain, I believe they have the potential to be used for other strains of purple bacteria, such as R. capsulatus and others.  3.2. Pigment composition of RCs with amino acid changes at position 214 on the M subunit. Absorption spectroscopy of RC pigment extractions provides a simple and remarkably accurate tool to analyze the pigment content of RCs using the extinction coefficients of the isolated pigments in a 7:2 acetone:methanol mixture (Van der Rest and Gingras 1974). I applied this method in my analysis of RCs in which residue 214 of the M protein had been changed from Leu, designated as (M)214 mutants. The BChl:BPhe ratios of all (M)L214 mutant RCs are listed in Table 3.1. These data indicate that RCs containing 76  aliphatic or sulfur-containing side chains at (M)214 have a chlorin pigment composition that is the same as or close to that of the wild type RC (i.e., 2 molecules of BChl for every  molecule of BPhe). Conversely, mutants with amide-containing side chains at HA (Asn, Gln) have a pigment composition that resembles that of the previously reported (M)L214H mutant (5 molecules of BChl for every molecule of BPhe (Kirmaier et al. 1991)).    77   Table 3.1. Pigment composition of (M)214 mutant RCs. Strain BChl:BPhe Ratio* wild type 2.03 ± 0.09 (M)L214G 2.60 ± 0.11 (M)L214A 2.20 ± 0.11 (M)L214C 2.19 ± 0.09 (M)L214V 2.19 ± 0.02 (M)L214I 2.15 ± 0.11 (M)L214M 2.39 ± 0.01 (M)L214H 4.74 ± 0.04 (M)L214N 4.70 ± 0.29 (M)L214Q 4.77 ± 0.08  * Values are averages of three extractions ± the standard deviation about the mean.   78  3.3 Ground-state analysis of mutant RCs by low temperature absorbance spectroscopy. In order to gain more insight into the cofactor environments of the (M)L214 mutants, and to confirm the results of RC pigment extractions, absorption measurements of samples cooled using a helium cryostat were employed. Such techniques restrict protein dynamic heterogeneity, yielding high resolution spectra in which some, although not all, individual pigments in the RC are revealed as absorption peaks.  3.3.1 Effect of replacement of (M)L214 with nonpolar or S-containing amino acid residues on the Qy and Qx transitions of RC BPhes. The 11 K electronic absorbance spectra of RC mutants bearing nonpolar residues at (M)214 are compared in Figure 3.4. At this temperature, spectroscopic differences that were subtle in room temperature absorbance spectra became more evident. Nevertheless at 760 nm, for example, the Qy transitions associated with the two BPhes in the wild type RC overlapped considerably, although absorbance on the red and blue edges of the peak corresponds to HA and HB, respectively (Figure 3.4A (Watson et al. 2005b)). All of the nonpolar (M)214 substitutions created in this work resulted in changes in the BPhe Qy absorbance region, ranging from almost imperceptible (in the (M)L214A RC) to most obvious in the (M)L214G/I/M RCs. Replacement of (M)L214 with Gly resulted in a 5 nm blue-shift of the BPhe Qy absorbance peak, from 759 to 754 nm, evidently due to a change in the HA band. In the spectra of the (M)L214M and (M)L214A mutants the corresponding peak shifts were 3.9 nm and 0.4 nm, respectively; the overlap of the HA and HB Qy transitions was enhanced as a result, and the combined absorbance bands appeared more symmetrical. In contrast, the absorbance spectra of the RC mutants with Cys, Val or Ile in place of (M)L214 exhibited red-shifts of the HA 79  BPhe Qy transition maximum, thereby accentuating the difference between the Qy absorbances of HA and HB.  Together, the above observations reflect primarily environmental perturbations of the HA chlorin when (M)L214 is replaced with another nonpolar residue, which affect the wavelength of light that is absorbed. Unlike the changes observed in the spectrum of the (M)L214H mutant (see below), the magnitudes of these shifts are much smaller, indicating that the pigment at this site in each mutant RC was a BPhe rather than a BChl.  As shown in Figure 3.4B, the low temperature Qx transitions of the wild type RC yielded three major absorbance peaks, and a long-wavelength shoulder. The peak near 535 nm is due to the HB BPhe and the ~545 nm peak corresponds to the absorbance maximum of the HA BPhe, whereas the peak and shoulder near 600 nm are due to BChls (Shochat et al. 1994; Kirmaier et al. 1985).    The (M)L214G and (M)L214A RCs exhibited a blue-shifting of the HA BPhe peak, such that the two Qx transitions merged to form a peak with a shoulder (Figure 3.4B). In contrast, the BChl region around 600 nm was not greatly affected. These data confirm our interpretation of the Qy data: although the electronic properties of HA are affected when (M)L214 is substituted by small, nonpolar moieties such as Gly and Ala, the chlorin in the HA site remains a BPhe. The HB and HA Qx transitions of the (M)L214C/V/I/M mutants were relatively similar to each other and to the wild type RC, in peak amplitude and wavelength, indicating that the HA site of these mutants contains BPhe.    80   Figure 3.4. Low temperature (~11 K) ground-state absorption spectra of (A) the Qy and (B) the Qx transitions of the nonpolar (M)214 mutant series. Spectra were normalized to the Qy absorption maximum of the dimer (P) BChl peak at ~895 nm. Peak assignments for the spectra can be found in Figure 1.8. The dashed lines mark the Qx and Qy absorption peak wavelengths of HA, as well as the peak wavelengths for the Qy transition of P and BA/BB, in the spectrum of wild type RC. Reprinted with permission from Saer et al. Copyright 2013 American Chemical Society (Saer et al. 2013a).  81  3.3.2 Effect of replacement of (M)L214 with nonpolar or S-containing amino acid residues on the Qy and Qx transitions of RC BChls. Because the (M)214 side chain is relatively distant from the RC BChl macrocycles, it was surprising to discover striking differences between the absorbance of wild type and some nonpolar RCs in the ~802 nm Qy transition region of the BA/BB accessory BChls. In the low temperature absorbance spectrum of the wild type RC, there is a major peak (attributed to a combination of BA and BB) that has a distinct shoulder on the long wavelength side, which has been attributed to BB, the BChl in the inactive B branch (Kirmaier et al. 1985; Neupane et al. 2012). It is interesting to note that the magnitude of this shoulder varies between low temperature ground state absorbance spectra of the wild type RC obtained by different groups (Mattioli et al. 1991; Middendorf et al. 1993; Frolov et al. 2010). In my work, the prominence of this shoulder was influenced by the bandwidth of the larger peak, which was correlated with the volume of the side chain at (M)214. RCs with nonpolar (M)214 side chains containing a δ or ε carbon (Leu, Ile, Met) exhibited a more pronounced shoulder than in the (M)L214G/A/C/V RCs (Figure 3.4A). Note that in the case of the (M)L214C and (M)L214V RCs the bandwidth increase of the larger, BA peak was accompanied by an increase in amplitude, in contrast to the (M)L214A RC, which showed no amplitude increase (also see Figure 3.5). The (M)L214G mutant is a special case, as the significantly broadened BA/BB region indicates an additional component at ~795 nm in addition to a large overall decrease in the amplitude of this region relative to the wild type RC (Figures 3.4A and 3.5).  The Qy transition of the P BChls of nonpolar mutants did not differ greatly. The P BChls in the (M)L214G mutant absorbed slightly to the blue compared to the wild type RC. 82  The (M)L214A and (M)L214C mutants also exhibited this change, although to a lesser degree than the (M)L214G mutant.   83             Figure 3.5. Overlay of the BA,B Qy absorbance band of wild type and (M)L214G/A/C/V mutant RCs from the 11 K absorbance spectra. Spectra were normalized to the maximum absorbance for the dimer (P) BChls. Reprinted with permission from Saer et al. Copyright 2013 American Chemical Society (Saer et al. 2013a).   84  Turning to the Qx region, I focus on the overlapping ~595/605 nm peaks, which are composed of contributions from all BChl molecules present in the RC (Shochat et al. 1994). In contrast to the BPhe Qx region (~535/545 nm), the BChl Qx region was relatively unchanged (Figure 3.4B). Because the  ~595/605 nm peaks were not greatly affected by the nonpolar mutations, relative to the ~535/545 nm peaks, the Qx absorbance data support the interpretation of the Qy data, that these mutations did not result in the presence of BChl in the HA pocket. 3.3.3 The Qy and Qx regions in the absorbance spectra of (M)214 polar mutants. The 11 K absorbance spectra of the RC mutants containing a polar residue (i.e. (M)L214 -> H, N, Q) at (M)214 differed significantly from the spectrum of the wild type protein, resembling that of the “-type” RC mutant (M)L214H (Figure 3.6). For example, in the spectrum of the (M)L214N mutant, a new absorbance band was observed at 785 nm (denoted ), and the absorbance that remained at 756 nm decreased relative to the amplitude of the PA,B absorbance band. In other words, it appeared that the absorbance intensity associated with the Qy transition of a BPhe group was replaced with the corresponding absorbance of a new BChl group. In addition, there was an increase in the intensity of the BA BChl absorbance band of the (M)L214H and (M)L214N mutants, where the β and BA absorbance bands overlap. In the case of the (M)L214Q RC, the relatively small red-shift of the HA peak resulted in BA/BB amplitudes similar to the wild type RC (Figure 3.7). Like the nonpolar and S-containing mutants (except the (M)L214A and (M)L214G mutants), these amplitude increases were also associated with a masking of the BB shoulder present in the Qy region of the spectrum.   85   Figure 3.6. Low temperature (~11 K) steady-state absorption spectra of (A) the Qy and (B) the Qx transitions of the polar (M)214 mutant series. Spectra were normalized to the Qy absorption maximum of the dimer (P) BChl peaks at ~895 nm. The dashed line indicates the absorption peak wavelength of the BChl Qx transition in the spectrum of the wild type RC. Reprinted with permission from Saer et al. Copyright 2013 American Chemical Society (Saer et al. 2013a).   86    Figure 3.7. Overlay of the Qy transitions corresponding to the BA,B and HA,B regions in the ~11 K absorbance spectrum of the wild type and (M)L214H/N/Q RCs. The peaks denoted “β” correspond to the HA BChl absorbance in the (M)L214H/N/Q RCs. Spectra were normalized to the maximum absorbance for the dimer (P) BChls. Reprinted with permission from Saer et al. Copyright 2013 American Chemical Society (Saer et al. 2013a).    87   The Qx transition spectra of the (M)L214N and (M)L214Q mutants confirm the conclusion that these RCs are β-type (Figure 3.4B). As in the (M)L214H control, the peak around 540 nm was absent from the absorbance spectra of the (M)L214N and (M)L214Q mutants, indicative of a loss of the HA BPhe from these RCs. There was a corresponding increase in the (M)L214H/Q RC absorbance around 595 nm, indicating the presence of BChl that was absent from the wild type RC.  In the wild type RC the BChl Qx region is composed of two components: a peak at ~595 nm corresponding to the accessory BA and BB BChls, and a shoulder at ~605 nm, which represents the P BChls, as well as a possible contribution from BA (Shochat et al. 1994; Kirmaier et al. 1985). In the (M)L214H mutant, the BChl Qx region had an increased amplitude at ~595 nm while retaining the ~605 nm shoulder (compare the relative magnitudes of the ~595 and ~535 nm peaks in the wild type vs. the (M)L2124H RC). This Qx absorbance profile was thought to be representative of the additional BChl, and the corresponding loss of BPhe, at HA (Kirmaier et al. 1991; Heller et al. 1995). However, in the amide-containing (M)L214N/Q RCs, this ~595 nm peak was blue-shifted relative to the wild type and (M)L214H mutant RCs. The shift was roughly 5 nm (the (M)L214N RC) and  ~1-2 nm (the (M)L214Q RC).  In summary, the ground state absorption spectroscopy results suggest that direct ligation of the HA chlorin to an amino acid side chain is required for the incorporation of BChl into HA. The use of nonpolar or sulfur-containing moieties at position (M)214 result in the presence of BPhe instead. Based on these findings, it appears that other species, such as water, are unable to act as a fifth ligand, even when a cavity is created in the region corresponding to the center of the chlorin macrocycle, as is the case in the (M)L214G and 88  (M)L214A mutants. When the HA chlorin is ligated to a competent residue, the nature of the ligating atom, as well as the volume of the ligating side chain, cause the absorption maximum of the HA band to shift considerably. 3.4 Functional analysis of (M)214 mutant RCs in photoheterotrophic growth experiments. The in vivo function of (M)214 mutant RCs was evaluated by comparing the RC-dependent phototrophic growth rates of strains containing RC mutations to an otherwise isogenic strain that contains the wild type RC. Under the low photon flux used in this study (5 µE.m-2.s-1), light intensity is a limiting growth factor, and cultures require a significantly longer time to reach stationary phase than when grown under a high light intensity (~500 h vs. ~50 h, data not shown). I postulated that under this low light intensity, RC performance in vivo may be measured indirectly as a phototrophic growth rate.  As shown in Figure 3.8, the control strain expressing the (M)L214H (β) mutant RC grew much more slowly than the strain containing the wild type RC. It had been reported that the (M)L214H mutant RC is impaired in electron transfer because replacement of the HA BPhe with a BChl results in an increase in the free energy of the pigment by up to 300 meV, and a lower midpoint potential (Kirmaier et al. 1991; Heller et al. 1995; Fajer et al. 1975; Cotton and Vanduyne 1979). Therefore, my in vivo measurement of culture growth rate under low light intensity corresponds to the in vitro measurements of electron transfer rates in purified RCs.  The growth of strains expressing the (M)L214N or (M)L214Q (β-like) mutant RCs was similarly impaired by comparison to strains expressing the wild type, nonpolar or S-containing (M)214 mutants. These results indicate that the (M)L214N and (M)L214Q RCs are impaired in electron transfer similarly to the (M)L214H RC. 89  In contrast to the growth of the polar mutants, all of the strains containing nonpolar or S-containing substitutions at (M)214 grew with approximately the same kinetics as the wild type strain (Figure 3.8). The average growth of the nonpolar or S-containing mutants (0.5 Klett units per hour, Appendix 1) was faster than that of the polar mutants (0.1 Klett units per hour, Appendix 1). The similar growth kinetics of the nonpolar or S-containing mutants was surprising, specifically because the (M)L214G/A mutants were found to be impaired in electron transfer from HA– → QA, such that charge recombination competes with the forward reaction, resulting in a drop in the overall yield of charge separation (Pan et al. 2013). These contrasting in vitro results are considered in more detail in Section 3.6.  90   Figure 3.8. Photoheterotrophic growth of R. sphaeroides strains containing mutant RCs cultured with 5 µE.m-2.s-1 illumination.  The ΔpuhA mutant, which lacks the RC H protein illustrates zero-growth. WT, wild type RC; single letter amino acid code identifies the residue at (M)214 in the RC mutants. Curves are representative of three independent measurements. Reprinted with permission from Saer et al. Copyright 2013 American Chemical Society (Saer et al. 2013a).   91  3.5 X-ray crystallography of (M)L214G, (M)L214A, and (M)L214N mutant RCs. The (M)L214G, (M)L214A and (M)L214N mutants exhibited notable differences in their low-temperature absorbance spectra compared to the wild type RC and to each other (Figures 3.4 and 3.6). In order to identify structural changes in the RC, such as changes in the positions of HA and BA, that might account for these spectroscopic changes, the crystal structures of these three mutant RCs were determined to resolutions of 2.20 Å, 2.70 Å, and 2.85 Å, respectively (see Appendix 1). The protein backbones of all three mutant RCs were almost identical to that of the wild type protein (PDB: 2J8C, RMSDs < 0.15 Å for all CA atoms).   In the crystal structures of the (M)L214G and (M)L214A mutant RCs, no difference from the wild type RC in the location of the HA BPhe was observed within coordinate error. The HA chlorin was well-defined but lacked electron density indicative of a metal chelated in the macrocycle, consistent with the interpretation that the HA pocket contains a BPhe.   However, there were clear differences in electron density in the vicinity of (M)214 in addition to the mutations themselves. These differences were clearest in the highest resolution structure of the (M)L214G mutant RC. In this structure, electron density around the phytyl tail of BA was discontinuous in the region where the tail bends, near the C4 methyl group (Figure 3.9A), implying that the BA tail is disordered. Notably, the bend in the phytyl tail is in close proximity to residue (M)L214 in the wild type RC (Figure 3.9B). The volume made available by the deletion of the leucine side chain in the (M)L214G mutant RC is occupied in part by the rotation around the phytyl tail bend. By inspection of difference omit maps, the BA phytyl tail was modeled in two conformations with equal occupancy. The "right" conformer depicted in Figure 3.9A is equivalent to the conformation observed in the 92  wild type R. sphaeroides RC (Figure 3.9B). In the “left” conformation, also shown in Figure 3.9A, the BA phytyl tail is directed away from the HA macrocycle such that a portion of HA ring 1 is exposed to the solvent (see Discussion). In this conformer, the phytyl chain and a glycerol (cryoprotectant) molecule displace the molecule of LDAO detergent that occupies this space in the structure of the wild type RC (compare Figures 3.9A and 3.9B). The two conformers and glycerol molecule do not completely account for the difference density present in sigma weighted and unweighted maps at the mutation site or in the region of the two phytyl tail conformations, suggesting that partially occupied solvent molecules and perhaps additional conformations of the phytyl tail are present. The difference maps derived from the (M)L214A crystal diffraction data are consistent with the two conformations of the phytyl tail (Figure 3.9C), although the lower quality of the (M)L214A data restrict structural comparisons with the (M)L214G structure to the mutation site.    93   Figure 3.9. Omit difference electron density maps of: (A), the (M)L214G RC; (B), the wild type RC (PDB:2J8C); and (C), the (M)L214A RC. These maps were generated by setting the occupancy of the phytyl tail and the nearby LDAO (if present) to zero, followed by 8 cycles of refinement.  Discontinuous density is visible in the “right” conformation of the (M)L214G mutant, and continuous density for the “left” conformation can also be seen (A, green mesh). Electron density meshes are contoured at 3.0 σ and carved in a 2.5 Å radius. Oxygen atoms are colored red and nitrogen atoms blue. Carbon atoms are colored orange for LDAO, grey for glycerol and HA, yellow for residue (M)214, and green, cyan, and magenta, respectively, for the (M)L214G, wild type, and (M)L214A RCs. Reprinted with permission from Saer et al. Copyright 2013 American Chemical Society (Saer et al. 2013a).  94  In the (M)L214N crystal structure, positive difference electron density was obvious at the center of the chlorin macrocycle in the HA position (Figure 3.10). This density was modeled as a Mg2+ ion, confirming that the chlorin at this position is a BChl group, in agreement with the results obtained in Table 3.1. After subsequent refinement, the B-factor associated with the Mg2+ ion was commensurate with the BChl ring, and no residual difference density was present. At residue (M)214, adjacent to the Mg2+, density for an Asn side chain was present and modeled accordingly. The (M)N214 OD2 atom is located approximately 2.0 Å from the Mg2+, the Mg2+ to OD2 bond angle is approximately perpendicular to the plane of the macrocycle, and the Mg2+ ion is displaced ~0.4 Å from the plane formed by the tetrapyrrole N atoms towards the new ligand, supporting the assignment of a metal-ligand bond. A similar ~0.4 Å displacement of Mg2+ from the plane was observed in the equivalent N to Mg2+ bond between (L)H153 and the BA BChl (not shown). The ND1 atom of (M)N214 is within H-bond distance (3.1 Å) of the main chain carbonyl of (M)Y210. The alternative amide conformation of the Asn side chain with Mg2+ coordination by the ND1 atom of Asn214 is unlikely because of the inability of the OD2 to establish a hydrogen bond with this main chain carbonyl group, and because of the requirement for the metal to displace a proton of the NH2 group to form a metal-N ligand. Furthermore, the lone electron pair of the amide nitrogen atom should participate in a resonance structure of the amide group, resulting in a partial positive charge on the nitrogen.   95    Figure 3.10. Omit difference electron density map of the (M)L214N RC, illustrating positive difference density for the Mg2+ ion at the center of the chlorin ring of BChl in the HA site. The electron density mesh is contoured at 4.0 σ and carved in a 2.0 Å radius. Oxygen atoms are colored red and nitrogen atoms are in dark blue. HA carbon atoms are in lighter blue and (M)L214N carbon atoms are shown in yellow. Reprinted with permission from Saer et al. Copyright 2013 American Chemical Society (Saer et al. 2013a).    96   The results from section 3.5 suggest that the protein component of the RC is largely undisturbed from mutation of the (M)214 residue to amino acids containing small, nonpolar side chains. In contrast to the protein component, the phytyl tail of the nearby BChl residue, BA, appears to be disordered, as demonstrated by the electron density maps (Figure 3.9). Furthermore, the structure of the (M)L214N mutant revealed that amide-containing side chains are indeed capable of acting as fifth coordinates to the HA chlorin. I suggest that this ligating residue is the amide oxygen, and not the nitrogen.   3.6 Kinetic analysis of wild type and mutant RCs with nonpolar or sulfur-containing side chains at (M)214. Transient difference spectroscopy was used to evaluate the in vitro function of the (M)L214 nonpolar or S-containing mutant RC series on several levels, including the primary events of charge separation, the subsequent electron transfer steps, as well as the quantum yield of charge separation (calculated within a 6 ns time window).  In this form of spectroscopy, the obtained data are interpreted as 2D datasets, representing a set of difference spectra (excited minus non-excited) spanning from 500 nm to 960 nm, measured over a time window from 150 fs (0.15 ps) to 6 ns (6000 ps). Thus, data can be interpreted as changes in absorbance (differences) measured at a single wavelength over time (a kinetic trace), or entire difference spectra displayed at single points in time (transient absorbance spectra). One method of further analyzing such spectra is to perform what is called a global analysis. In this technique, the datasets are modeled to fit a series of exponential decays using a least squares approach, with the time constant of each decay being representative of the disappearance of an entire transient difference spectrum. The values of the time constants signify the point in time in which the transient difference spectrum has decayed to 1/e (~36.8%) of its maximum value. These transient difference 97  spectra have been called “decay associated difference spectra” (van Stokkum et al. 2004). The total number of decay associated difference spectra present in a dataset is equal to the number of exponentials used to fit the dataset. The sum of all decay associated difference spectra at a time point that corresponds to the time constant of only one of these spectra is termed the "evolution associated difference spectrum" (EADS (van Stokkum et al. 2004)). When referring to the RC, the different EADSs are by no means representative of the different electronic states of the protein, however they provide a simple method of visualizing the temporal progression of the difference spectra dataset. When the different EADSs of the wild type protein are calculated, a meaningful comparative analysis can be made to mutant RCs, providing insight into the electron transfer behavior of these proteins in the ultrafast time domain.   Shortly following excitation of the RC sample (0.5 ps post excitation), the transient absorbance spectrum becomes dominated by the excited state special pair, or P*. This spectrum is characterized by a large broadband bleaching (negative amplitude) of the spectral region between 800 nm to 950 nm, and represents a large degree of stimulated emission from the excited state. The decay of this stimulated emission indicates a decrease, but not a complete disappearance, of the bleaching amplitude around 800 nm to 950 nm. Because other transient signals of the RC (such as P+) eventually overlap with this bleaching signal within a few picoseconds, the decay of stimulated emission from P* is commonly measured as a kinetic trace from 925 nm to 930 nm, because stimulated emission represents the only contribution to this spectral region (note how the bleaching in this region disappears after 10 ps in the inset of Figure 3.11). Because the rate of non-productive fluorescence (i.e., fluorescence emission in the absence of electron transfer) from P* occurs on a timescale of 98  ~500 ps (Jones 2009), any observed changes in P* stimulated emission within 10 ps result from direct electron transfer from P* to a nearby cofactor, as seen in the wild type RC where P* almost always decays via forward electron transfer (Figure 3.11).    99    Figure 3.11. Normalized kinetics of stimulated emission decay in wild type and (M)L214 mutant RCs measured at 926 nm with an excitation pulse at 865 nm. Letters in the legend represent: WT, wild type; LM, (M)L214M; LC, (M)L214C; and so on. The inset shows the transient absorption difference spectra of the wild type RC 0.5 ps (white circles) and 10 ps (red circles) post excitation. The upward pointing arrows are representative of wavelengths in which kinetic traces have been described in Section 3.6 of this work. Reprinted with permission from Pan et al. Copyright 2013 American Chemical Society (Pan et al. 2013).   100  The rate of stimulated emission decay at 926 nm from RCs containing Cys, Met, Ala, or Gly at (M)214 closely resemble that of the wild type RC, in that the signal decays almost completely within 10 ps post excitation. Thus, it appears that these RCs are able to mobilize an electron from P* to a nearby cofactor just as efficiently as the wild type RC.  Following the oxidation of P*, the RC enters a series of charge-separated states via electron transfer reactions along the A branch, increasing the distance between P+ and its electron. The nearest cofactor to P+ is BA, which is reduced with a time constant of ~3 ps post excitation. This P+BA- state quickly decays to P+HA- with a time constant of ~1 ps, The speed at which P+BA- decays into P+HA- made the detection of the BA- anion difficult, with some researchers even questioning its presence (Martin et al. 1986), Nevertheless, spectroscopic studies in the near IR region have shown that BA is a genuine electron carrier with a spectroscopic transient at 1020 nm (Arlt et al. 1993). Because my work did not capture data on wavelengths to the red of 960 nm, the kinetics of BA- are excluded from the study. Instead, I focus on the formation of HA-, which has two characteristic transient signatures. The first transient signature is an electrochromic shift of the ground state absorbance of BA upon the RC entering the P+HA- charge separated state. Specifically, the 802 nm absorbance band of BA shifts to a higher energy. Measurement of this shift is observed at 785 nm (Figure 3.12).   101    Figure 3.12. Normalized kinetic traces of the electrochromic shift of the BA absorbance, measured at 785 nm, for wild type and (M)L214 mutant RCs. Solid lines are fits to the data points (circles). Reprinted with permission from Pan et al. Copyright 2013 American Chemical Society (Pan et al. 2013). LC = (M)L214C, LA = (M)L214A, LG = (M)L214G. The horizontal black line is a zero absorbance reference line. Data points to the left of the vertical black line are on a linear time scale, and those to the right of the line are on a logarithmic time scale.    102   Within 0.5 ps, there was a sharp absorbance increase at 785 nm in all RCs measured (the (M)L214M kinetics were practically identical to wild type, and are thus not shown in Figure 3.12 or any subsequent kinetic traces). This increase represents a broadband absorption increase due to formation of P* (Martin et al. 1986). Following this increase, the amplitude of the absorbance in this region continued to increase, which is attributed to a shift from the ground state absorbance of BA due to the emergence of the HA- anion.  Interestingly, the amplitude of this absorbance increase is proportional to the volume of the side chain at (M)214, such that larger side chain volumes are associated with a larger amplitude increase at 785 nm. Although the cause of these amplitude differences is not clear, it is possible that a higher degree of static heterogeneity (i.e. multiple subpopulations of RCs in a single sample) present in the RCs with lower (M)L214 side chain volumes accounts for this difference. This is because the increased bandwidth of an absorption band, arising from static heterogeneity (most evident in the BA absorption band in the Qy transition of the (M)L214G mutant), concurs with a decrease in peak amplitude. Nevertheless, it is clearly evident from the amplitude increases at 785 nm that the P+HA- state is attained in all the measured RCs.   The second spectral signature that is characteristic of the HA- anion is an increase in absorbance at 675 nm. Although this signal is broadband, detection at 675 nm has been classically used in the literature as the wavelength of measurement. The kinetic traces measured at 675 nm for the wild type and RCs containing Cys, Ala, or Gly at (M)214 are shown in Figure 3.13.    103   Figure 3.13. Normalized kinetic traces measured at 675 nm for wild type and (M)L214G/A/C mutant RCs. A quinone-lacking RC (WT-deQ) is used as a negative control. Reprinted with permission from Pan et al. Copyright 2013 American Chemical Society (Pan et al. 2013). The vertical black line represents the shift from a linear to logarithmic time scale. The horizontal black line is a zero ΔA reference line.     104  In Figure 3.13, the absorbance amplitude at 675 nm increases to a maximum within 10 to 20 ps post excitation. In my work, we assumed that the majority of the population of RCs in a sample were capable of achieving the charge separated P+HA- state. Thus, all spectra in Figure 3.13 were normalized to the Qx bleaching of P at 600 nm at 0.5 ps post excitation. In the wild type RC, this 675 nm absorbance signature decays with a time constant of approximately 200 ps, which is indicative of forward electron transfer to the primary quinone, QA. In the mutant RCs, however, one can see that this increase in amplitude decays with kinetics different from the wild type RC. Specifically, the decay time is increased to the point where none of the RCs with small, nonpolar or S-containing moieties at HA ((M)L214G/A/C) show signs of complete decay within the 6 ns time window of measurement. In addition, the 675 nm transient decay kinetics of the (M)L214G/A/C RCs require multiple exponential components to fully fit to the data (compared to a single exponential in the wild type RC). My interpretation is that there exists a sub-population of these three mutant RCs with a long-lived P+HA- state. This means that either: (i) the forward electron transfer event in the (M)L214G/A/C RCs occurrs with more than one rate; (ii) the P+HA- state begins to decay via the non-productive pathway of charge recombination, resulting in a return to the ground state; or (iii) both i) and ii) happen at the same time. To begin to parse this issue, the kinetics of ground state recombination was measured. Figure 3.14 compares the normalized kinetics of wild type and (M)L214G/A/C RCs at 837 nm. Because the negative bleaching amplitude from P+ dominates the absorption changes at this wavelength , a decrease in bleaching amplitude at 837 nm after ~15 ps can be attributed to charge recombination of P+ from some charge separated state, whether it is P+BA-, P+HA-, or a combination of both states.   105    Figure 3.14. Normalized kinetic traces (circles) and their respective fits (solid lines) measured at 837 nm for wild type and (M)L214G/A/C mutant RCs. A quinone-lacking RC is used as a negative control. Black circles: wild type, blue circles: (M)L214C, teal Circles: (M)L214A, pink circles: (M)L214G, white circles: quinone-removed RCs. The vertical black line represents the shift from a linear to logarithmic time scale. The horizontal black line is a zero ΔA reference line. Reprinted with permission from Pan et al. Copyright 2013 American Chemical Society (Pan et al. 2013).    106  It is clear from Figure 3.14 that the wild type RC does not exhibit charge recombination on this time scale because, after the initial bleaching from sample excitation, there was no decay of bleaching throughout the window of measurement. In contrast, the bleaching of the (M)L214G/A/C RCs decays in the nanosecond time range. The degree of decay was most pronounced in the (M)L214G mutant, where the curve decayed almost halfway within the measurement window, similar in magnitude to an RC in which the QA quinone was removed by chemical means. By comparison of the bleaching kinetics, the behaviour of RCs with small, nonpolar or S-containing moieties at HA in the nanosecond time domain becomes clearer; although some forward electron transfer occurs on the hundreds of ps time scale, the population of RCs with long-lived P+HA- states is poised in a state of competition between a slow forward electron transfer and charge recombination. The degree of forward electron transfer vs. charge recombination appears to be correlated with the volume of the (M)214 side chain, because the (M)L214G mutant exhibited the largest amount of recombination (Figure 3.14).   The similar recombination kinetics of the (M)L214G and the QA-lacking RCs led me to investigate the possibility that some of the QA quinones were absent in this RC sample. To address this, a saturated solution of ubiquinone-10 in ethanol was added to both the wild type RC and the (M)L214G RC at 1/10 the sample volume. The RCs were left to incubate for 1 hour in the dark at room temperature and an absorption spectrum was measured. Because the addition of quinone solution changed the baseline of the absorption spectrum for both wild type and (M)L214G RCs, a light minus dark difference spectrum was taken for each sample before and after the addition of quinone. Any increases in quinone occupancy were interpreted as increases in bleaching amplitude at 865 nm. The spectra were not normalized, 107  however the samples were maintained at the same concentration (42 µM). The results, shown in Appendix 5, show no major changes in the difference spectra in both the wild type and (M)L214G RCs before or after the addition of ubiquinone. Thus, it appears that the (M)L214G sample is QA-replete. These data are in agreement with our crystallographic data for the (M)L214G RC, which show no significant deviations in B-factors for QA head atoms (~31-39) compared to the rest of the RC (overall B-factor is 40, the tail atoms have higher B-factors, but this is feature common to all RC crystallographic datasets).   In contrast to the in vivo photoheterotrophic growth experiments, the results from the work in this section suggest that RCs with smaller amino acids at (M)214 are associated with more complex kinetics, as well as a lower degree of charge separation ability. These RCs are unable to maintain the ~100% degree of charge separation observed in the wild type RC over a 6 ns time window. Furthermore, the decay of the P+HA- state appears to be multiexponential in the (M)L214G/A/C mutants. An explanation of these kinetics is given in the Discussion.  3.7 Pigment substitutions of wild type and (M)L214G RCs. When heated in the presence of a molar excess of a heterologous chlorin-type of pigment, the R. sphaeroides RC has been shown to substitute native pigments for a heterologous pigment. Such experiments have demonstrated selectivity in the binding pockets of the RC, because only certain heterologous pigments substitute for the native pigment in only certain RC pockets. These experiments have also provided insight into the effects of pigment substitution on neighboring chromophores (Scheer and Hartwich 1995), as well as the effects of substituted pigments on the electron transfer properties of bacterial RCs (Huber et al. 1995). In my 108  research, pigment substitution experiments were used to investigate the role of (M)214 in pigment selectivity by comparing the results obtained from wild type and (M)L214G RC pigment substitutions. I investigated several temperatures, lengths of time, and three pigments were used in substitution experiments: BChl, plant Pheo, and plant Chl. Both plant Chl and Pheo differ from BChl and BPhe in that they contain a vinyl group instead of an acetyl group on ring A, as well as an additional double bond on ring B. Plant Pheo differs from Chl in that, like BPhe it does not contain a central Mg2+ ion. The results were analyzed using room temperature absorption spectroscopy, and some are depicted in Figure 3.15.    109       Figure 3.15. Absorption spectra of wild type (left) and (M)L214G (right) RCs after incubation at 37o C with a molar excess of heterologous pigment. The spectra were normalized to the maximum absorbance of P. The absorbance peaks are labeled according to the convention in Figure 1.3. The Chl/Pheo peak corresponds to the absorption band of plant chlorophyll a and pheophytin a.   110  My work on substitution of wild type and (M)L214G RCs with plant Pheo demonstrates that chromophore substitution occurs at 37o C, although the substitution is by no means complete when compared with previous results for the WT RC, in which the incubation temperature was 42.7 o C (Franken et al. 1997). Nevertheless, the amount of substitution was sufficient for a comparative analysis using absorbance spectroscopy. The absorbance spectra of wild type and (M)L214G RC substituted with Pheo show a decrease of absorbance amplitude in the 760 nm band, with an associated increase in 670 nm absorbance. Both RCs appear to show an equal degree of substitution, although the shapes of the peaks were different. The 670 nm peak in the Pheo-substituted wild type RC contained a small shoulder on the red side. In addition, the 760 nm peak in the (M)L214G mutant RC appeared as a shoulder to the blue of the BA,B peak, lacking a discrete apex. Although the shapes of both the 760 nm and 670 nm peaks differ between the two RCs, I suggest that these changes are a consequence of the different binding pocket environments induced by the amino acid change. The amplitudes of both the 670 nm peak and 760 nm peak are roughly equal between the wild type and (M)L214G RC. Specifically, the broadened BA,B absorption band of the (M)L214G RC (see Section 3.3) allows for a larger degree of overlap between this band and the smaller neighboring HA,B band, thereby making this peak appear more like a shoulder when its amplitude was decreased due to the (M)L214G substitution (Figure 3.15). The shoulder of the Pheo band observed in the wild type RC is more difficult to explain, as the absorbance properties of Pheo in bacterial RCs is poorly understood. However this shoulder may be the result of a smaller fraction of Pheo incorporating into HA compared to HB. In the low temperature absorbance spectroscopy results (Fig 3.4), the HA Qx transition band is red-shifted relative to the HB band, and so if the same effect were to occur when Pheo is inserted 111  into the HA pocket this shoulder would represent the population of Pheo present at HA. In the (M)L214G RC the spectral shifting of the Qx transition is less pronounced (Fig 3.4), and therefore a shoulder on the 670 nm peak would not be expected.    The use of BChl a in substitution experiments on the wild type RC yielded no discernible changes in the absorbance spectrum, as shown in Figure 3.15. In contrast, the (M)L214G mutant RC produced a different spectral phenotype post-substitution. Firstly, the amplitude of the ~760 nm HA,B peak in the BChl a-substituted (M)L214G RC was decreased relative to the nonsubstituted RC (Figure 3.15). This was accompanied by an increase in absorbance in the region corresponding to the trough between the ~760 nm HA,B peak and the ~802 nm BA,B peak. Taken together, these results indicate that a spectral red-shift at one of the chromophores in the (M)L214G RC resulted from pigment substitution using BChl a. Thus, it appears that incubating the (M)L214G with a molar excess of purified BChl a results in the incorporation of the metal-containing chlorin BChl a  in this HA pocket, although further experiments are required to provide a more definitive answer.   Because the absorption spectrum of the Qy transition of free BChl a overlaps with that of the RC to a large degree, I sought to perform a similar pigment substitution experiment using a chromophore that does not spectrally overlap the absorption bands of the RC.  It was thought that this approach might paint a clearer picture with regards to the HA pocket differences between the (M)L214G and wild type RC. Thus, the wild type and (M)L214G RC were used in substitution experiments employing the heterologous pigment Chl a. The Q band of free Chl a is similar to that of Pheo in the sense that it absorbs maximally at ~675 nm, in a region where the RC absorption bands exhibit no major contributions. Like the experiments with plant Pheo, I would expect the outcome of a successful substitution with 112  Chl a to result in both a decrease in one of the Qy absorption band of the RC, as well as the presence of a new absorption band near ~675 nm since the Q band of Chl a peaks at a shorter wavelength than the Qy band of BChl or BPhe. The results of pigment substitution with Chl a are also shown in Figure 3.15. In the wild type RC, the Qy transition of the RC absorption spectrum remained largely unchanged, although there was a new peak at ~675 nm present. This result was not expected because the appearance of a new absorbance band without an accompanying decrease in the intensity of one of the RC absorbance bands suggests that pigments were binding to the sample without substitution. Subsequent re-purification of the RC using DEAE anion exchange chromatography failed to remove this peak from the spectrum. Because there were no changes to the three Qy peaks of the RC spectrum, it appears that Chl a binds the RC in a non-specific manner more tightly than the other tested chromophores (Pheo and BChl a). Repeating this experiment, but now with the (M)L214G RC produced an absorption spectrum that was largely similar to the Chl a-substituted wild type RC, although there was a noticeable decrease in ~760 nm peak amplitude. In addition, the amplitude of the ~802 nm BA,B peak decreased. Although this decrease could be in part a result of the spectral overlap between the ~760 nm and ~802 nm peaks in the RC absorption spectrum, I do not believe that this explanation is sufficient to account for the entire ~802 nm absorbance decrease. This is because the experiment with Pheo substitution produced a ~760 nm peak with a peak amplitude similar to that of the Chl a-substituted (M)L214G RC, whereas the ~802 nm peak did not decrease by a similar amount. Thus, I suggest that in the Chl a substitution experiment on the (M)L214G RC the BA site was affected. Because the ~760 nm peak represents contributions from BPhe in the HA,B pockets, it could be that the pigments at HA,B are either removed from the RC or are substituted with Chl a in these 113  experiments. Because the process of heating the (M)L214G RC does not result in loss of pigments, as evidenced by a lack of change in the RC absorption spectrum, the latter explanation (that Chl a  is incorporated into the HA,B pockets) seems more likely. Taken together, the experiments described above illustrate that a bulky residue like Leu at position (M)214 in the RC plays a role in the structural integrity and pigment specificity of this pigment-protein complex. Changing this side chain to a smaller moiety appears to allow a wider variety of heterologous chlorins, including those that contain central metal atoms, to incorporate into the RC under conditions of prolonged exposure to higher temperatures.   The above results hint at a novel role for residue (M)214 in providing pigment selectivity at the HA site. In contrast to the wild type RC, whose Qy absorption spectrum remains largely unchanged after incubation with a molar excess of either Chl a or BChl, the (M)L214G mutant does incur spectral changes. These changes are localized on the Qy absorption band of the RC BPhes, and suggest that the (M)L214G RC is indeed capable of accepting metallo-(bacterio)chlorins in the HA pocket. 3.8. Bacteriopheophytin incorporation into the RC in the presence of adventitious ligands. The results of pigment substitution experiments on the wild type and (M)L214G mutant RCs appeared to complicate the issue of the origin of BPhe, in addition to the question of HA pocket selectivity in the RC. If metal-containing chlorins were incorporated into the (M)L214G mutant in vitro, then why did purification of this RC not yield an appreciable population of RCs with a molecule of BChl at HA? One possible answer is that BChls may bind to an axial ligand during the folding of the RC and, if such a ligand is not present, the Mg2+ ion leaves the chlorin macrocycle. Thus, in order to test whether the wild type or (M)L214G mutant RC can accommodate a molecule of BChl at HA in vivo, these 114  two RCs were purified in the presence of an adventitious ligand which could stabilize the BChl molecule in a pentacoordinate state during the RC folding process. Previous reports have shown that imidazole can provide such a function in both heme-containing proteins, as well as the RC (Depillis et al. 1994; Goldsmith et al. 1996). Because it was reported in Goldsmith et al. that concentrations of imidazole in excess of 5 mM produced inhibitory effects on cell growth (Goldsmith et al. 1996), this concentration was chosen as a starting point for the subsequent experiments. In order to prevent dilution of the solute until the final purified protein was obtained, imidazole was added to all the buffers used in the RC purification procedure, at a final concentration of 5 mM. Interestingly, the concentration of imidazole used (5 mM) proved to be inhibitory for culture growth when cells were grown under chemotrophic conditions (aerobically in the absence of illumination). Under photoheterotrophic conditions (anaerobically with illumination), however, the cultures appeared to be tolerant of this concentration of imidazole however they required several more days to reach stationary phase. This was undesirable because the extra growth time may give rise to faster-growing spontaneous mutants. It was thus necessary to find the maximum allowable concentration of imidazole that was permissive for chemotrophic growth in the dark. R. sphaeroides strain ΔRCLH containing plasmid pRS1 was grown in RLB media containing concentrations of imidazole ranging from 1 to 5 mM, and optical density readings were obtained after 72 hours. The results of this experiment are shown in Figure 3.16. Under conditions of chemotrophic growth in the dark, the results from Figure 3.16 suggest that R. sphaeroides loses its ability to grow to an appreciable culture density when the concentration of imidazole in the medium is greater than 2 mM. This concentration was thus used to for a 115  comparative analysis of RCs purified from cultures grown with or without imidazole in the medium.    116        Figure 3.16. Optical density of chemotrophically grown R. sphaeroides strain ΔRCLH containing plasmid pRS1 under varying concentrations of imidazole. Measurements were taken at 700 nm after 72 hours of growth at 30o C in the dark. Error bars represent standard deviations about the mean from triplicate measurements.    00.511.522.51 2 3 4 5OD at 700 nmImidazole concentration (mM)117  Following the determination of the maximum allowable concentration of imidazole that was permissive of chemotrophic growth, the wild type and (M)L214G RCs were purified from cultures grown in the presence of 2 mM imidazole.  Figure 3.17 shows an overlay of the wild type and (M)L214G RC absorption spectra after purification with or without the addition of 2 mM imidazole to the growth medium and buffers used for purification.   118   Figure 3.17. Absorbance spectra of wild type and (M)L214G RCs from cultures grown photoheterotrophically in the presence or absence of 2 mM imidazole, and purified in the presence or absence of 2 mM imidazole. The four spectra were normalized to the maximum absorbance of the PA,B band at ~865 nm. The absorbance peaks are labeled according to the convention in Figure 1.3.   119  The results obtained from imidazole-purified RCs demonstrate that there was little to no difference in the absorption spectra of the wild type RC from cultures grown chemotrophically in the presence or absence of 2 mM imidazole, and purified in the presence or absence of imidazole. The absorption spectra of the two wild type RC samples almost overlap throughout the entire Qy transition region, and so I conclude that imidazole does not have an effect on the pigment composition of this RC.  The (M)L214G RC, on the other hand, appears to have been affected in cofactor composition when imidazole was added to the growth medium. Specifically, the Qy transition bands corresponding to the BA,B and HA,B chromophores of the imidazole-purified (M)L214G RC showed decreases in amplitude relative to that of the (M)L214G RC purified in the absence of this solute, indicating that one or more cofactors were not present in the (M)L214G RC purified with imidazole. The Qx transition also exhibits an absorption intensity decrease, however the decrease is localized on the Qx absorption of the BPhes alone. Pigments were then extracted from the purified RCs and the BChl:BPhe ratio was calculated according to the procedure outlined in the Materials and Methods. The calculated BChl:BPhe ratios are listed in Table 3.2. The results shown in Table 3.2 reveal that the BChl:BPhe ratio of the (M)L214G RC purified in 2 mM imidazole was close to 3.3, whereas the ratios for the other three measured samples were between 2 and 2.6.   120   Table 3.2. Calculated BChl:BPhe ratios of wild type and (M)L214G RCs purified in the presence and absence of 2 mM Imidazole RC BChl:BPhe Ratio* wild type 2.03 ± 0.09 wild type + Imidazole 2.29 ± 0.01 (M)L214G 2.60 ± 0.11 (M)L214G + Imidazole 3.27 ± 0.02  * Values are averages of three extractions ± the standard deviation about the mean.    121  Taken together with the results from Figure 3.17, the data in Table 3.2 indicate that one of the BPhe molecules was partially absent from the imidazole-purified (M)L214G RC. Thus, it appears that the (M)214 residue plays a role in the efficient incorporation of pigments into the RC in vivo when solutes which can act as adventitious ligands are present in the growth medium.  A detailed interpretation of these results is presented in the Section 4.1 of the Discussion.  3.9 Structural, spectroscopic, and functional studies of RCs containing zinc-bacteriochlorophyll (Zn-BChl) in place of both BChl and BPhe. The results obtained from Section 3.6 highlighted an important role for residue (M)214 in electron transfer by showing that changes in the protein environment near the photoactive HA cofactor of the RC changed the rate of electron transfer without affecting the cofactor composition. In Sections 3.7 and 3.8, the effects of cofactor substitution and growth in the presence of adventitious ligands on the wild type and (M)L214G mutant RCs are described. In related research, the electron transfer kinetics of RCs containing Zn-BChls (in place of BChls and Bphes) were studied. The initial research demonstrated that an otherwise wild type RC containing six Zn-BChls in place of the four BChls and two Bphes of the wild type RC was capable of charge separation with yields similar to the wild type RC (Lin et al. 2009). This was surprising, especially considering the fact that a Zn-BChl was present as the HA cofactor. The reason this was surprising was because of the observed low charge separation yields of the (M)L214H RC, which, like the Zn-RC studied by Lin et al., also contained a metallo-chlorin at HA (Kirmaier et al. 1995a, b; Kirmaier et al. 1991). Lin et al. postulated that electron transfer through HA is strongly affected by the coordination state, and less so by the presence or absence of a metal 122  center in the chlorin ring. The results from Section 3.6, however, show that the protein component of the RC plays a role in controlling electron transfer rates irrespective of the HA coordination state, because of the kinetic disparity observed between RCs with identical pigment compositions and ligation states. To gain a more comprehensive understanding of the different factors that govern electron transfer rates through HA, I used transient spectroscopy to compare RCs containing six Zn-BChls, in which the coordination state of the Ha Zn-BChl was changed from four to five by changing (M)L214 (denoted the Zn-RC) to His (denoted the Zn-β-RC).  These two types of Zn-RC were compared to the wild type RC and the (M)L214H mutant. Using these comparisons, I attempted to parse the electron transfer effects related to coordination state of the HA cofactor from those related to the presence or absence of a metal center at HA.  From my analysis of the measurements of electron transfer kinetics and the corresponding structural data of Zn-BChl- containing RCs I suggest that the coordination state of a chlorin cofactor in a protein governs the redox midpoint potential, and therefore the electron transfer properties in the RC protein.  3.9.1 Comparison of the wild type RC, the (M)L214H RC, the Zn-RC, and the Zn-β-RC in low temperature absorption spectroscopy. The ground state absorption spectra of the wild type, (M)L214H mutant, Zn-RC and Zn-β-RC at 10 K are shown in Figure 3.18. The spectra were normalized to the amplitude of the Qy band of the dimeric BChls (PA,B). The Qy transition differences in absorption between the Zn-β-RC and the Zn-RC are more subtle than the differences between the (M)L214H mutant and the wild type RC (described in Section 3.3.3). The PA,B band in the Zn-β-RC is slightly broadened relative to the Zn-RC, and the BA,B and HA,B Qy bands of the Zn-β-RC appear to be shifted ~2 nm to the blue. In the Qx 123  transition region of the Zn-β-RC spectrum the conversion of the 4-coordinate Zn-BChl at HA in the Zn-RC to 5-coordinate in the Zn-β-RC is indicated by the increased amplitude of the ~590 nm peak, corresponding to a 5-coordinate Zn-BChl, accompanied by the disappearance of a peak at ~560 nm, where the 4-coordinate Zn-BChl absorbs in the Zn-RC (Lin et al. 2009; Neupane et al. 2012). The HB Zn-BChl content in the Zn-RC appears to be low, because the peak at ~560 nm seems too small to represent contributions from two Zn-BChls (compare this region in the Zn-RC to the wild type RC, which shows two distinct peaks of roughly the same size). In the Zn-β-RC, this the ~560 nm peak is almost completely absent, suggesting that little to no 4-coordinate Zn-BChls are present in this RC. This is in contrast to the (M)L214H mutant, in which the corresponding BPhe peak at ~530 nm is still present despite the absence of one of the BPhe peaks.   124   Figure 3.18. Ground state absorption spectra obtained at 10 K for (A) the wild type (WT, black), and (M)L214H mutant RCs (green), and (B) the Zn-RC (red) and Zn-β RCs (blue). The spectra were normalized to the P865 absorption maximum. Peaks are labeled according to the convention on Figure 1.3. The peak denoted “β” refers to the HA BChl in the (M)L214H mutant. Tetra- and Penta- Zn-BChl peak labels refer to the absorption bands of 4- and 5-coordinate Zn-BChl, respectively The spectrum of the Zn-RC was corrected for scatter by subtracting a straight line from 500 to 730 nm. Figure modified with permission from Saer et al. (Saer et al. 2013b).  125  The bandwidth of PA,B absorption in the Qy transition for both the (M)L214H and Zn-β-RC appears slightly broadened on the shorter wavelength side relative to their non-β counterparts. Therefore, although (M)H214 should not be in the immediate vicinity of the macrocycle component of the dimeric (Zn-)BChls, it appears that this change slightly perturbs this region. The highest resolution crystal structure available for the RC shows that (M)L214 lies close to the phytyl tail of the PA BChl (Koepke et al. 2007). Thus, it appears that this bandwidth increase may be a consequence of a direct interaction between the histidine residue at (M)214 and the PA tail, or an indirect effect. Alternatively, the (M)L214H mutation may induce a perturbation in the membrane-spanning helix (helix D) in which this residue resides. Because this helix also contains residue (M)H202, which acts as the fifth coordinate to PB(Yeates et al. 1988), changes in (M)214 may very well contribute to changes in PA,B Qy absorbance through minor variations of the PB-(M)H202 bond.  3.9.2 X-ray crystallography and anomalous scattering of Zn- and Zn-β-RCs. To elucidate the type, coordination state and occupancy of the bacteriochlorin cofactors in Zn-BChl-containing RCs, the crystal structures of the Zn-RC and Zn-β-RC were solved by X-ray diffraction to a resolution of 2.85 Å. The refinement statistics are listed in Appendix 2. Overall, the structures resemble that of the native RC (PDB entry 2J8C). In both the Zn-RC and Zn-β-RC structures, the iron, carotenoid and all the bacteriochlorin cofactors are present. An electron density map was computed from the anomalous difference data collected above the Zn-edge of the two RC structures to reveal the location of Zn2+ ions within the two structures (Table 3.3 (Bearden 1967)). A peak greater than 4 was observed in the 126  anomalous maps at the center of all bacteriochlorin-type cofactors indicative of the presence of Zn2+ ions at these sites. At the wavelength used for data collection (1.12709 Å), the anomalous signal from Mg2+ is weak and would not significantly contribute to the density observed in the map. The signal from iron can be used as an internal standard, and indeed peaks at 8.7 and 10 were observed at the iron site in the Zn-RC and Zn-β-RC structures, respectively. In the Zn-RC structure, five of the bacteriochlorin cofactors appear to be bound at full occupancy based on refined Zn2+ ion B-factors of 58 Å2 or less and anomalous map peaks >12In contrast, the larger B-factor of the Zn atom at HB site (86 Å2), in combination with the lower anomalous map peak (7), imply a lower occupancy and/or a higher degree of disorder of this cofactor. The HB cofactor in the Zn-β-RC structure exhibits even lower occupancy and/or greater disorder (Table 3.3). These values for the HB cofactor in both the Zn-RC and Zn-β-RC, relative to the values for the other bacterochlorin cofactors, indicate that Zn-BChl is not well-occupied in the HB pocket, and perhaps less well-occupied at this site in the Zn-β-RC than in the Zn-RC. This decrease in occupancy may provide a rationale for the absence of a Qx transition HB peak (Figure 3.17 (Neupane et al. 2012)), by showing that there is some degree of structural perturbation present at the HB site of the Zn-RC and Zn-β-RC. Interestingly, the B-factors of the amino acid residues surrounding the HB cofactor were similar to those in the rest of the protein, indicating that the observed disorder was centered on the Zn-BChl molecule itself, and not the surrounding protein.    127  Table 3.3. Anomalous map statistics for metals in the Zn-RC and Zn-β-RC from x-ray diffraction datasets. Site Zn-RC Zn-β RC  OmitFo-Fc peak () Anomalous map peak () Zn B-factor (Å2) Leu/His-Zn distance (Å) Omit Fo-Fc peak () Anomalous map peak () Zn B-factor (Å2) Leu/His-Zn distance (Å) PA 25 13 50 2.1 25 16 43 2.2 PB 25 13 50 2.1 24 14 44 2.2 BA 25 14 44 2.2 25 15 39 2.3 BB 24 14 39 2.4 25 16 37 2.5 HA 20 12 58 3.2 26 15 43 2.3 HB 13 7 86 3.6 11 4 113 3.9 Fe 24 9 45 -- 24 10 41 --    128   In comparison to the HB site, the BB, PA/B, and A-branch Zn-BChls of the Zn-β-RC and Zn-RC appear to be relatively well-ordered and fully occupied, based on the anomalous signal intensity and B-factor values, which are similar to the mean values of each respective structure (Table 3.3). A side by side comparison of the HA cofactors of the Zn-RC and Zn-β-RC is shown in Figure 3.19. For Zn-BChl in sites with a His residue available for coordination, the imidazole ring is observed 2.1 to 2.4 Å from the Zn2+ ion. At the Zn-RC HA site (with a Leu present), the side chain is separated by 3.2 Å from the Zn2+ ion of the Zn-BChl. Although at 2.85 Å resolution for these structures the error estimate of these distances is ±0.3 Å, at all of the Zn-BChls the electron density maps indicate the absence of an externally-derived metal ligand, such as water. Therefore, I suggest that the major structural difference between the A-branch electron transfer carriers in the Zn-RC and Zn-β-RC is the coordination state of the Zn2+ ion in the HA Zn-BChl, which is 4-coordinate in the Zn-RC and 5-coordinate in the Zn-β-RC.   129   Figure 3.19. Stick models and electron density (mesh) of the HA cofactor and axial (M)214 residues from the crystal structures of the (A) Zn-RC and (B) Zn-β-RC. The distances from the axial residues to the macrocycle centers are 2.1-2.4 Å and 3.9 Å for the Zn-RC and Zn-β-RC, respectively. Color code: teal, carbon atoms; blue, nitrogen atoms; red, oxygen atoms. Figure reprinted with permission from Saer et al. (Saer et al. 2013b).  130  3.9.3 Time-resolved spectroscopy of wild type, (M)L214H, Zn-, and Zn-β-RCs. The fit in the Qy region for the decay of P*, representative of forward electron transfer to the nearby BA cofactor (and subsequently to HA), yielded time constants of 3.8 ps for the wild type RC, 3.9 ps for the Zn-RC, 4.8 ps for the (M)L214H RC, and 5.5 ps for Zn-β-RC (Table 3). Thus, initial electron transfer takes place on the picosecond time scale for all four RCs, although the rates of the wild type RC and Zn-RC group together, separately from the rates of the (M)L214H RC and Zn-β-RC. A table of all the time constants and charge separation yields for the RCs used in this study is presented in Table 3.4.   131  Table 3.4. Calculated transient state lifetimes, reaction rates and final yields of charge separation for RCs in this study based on global analysis. Sample P*  (fast) P+IA- τIQa τrecb ϕc (P+QA-) Wild type 3.8 ± 0.2 ps 210 ± 30 ps 0.2 ns 20 nsd 100 % (M)L214H 4.8 ± 0.5 ps 560 ± 100 ps 4.0 ± 0.7 ns 0.65 ± 0.1 ns 14 % Zn-RC 3.9 ± 0.7 ps 290 ± 50 ps 0.4 ± 0.1 ns 1.1 ± 0.2 ns 74 % Zn-β-RC 5.5 ± 0.2 ps 640 ± 90 ps 6.4 ± 0.9ns 0.71 ± 0.1 ns 10 %  aCalculated rate of forward electron transfer from IA- to QA using Equation 3 in Section 4.6 . bCalculated rate of recombination using Equation 2 in Section 4.6. cCalculated yield of P+QA- based on the amount of P+ bleaching amplitude (measured at 860 nm) at the end of the measurement window compared to 45 ps post excitation. dBased on measurements in which QA was removed from the RC.   132  Transient absorption difference spectra from 500 to 725 nm for all four RCs at 25 ps after excitation are shown in Figure 3.20. The spectra were normalized to the P* bleaching maximum near 600 nm at 0.5 ps. In the wild type RC, this spectrum is representative of the P+HA- state, however, as the other RCs in this study may exhibit different electron transfer kinetics, the spectrum at 25 ps may include features of other RC intermediate states, and thus this state is denoted P+IA- in the mutants (i.e. IA- denotes a combination of HA- and BA- anions). In the Zn-RC, there is an absorption decrease near 560 nm, attributed to bleaching of the Zn-BChl in the HA site because this wavelength corresponds to the Qx absorption maximum of this cofactor in the ground state (Figure 3.18). Like the ground state transition, the magnitude of this bleaching is substantially less than that of the corresponding HA bleaching in the wild type RC at ~545 nm. The difference in Qx transition extinction coefficients between the wild type and Zn-RC may in part account for this amplitude difference, however the exact reason remains unclear. In contrast, the magnitude of absorbance decrease near 600 nm, which represents ground state bleaching of P on this time scale, is nearly identical in the wild type RC and Zn-RC.    133   Figure 3.20. Transient absorption difference spectra of the Qx transition of wild type (WT) (M)L214H, Zn-, and Zn-β-RCs 25 ps after excitation, representative of the P+IA- state. The data were normalized to the Qx bleaching maximum of P* (~ 600 nm) at 0.5 ps post-excitation. Figure reprinted with permission from Saer et al. (Saer et al. 2013b).     134  In the Zn-β-RC there is no ground state absorbance in the 545 to 560 nm region (Figure 3.18), and therefore no ground state bleaching in this region of the difference spectrum (Figure 3.18). However the bleaching at ~600 nm is increased compared to both the wild type RC and Zn-RC. This is consistent with ground state absorption measurements of the Zn-β-RC at low temperatures, which show that the HA Zn-BChl absorbs maximally at 598 nm, overlapping the ground state bands of P, BA and BB (Figure 3.18 (Neupane et al. 2012)). The red shift from ~545 to ~600 nm in both the ground state absorption spectrum and the difference spectrum of the (M)L214H RC results from conversion of BPhe to His-coordinated BChl, whereas in the Zn-β-RC the analogous shift is consistent with a Zn-BChl coordination state change at HA from 4- to 5-coordinate.   A prominent feature of the P+IA- absorption spectrum of all these RCs is the large broadband absorption increase in the ~620 to 725 nm region (Figure 3.20), primarily due to anion absorption in this region. The peak of this absorbance increase shifts from ~660 nm in the wild type RC to ~680 nm in the (M)L214H RC, as shown previously (Kirmaier et al. 1995a). In both the Zn-RC and Zn-β-RC, the portion of this maximum is intermediate in wavelength between that of the wild type RC and the (M)L214H RC. The fact that the two Zn-BChl-containing RCs share a P+IA- absorption maximum at a similar wavelength, and which differs from that of both the wild type RC and the (M)L214H RC indicates that the position of the anion absorption peak is sensitive to both the presence and identity of the metal associated with the HA bacteriochlorin, but insensitive to its coordination state. The absorbance between ~620 and 725 nm associated with the IA-  anion decays in the wild type RC with a time constant of 200 to 250 ps, as shown for the 675 nm wavelength in Figure 3.21, in agreement with previous work (Lin et al. 1996; Lin et al. 2009; Kirmaier et al. 135  1995b; Martin et al. 1986). The decay of the IA- anion band in the Zn-RC has a time constant of ~290 ps (Fig. 3.20, Table 3.4), as seen previously (Lin et al. 2009). In both the (M)L214H mutant and Zn-β RCs, much longer time constants of ~560 ps and ~640 ps, respectively, were resolved from the fitting (Table 3.4). Note that minima of larger uncertainties (+ 50 ps) were found in the fits for the (M)L214H RC and Zn-β-RC compared to the wild type and Zn-RCs. The analysis of this complication will be given in the Discussion section.   136   Figure 3.21. Kinetics of P+IA- state decay at 675 nm for wild type (WT), (M)L214H, Zn-, and Zn-β-RCs measured up to 5500 ps. Spectra were normalized to the maximum amplitude increase at 675 nm. Open circles represent raw data points, and solid lines are representative of fits to the data. Figure reprinted with permission from Saer et al. (Saer et al. 2013b)   137  In Figure 3.22, the fitting results for each type of RC are represented as EADS (van Stokkum et al. 2004). Once again, as in section 3.6, this representation allows simplified visualization of the spectral progression associated with the decay times of the fit, and is not meant to depict the spectra of the actual excited and charge separated state populations (which would require the application of a specific kinetic model). As can be seen, the values of the time constants for the hundreds of picoseconds EADS for the wild type RC and Zn-RC (210 ps and 290 ps, respectively) are more similar to each other than to the time constants found for the (M)L214H RC and the Zn-β-RC (560 ps and 640 ps, respectively). Furthermore, the EADS with the longest time constant in both the wild type RC and Zn-RC was non-decaying (i.e., greater than 3 times the time window of measurement), whereas the (M)L214H RC and the Zn-β-RC did not contain such a long-lived component. Thus, the kinetic behavior of the 4-coordinate Zn-BChl at HA is more like that of a BPhe than a 5-coordinate (Zn-)BChl. All of the mutants, including the Zn-RC, however, appear to have lower P+QA- yields than the wild type RC on the nanosecond time scale. This can be seen in Figure 3.22 by comparing the bleaching of P near 860 nm in the green EADS component that has a lifetime of hundreds of picoseconds, with that of the red (nanosecond, or non-decaying) component. In the wild type RC, there is no significant decrease in the amplitude of the bleaching signal over this time, as would be expected for 100% yield of charge separation. In contrast, each of the mutants shows a partial decrease in bleaching amplitude. In the case of the Zn-RC, this decrease in bleaching amplitude is relatively small (74% of the bleaching amplitude remains; Table 3.3), however this amplitude is still much larger than those of the (M)L214H RC and the Zn-β-RC (14% and 10%, respectively; Table 3.3). In our experiments, the degree of  P+ remaining in the (M)L214H RC at the end of the measurement, as indicated by the difference in bleaching 138  amplitude at 970 nm between 45 ps post excitation and the end of the measurement window, was less than reported previously by Kirmaier et al.(Kirmaier et al. 1995b) because our experiments used a longer measurement window (5.5 ns vs. 2.7 ns), which revealed additional amounts of ground state recombination. Note that there is no evidence for any yield loss on the picosecond time scale in either the wild type or mutant RCs.  This can be seen by the invariance between the blue (few picoseconds) and green (hundreds of picoseconds) EADS near 830 nm in Figure 3.22. At this wavelength, there is essentially no contribution in the picosecond spectrum from stimulated emission, and so one can monitor the decay of ground state bleaching independently.  In summary, the studies on Zn-BChl-containing RCs presented herein highlight their usefulness as an investigative tool in understanding the properties of the RC pigments. The above results show that the presence of a metal center in the HA chlorin results in a shifting of the transient absorption spectrum associated with the P+IA- state. Furthermore, the presence of an HA metal center is associated with RC ground state recombination within a 6 ns measurement window.  Introducing a fifth coordinate to the HA Zn-BChl via site-directed mutagenesis results in an RC in which ground state recombination appears to be the dominant fate of the P+HA- state. In addition to these kinetic differences, the ground state absorbance spectroscopy and x-ray crystallography data suggest that Zn-BChl-containing RC assemble with a partial chlorin occupancy of the HB site.    139   Figure 3.22. Evolution associated difference spectra (EADS) of the Qx and Qy transitions of wild type (WT), (M)L214H, Zn-, and Zn-β-RCs. The solid black line is a zero absorbance reference line. Each RC was fit to a fast < 10 ps component, a hundreds of ps component, and a nanosecond or non-decaying component. Component lifetimes are representative of best fits to the transient absorption spectrum datasets using a least squares fitting approach. Deviations represent the overlap of 5% variations in the χ2 (goodness of fit) values between independent fittings of the Qx and Qy datasets. Figure reprinted with permission from Saer et al. (Saer et al. 2013b).   140  4. DISCUSSION  4.1 Pigment discrimination at HA. The foregoing low temperature spectroscopy data (Section 3.3) support a hypothesis that the presence of BChl or BPhe in the HA pocket is determined solely by the presence or absence of an axial residue capable of Mg2+ coordination. Although it was plausible that the RC HA pocket could de-chelate a BChl Mg2+ ion through steric exclusion via the bulky (M)L214 side chain, the absorbance spectra of RCs containing the (M)L214G/A substitutions, which are neither bulky nor able to coordinate a Mg2+ ion, resemble the wild type spectrum more closely than the β mutant spectrum. Furthermore the crystal structures showed that the bacteriochlorin composition in the P, B and H sites of the (M)L214G/A RCs is the same as in the wild type protein.  In the low temperature spectrum of the (M)L214G mutant (Figure 3.4), however, small decreases in BPhe absorbance are observed for both the Qx and Qy transitions. These data are in agreement with the extraction experiment, which showed a slight increase in the BChl/BPhe ratio for the (M)L214G RC compared to the wild type (~2.6 vs. ~2.0). Thus, it is plausible that some RCs in this population may lack a BPhe cofactor at HA. The data disagree with the (M)L214G X-ray crystallographic dataset, however, because the B-factors associated with the atoms of the BPhe cofactor at HA do not differ from the average of the RC. One possible explanation for this is that the (M)L214G crystal was enriching for RCs containing a BPhe molecule at HA during formation. Experiments aimed at supplementing the (M)L214G RC with excess BPhe may shed light on this matter. In addition to a missing BPhe cofactor, it is possible that the QA cofactor was absent in (M)L214G RC samples because this RC’s recombination kinetics were similar an RC which the QA quinone was 141  removed. This possibility does not seem likely, however, because an incubation of the (M)L214G RC with excess ubiquinone 10 did not significantly change the light minus dark difference spectrum of this RC (Appendix 5).  Our results are complementary to and extend previous studies on "cavity mutants" of the R. sphaeroides RC, in which His axial ligands to the P BChls (at positions (M)202 and (L)173), were changed to Gly residues, and proposed to introduce a cavity in the vicinity of the P BChls. Although crystal structures were not presented, Stark spectroscopy studies and pigment analysis demonstrated that those RCs assemble with a native pigment composition (Goldsmith et al. 1996). Furthermore, water molecules were proposed to act as a fifth ligand to the P BChls in the cavity mutants, and these waters appeared to be switched to other coordinating small molecules by incubation of the RC in the appropriate solute (Goldsmith et al. 1996). The results with mutations at HA enrich our understanding of the complexity in pigment discrimination within RCs, by indicating that not all binding pockets conform to a single set of rules when it comes to populating these cofactor binding sites. Although adventitious ligands such as water may compensate for the loss of an imidazole group in the P BChls, such polar ligands do not appear to be suitable as BChl Mg2+ ligands when the (M)214 Leu side chain is absent from the HA pocket. This difference may stem from the fact that the HA pocket is imbedded more deeply in the membrane bilayer than the P region, and hence less likely to incorporate water (Yeates et al. 1987). Instead, the void is occupied in part by an alternative conformation of the phytyl tail of BA.   An analogous structural plasticity in the P region of the RC was found in "heterodimer" mutations in R. capsulatus in which one of the P BChls was changed to BPhe by substitution of Leu for a His ligand at position (M)200 (the analogous residue in R. 142  sphaeroides is (M)202) of the RC (Bylina and Youvan 1988; Kirmaier et al. 1988). In an R. sphaeroides (M)L214H/(M)H202L double mutant, however, substitution of Leu for a His ligand at this/these position(s) failed to yield a heterodimer, and the resultant RC incorporated two BChls at P for unknown reasons (Heller et al. 1995). Those results demonstrate that even when the coordinating axial (M)H202 residue is substituted with a Leu residue, the P binding pocket is capable of incorporating BChl. Although it is conceivable that an adventitious ligand was incorporated into the binding pocket of the P BChls (in the (M)H202/(M)L214H double mutant), those results raise additional questions about the nature of pigment selectivity in RCs as well as in the biogenesis of (B)Phes.   With respect to pigment selectivity, it would be interesting to investigate how pigment incorporation into the RC is affected by changes in the binding pockets of nearby cofactors, as was demonstrated by Heller et al. (Heller et al. 1995). One possibility is that, because residues (M)202 and (M)214 lie on the same helix, (Yeates et al. 1988) a perturbation of the protein backbone resulting from the combination of the (M)H202/(M)L214H changes may allow for the incorporation of adventitious ligands into the P pocket. The fact that the RC can incorporate BChl in the presence of imidazole also raises the question: how does the presence of these ligands prevent the demetallation of BChl in BPhe binding pockets?    Although the mechanisms underlying BPhe production are unknown, several groups have investigated the production of pheophytin (Phe) from Chl molecules in cyanobacteria and higher plants. The biological production of a Phe from a chlorophyll (Chl) is best understood in senescent plant materials. Shioi and coworkers partially purified a heat-stable “magnesium de-chelating substance”, that catalyzed the conversion of chlorophyllide to 143  pheophorbide (Shioi et al. 1996). The name of this compound was then changed to “metal-chelating substance”, because it was found that this substance was not specific for Mg2+ ions, and was able to chelate other divalent cations (Suzuki and Shioi 1999; Kunieda et al. 2005). Although this as yet poorly defined low molecular weight substance was implicated as functioning during leaf senescence, and therefore Chl degradation, it cannot be excluded that such a substance could function during Chl synthesis, to yield Phe for incorporation into RCs.   Other work on plant material, in this case on the source of Phe in the photosystem II RC (Ignatov and Litvin 1994), pointed to a pre-Chl origin of Phe in which a branch-point in the Chl biosynthetic pathway was proposed as a possible source of Phe in etiolated leaves. The authors suggested that the magnesium chelatase enzyme itself was responsible for Mg2+ removal (Ignatov and Litvin 1994). Such a branch-point has not been reported in the BChl biosynthesis pathway of purple phototrophic bacteria despite extensive mutagenic analyses (Willows and Kriegel 2009). Although it is possible that a dedicated de-chelating enzyme exists for the synthesis of RC (B)Phe , a mutant strain deficient in this activity has not been discovered thus far. Moreover, (B)Phe is functionally present only in RCs, and free (B)Phe is not found in non-senescent photosynthetic organisms. Alternatively, it is possible that the local environment of the (B)Phe-binding pockets within RCs, in general, results in the loss of Mg2+ from (B)Chl unless a coordinating ligand specific for the central metal is present. Because of the tendency for this metal ion to exist only in the penta- or hexacoordinated state in BChl (Callahan and Cotton 1987; Goldsmith et al. 1996; Evans and Katz 1975), it is conceivable that the Mg2+ is removed from the macrocycle in the HA site in part because BChl is simply not stable in the tetracoordinate state (Evans and Katz 1975; Katz 1968).  144   The issue about BPhe’s origin was further explored in my thesis work through experiments involving biochemical pigment substitution in purified RCs. One interesting result derived from the substitution experiments was the capability of the (M)L214G mutant RC to seemingly incorporate metal-containing chlorins. Although it is not entirely clear, the fact that the absorption spectrum of the (M)L214G RC, but not the wild type RC, changed upon substitution with both Chl and BChl supports this conclusion (Figure 3.15). When incubated in a molar excess of BChl a, the (M)L214G RC exhibits a small absorbance decrease at ~765 nm with a concomitant increase in ~775 nm absorbance. Furthermore, the (M)L214G RC demonstrated a large absorbance decrease at ~765 nm when incubated with a molar excess of Chl a. Based on these findings, it appears that a bulky residue at (M)214 contributes to the selectivity (at least in vitro) of BPhe over BChl in the HA pocket when no axial ligand to the chlorin is provided. Because the (M)L214G mutant RC assembles exclusively with BPhe at HA in vivo, it appears that axial ligation of Mg2+ is the most important determinant for BChl incorporation.  4.2 Spectroscopic and structural consequences of nonpolar (M)L214 mutations. An interesting consequence of substituting (M)L214 with small-volume, nonpolar side chain residues in the R. sphaeroides RC is the shifting of the Qx transition corresponding to the HA BPhe to higher energy. It was previously suggested by Bylina et al. that an analogous blue-shifting may be attributed to changes in the hydrogen bonding interaction between a carbonyl group on the HA BPhe macrocycle and a nearby Glu at position (L)104 (Bylina et al. 1988; Yeates et al. 1988). Specifically, when (L)E104 was changed to a weaker proton donor (Gln), or to a Leu incapable of hydrogen-bonding to the BPhe, the Qx absorbance maximum of HA became more similar to that of HB (a shifting to higher energy was found). In the case of a 145  RC (L)E104L mutant, the Qx transitions of HA and HB were almost overlapping. A blue shift similar to what was reported for the (L)E104Q mutant was observed in my (M)L214G and (M)L214A mutants (Figure 3.4B), but the crystal structures show that these shifts are independent of changes in macrocycle hydrogen-bonding interactions. I speculate that the changes in the Qx transition of HA observed in the (M)L214G and (M)L214A mutants are induced by the disorder in the BA phytyl tail, related changes in the binding of detergent, and possibly disordered water molecules.  The X-ray structures of the (M)L214G/A mutants reveal that this residue affects the orientation of the BA phytyl tail (Figure 3.9), and offer an explanation for the changes in the low temperature absorbance spectra of the RCs with these low-volume side chains. Although the influence of the phytyl tail in the direct tuning of the electrochemical midpoint potential of a chlorin is not well understood and considered to be negligible, the tail was thought to affect the packing of protein residues and nearby chlorins (Scheer 2006). Our crystal structures indicate that, other than the substituted side chain itself, there is not a direct effect of the (M)214 side chain substitution on the electronic properties of HA (i.e., there appear to be no changes in the distance between the HA macrocycle and protein atoms that would give rise to changes in the electronic properties of the HA BPhe). Therefore, the absorbance changes in the Qx and Qy transitions of HA appear to stem predominantly from changes in the BA phytyl tail and perhaps the presence of disordered water, as an indirect effect of the change of (M)L214 to smaller, nonpolar moieties. These mutational changes are also accompanied by a broadening of the Qy absorbance band of BA in the (M)L214G mutant (Figure 3.4A), perhaps reflecting a population of mutant RCs that assume a range of alternative BA phytyl tail configurations, as indicated by the crystal structure.  146  Although the X-ray diffraction data from (M)L214A RC crystals indicate that some degree of BA phytyl tail disorder is present (Figure 3.9C), this disorder appears to be less significant than in the (M)L214G RC crystal, because electron density for the “right” conformation is continuous whereas electron density for the alternative conformation is weaker. It could be argued that this difference in electron density arises solely from the lower resolution of the (M)L214A data relative to the (M)L214G structure, but in the absorbance spectra the broadening of the BChl BA Qy band is seen most prominently in the (M)L214G mutant (Figure 3.4A). Although to a lesser extent, the BChl BA Qy band is also broadened in the (M)L214A RC without a corresponding increase in the amplitude of the band (Figure 3.5). In combination with the similarities between the (M)L214A and (M)L214G Qx transitions, these data indicate some disorder or motion in the (M)L214A BA phytyl tail, although it appears to be insufficient to allow the adoption of alternative conformations to the same degree as in the (M)L214G mutant.  The RC contains the electronically active cofactors within well-insulated tunneling pathways that prevent molecules such as intramembrane quinones from interacting with RC pigments and thus potentially affecting rates of electron transfer. In vivo, the RC is surrounded by light-harvesting complex 1 (LH1), which may act as a barrier between the quinone pool and the RC (Bahatyrova et al. 2004; Lilburn et al. 1995; Qian et al. 2008). However, it was shown that 25 to 30% of the native quinone pool is retained in an isolated RC/LH1/PufX preparation (Comayras et al. 2005), evidently because quinones bind to these proteins and may be located within the space between the RC and LH1. In fact, a recent report on the structure of the RC/LH1/PufX core complex revealed that multiple quinones could be accommodated in this space (Qian et al. 2013). In addition to the insulating shield 147  provided by the RC protein matrix, the long, aliphatic phytyl tails esterified to the BChl and BPhe macrocycles may also serve a similar role, at least in vitro. This possibility becomes apparent when the RC crystal structure is viewed parallel to the plane of the membrane, along the side of the A branch, which reveals a cofactor-filled hole (Figure 4.1A). From this perspective, the phytyl tails of the BChls (yellow, green) and BPhes (cyan), as well as the isoprenoid tail of the QA ubiquinone (blue), appear to insulate the macrocycles of BA and HA from the external milieu (Figure  4.1B). However, my low light intensity phototrophic growth experiments indicated that BA tail conformation heterogeneity does not have a great effect on the in vivo efficiency of RC function, because the (M)L214G/A/C mutants grew as well as the wild type control strain (Section 3.4). Although it is plausible that these cofactor hydrophobic tails spontaneously adopt alternative configurations, there has been little account of the mechanisms by which these tails assume their native positions, or what factors might govern their 3-dimensional configurations. However the R. sphaeroides BA phytyl tail “right” configuration is conserved amongst the other three type 2 RC crystal structures (Figure 4.2).  The in vivo efficient functionality of R. sphaeroides RCs containing these structural perturbations (i.e., the wild type-like phototrophic growth of mutants under extremely low intensity illumination) testifies to the robustness of this protein in the catalysis of light-driven excitation energy transduction.  However, kinetic analysis of electron transfer reactions in purified RCs reveals that the (M)L214G RC is significantly impaired in electron transfer from HA to QA, as is the (M)L214A RC to a lesser extent (Pan et al. 2013).   148     Figure 4.1. Organization of the pigment cofactors in crystal structures of the RC. (A) Overview of the wild type RC (PDB 2J8C) with the region of interest enclosed by the green circle. (B) Close-up of the area indicated by the green circle in Panel A; the phytyl and isoprenoid tails of the BChls, BPhe and quinone on the A branch of the RC insulate electronically active pigment macrocycles by acting as a barrier against the external environment. Colour code: Green, PA tail; yellow, BA tail; orange, BA macrocycle; cyan, HA tail; magenta, HA macrocycle; blue, QA tail; gray, protein. (C) View of the equivalent region highlighted in Panel B, but now in the (M)L214G mutant, with the BA phytyl tail modeled in the “left” orientation. The ring 1 acetyl group of the HA macrocycle (magenta) and ring 4 of the BA macrocycle (orange) are exposed to the RC-external milieu. Figure modified with permission from Saer et al. Copyright 2013 American Chemical Society. (Saer et al. 2013a).  149      Figure 4.2. Orientation of the BA phytyl tails in RCs for which high-resolution data are obtainable. A, R. sphaeroides (PDB 2UWV). B, Blastochloris viridis (PDB 3T6D). C, Thermochromatium tepidum (PDB 1EYS). D, Cyanobacterial Photosystem II (PDB 3BZ1). E, the (M)L214G RC. Figure reprinted with permission from Saer et al. Copyright 2013 Amercian Chemical Society. (Saer et al. 2013a).    150  In contrast to the (M)L214G RC, the (M)L214C, (M)L214V and (M)L214I RCs all exhibited red shifts of the HA Qy transition (Figure 3.4A). It appears that the presence of a bulky moiety at the γ position of (M)214 (either SH in the case of Cys, or two methyl groups in the case of Val and Ile) is sufficient to cause a detectable shift. Replacement of (M)L214 with Met has the opposite effect despite presenting a bulky moiety predicted to be at the same position as Leu side chain in the wild type RC. It is possible in the case of the (M)L214M mutant that the blue-shift of the HA Qy transition stems from the establishment of hydrogen bonds between the Met residue SD atom and the pyrrole ring nitrogen protons of the BPhe macrocycle. X-ray crystallography of the (M)L214M RC may shed more light on whether the Met residue and the BPhe are able to hydrogen bond.  4.3 Spectroscopic and structural consequences of polar (M)L214 mutations. The mutant RCs containing Asn and Gln amide side chains at (M)214 incorporated BChl into the HA pocket. These proteins resemble the (M)L214H (-type) mutant RC originally described by Kirmaier et al. (Heller et al. 1995; Kirmaier et al. 1991). In terms of in vivo activity, strains harboring these RCs exhibited similarly impaired phototrophic growth kinetics under a low light intensity (Figure 3.8). However, the absorbance band shifts induced by these three β-type mutations were very different from each other (Figure 3.6). Because the Qy bandhifts of the (M)L214N and (M)L214Q RCs differed from each other more than from the (M)L214H RC (Figure 3.6A), I suggest that the absorbance spectra of HA BChls are not greatly affected by whether the Mg2+-ligating atom is an imidazole nitrogen or an amide oxygen. In these cases, the possibility that the amide nitrogen atom acts as a ligand is excluded due to the participation of the nitrogen’s lone electron pair in a resonance structure of the amide group, the location of the ND1 atom of (M)N214 within H-bonding distance of 151  the main chain carbonyl of (M)Y210, and because of the need for loss of a proton from the amide group to form a Mg2+-N ligand. The side chain of the residue replacing the Leu in the (M)L214N, (M)L214Q, and (M)L214H RCs differs in length and volume. The Asn side chain is shorter than the Gln side chain, but it offers equivalent atoms to coordinate the Mg2+ ion. The (M)L214N and (M)L214Q RC HA Qy absorbance bands differed by about 25 nm (Figure 3.6A), and so the length and/or volume of the Mg2+-coordinating side chain appears to have a major effect on the energy of the HA Qy absorbance band, with the longer Gln side chain resulting in a higher energy HA band in the Qy region. Indeed, studies of BChls in a variety of organic solvents demonstrated that the Qy transition is sensitive to the nature of the axial ligand to the pigment. Specifically, the Qy transition shifts to a higher energy when the Mg2+ ion approaches the plane of the macrocycle due to a stronger metal-macrocycle interaction (this interaction being strongest in an in-plane Mg2+, as is the case in hexacoordinated BChls), and vice versa (Fiedor et al. 2008; Hanson 1988). The distance between the HA Mg2+-coordinating atom in the (M)L214H side chain and the BChl Mg2+ may be intermediate between the distances in the (M)L214N and (M)L214Q RCs; if so, this difference would account for the intermediate position of the HA Qy absorbance band relative the HA Qy bands of the (M)L214N and (M)L214Q RCs, but the coordinates of  the (M)L214H mutant RC are unavailable. The atomic structures of Gln and His show that the Mg-ligating atom and the α carbon of the amino acid are separated by an equal number of bonds, thus a structural dataset of the (M)214H mutant is a must before an accurate comparison can be made.  Based on the available data, I suggest that the distance between the metal-coordinating atom of the RC protein and the Mg2+ ion of (B)Chls, and therefore the position 152  of the Mg2+ relative to the plane of the (B)Chl macrocycle, is the major determinant of the Qy band energy in absorbance spectra, rather than the identity of the coordinating atom (i.e., a His nitrogen atom or an oxygen atom from Asn or Gln). Greater distances of coordinating atoms from the Mg2+, as is the case with Asn, result in a Qy absorbance maximum at a longer wavelength, whereas lesser distances (as with His or Glu) result in correspondingly shorter wavelength of the Qy absorbance peaks.  4.4 Effects of changing the volume of (M)214 on the structure of RCs in vivo. The fact that the pigment composition of the (M)L214G RC is the same as the wild type, in addition to its similar rate of photoheterotrophic growth, led me to investigate the question of whether other phototrophs naturally exist with small, nonpolar moieties axial to BPhe molecules in their RC. One hundred different organisms were chosen by using the R. sphaeroides pufM gene in a BLAST search (excluding R. sphaeroides from the results). From there, a multiple sequence alignment (using Clustal Omega) of the pufM gene from the 100 sequenced organisms was performed (Sievers et al. 2011; Goujon et al. 2010). While not completely conserved among the 100 predicted pufM genes, all but five organisms contained a Leu residue at the position analogous to (M)214 in R. sphaeroides. The five exceptions, Met or Val, are listed in Table 4.1 below.    153   Table 4.1 Organisms that do not encode a Leu at the analogous (M)214 position from a search of 100 genome sequences that contain a pufM homologue Organism Name NCBI Accession No. Residue at position analogous to (M)214 Rhodobacter veldkampii DSM 11550 BAC54030_1 Methionine Erythrobacter sp. NAP1 WP_007164992_1   Valine Porphyrobacter sp. AAP82 WP_017664717_1 Valine Rhodovulum sp. PH10 WP_008383958_1 Valine Lamprocystis purpurea WP_020503997_1 Valine    154  The sequence alignment of 100 representative organisms (Appendix 3) showed that there is a strong conservation of a Leu residue at a position that would be axial to the photoactive BPhe in the RC of various photosynthetic bacteria. When this residue is not Leu, another bulky hydrophobic residue takes its place (i.e., Met or Val). One reason for incorporating a bulky residue at (M)214 could be explained in part by my results in the adventitious ligand experiments on cells grown in the presence of imidazole. In the (M)L214G mutant, purification of the RC in the presence of imidazole resulted in the apparent loss of one of the cofactors, whereas the wild type RC appeared resistant to this effect. Thus, the bulky (M)214 residue appeared to provide a means by which RCs can properly assemble in the presence of solutes such as imidazole, which otherwise may act as adventitious ligands to BChl. Although it is unlikely that these solutes are present in high concentrations in laboratory growth media (unless deliberately added), one cannot exclude the presence of such substances in the habitats in which these bacteria live and grow. High-throughput analyses (i.e., mass spectrometry) of the prevalence of different solutes found in common habitats of purple bacteria could provide a clearer picture on this matter.  4.5 In vitro performance of RCs with nonpolar moieties at HA. The results described in section 3.6 of this thesis demonstrate that the (M)214 residue plays a role in modulating the rate of electron transfer through the HA cofactor. In summary, the primary events of charge separation leading to the formation of the P+HA- (or P+IA-) state are not significantly altered in any of the mutants. This is in contrast to the (M)L214H mutant, in which the formation of P* → P+IA- takes approximately twice as long as the WT RC (~ 6 ps in the (M)L214H mutant (Kirmaier et al. 1995b)). Thus, the formation of P+HA- appears to be affected primarily by the presence or absence of a pentacoordinate metal center in the HA 155  cofactor. The fate of the P+HA- state, however, appears to be highly dependent on the volume of the (M)214 side chain. The (M)L214M mutant, with a side chain similar in bulkiness to the wild type Leu residue, exhibits efficient HA to QA electron transfer dynamics, whereas the other mutant RCs with smaller side chains at (M)214 do not. In fact, the (M)L214G/A/C mutants appear to contain traces of the P+HA- state even ~6 ns after the excitation flash. This long-lived state is associated with recombination of P+HA- to the ground state, resulting in a partial loss of charge separation among a given population of RCs, as indicated by the data in Figures 3.13 and 3.14, where recovery of P band bleaching is evident in mutants with long-lived 675 nm P+HA- transients.  A fitting of the time-resolved spectral datasets was performed as outlined in Pan et al (Pan et al. 2013). These data were fitted to a set of exponential decays using a sequential model (i.e., A → B → C). The resultant EADS obtained from these fits, in addition to a set of proposed models to explain the kinetic complexities (explained further in the following paragraph), are shown in Figure 4.3.    156   Figure 4.3. Evolution associated difference spectra (EADS) and kinetic models of the wild type (WT) and (M)L214G/A/C RCs. Middle panels: EADS obtained from global fitting analysis of time-resolved spectral datasets for the WT and (M)L214G/A/C RCs. Panels on the left column represent reaction schemes of the WT and mutant RCs based on a static heterogeneity model. Panels on the right column represent reaction schemes based on a dynamic heterogeneity model. Solid black arrows represent dominant reaction pathways, whereas solid gray arrows represent pathways that are unlikely to be a dominant one. Wavy arrows indicate P+HA- relaxation. Orange numbers represent time constants from the fits, as well as the calculated yields for each measured RC. Black numbers in the reaction schemes represent recombination rates obtained from the literature. Figure reprinted with permission from Pan et al. Copyright 2013 Amercian Chemical Society. (Pan et al. 2013).   157   The fitting results revealed the presence of extra decay components in the (M)L214G/A/C mutant RCs compared to the  wild type RC. In all of the RCs studied, all of three components were required to explain the fits to the experimental data during the initial phase of charge separation (i.e., P* → P+BA- → P+HA-). As mentioned in the Introduction (Section 1.4), fast components of 2.8 to 3.5, and 0.9 ps were resolved in all the RCs. These two components correspond to an electron transfer from P* to BA, and subsequently from BA- to HA, generating the P+HA- state. The third component, with a lifetime of 10 to 12 ps, arises from a slower decay phase of the P* state, as has been discovered in RCs of R. sphaeroides, R. capsulatus as well as Blastochloris viridis (Holzwarth and Muller 1996; Huppman et al. 2002; Jia et al. 1993; van Stokkum et al. 1997; Lin et al. 1996). The nature of this slower decay phase has been previously suggested to be charge separation occurring from a more vibrationally relaxed state of P*, however, this proposal has never been examined in closer detail (Lin et al. 1996). As for the two faster phases, all the (M)L214 mutant RCs show evidence of this decay phase with a similar lifetime. In addition to these three components, all RCs exhibit a slower, hundreds of picoseconds decay component which is characteristic of electron transfer from HA- to either QA, or back to P+ (recombination; Figure 4.3e-h). This decay rate appears to be correlated with the volume of the side chain at (M)214, thus acting as a kinetic branching point at which the behavior of the (M)L214 mutants begins to differ. The lifetime of this rate increases as the volume of the (M)214 side chain decreases. The time constants of the hundreds of picoseconds components were calculated to be 220 ps (for the wild type RC), 250 ps ((M)L214C), 290 ps ((M)L214A), and 330 ps ((M)L214G). Furthermore, a fourth decay component in the nanosecond time domain was required to adequately fit the datasets for the (M)L214G/A/C mutants (Figure 4.3f-h). Values of 6 to 10 158  ns (for the (M)L214C RC), 6 to 8 ns ((M)L214A), and 4 to 6 ns ((M)L214G) were obtained for this component,. In the (M)L214C and (M)L214A RC, this component is characterized by a slower electron transfer event from HA- to QA, as the 675 nm transient signal decreased in amplitude with no appreciable decrease in P+ bleaching amplitude. In the (M)L214G mutant, however, there appears to be some degree of recombination to the ground state within this time domain, suggesting that the recombination reaction occurs with a faster rate in this RC than the wild type, (M)L214A, or (M)L214C RC. Finally, all RCs exhibited a non-decaying component with a lifetime greater than 3 times the time window of the experiment. In the wild type RC, this component contains no trace of the HA- signal at 675 nm, however the P+ bleaching at 865 nm is still present, implying that this component is dominated by RCs in the P+QA- state. In the (M)L214G/A/C RCs, the non-decaying component spectrum contains traces of the HA- anion signal, as well as decreases in 865 nm bleaching amplitude. The degree of recombination, as evidenced by the loss of negative amplitude at 865 nm, is once again correlated with the volume of the side chain at (M)214, with the (M)L214G mutant showing the most pronounced amplitude loss. Thus, RCs with smaller (M)214 side chains are more likely to exhibit P+HA- → ground state recombination from a state of P+HA that is relatively long-lived compared to that of the wild type RC.   To calculate the population of P+HA- and P+QA- in the EADS for the measured RCs, the spectra were fitted to a linear combination of the P+HA- and P+QA- spectra from the wild type RC using the expression EADS = h[P+HA-] + q[P+QA-], where h and q are values that specify the amount of P+HA- and P+QA- in the EADS, respectively (Pan et al. 2013). For this calculation, it was assumed that there was a negligible amount of P+QA- present in the 159  hundreds of picoseconds EADS. The fits of the hundreds of picoseconds EADS from the (M)L214G/A/C RCs to that of the wild type RC are shown in Figure 4.4.    160            Figure 4.4. Overlay of the hundreds of picoseconds EADS for the (M)L214G, (M)L214A, and (M)L214C RCs (black dotted lines) with the 220 ps EADS of the wild type RC (gray solid lines). Figure reprinted with permission from Pan et al. Copyright 2013 Amercian Chemical Society. (Pan et al. 2013).   161  Fitting the hundreds of picoseconds EADS from the (M)L214C and (M)L214A RCs to that of the wild type RC resulted in deviations (i.e. a difference in amplitude) of less than 5% from each other. Thus, it was assumed that these two RCs contain 100% P+HA- after the initial steps of ET. In the (M)L214G mutant, the deviation was approximately 85%. Because the hundreds of picoseconds EADS of the (M)L214G RC exhibits a larger bleaching amplitude at ~600 nm, the remaining 15% deviation was assigned to P+BA-. Application of this approach to the nanosecond-scale EADS revealed that the three mutant RCs contained mixed population of P+HA- and P+QA- (compared to a ~100% population of P+QA- in the wild type RC). In the (M)L214G RC, it appears that there is some evidence of recombination to the ground state, because it was found that 70% of the population of RCs was in the P+HA- state, and 10% were in the P+QA- state. For the non-decaying EADS of the mutant RCs, the populations were fitted to a linear combination of the 220 ps and non-decaying EADS of the wild type RC. Note that the values of h and q do not necessarily add up to 1 if recombination of P+HA- to the ground state in a sufficiently large fraction of the RC population has occurred. The results of h and q for the hundreds of picoseconds, nanosecond, and non-decaying EADS are compiled in Table 4.2.   162   Table 4.2. Populations (in %) of charge-separated states in (M)L214 mutant RCs derived from linear combinations of the wild type hundreds of picoseconds and non-decaying EADS.  Wild type (M)L214G (M)L214A (M)L214C  P+HA- P+QA- P+BA- P+HA- P+QA- P+HA- P+QA- P+HA- P+QA- hundreds of picoseconds 100 0 15 85 0 100 0 100 0 nanosecond NA* NA* 5 68 10 79 19 48 52 non-decaying 0 100 0 40 10 36 24 14 62  * No nanosecond EADS was obtained for the WT RC.    163  From the population values obtained in Table 4.2, in addition to the X-ray crystallography results, two reaction schemes, shown in Figure 4.3, were devised to explain the kinetic heterogeneity present in the (M)L214 mutants. Because the kinetic differences between the mutant RCs begin to be manifested once the P+HA- state is reached, the two models are displayed as beginning from this state.   In a static model (Figure 4.3a-d), it is assumed that different populations of RCs in the same sample are behaving differently. For the (M)L214A and (M)L214C mutants, the application of this model would suggest that one population of RCs is behaving like the wild type RC with fast forward ET from HA- to QA. Two other populations are associated with a long-lived P+HA- state. One of these populations decays in 15 to 20 ns solely via recombination to the ground state, and is represented by the P+HA- contributions to the non-decaying EADS. A third population of RCs decays from the P+HA- state on the nanosecond time scale via two possible pathways. In one pathway, the RCs recombine to the ground state in 10 ns. In the other pathway P+HA- decays via a slow phase of forward ET, generating P+QA-. For the calculation of the time constants in Figure 4.3b-d, the time constant of the appropriate EADS obtained in the fitting was divided by the calculated population of P+HA- in the EADS. As an example, I explain in the following paragraph the calculation of the time constant for slow forward electron transfer in the (M)L214C mutant from P+HA-, yielding P+QA-.   To determine the rate of slow forward electron transfer from P+HA- to QA for the (M)L214C mutant (with a time constant of 24 ns), the 6 to 10 ns time constant for the nanosecond EADS was divided by a fraction of the subpopulation of RCs undergoing the slow phase of HA- to QA electron transfer. When moving from the nanosecond to non-164  decaying EADS for the (M)L214C mutant, it is evident that 10% of the RCs in the entire measured sample have entered the P+QA- state anew (Table 4.2). However, in this static model, 3 different populations of RC behave according to different kinetic parameters, and so we need to find the fraction of the subpopulation that this 10% of the entire sample represents. Because this subpopulation would be included in the fraction of RCs that are in the P+HA- state in the nanosecond EADS, but not in the non-decaying EADS, we can look for the P+HA- populations that differ between these two EADS (Table 4.2) to find the fraction of RCs that this subpopulation represents; namely, 48% - 14% = 34% (35% when rounded to the nearest 5%). Thus we have 6 to 10 ns divided by (10 / 35) yielding 21 to 35 ns (the 24 ns value in Figure 4.3b was a result of using the least squares fitting value, however error in the fit produces the 21 to 35 ns range of values). Similarly, for the fraction of this subpopulation that recombines to the ground state, we can calculate the time constant as 6 to 10 ns divided by (25 / 35) = 8 to 14 ns. These calculations demonstrate that: (i) the (M)L214C RC contains a higher fraction of proteins capable of fast forward electron transfer from HA- to QA when compared to the (M)L214A mutant; (ii) the (M)L214C mutant is associated with fewer RCs that recombine to the ground state via the slower (15 to 20 ns) decay phase of P+HA-; and (iii) the (M)L214C mutant P+HA- is also more likely to decay via slow forward electron transfer to QA than the (M)L214A mutant among the population of RCs with P+HA- states that decay in the nanosecond time domain (note that in both the (M)L214C and (M)L214A RCs, ground state recombination is still the dominant decay pathway for this subpopulation).  Using the static model on the (M)L214G mutant results in a total of four different subpopulations exhibiting kinetic heterogeneity. Unlike the (M)L214C and (M)L214A mutants, this RC lacks the slow decay phase of P+HA- to P+QA- (Figure 4.3h). In this RC, sub-165  nanosecond electron transfer is the only means of achieving the P+QA- state, and all other long-lived subpopulations of P+HA- return to the ground state. In addition, there appears to be a subpopulation of (M)L214G RCs that exhibit a relatively rapid rate of ground state recovery, as evidenced by the small decrease in P+ bleaching amplitude in the nanosecond EADS. Because this RC is the only one of the tested mutants exhibiting P+BA- character in the hundreds of picoseconds EADS (Table 4.2), this subpopulation of RCs was assigned to those which decay to the ground state via P+BA-.   A comparison of the kinetics of the three mutant RCs using the static model reveals certain key points with regard to the volume of the (M)214 side chain. Firstly, it appears that the rate of forward electron transfer from P+HA- to QA in the hundreds of picosecond domain decreases as (M)214 side chain volume increases. Secondly, substitution of (M)214 with small, nonpolar moieties (cysteine included) results in two subpopulations of RCs containing a nanosecond-long P+HA- state. One of these populations behaves similarly among all RCs, and presumably exhibits similar kinetics to QA-depleted RCs, in which the RC recombines to the ground state in ~15 to 20 ns. The second of these subpopulations decays mainly via recombination to the ground state, although it is also capable of forward electron transfer to QA in the nanosecond time domain if the (M)214 side chain is a methyl group (Ala) or larger, with larger side chains exhibiting a higher degree of forward electron transfer. Combining this model with the x-ray crystallography data of the (M)L214G and (M)L214A RCs, the data indicate that the structural disorder in the BA phytyl tail of RC mutants with smaller (M)214 side chains gives rise to populations of RCs that cannot efficiently undergo HA- to QA- electron transfer. Exactly how much of this kinetic complexity can be explained by the 166  static model, as well as the underlying mechanism by which structural differences in BA give rise to this complexity, is not clear.  In Figure 4.3i-l, the (M)L214C/A/G RCs are modeled in terms of dynamic heterogeneity. In a dynamic model, the different kinetics are ascribed to a relaxation in the energy of the P+HA- charge-separated state on the hundreds of picoseconds to nanosecond time scale. The application of this model arises from previous transient absorbance studies which suggested that the speed of an electron transfer event in the RC is faster than the energetic relaxation of charge-separated intermediates (i.e., the higher energies of the transient excited states and anionic radicals in the RC do not fully relax before electron transfer takes place (Jia et al. 1995; Lin et al. 1996)) The general assumption of this model is that the P+HA- initially formed from P* (termed (P+HA-)i) can relax to form two different intermediate states in the mutant RCs: (P+HA-)m and (P+HA-)f. In the wild type RC, only (P+HA-)i is formed, and it undergoes fast forward electron transfer to QA. In the mutants, however, the rate of electron transfer to QA is slow enough that relaxation to (P+HA-)m and (P+HA-)f compete with forward electron transfer, generating significant populations of these more relaxed charge-separated states. Electron transfer from these relaxed charge-separated states is associated with both a slower rate of forward electron transfer to QA, as well as an increased rate of recombination to the ground state. In the dynamic model (Figure 4.3i-l), the yield losses for recombination from (P+HA-)m and (P+HA-)f  are calculated by measuring the decrease in P+ bleaching amplitude when transitioning from the hundreds of picoseconds EADS to the nanosecond EADS (for (P+HA-)m →  ground state), and from the nanosecond to the non-decaying EADS (for (P+HA-)f →  ground state). The different populations of mutant RCs assigned to the charge-separated states were estimated by calculating the amount of 167  P+HA- relative to P+QA- for each EADS in the same fashion as in the static model. According to the dynamic model, the substitution of (M)L214 with smaller, nonpolar moieties results in slower HA- →  QA electron transfer, with smaller volume side chains associated with slower rates of forward electron transfer. These slowed electron transfer rates allow P+HA- to relax to lower energy states, resulting in recombination to the ground state, as well a relatively slow phase of forward electron transfer from HA- to QA.  The two models summarized above are useful in accounting for the observed kinetic heterogeneity in the (M)L214 mutant RCs, although the genuine behavior of these RCs is likely to borrow aspects of both models. Large molecular movements, such as the BA tail conformers observed in the crystal structure of the (M)L214G mutant, may give rise to static heterogeneity among RC populations, as these types of movements would occur much more slowly than the electron transfer events studied here. In addition, the observed lack of continuous electron density at the BA tail in the (M)L214G and (M)L214A structures suggests that more rapid and disordered movements also occur within these proteins, and may influence electron transfer rates on the picosecond and nanosecond timescales. This is because the rapid tail movements may either cause small displacements the HA macrocycle, or they may favor electron transfer reactions through alternate tunneling pathways which are associated with slower rates of electron transfer.  Furthermore, although the (M)214 mutations have little effect on the energetics of the HA pigment in the ground state, these mutations may give rise to opportunities for cofactor relaxation when entering an anionic state, giving rise to changes in electron transfer rates once P+HA- is attained, as the dynamic model indicates. Whatever the underlying reason accounting for these changes in electron transfer rates, it appears that there is a correlation between the volume of the (M)214 residue, 168  the degree of structural disorder in the BA tail, and the rate of electron transfer from HA to QA. 4.6 Functional analysis of Zn- and Zn-β RCs. I exploited Zn-BChl-containing RCs to evaluate the effects of changing the HA cofactor coordination state on electron transfer in the RC. This comparison would be impossible in a (Mg2+-) BChl-containing RC because the HA pocket contains BPhe when a fifth ligand is absent (Saer et al. 2013a). Evidently the Zn2+ ion is more tightly bound to the bacteriochlorin macrocycle than Mg2+, and so Zn-BChl is resistant to whatever mechanism is responsible for the dechelation of BChl to yield BPhe in the HA pocket (Kobayashi et al. 1998; Saer et al. 2013a; Jaschke and Beatty 2007). The kinetic consequences of substituting Zn-BChl into the RC in an otherwise wild type background were previously studied, and it was suggested that the Zn-RC functions similarly to the wild type RC (Lin et al. 2009). However, the EADS data (Figure 3.21) obtained as part of this thesis research indicate a slower rate of electron transfer from HA to QA for the Zn-RC. The EADS with a hundreds of picoseconds time constant for the Zn-RC fitted best to ~290 ps, compared to ~210 ps in the wild type RC. The difference between our experimental values and those previously obtained by Lin et al. (Lin et al. 2009) likely stems from our use of a broadband detection setup, in contrast to a single wavelength detector. Furthermore, the final charge separation yields of the Zn-RC in our work were also lower, with evidence of some P+IA- recombination in the non-decaying EADS.  In this respect, the Zn-RC is more similar to the (M)L214H mutant and Zn-β-RC because the presence of ground state recombination on this time scale indicates that the rate of recombination from P+IA- is increased.  The rate of recombination in the Zn-RC was thus calculated based on the 169  observed lifetime of the P+IA- state and the amount of P+ remaining at the end of the time window of the experiment according to the formula:  τrec = τobs/((100 – ϕ)/100)     [2] where τrec is the rate of recombination, τobs is the P+IA- lifetime, and ϕ is the final yield of P+ (Kirmaier et al. 1995b). For the Zn-RC, this yielded a value of 1.1 ns, a value closer to that of the (M)L214H mutant and Zn-β-RC (651 ps and 711 ps, respectively) than the 20 ns value obtained for the wild type RC from which QA had been removed (Table 3.3).(Parson et al. 1975)   The relatively long-lived hundreds of picoseconds EADS obtained for the (M)L214H mutant from our datasets compared to previously published P+IA- lifetimes suggests a degree of heterogeneity in the P+HA- decay (Kirmaier et al. 1995b; Pawlowicz et al. 2009). When the transient absorption datasets were truncated from 6 ns to a shorter time window (2 ns), the P+HA- decay rates were found to be closer to the previously reported rates. The slower rates observed in the (M)L214H mutant and the Zn-β-RC likely reflect more complex kinetic heterogeneity than can be described by four exponential components, reflecting an average of forward electron transfer as well as recombination to the ground state in RCs in the P+IA- state. A previous analysis of the kinetics of the (M)L214H mutant in which electron transfer to QA was blocked demonstrated that the recombination rate from P+I- was approximately 1 ns (compared to ~20 ns in the wild type RC (Kirmaier et al. 1995b). Furthermore, the calculated charge recombination rates in the (M)L214H, and Zn-β-RC indicate that charge recombination occurs much faster than in the wild type RC. This difference likely contributes to the relatively shallow fitting minima obtained in the (M)L214H mutant and Zn-β-RC for 170  this EADS (compare the deviations between the wild type RC/Zn-RC and (M)L214H mutant/Zn-β-RC in the green spectra of Figure 3.22). Because the Zn-RC also has a more rapid rate rate of charge recombination than the wild type RC, I suggest that the presence of the metal center at HA in the Zn-RC results in an HA- anion that is closer in energy to P+BA- compared to the wild type RC. This suggestion is consistent with previous mutagenic studies on RCs in which a strong correlation was found between the rate of P+HA- recombination and the similarity of BA- and HA- energies (Wang et al. 2012). Clearly, this energetic proximity of HA- to BA- is less in the Zn-RC than in the (M)L214H mutant and Zn-β-RC, because the P+HA- recombination rate of the Zn-RC is relatively slow.  In addition to the rate of recombination, it was also possible to extract information on the rates of forward electron transfer from IA- to QA, by taking into account the lifetime of the hundreds of picoseconds EADS, as well as the final P+QA- yield of each RC.  The rate of electron transfer was calculated according to the equation: τQT = τobs/(ϕ/100)                                                            [3] where τQT is the rate of electron transfer, τobs is the P+IA- lifetime, and ϕ is the final yield of P+ (Kirmaier et al. 1995b). The values of τQT are 0.2 ns for the wild type RC, 4 ns ((M)L214H mutant), 0.4 ns (Zn-RC), and 6.4 ns (Zn-β-RC). From these data, it appears that the presence of a fifth coordinate at the HA metal center is associated with a much reduced rate of forward electron transfer from IA- to QA.  The greater similarity of Zn-RC kinetic behavior to that of the wild type RC, than to the (M)L214H mutant and Zn-β-RC, evidently is because the HA cofactor is a 4-coordinate Zn-BChl. This indicates that the five-coordinate center present in the latter two RCs is the 171  key player in the observed low yields of charge separation, and that the presence or absence of a metal in the HA bacterochlorin is of less consequence. One explanation for the observed results stems from the Qx transition energies of the HA cofactors among these RCs. Previous studies on metal-substituted bacteriochlorins in vitro have reported a linear correlation between Qx transition energies, redox potential, and electronegativity (Pauling values) of the bacteriochlorin metal center (Hartwich et al. 1998; Noy et al. 1998). I suggest that the 4-coordinate Zn-BChl at HA functions similarly to a BPhe molecule because the Zn-BChl transition energy, redox potential, and central ion electronegativity are closer to that of BPhe (as indicated by the Qx aborption peak) than a 5-coordinate BChl or Zn-BChl. In the case of central ion electronegativity, although the Pauling value for Zn2+ (~ 1.6) is closer to Mg2+ (~ 1.3) than H (~ 2), this value is similar to both and lies between Mg2+ and H. Additionally, an increase in the coordination state of the metal center of a bacterochlorin has been associated with a lower electronegativity value for the metal (Hartwich et al. 1998). Thus, a 5-coordinate Zn-BChl should be less electronegative than both the 4-coordinate counterpart and BPhe.  In addition to the observed kinetic differences upon entering the P+HA- state, the presence of a metal at HA imbues different spectroscopic properties on the IA- anion. The transient absorption spectra of the P+IA- state for Zn-BChl-containing RCs are more similar to each other (and the (M)L214H mutant) than the wild type RC in the 650 to 730 nm region. As noted in the Results (section 3.9.3), the wild type RC has an absorption maximum at higher energy in this region compared to the other 3 RCs. Thus, the red-shifted peak in these three RCs appears to be correlated with the presence of a metal at HA, regardless of coordination state or the type of metal present.  172  Interestingly, the RC of the photosynthetic bacterium Acidiphilium rubrum contains Zn-BChl in place of BChl, but this organism contains BPhe at HA. In this organism, Zn-BChl is presumably synthesized by Mg-containing BChl synthesis intermediates (Masuda et al. 1999). After the final step of BChl synthesis, it is thought that the low environmental pH promotes pheophytinization BChl, at which point a Zn2+ atom incorporates into the chlorin macrocycle by an unknown mechanism (Masuda et al. 1999). This theoretically provides a pool of both Zn-BChl and BPhe with which the apo-RC can selectively incorporate. This mechanism of Zn-BChl biosynthesis is different from that of the R. sphaeroides bchD mutant, because this mutant utilizes Zn-containing synthesis intermediates. Although electron transfer has been studied only to the P+HA- state in A. rubrum (Tomi et al. 2007), we speculate that in this RC the rate of electron transfer to QA is like that in the wild type RC. Based on our results, it appears that natural selection would favor the presence of BPhe at HA in evolution because it results in a much slower decay of P+IA- via charge recombination. 4.7 Consequences of the absence of Mg-BChl in RCs. In the ground state absorption spectra of the Zn-RC, as well as in previous low temperature measurements, a single peak was present in the Qx region near 560 nm, originally attributed to overlapping absorption of 4-coordinate Zn-BChls in the HA,B pockets (Neupane et al. 2012; Lin et al. 2009). However, as reported by Neupane et al. (Neupane et al. 2012) and confirmed here, the Zn-β-RC has little or no absorption around 560 nm, whereas it would be thought that only the component due to the HA Zn-BChl would be shifted as a result of the (M) L214H mutation, as is seen when this mutation is introduced into the wild type RC (Figure 3.18). That is, a ~50% decrease in the ~560 nm peak amplitude was expected, whereas a nearly complete loss of this peak was found. Neupane et al. offered three possible explanations for the absence of a 173  defined HB peak in the Qx region of the Zn-β RC: (i) the Zn-RC and the Zn-β-RC lack an HB cofactor; (ii) the Zn-β-RC contains two penta-coordinated Zn-BChls at HA,B; (iii) absorption intensity is being redistributed among the bands of the cofactors (a phenomenon known as intensity borrowing .(Orlandi and Siebrand 1973; Neupane et al. 2012)). The Zn-RC and Zn-β-RC crystal structures show that, relative to the wild type RC, the Zn-BChl in the HB pocket of the Zn-RC is bound at less than full occupancy or is highly disordered, and perhaps the occupancy is even lower or disorder greater in the Zn-β-RC (Table 3.3). These conclusions are based on weaker density in omit difference maps of the HB Zn-BChl, the large B-factors associated with the Zn2+ in the HB pocket, and the low anomalous signal intensities for that Zn2+ atom (compare the values for HB to those of the other cofactors in Table 3.3). Interestingly, the average B-factor of the HB BPhe from the highest resolution RC structure currently available in the PDB (2J8C) database is greater than all of the other cofactors (Koepke et al. 2007), perhaps indicating weaker binding to this pocket in the wild type RC as well. As mentioned in Section 3.9.2, the B-factor for the protein component surrounding the HB Zn-BChl is comparable to the average of the structure, consistent with the idea that the RC is a relatively rigid protein, and does not rearrange its folding due to changes in cofactor composition or even when a cavity is created by the lack of a cofactor, as was observed in pigment exclusion experiments on the HB BPhe (Jones 2008; Watson et al. 2005a). The Qx larger spectral bandwidth of the HB cofactor compared to HA also supports a greater HB site heterogeneity (Figure 3.18). Additional evidence for relatively weak binding of BPhe in the wild type RC HB pocket comes from experiments involving substitution of BPhe with plant pheophytin (as well as a variety of other pigments). These experiments demonstrated that substitution is preferential in the HB over the HA site, consistent with a weaker binding of 174  BPhe in the HB pocket (Meyer and Scheer 1995; Franken et al. 1997). In vitro, the binding pockets for HA,B are selective for pheophytins, as incubation of RCs in a molar excess of metal-containing bacteriochlorins (the standard procedure for pigment exchange) resulted in substitution of pigments at the BA,B binding pockets instead of HA,B (Struck and Scheer 1990; Scheer and Hartwich 1995; Jones 2008). The greater hydrophobicity of BPhe over BChl and Zn-BChl, as indicated by the longer retention time for BPhe in reversed-phase HPLC using a C18 column (Jaschke and Beatty 2007), may be part of the reason why binding of metal-containing bacteriochlorins to the HA,B pockets is weak, although it is not clear why the HB Zn-BChl may be lost more readily or less efficiently inserted into the RC than the HA Zn-BChl. Alternatively, it is possible that the Zn-RCs contain partially occupied HB sites due to a lack of a sufficient quantity of available Zn-BChls in the cell. One important consequence of the bchD mutation yielding Zn-BChl-producing strains of R. sphaeroides is the severely impaired ability to synthesize the pigment (Jaschke and Beatty 2007; Jaschke et al. 2011). Therefore the inability of the organism to grow phototrophically is because of a paucity in the cellular content of light-harvesting and RC complexes (Jaschke and Beatty 2007).  Unlike the HB cofactors of the Zn-RC and Zn-β-RC, the HA cofactors appear less disordered, suggesting there is no issue with the occupancy of the pocket. From a structural perspective, the HA cofactors appear to assume a similar configuration in the two Zn-BChl-containing RCs (Fig. 3.19), with a possible macrocycle displacement of 0.6 Å between the two RCs. One cannot reach a firm conclusion on this geometric displacement between the cofactors because of the poor resolution of the data obtained for these two RCs; however the small degree of displacement is consistent with the interpretation that the differences in the 175  functional characteristics of the two Zn-BChl-containing RCs stem predominantly from changes in HA cofactor coordination state.  4.8 Biological implications of unnatural cofactor incorporation. Our time-resolved data demonstrate that in all 4 of these RCs P+IA- is achieved rapidly (within a few ps), and that differences between RCs become apparent after this state is attained. The decay kinetics of the P+HA- transient at 675 nm indicate that this state is relatively long-lived in the Zn-β-RC, as well as in the (M)L214H mutant, in agreement with previous results (Kirmaier et al. 1991; Kirmaier et al. 1995a, b). Furthermore, the global analysis indicates a similar degree of special pair charge recombination associated with this long-lived P+HA- state in the Zn-β-RC, much like in the (M)L214H mutant. The charge separation yields are consistent with this interpretation (i.e., the (M)L214H mutant and the Zn-β-RC yields were 14% and 10% P+QA-, respectively, whereas the wild type RC and Zn-RC yields were 100% and 74% P+QA-, respectively). These results indicate that RCs have a degree of functional tolerance for different cofactors, and that the protein component is a key instrument in the design of an efficient charge separation pathway. The Zn-RC studied here is a case in point, as it contains a cofactor arrangement supportive of a faster rate of ground state recombination from the P+IA- state, but nevertheless is capable of attaining a relatively high charge separation efficiency. This likely is because the protein scaffold of the Zn-RC has remained unchanged from the wild type RC counterpart, granting this mutant RC the same favorable cofactor geometries and tunneling pathways available in the wild type RC. From an evolutionary perspective, this could be advantageous to photosynthetic organisms during periods of environmental change, such as a change in the wavelengths of light available, for it would allow for the continued utilization of existing protein machinery when the production of a 176  new type of (bacterio)chlorin would provide a selective advantage. In such a case, the new pigment could be incorporated into a protein “template” with suitable geometries, which would grant the organism some degree of charge separation capability, enabling photosynthetic growth. Mutations in protein side chains may later provide further energetic optimization, resulting in a high-efficiency RC that assembles with a new pigment. For this to happen, however, the photosynthetic organism must be able to survive for some time with an inefficient RC to allow beneficial mutations to accumulate. The results from Section 3.4 show that LH2-lacking R. sphaeroides strains containing RCs with small-nonpolar side chains at HA retain their ability to grow at low photon fluxes, despite the relatively low efficiencies of these RCs (compared to the wild type RC), indicating that purple photoheterotrophs are indeed quite capable of surviving with lower efficiency RCs.  4.9 Why do natural type II RCs contain a BPhe in the HA or analogous pocket, regardless of the presence of Mg2+ versus Zn2+ in their (B)Chls? The type II RC found in many anoxygenic phototrophs, as well as all species of cyanobacteria and photosynthetic eukaryotes, contains a (B)Phe molecule as an intermediate electron acceptor followed by a quinone molecule (Blankenship 2010). The use of this pigment is rather interesting, because type I RCs such as photosystem I of oxygenic phototrophs and the RCs of green sulfur bacteria and heliobacteria lack such a pigment, yet function with high quantum efficiencies (Chauvet et al. 2013). Why, then, has evolution chosen pheophytins as intermediate electron acceptors in type II RCs when other RCs that don’t require pheophytin exist?   Regardless of the subsequent uses of captured light energy, charge separation in the RC relies on the separation of an electron-hole pair. This result is ultimately accomplished 177  using a cofactor cascade capable of reducing some entity (quinone or ferredoxin) which can physically remove the electron from the RC to prevent recombination. For this to happen, the rate of recombination between the diffusible electron acceptor and primary donor must be slow enough to allow for the electron acceptor to dissociate from the protein; and so the relatively slow rate of recombination is obtained by a sufficient number of electron transfer intermediates to create a large distance between the terminal electron acceptor and the primary donor. Because the back reaction does not involve the electron transfer intermediates (it can occur only between the oxidized donor and an anionic radical), creating a large distance between the donor and acceptor significantly decreases the rate of the back reaction (Moser et al. 1992). I suggest that exclusive use of (B)Chl as electron transfer intermediates in type II RCs would not generate a sufficient energy barrier between the intermediates for electron transfer to proceed efficiently (essentially unidirectionally) in the forward direction. Nature has solved this problem by using a quinone (QA) as an electron transfer intermediate in type II RCs. The use of a quinone, however, results in a driving force for forward electron transfer that would be too large if (B)Chl were used as the electron donor. Instead of having to synthesize a novel compound with the appropriate energetic properties, Mg2+ is removed from some of the (B)Chl, generating (B)Phe that can be reduced by a (B)Chl as well as efficiently transfer an electron to a quinone.  The energetic landscape of the RC’s photoactive pigments only partly answers the question about the use of (B)Phe in type II RCs, however. Studies in which hydrogen bonds are introduced into or removed from the BChls of the RC from R. sphaeroides have shown that the electron transfer properties of the pigment are highly tunable by its protein environment (Mattioli et al. 1995; Mattioli et al. 1994). Thus it should be feasible to simply 178  tune a (B)Chl with the surrounding protein component for the (B)Chl to act similarly to (B)Phe in the context of electron transfer (although this has yet to have been performed). Furthermore, the use of different quinones in the RC can achieve a similar effect (Gunner and Dutton 1989), although the use of multiple quinones in an electron transfer cascade has not been observed in RCs in nature. The fact that an HA BChl in the (M)L214H mutant RC has similar energetic properties to the BA BChl suggests that multiple mutations must be made over evolutionary timescales to properly tune the HA BChl for efficient electron transfer (Kirmaier et al. 1995b). If a different quinone were to be used, then an ensemble of new biosynthetic genes would be needed for the organism to make the novel molecule, as well as modification of all the quinone-utilizing proteins for the cell to function properly with the different quinone molecule.    179  5. SUMMARY AND FUTURE DIRECTIONS   The work presented here was begun with the intention of advancing the understanding of the function of the RC of Rhodobacter sphaeroides, a model photosynthetic organism. A mutagenic approach was employed to change the environment of the photoactive BPhe molecule located in the HA binding pocket, to investigate questions relating to how a BPhe comes to be present in this pocket, instead of the BChl present in other RC bacteriochlorin-containing pockets. The results show that: (i) BChl incorporation into HA requires the presence of fifth coordinate that may be provided by a histidine, asparagine or glutamine side chain ligand; (ii) the large side chain of a leucine residue at (M)214 is important in maintaining the positioning of the BA phytyl tail, as well as (iii) maintaining a fast electron transfer rate from HA to QA and (iv) a slow rate of recombination from P+HA- to the ground state. The interplay between (M)214 and electron transfer is likely to be the result of structural perturbations, giving rise to a variety of static populations of RCs that result in decreased electron transfer rates, as well as increasing opportunities for relaxation of the HA- state, thereby increasing the rate of recombination from P+ HA- to the ground state.   One key unknown that remains in spite of the results of this work is the remarkable ability of the (M)L214G mutant strain to grow photoheterotrophically despite its RC having a lower yield of charge separation than the (M)L214H mutant, which suffers a detriment to photoheterotrophic growth. Although the (M)L214G mutant strain does not suffer a penalty in photoheterotrophic growth, no natural (M)L214G mutants were found in the 100-organism sequence alignment. It would be interesting to study whether the (M)L214G RC incurs the 180  same structural perturbations in the native membrane that were observed in vitro. Because this mutant RC is associated with changes in the absorption spectrum, spectroscopic studies on the chromatophores of the (M)L214G mutant lacking LH1 could provide insight into the behavior of this RC in vivo.   Throughout this work, attempts were made to elucidate the origin of the molecule BPhe. However the precise locus or mechanism of the origin of BPhe remains a mystery. Based on my work, it appears that the RC does not act as a dedicated de-chelator of BChl, and that axial ligands to BChl provide the selectivity that distinguishes a BChl-binding pocket from a BPhe-binding pocket. The isolation of a mutant that is unable to produce BPhe would provide valuable information on this molecule’s origin. Alternatively, it is possible that BChl self-pheophytinizes during RC assembly due to the lack of a fifth coordinate. The de novo assembly of a functional RC using a cell-free system could resolve this question, as one could theoretically control the pigment input into the reaction mixture (i.e., add only BChl to a reaction mixture and see if BPhe is incorporated into the RC). Although bacterial light-harvesting complexes have been assembled from purified proteins and BChl (Recchia et al. 1998; Schwarz et al. 2008), membrane protein complex assembly in vitro is largely an emergent technology, and so attempts to employ these methods on the RC would be challenging.     181  REFERENCES  Allen JP, Feher G, Yeates TO, Komiya H, Rees DC (1987a) Structure of the reaction center from Rhodobacter sphaeroides R-26: the cofactors. Proc Natl Acad Sci U S A 84 (16):5730-5734. Allen JP, Feher G, Yeates TO, Komiya H, Rees DC (1987b) Structure of the reaction center from Rhodobacter sphaeroides R-26: the protein subunits. Proc Natl Acad Sci U S A 84 (17):6162-6166. Arlt T, Schmidt S, Kaiser W, Lauterwasser C, Meyer M, Scheer H, Zinth W (1993) The accessory bacteriochlorophyll: a real electron carrier in primary photosynthesis. Proc Natl Acad Sci U S A 90 (24):11757-11761. Bahatyrova S, Frese RN, Siebert CA, Olsen JD, Van Der Werf KO, Van Grondelle R, Niederman RA, Bullough PA, Otto C, Hunter CN (2004) The native architecture of a photosynthetic membrane. Nature 430 (7003):1058-1062. Bearden JA (1967) X-Ray Wavelengths. Reviews of Modern Physics 39 (1):78-&. Beatty JT, Gest H (1981) Generation of succinyl coenzyme-A in photosynthetic bacteria. Arch Microbiol 129 (5):335-340. Bjorn LO, Govindjee (2009) The evolution of photosynthesis and chloroplasts. Curr Sci 96 (11):1466-1474. Blankenship RE (2002) Molecular Mechanisms of Photosynthesis. Wiley-Blackwell. Blankenship RE (2010) Early evolution of photosynthesis. Plant Physiol 154 (2):434-438. Bolton JR, Hall DO (1979) Photochemical conversion and storage of solar energy. Annual Review of Energy 4:353-401. Boso B, Debrunner P, Okamura MY, Feher G (1981) Mössbauer spectroscopy studies of photosynthetic reaction centers from Rhodopseudomonas sphaeroides R-26. Biochim Biophys Acta 638 (1):173-177. Brettel K (1997) Electron transfer and arrangement of the redox cofactors in photosystem I. Biochim Biophys Acta-Bioenergetics 1318 (3):322-373. Brimacombe CA, Stevens A, Jun D, Mercer R, Lang AS, Beatty JT (2013) Quorum-sensing regulation of a capsular polysaccharide receptor for the Rhodobacter capsulatus gene transfer agent (RcGTA). Mol Microbiol 87 (4):802-817. Bryant DA, Costas AMG, Maresca JA, Chew AGM, Klatt CG, Bateson MM, Tallon LJ, Hostetler J, Nelson WC, Heidelberg JF, Ward DM (2007) Candidatus Chloracidobacterium thermophilum: An aerobic phototrophic acidobacterium. Science 317 (5837):523-526. Buttner M, Xie DL, Nelson H, Pinther W, Hauska G, Nelson N (1992) Photosynthetic Reaction Center Genes in Green Sulfur Bacteria and in Photosystem-1 Are Related. Proc Natl Acad Sci U S A 89 (17):8135-8139. Bylina EJ, Kirmaier C, Mcdowell L, Holten D, Youvan DC (1988) Influence of an amino-acid residue on the optical properties and electron transfer dynamics of a photosynthetic reaction centre complex. Nature 336 (6195):182-184. Bylina EJ, Youvan DC (1988) Directed mutations affecting spectroscopic and electron transfer properties of the primary donor in the photosynthetic reaction center. Proc Natl Acad Sci U S A 85 (19):7226-7230. 182  Byrappa S, Gavin DK, Gupta KC (1995) A highly efficient procedure for site-specific mutagenesis of full-length plasmids using Vent DNA polymerase. Genome Res 5 (4):404-407. Callahan PM, Cotton TM (1987) Assignment of bacteriochlorophyll a ligation state from absorption and resonance Raman spectra. J Am Chem Soc 109 (23):7001-7007. Chandler DE, Gumbart J, Stack JD, Chipot C, Schulten K (2009) Membrane curvature induced by aggregates of LH2s and monomeric LH1s. Biophys J 97 (11):2978-2984. Chauvet A, Sarrou J, Lin S, Romberger SP, Golbeck JH, Savikhin S, Redding KE (2013) Temporal and spectral characterization of the photosynthetic reaction center from Heliobacterium modesticaldum. Photosynth Res 116 (1):1-9. Chen XY, Yurkov V, Paddock ML, Okamura MY, Beatty JT (1998) A puhA gene deletion and plasmid complementation system for facile site directed mutagenesis studies of the reaction center H protein of Rhodobacter sphaeroides. Photosynth Res 55 (2-3):369-373. Chiu J, March PE, Lee R, Tillett D (2004) Site-directed, Ligase-Independent Mutagenesis (SLIM): a single-tube methodology approaching 100% efficiency in 4 h. Nucleic Acids Res 32 (21):e174. Choudhary M, Kaplan S (2000) DNA sequence analysis of the photosynthesis region of Rhodobacter sphaeroides 2.4.1. Nucleic Acids Res 28 (4):862-867. Cogdell RJ, Frank HA (1987) How carotenoids function in photosynthetic bacteria. Biochim Biophys Acta 895 (2):63-79. Comayras R, Jungas C, Lavergne J (2005) Functional consequences of the organization of the photosynthetic apparatus in Rhodobacter sphaeroides - I. Quinone domains and excitation transfer in chromatophores and reaction center center dot antenna complexes. J Biol Chem 280 (12):11203-11213. Cotton TM, Vanduyne RP (1979) An electrochemical investigation of the redox properties of bacteriochlorophyll and bacteriopheophytin in aprotic solvents. J Am Chem Soc 101 (25):7605-7612. Crowe SA, Dossing LN, Beukes NJ, Bau M, Kruger SJ, Frei R, Canfield DE (2013) Atmospheric oxygenation three billion years ago. Nature 501 (7468):535-538. Dalrymple GB (2001) The age of the Earth in the twentieth century: a problem (mostly) solved. Age of the Earth: From 4004 Bc to Ad 2002 190:205-221. Dauphas N, van Zuilen M, Wadhwa M, Davis AM, Marty B, Janney PE (2004) Clues from Fe isotope variations on the origin of early Archean BIFs from Greenland. Science 306 (5704):2077-2080. Deisenhofer J, Epp O, Miki K, Huber R, Michel H (1985) Structure of the protein subunits in the photosynthetic reaction center of Rhodopseudomonas viridis at 3 Å resolution. Nature 318 (6047):618-624. Depillis GD, Decatur SM, Barrick D, Boxer SG (1994) Functional Cavities in Proteins - a General-Method for Proximal Ligand Substitution in Myoglobin. J Am Chem Soc 116 (15):6981-6982. Ditta G, Schmidhauser T, Yakobson E, Lu P, Liang XW, Finlay DR, Guiney D, Helinski DR (1985) Plasmids related to the broad host range vector, pRK290, useful for gene cloning and for monitoring gene expression. Plasmid 13 (2):149-153. Duysens LNM (1951) Transfer of Light Energy within the Pigment Systems Present in Photosynthesizing Cells. Nature 168 (4274):548-550. 183  Eichacker LA, Helfrich M, Rudiger W, Muller B (1996) Stabilization of chlorophyll a-binding apoproteins P700, CP47, CP43, D2, and D1 by chlorophyll a or Zn-pheophytin a. J Biol Chem 271 (50):32174-32179. Emsley P, Cowtan K (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr 60 (Pt 12 Pt 1):2126-2132. Evans TA, Katz JJ (1975) Evidence for 5- and 6-coordinated magnesium in bacteriochlorophyll a from visible absorption spectroscopy. Biochim Biophys Acta 396 (3):414-426. Fajer J, Brune DC, Davis MS, Forman A, Spaulding LD (1975) Primary charge separation in bacterial photosynthesis: oxidized chlorophylls and reduced pheophytin. Proc Natl Acad Sci U S A 72 (12):4956-4960. Falkowski P, Scholes RJ, Boyle E, Canadell J, Canfield D, Elser J, Gruber N, Hibbard K, Hogberg P, Linder S, Mackenzie FT, Moore B, 3rd, Pedersen T, Rosenthal Y, Seitzinger S, Smetacek V, Steffen W (2000) The global carbon cycle: a test of our knowledge of earth as a system. Science 290 (5490):291-296. Fiedor L, Kania A, Mysliwa-Kurdziel B, Orzel L, Stochel G (2008) Understanding chlorophylls: Central magnesium ion and phytyl as structural determinants. Biochim Biophys Acta-Bioenergetics 1777 (12):1491-1500. Franken EM, Shkuropatov AY, Francke C, Neerken S, Gast P, Shuvalov VA, Hoff AJ, Aartsma TJ (1997) Reaction centers of Rhodobacter sphaeroides R-26 with selective replacement of bacteriopheophytin by pheophytin a: I. Characterisation of steady-state absorbance and circular dichroism, and of the P+QA- state. Biochim Biophys Acta-Bioenergetics 1319 (2-3):242-250. Frolov D, Marsh M, Crouch LI, Fyfe PK, Robert B, van Grondelle R, Hadfield A, Jones MR (2010) Structural and spectroscopic consequences of hexacoordination of a bacteriochlorophyll cofactor in the Rhodobacter sphaeroides reaction center. Biochemistry 49 (9):1882-1892. Gauthier-Jaques A, Bortlik K, Hau J, Fay LB (2001) Improved method to track chlorophyll degradation. J Agric Food Chem 49 (3):1117-1122. Gest H (1993) Photosynthetic and Quasi-Photosynthetic Bacteria. FEMS Microbiol Lett 112 (1):1-6. Gest H (2005) History of the word photosynthesis and evolution of its definition. In: Govindjee, Beatty JT, Gest H, Allen J (eds) Discoveries in Photosynthesis, vol 20. Advances in Photosynthesis and Respiration. Springer Netherlands, pp 39-42. Goldsmith JO, Boxer SG (1996) Rapid isolation of bacterial photosynthetic reaction centers with an engineered poly-histidine tag. Biochim Biophys Acta-Bioenergetics 1276 (3):171-175. Goldsmith JO, King B, Boxer SG (1996) Mg coordination by amino acid side chains is not required for assembly and function of the special pair in bacterial photosynthetic reaction centers. Biochemistry 35 (7):2421-2428. Goujon M, McWilliam H, Li W, Valentin F, Squizzato S, Paern J, Lopez R (2010) A new bioinformatics analysis tools framework at EMBL-EBI. Nucleic Acids Res 38 (Web Server issue):W695-699. Grabolle M, Dau H (2005) Energetics of primary and secondary electron transfer in Photosystem II membrane particles of spinach revisited on basis of recombination-fluorescence measurements. Biochim Biophys Acta 1708 (2):209-218. 184  Griffiths M, Sistrom WR, Cohenbazire G, Stanier RY, Calvin M (1955) Function of carotenoids in photosynthesis. Nature 176 (4495):1211-1215. Gruber AR, Lorenz R, Bernhart SH, Neubock R, Hofacker IL (2008) The Vienna RNA websuite. Nucleic Acids Res 36 (Web Server issue):W70-74. Gunner MR, Dutton PL (1989) Temperature and -ΔG0 dependence of the electron transfer from BPh.- to QA in reaction center protein from Rhodobacter sphaeroides with different quinones as QA. J Am Chem Soc 111 (9):3400-3412. Hanson LK (1988) Theoretical calculations of photosynthetic pigments. Photochem Photobiol 47 (6):903-921. Hartwich G, Fiedor L, Simonin I, Cmiel E, Schafer W, Noy D, Scherz A, Scheer H (1998) Metal-substituted bacteriochlorophylls. 1. Preparation and influence of metal and coordination on spectra. J Am Chem Soc 120 (15):3675-3683. Heimdal J, Jensen KP, Devarajan A, Ryde U (2007) The role of axial ligands for the structure and function of chlorophylls. J Biol Inorg Chem 12 (1):49-61. Heller BA, Holten D, Kirmaier C (1995) Characterization of bacterial reaction centers having mutations of aromatic residues in the binding site of the bacteriopheophytin intermediary electron carrier. Biochemistry 34 (15):5294-5302. Hofacker IL, Fontana W, Stadler PF, Bonhoeffer LS, Tacker M, Schuster P (1994) Fast folding and comparison of RNA secondary structures. Monatshefte Fur Chemie 125 (2):167-188. Hohmann-Marriott MF, Blankenship RE (2011) Evolution of photosynthesis. Annu Rev Plant Biol 62:515-548. Holzwarth AR, Muller MG (1996) Energetics and kinetics of radical pairs in reaction centers from Rhodobacter sphaeroides. A femtosecond transient absorption study. Biochemistry 35 (36):11820-11831. Huber H, Meyer M, Nägele T, Hartl I, Scheer H, Zinth W, Wachtveitl J (1995) Primary photosynthesis in reaction centers containing four different types of electron acceptors at site HA. Chem Phys 197 (3):297-305. Huppman P, Arlt T, Penzkofer H, Schmidt S, Bibikova M, Dohse B, Oesterhelt D, Wachtveit J, Zinth W (2002) Kinetics, energetics, and electronic coupling of the primary electron transfer reactions in mutated reaction centers of Blastochloris viridis. Biophys J 82 (6):3186-3197. Ignatov NV, Litvin FF (1994) Photoinduced formation of pheophytin/chlorophyll-containing complexes during the greening of plant leaves. Photosynth Res 42 (1):27-35. Iriyama K, Ogura N, Takamiya A (1974) A simple method for extraction and partial purification of chlorophyll from plant material, using dioxane. J Biochem 76 (4):901-904. Ishikita H, Knapp EW (2006) Electrostatic role of the non-heme iron complex in bacterial photosynthetic reaction center. FEBS Lett 580 (18):4567-4570. Jaschke PR, Beatty JT (2007) The photosystem of Rhodobacter sphaeroides assembles with zinc bacteriochlorophyll in a bchD (magnesium chelatase) mutant. Biochemistry 46 (43):12491-12500. Jaschke PR, Hardjasa A, Digby EL, Hunter CN, Beatty JT (2011) A bchD (magnesium chelatase) mutant of Rhodobacter sphaeroides synthesizes zinc-bacteriochlorophyll through novel zinc-containing intermediates. J Biol Chem 286 (23):20313-20322. 185  Jia YW, Dimagno TJ, Chan CK, Wang ZY, Du M, Hanson DK, Schiffer M, Norris JR, Fleming GR, Popov MS (1993) Primary Charge Separation in Mutant Reaction Centers of Rhodobacter-Capsulatus. J Phys Chem 97 (50):13180-13191. Jia YW, Jonas DM, Joo TH, Nagasawa Y, Lang MJ, Fleming GR (1995) Observation of Ultrafast Energy-Transfer from the Accessory Bacteriochlorophylls to the Special Pair in Photosynthetic Reaction Centers. J Phys Chem 99 (17):6263-6266. Jones MR (2008) Structural plasticity of reaction centers from purple bacteria. In: Hunter CN, Daldal F, Thurnauer M, Beatty JT (eds) The Purple Phototrophic Bacteria, vol 28. Advances in Photosynthesis and Respiration. Springer Netherlands, pp 295-321. Jones MR (2009) The petite purple photosynthetic powerpack. Biochem Soc Trans 37 (Pt 2):400-407. Jun D, Saer RG, Madden JD, Beatty JT (2013) Use of new strains of Rhodobacter sphaeroides and a modified simple culture medium to increase yield and facilitate purification of the reaction centre. Photosynth Res:1-9. Kabsch W (2010) Integration, scaling, space-group assignment and post-refinement. Acta Crystallogr D Biol Crystallogr 66 (Pt 2):133-144. Katilius E, Babendure JL, Lin S, Woodbury NW (2004) Electron transfer dynamics in Rhodobacter sphaeroides reaction center mutants with a modified ligand for the monomer bacteriochlorophyll on the active side. Photosynth Res 81 (2):165-180. Katz JJ (1968) Coordination properties of magnesium in chlorophyll from IR and NMR spectra. In: Baer WK, Perkins AJ, Grove EL (eds) Developments in Applied Spectroscopy, vol 6. Developments in Applied Spectroscopy. Springer US, pp 201-218. Khatypov RA, Khmelnitskiy AY, Khristin AM, Fufina TY, Vasilieva LG, Shuvalov VA (2012) Primary charge separation within P870* in wild type and heterodimer mutants in femtosecond time domain. Biochim Biophys Acta 1817 (8):1392-1398. Kirmaier C, Gaul D, DeBey R, Holten D, Schenck CC (1991) Charge separation in a reaction center incorporating bacteriochlorophyll for photoactive bacteriopheophytin. Science 251 (4996):922-927. Kirmaier C, Holten D, Bylina EJ, Youvan DC (1988) Electron transfer in a genetically modified bacterial reaction center containing a heterodimer. Proc Natl Acad Sci U S A 85 (20):7562-7566. Kirmaier C, Holten D, Debus RJ, Feher G, Okamura MY (1986) Primary photochemistry of iron-depleted and zinc-reconstituted reaction centers from Rhodopseudomonas sphaeroides. Proc Natl Acad Sci U S A 83 (17):6407-6411. Kirmaier C, Holten D, Parson WW (1985) Picosecond-photodichroism studies of the transient states in Rhodopseudomonas sphaeroides reaction centers at 5 K. Effects of electron transfer on the 6 bacteriochlorin pigments. Biochim Biophys Acta 810 (1):49-61. Kirmaier C, Laporte L, Schenck CC, Holten D (1995a) The nature and dynamics of the charge-separated Intermediate in reaction centers in which bacteriochlorophyll replaces the photoactive bacteriopheophytin .1. Spectral characterization of the transient state. J Phys Chem 99 (21):8903-8909. Kirmaier C, Laporte L, Schenck CC, Holten D (1995b) The nature and dynamics of the charge-separated intermediate in reaction centers in which bacteriochlorophyll 186  replaces the photoactive bacteriopheophytin .2. The rates and yields of charge separation and recombination. J Phys Chem 99 (21):8910-8917. Kobayashi M, Yamamura M, Akiyama M, Kise H, Inoue K, Hara M, Wakao N, Yahara K, Watanabe T (1998) Acid resistance of Zn-Bacteriochlorophyll a from an acidophilic bacteria Acidiphilium rubrum. Anal Sci 14:1149-1152. Koepke J, Krammer EM, Klingen AR, Sebban P, Ullmann GM, Fritzsch G (2007) pH modulates the quinone position in the photosynthetic reaction center from Rhodobacter sphaeroides in the neutral and charge separated states. J Mol Biol 371 (2):396-409. Kontur WS, Schackwitz WS, Ivanova N, Martin J, LaButti K, Deshpande S, Tice HN, Pennacchio C, Sodergren E, Weinstock GM, Noguera DR, Donohue TJ (2012) Revised Sequence and Annotation of the Rhodobacter sphaeroides 2.4.1 Genome. J Bacteriol 194 (24):7016-7017. Kunieda T, Amano T, Shioi Y (2005) Search for chlorophyll degradation enzyme, Mg-dechelatase, from extracts of Chenopodium album with native and artificial substrates. Plant Science 169 (1):177-183. Leslie AGW, Powell HR (2007) Processing diffraction data with MOSFLM. Nato Sci Ser II Math:41-51. Liebl U, Mockensturm-Wilson M, Trost JT, Brune DC, Blankenship RE, Vermaas W (1993) Single core polypeptide in the reaction center of the photosynthetic bacterium Heliobacillus mobilis: structural implications and relations to other photosystems. Proc Natl Acad Sci U S A 90 (15):7124-7128. Lilburn TG, Prince RC, Beatty JT (1995) Mutation of the Ser2 codon of the light-harvesting B870 alpha polypeptide of Rhodobacter capsulatus partially suppresses the pufX phenotype. J Bacteriol 177 (16):4593-4600. Lin S, Jaschke PR, Wang H, Paddock M, Tufts A, Allen JP, Rosell FI, Mauk AG, Woodbury NW, Beatty JT (2009) Electron transfer in the Rhodobacter sphaeroides reaction center assembled with zinc bacteriochlorophyll. Proc Natl Acad Sci U S A 106 (21):8537-8542. Lin S, Taguchi AKW, Woodbury NW (1996) Excitation wavelength dependence of energy transfer and charge separation in reaction centers from Rhodobacter sphaeroides: Evidence for adiabatic electron transfer. J Phys Chem 100 (42):17067-17078. Lin X, Murchison HA, Nagarajan V, Parson WW, Allen JP, Williams JC (1994) Specific Alteration of the Oxidation Potential of the Electron-Donor in Reaction Centers from Rhodobacter sphaeroides. Proc Natl Acad Sci U S A 91 (22):10265-10269. Mackenzie C, Choudhary M, Larimer FW, Predki PF, Stilwagen S, Armitage JP, Barber RD, Donohue TJ, Hosler JP, Newman JE, Shapleigh JP, Sockett RE, Zeilstra-Ryalls J, Kaplan S (2001) The home stretch, a first analysis of the nearly completed genome of Rhodobacter sphaeroides 2.4.1. Photosynth Res 70 (1):19-41. Mahmoudzadeh A, Saer R, Jun D, Mirvakili SM, Takshi A, Iranpour B, Ouellet E, Lagally ET, Madden JDW, Beatty JT (2011) Photocurrent generation by direct electron transfer using photosynthetic reaction centres. Smart Materials & Structures 20 (9). Martin JL, Breton J, Hoff AJ, Migus A, Antonetti A (1986) Femtosecond spectroscopy of electron transfer in the reaction center of the photosynthetic bacterium Rhodopseudomonas sphaeroides R-26: Direct electron transfer from the dimeric 187  bacteriochlorophyll primary donor to the bacteriopheophytin acceptor with a time constant of 2.8 +/- 0.2 psec. Proc Natl Acad Sci U S A 83 (4):957-961. Masuda T, Inoue K, Masuda M, Nagayama M, Tamaki A, Ohta H, Shimada H, Takamiya K (1999) Magnesium insertion by magnesium chelatase in the biosynthesis of zinc bacteriochlorophyll a in an aerobic acidophilic bacterium Acidiphilium rubrum. J Biol Chem 274 (47):33594-33600. Mattioli TA, Gray KA, Lutz M, Oesterhelt D, Robert B (1991) Resonance raman characterization of Rhodobacter sphaeroides reaction centers bearing site-directed mutations at tyrosine M210. Biochemistry 30 (6):1715-1722. Mattioli TA, Lin X, Allen JP, Williams JC (1995) Correlation between multiple hydrogen bonding and alteration of the oxidation potential of the bacteriochlorophyll dimer of reaction centers from Rhodobacter sphaeroides. Biochemistry 34 (18):6142-6152. Mattioli TA, Williams JC, Allen JP, Robert B (1994) Changes in primary donor hydrogen-bonding interactions in mutant reaction centers from Rhodobacter sphaeroides - identification of the vibrational frequencies of all the conjugated carbonyl groups. Biochemistry 33 (7):1636-1643. McCaskill JS (1990) The equilibrium partition function and base pair binding probabilities for RNA secondary structure. Biopolymers 29 (6-7):1105-1119. Meyer M, Scheer H (1995) Reaction centers of Rhodobacter sphaeroides R26 containing C-3 acetyl and vinyl (bacterio)pheophytins at sites HA,B. Photosynth Res 44 (1-2):55-65. Middendorf TR, Mazzola LT, Lao KQ, Steffen MA, Boxer SG (1993) Stark-effect (electroabsorption) spectroscopy of photosynthetic reaction centers at 1.5K: Evidence that the special pair has a large excited-state polarizability. Biochim Biophys Acta 1143 (2):223-234. Molisch H (1907) Die Purpurbakterien nach neuen Untersuchungen, eine mikrobiologische Studie. Jena, Fischer. Moser CC, Keske JM, Warncke K, Farid RS, Dutton PL (1992) Nature of biological electron transfer. Nature 355 (6363):796-802. Muller B, Eichacker LA (1999) Assembly of the D1 precursor in monomeric photosystem II reaction center precomplexes precedes chlorophyll a-triggered accumulation of reaction center II in barley etioplasts. Plant Cell 11 (12):2365-2377. Mullet JE, Klein PG, Klein RR (1990) Chlorophyll regulates accumulation of the plastid-encoded chlorophyll apoproteins CP43 and D1 by increasing apoprotein stability. Proc Natl Acad Sci U S A 87 (11):4038-4042. Murshudov GN, Vagin AA, Dodson EJ (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr 53 (Pt 3):240-255. Nabedryk E, Schulz C, Muh F, Lubitz W, Breton J (2000) Heterodimeric versus homodimeric structure of the primary electron donor in Rhodobacter sphaeroides reaction centers genetically modified at position M202. Photochem Photobiol 71 (5):582-588. Namsaraev ZB (2009) Application of extinction coefficients for quantification of chlorophylls and bacteriochlorophylls. Microbiology 78 (6):794-797. Neupane B, Jaschke P, Saer R, Beatty JT, Reppert M, Jankowiak R (2012) Electron transfer in Rhodobacter sphaeroides reaction centers containing Zn-bacteriochlorophylls: a hole-burning study. J Phys Chem B 116 (10):3457-3466. 188  Nogi T, Fathir I, Kobayashi M, Nozawa T, Miki K (2000) Crystal structures of photosynthetic reaction center and high-potential iron-sulfur protein from Thermochromatium tepidum: Thermostability and electron transfer. Proc Natl Acad Sci U S A 97 (25):13561-13566. Noy D, Fiedor L, Hartwich G, Scheer H, Scherz A (1998) Metal-substituted bacteriochlorophylls. 2. Changes in redox potentials and electronic transition energies are dominated by intramolecular electrostatic interactions. J Am Chem Soc 120 (15):3684-3693. Okamura MY, Isaacson RA, Feher G (1975) Primary acceptor in bacterial Photosynthesis: Obligatory role of ubiquinone in photoactive reaction centers of Rhodopseudomonas spheroides. Proc Natl Acad Sci U S A 72 (9):3491-3495. Orlandi G, Siebrand W (1973) Theory of vibronic intensity borrowing. Comparison of Herzberg-Teller and Born-Oppenheimer coupling. J Chem Phys 58 (10):4513-4523. Oster U, Bauer CE, Rudiger W (1997) Characterization of chlorophyll a and bacteriochlorophyll a synthases by heterologous expression in Escherichia coli. J Biol Chem 272 (15):9671-9676. Otwinowski Z, Minor W (1997) Processing of X-ray diffraction data. Methods Enzymol 276:307-326. Paddock ML, Feher G, Okamura MY (2003) Proton transfer pathways and mechanism in bacterial reaction centers. FEBS Lett 555 (1):45-50. Pan J, Lin S, Allen JP, Williams JC, Frank HA, Woodbury NW (2011) Carotenoid excited-state properties in photosynthetic purple bacterial reaction centers: effects of the protein environment. J Phys Chem B 115 (21):7058-7068. Pan J, Saer RG, Lin S, Guo Z, Beatty JT, Woodbury NW (2013) The protein environment of the bacteriopheophytin anion modulates charge separation and charge recombination in bacterial reaction centers. J Phys Chem B 117 (24):7179-7189. Parson WW, Clayton RK, Cogdell RJ (1975) Excited-States of Photosynthetic Reaction Centers at Low Redox Potentials. Biochim Biophys Acta 387 (2):265-278. Pawlowicz NP, van Grondelle R, van Stokkum IH, Breton J, Jones MR, Groot ML (2008) Identification of the first steps in charge separation in bacterial photosynthetic reaction centers of Rhodobacter sphaeroides by ultrafast mid-infrared spectroscopy: electron transfer and protein dynamics. Biophys J 95 (3):1268-1284. Pawlowicz NP, van Stokkum IH, Breton J, van Grondelle R, Jones MR (2009) Identification of the intermediate charge-separated state P+βL- in a leucine M214 to histidine mutant of the Rhodobacter sphaeroides reaction center using femtosecond midinfrared spectroscopy. Biophys J 96 (12):4956-4965. Popot JL, Engelman DM (2000) Helical membrane protein folding, stability, and evolution. Annu Rev Biochem 69:881-922. Qian P, Bullough PA, Hunter CN (2008) Three-dimensional reconstruction of a membrane-bending complex - The RC-LH1-PufX core dimer of Rhodobacter sphaeroides. J Biol Chem 283 (20):14002-14011. Qian P, Hunter CN, Bullough PA (2005) The 8.5 angstrom projection structure of the core RC-LH1-PufX dimer of Rhodobacter sphaeroides. J Mol Biol 349 (5):948-960. Qian P, Papiz MZ, Jackson PJ, Brindley AA, Ng IW, Olsen JD, Dickman MJ, Bullough PA, Hunter CN (2013) Three-Dimensional Structure of the Rhodobacter sphaeroides RC-189  LH1-PufX Complex: Dimerization and Quinone Channels Promoted by PufX. Biochemistry. Recchia PA, Davis CM, Lilburn TG, Beatty JT, Parkes-Loach PS, Hunter CN, Loach PA (1998) Isolation of the PufX protein from Rhodobacter capsulatus and Rhodobacter sphaeroides: evidence for its interaction with the α-polypeptide of the core light-harvesting complex. Biochemistry 37 (31):11055-11063. Rohmer M (1999) The discovery of a mevalonate-independent pathway for isoprenoid biosynthesis in bacteria, algae and higher plants. Nat Prod Rep 16 (5):565-574. Saer RG, Hardjasa A, Rosell FI, Mauk AG, Murphy ME, Beatty JT (2013a) Role of Rhodobacter sphaeroides photosynthetic reaction center residue M214 in the composition, absorbance properties, and conformations of H(A) and B(A) cofactors. Biochemistry 52 (13):2206-2217. Saer RG, Pan J, Hardjasa A, Lin S, Rosell F, Mauk AG, Woodbury NW, Murphy ME, Beatty JT (2013b) Structural and kinetic properties of Rhodobacter sphaeroides photosynthetic reaction centers containing exclusively Zn-coordinated bacteriochlorophyll as bacteriochlorin cofactors. Biochim Biophys Acta. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning : a laboratory manual. 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Scheer H (2006) An overview of chlorophylls and bacteriochlorophylls: biochemistry, biophysics, functions and applications. In: Grimm B, Porra RJ, Rüdiger W, Scheer H (eds) Chlorophylls and Bacteriochlorophylls, vol 25. Advances in Photosynthesis and Respiration. Springer Netherlands, pp 1-26. Scheer H, Hartwich G (1995) Bacterial reaction centers with modified tetrapyrrole chromophores. In: Blankenship R, Madigan M, Bauer C (eds) Anoxygenic Photosynthetic Bacteria, vol 2. Advances in Photosynthesis and Respiration. Springer Netherlands, pp 649-663. Scheuring S, Seguin J, Marco S, Levy D, Breyton C, Robert B, Rigaud JL (2003) AFM characterization of tilt and intrinsic flexibility of Rhodobacter sphaeroides light harvesting complex 2 (LH2). J Mol Biol 325 (3):569-580. Schwarz D, Dotsch V, Bernhard F (2008) Production of membrane proteins using cell-free expression systems. Proteomics 8 (19):3933-3946. Shimada H, Iba K, Takamiya K (1992) Blue-light irradiation reduces the expression of puf and puc operons of Rhodobacter sphaeroides under semi-aerobic conditions. Plant Cell Physiol 33 (4):471-475. Shioi Y, Fukae R, Sasa T (1983) Chlorophyll Analysis by High-Performance Liquid-Chromatography. Biochim Biophys Acta 722 (1):72-79. Shioi Y, Tomita N, Tsuchiya T, Takamiya K (1996) Conversion of chlorophyllide to pheophorbide by Mg-dechelating substance in extracts of Chenopodium album. Plant Physiology and Biochemistry 34 (1):41-47. Shochat S, Arlt T, Francke C, Gast P, Vannoort PI, Otte SCM, Schelvis HPM, Schmidt S, Vijgenboom E, Vrieze J, Zinth W, Hoff AJ (1994) Spectroscopic characterization of reaction centers of the (M)Y210W mutant of the photosynthetic bacterium Rhodobacter sphaeroides. Photosynth Res 40 (1):55-66. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Soding J, Thompson JD, Higgins DG (2011) Fast, scalable generation of 190  high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 7:539. Sistrom WR (1960) A requirement for sodium in the growth of Rhodopseudomonas spheroides. J Gen Microbiol 22:778-785. Sobotka R (2013) Making proteins green; biosynthesis of chlorophyll-binding proteins in cyanobacteria. Photosynth Res. Struck A, Scheer H (1990) Modified reaction centers from Rhodobacter sphaeroides R26 - Exchange of monomeric bacteriochlorophyll with 132-hydroxy-bacteriochlorophyll. FEBS Lett 261 (2):385-388. Strumpfer J, Schulten K (2009) Light harvesting complex II B850 excitation dynamics. J Chem Phys 131 (22):225101. Sturgis J, Niederman R (2008) Organization and Assembly of Light-Harvesting Complexes in the Purple Bacterial Membrane. In: Hunter CN, Daldal F, Thurnauer M, Beatty JT (eds) The Purple Phototrophic Bacteria, vol 28. Advances in Photosynthesis and Respiration. Springer Netherlands, pp 253-273. Suzuki Y, Shioi Y (1999) Detection of chlorophyll breakdown products in the senescent leaves of higher plants. Plant Cell Physiol 40 (9):909-915. Takaichi S (2004) Carotenoids and carotenogenesis in anoxygenic photosynthetic bacteria. In: Frank H, Young A, Britton G, Cogdell R (eds) The Photochemistry of Carotenoids, vol 8. Advances in Photosynthesis and Respiration. Springer Netherlands, pp 39-69. Takaichi S (2008) Distribution and Biosynthesis of Carotenoids. In: Hunter CN, Daldal F, Thurnauer M, Beatty JT (eds) The Purple Phototrophic Bacteria, vol 28. Advances in Photosynthesis and Respiration. Springer Netherlands, pp 97-117. Tehrani A, Beatty JT (2004) Effects of precise deletions in Rhodobacter sphaeroides reaction center genes on steady-state levels of reaction center proteins: a revised model for reaction center assembly. Photosynth Res 79 (1):101-108. Tomi T, Shibata Y, Ikeda Y, Taniguchi S, Haik C, Mataga N, Shimada K, Itoh S (2007) Energy and electron transfer in the photosynthetic reaction center complex of Acidiphilium rubrum containing Zn-bacteriochlorophyll a studied by femtosecond up-conversion spectroscopy. Biochim Biophys Acta 1767 (1):22-30. Tucker JD, Siebert CA, Escalante M, Adams PG, Olsen JD, Otto C, Stokes DL, Hunter CN (2010) Membrane invagination in Rhodobacter sphaeroides is initiated at curved regions of the cytoplasmic membrane, then forms both budded and fully detached spherical vesicles. Mol Microbiol 76 (4):833-847. Vagin AA, Steiner RA, Lebedev AA, Potterton L, McNicholas S, Long F, Murshudov GN (2004) REFMAC5 dictionary: organization of prior chemical knowledge and guidelines for its use. Acta Crystallogr D Biol Crystallogr 60 (Pt 12 Pt 1):2184-2195. Van der Rest M, Gingras G (1974) The pigment complement of the photosynthetic reaction center isolated from Rhodospirillum rubrum. J Biol Chem 249 (20):6446-6453. van Niel CB (1944) The Culture, General Physiology, Morphology, and Classification of the Non-Sulfur Purple and Brown Bacteria. Bacteriol Rev 8 (1):1-118. van Stokkum IH, Larsen DS, van Grondelle R (2004) Global and target analysis of time-resolved spectra. Biochim Biophys Acta 1657 (2-3):82-104. 191  van Stokkum IHM, Beekman LMP, Jones MR, van Brederode ME, van Grondelle R (1997) Primary electron transfer kinetics in membrane-bound Rhodobacter sphaeroides reaction centers: a global and target analysis. Biochemistry 36 (38):11360-11368. Wang HY, Hao YW, Jiang Y, Lin S, Woodbury NW (2012) Role of Protein Dynamics in Guiding Electron-Transfer Pathways in Reaction Centers from Rhodobacter sphaeroides. J Phys Chem B 116 (1):711-717. Watson AJ, Fyfe PK, Frolov D, Wakeham MC, Nabedryk E, van Grondelle R, Breton J, Jones MR (2005a) Replacement or exclusion of the B-branch bacteriopheophytin in the purple bacterial reaction centre: the HB cofactor is not required for assembly or core function of the Rhodobacter sphaeroides complex. Biochim Biophys Acta 1710 (1):34-46. Watson AJ, Fyfe PK, Frolov D, Wakeham MC, Nabedryk E, van Grondelle R, Breton J, Jones MR (2005b) Replacement or exclusion of the B-branch bacteriopheophytin in the purple bacterial reaction centre: The HB cofactor is not required for assembly or core function of the Rhodobacter sphaeroides complex. Biochim Biophys Acta-Bioenergetics 1710 (1):34-46. Willows RD, Kriegel AM (2009) Biosynthesis of bacteriochlorophylls in purple bacteria. In: Hunter CN, Daldal F, Thurnauer MC, Beatty JT (eds) The Purple Phototrophic Bacteria, vol 28. Advances in Photosynthesis and Respiration. Springer, Po Box 17, 3300 Aa Dordrecht, Netherlands, pp 57-79. Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P, Evans PR, Keegan RM, Krissinel EB, Leslie AG, McCoy A, McNicholas SJ, Murshudov GN, Pannu NS, Potterton EA, Powell HR, Read RJ, Vagin A, Wilson KS (2011) Overview of the CCP4 suite and current developments. Acta Crystallogr D Biol Crystallogr 67 (Pt 4):235-242. Woronowicz K, Olubanjo OB, Sung HC, Lamptey JL, Niederman RA (2012) Differential assembly of polypeptides of the light-harvesting 2 complex encoded by distinct operons during acclimation of Rhodobacter sphaeroides to low light intensity. Photosynth Res 111 (1-2):125-138. Woronowicz K, Sha D, Frese RN, Niederman RA (2011) The accumulation of the light-harvesting 2 complex during remodeling of the Rhodobacter sphaeroides intracytoplasmic membrane results in a slowing of the electron transfer turnover rate of photochemical reaction centers. Biochemistry 50 (22):4819-4829. Xiong J, Fischer WM, Inoue K, Nakahara M, Bauer CE (2000) Molecular evidence for the early evolution of photosynthesis. Science 289 (5485):1724-1730. Yeates TO, Komiya H, Chirino A, Rees DC, Allen JP, Feher G (1988) Structure of the Reaction Center from Rhodobacter sphaeroides R-26 and 2.4.1: Protein-Cofactor (Bacteriochlorophyll, Bacteriopheophytin, and Carotenoid) Interactions. Proc Natl Acad Sci U S A 85 (21):7993-7997. Yeates TO, Komiya H, Rees DC, Allen JP, Feher G (1987) Structure of the reaction center from Rhodobacter sphaeroides R-26: membrane-protein interactions. Proc Natl Acad Sci U S A 84 (18):6438-6442. Zuker M, Stiegler P (1981) Optimal computer folding of large RNA sequences using thermodynamics and auxiliary information. Nucleic Acids Res 9 (1):133-148.   192  APPENDIX 1. Tabulated photoheterotrophic growth rates of wild type and (M)214 mutants at 5 µE.m-2.s-1 light intensity.  Mutant Average slope of growth curve* Standard deviation Wild type 0.54 0.04 ΔpuhA No growth No growth (M)L214G 0.47 0.16 (M)L214A 0.51 0.01 (M)L214C 0.49 0.22 (M)L214V 0.44 0.10 (M)L214I 0.68 0.18 (M)L214M 0.56 0.08 (M)L214N 0.13 0.03 (M)L214Q 0.11 0.06 (M)L214H 0.07 0.01  *In Klett units per hour. One Klett unit is approximately equal to 3 x 106 cells per mL.   193  APPENDIX 2. X-ray data collection and refinement statistics for (M)L214G, (M)L214A, and (M)L214N RCs.  Structure (M)L214G (M)L214A (M)L214N Accession Code 4IN5 4IN6 4IN7 Unit Cell Parameters a = b 139.11 Å 139.03 Å 139.14 Å c 184.69 Å 184.34 Å 185.14 Å Resolution (Å)* 60.24 – 2.20  (2.32 – 2.20) 60.00 - 2.70  (2.75 – 2.70) 38.54 – 2.85  ( 3.00 – 2.85) No. of reflections** 579188 (105008) 330265 (56265) 353391 (48933) Rmerge* 0.085 (0.590) 0.154 (0.545) 0.098 (0.498) I/σI* 11.3 (2.8) 25.7 (3.3) 17.2 (4.3) Multiplicity* 5.5 (5.4) 5.9 (3.9) 7.2 (7.3) Completeness* (%) 99.9 (100.0) 98.5 (97.8) 99.9 (99.9) Rwork 0.191 0.189 0.179 Rfree 0.212 0.225 0.207 Wilson B-factor (Å2) 38.8 53.2 60.9  Overall B-factor (Å2)  40.8 41.2 42.5  Bond length rmsd (Å) 0.018 0.019 0.020 Bond angle rmsd (°) 1.91 2.18 2.27  * Numbers in parentheses reflect statistics for the highest resolution shell. ** Number in parentheses corresponds to number of unique reflections   194  APPENDIX 3. X-ray data collection and refinement statistics for the Zn- and Zn-β-RCs.  Structure Zn-RC Zn-β-RC Accession Code   Unit Cell Parameters a = b 139.45 Å 139.59 Å c 184.18 Å 184.05 Å Resolution (Å)* 73.23 – 2.85 (2.90 – 2.85) 69.79 – 2.85  (2.90 – 2.85) Rmerge* 0.105 (0.972) 0.098 (0.705) I/σI* 10.5 (1.5) 12.6 (2.4) Multiplicity* 6.2 (4.7) 7.0 (6.2) Completeness* (%) 96 (95) 87.4 (85.6) Rwork 0.182 0.167 Rfree 0.242 0.216 Wilson B-factor (Å2) 53.1 47.0 Overall B-factor (Å2) 58.2 54.8 Coordinate error** (Å) 0.21 0.17 Bond length rmsd (Å) 0.011 0.013 Bond angle rmsd (°) 2.6 2.7   * Numbers in parentheses reflect statistics for the highest resolution shell. **Coordinate error is the estimated standard uncertainty from maximum likelihood refinement.  195  APPENDIX 4: Multiple sequence alignment of the PufM protein sequence  Accession No.*     Position                                         Protein Sequence**  WP_002720420_1  181      WTNNFSLVHGNLFYNPFHGLSIAFLYGSALLFAMHGATILAVSRFGGERELEQIADRGTA BAC54028_1      181      WTNNFSLVHGNLFYNPFHGLSIAFLYGSALLFAMHGATILAVSRFGGVRELEQIADRGTA CAR63836_1      181      WTNNFSLVHGNLFYNPFHGLSIAFLYGSALLFAMHGATILAVSRFGGERELEQIADRSRQ WP_008028573_1  181  WTNMFSLTYGNLFYNPFHGLCIGFLYGSALLFAMHGATIIAVSRFGGERELEQMVDRGTA YP_003576866_1  179  WTNQFSLDHGNLFYNPFHGLSIAALYGSALLFAMHGATILAVTRFGGERELEQIVDRGTA AEW50177_1      179  WTNQFSLDHGNLFYNPFHGLSIAALYGSALLFAMHGATILAVTRFGGERELEQIVDRGTA AAM48658_1      180  WTNLFSLTYGNLFYNPFHALSIAFLYGSALLFAMHGATILAVSRYGGEREIEQIVDRGTA WP_008231901_1  181  WTNLFSITYGNLFYNPFHALSIVFLYGSALLFAMHGATILAVSRFGGEREIEQIVDRGTA YP_006965282_1  181  WTNLFSLTYGNLFYNPFHALSIVFLYGSALLFAMHGATILAVSRYGGEREIEQIVDRGTA AAX48162_1      181  WTNLFSLTYGNLFYNPFHALSIAFLYGSALLFAMHGATILSVTRLGGDRELEQIVDRGTA WP_009159933_1  181  WTNLFSLTYGNLFYNPFHALSIAFLYGSALLFAMHGATILAVSRYGGEREIEQIVDRGTA WP_009831029_1  181  WTNLFSLTYGNLFYNPFHALSIAFLYGSALLFAMHGATILAVSRYGGEREIEQIVDRGTA WP_007206416_1  181  WTNLFSITHGNLFYNPFHALSIAFLYGSVLLFAMHGATILAVSRFGGEREIEQIVDRGTA WP_019954423_1  181  WTNLFSITHGNLFYNPFHALSIVFLYGSVLLFAMHGATILAVSRFGGEREIEQIVDRGTA WP_007164992_1  181  WTNNFSLIHGNLFYNPFHALSIVFLYGSAVLFAMHGATILAVGRYGGEREIEQITDRGTA BAA76948_1      180  WTTAFSIRYGNLYYNPFHALSIAFLYGSALLFAMHGGTILAVTRFGGDRELEQIYDRGTA BAC54030_1      181  WTNNFSLTHGNLFYNPFHGLSIAFLYGSTMLFAMHGATILAVSRLGGERELEQIVDRGTA WP_008280383_1  181  WTTAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILAVTRFGGDRELEQIVDRGTA WP_009819265_1  181  WTTAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILAVSRFGGDRELEQIVDRGTA WP_022573914_1  181  WTTAFSIRYGNLYYNPFHALSIVFLYGSVLLFAMHGATILAVTRFGGDRELEQIYDRGTA CAR63830_1      156  WTNNFSLVHGNLFYNPFHGLSIAFLYGSALLFAMHGATILAVSRFGGERELEQIADRGTA AAY85889_1      181  WTTAYSIRYGNLYYNPFHALSIVFLYGSVLLFAMHGGTILATTRFGGDRELEQIYDRGTA WP_007814239_1  181  WTTAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILAVTRFGGDRELEQIYDRGTA CAR63832_1      181  WTNMFSLTYGNLFYNPFHALSIAFLYGSALLFAMHGATILAVSPLRRRA WP_008962605_1  181  WTNNFSITHGNLFYNPFHMLSIAFLYGSALLFAMHGATVLAISRYGGERELEQIVDRGTA WP_006615516_1  181  WTNNFSITHGNLFYNPFHMLSIAFLYGSALLFAMHGATVLAISRYGGERELEQIVDRGTA YP_001203741_1  181  WTNNFSITHGNLFYNPFHMLSIAFLYGSALLFAMHGATVLAISRYGGERELEQIVDRGTA WP_020044323_1  181  WTTAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILATTRFGGDRELEQIYDRGTA YP_001534858_1  181  WTTAFSIRYGNLYYNPFHALSIVFLYGSVLLFAMHGATILAVTRFGGDRELEQIYDRGTA YP_007515475_1  181  WTNNFSITHGNLFYNPFHMLSIAFLYGSALLFAMHGATVLAISRYGGERELEQIVDRGTA 196  WP_017669607_1  181  WTAAFSIRYGNLFYNPFHALSIAFLYGSTLLFAMHGATILAVGRYGGEREIEQIYDRGTA BAA22642_1      181  WTNNFSLTYGNLFYNPFHGLSIAFLYGSALLFAMHGATILAVSRFGGDRELEQIVDRGTA WP_009758337_1  181  WTNNFSLTYGNLFYNPFHALSIVFLYGSALLFAMHGATILAVSRYGGEREIEQITDRGTA WP_009031272_1  181  WTNNFSITHGNLFYNPFHMLSIAFLYGSALLFAMHGATVLAISRYGGERELEQIVDRGTA YP_005438200_1  181  WTAAFSLRYGNLFYNPFHALSIAFLYGATLLFAMHGATILAVSRFGGERELEQIADRGTA AAO93122_1      181  WTAAFSLRYGNLFYNPFHALSIAFLYGATLLFAMHGATILAVSRFGGERELEQIADRGTA WP_020042497_1  181  WTTAFSIRYGNLYYNPFHCLSIVFLYGSCLLFAMHGATILAVSRFGGDRELEQIVDRGTA WP_010236075_1  181  WTNNFSLIYGNLFYNPFHALSIAFLYGSALLFAMHGGTILAVSRYGGEREIEQITDRGTA WP_019903654_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGATILAVSRFGGEREIDQIVDRGTA YP_001242246_1  181  WTNNFSITHGNLFYNPFHMLSIAFLYGSALLFAMHGATVLAISRYGGERELEQIVDRGTA WP_008296327_1  181  WTAAFSLRYGNLFYNPFHMLSIAFLYGSTLLFAMHGATILAVTRYGGDREIDQITDRGTA WP_018406831_1  180  WTAAFSIRYGNLFYNPFHMLSIAFLYGSALLFAMHGATILAVGRFGGEREIEQIIDRGTA YP_508114_1     180  WTTAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILAVSRYGGDRELEQIVDRGTA WP_018043295_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGATILAVSRFGGEREIDQIVDRGTA WP_019916989_1  181  WTAAFSIRYGNLFYNPFHMLSIAFLYGSTLLFAMHGATILAVSRFGGEREIEQIVDRGTA WP_020165686_1  181  WTAAFSIRYGNLFYNPFHMLSIAFLYGSTLLFAMHGATILAVSRFGGEREIEQIVDRGTA WP_008064766_1  181  WTAAFSIRYGNLFYNPFHMLSIAFLYGSTLLFAMHGATILAVSRFGGEREIEQIVDRGTA YP_001770519_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGATILATSRYGSEREIEQIVDRGTA YP_005449250_1  181  WTNNFSITHGNLFYNPFHMLSIAFLYGSALLFAMHGATILAVSRYGGEREIEQIVDRGTA WP_022677059_1  181  WTAAFSIRYGNLFYNPFHCFSIAFLYGSTLLFAMHGATILAVGRYGGEREIEQITDRGTA WP_009019700_1  181  WTAAFSLRYGNLFYNPFHMLSIAFLYGSTLLFAMHAGTILATSRFGGDREIDQITDRGTA WP_006787498_1  181  WTTAFSLRYGNLFYNPFHMLSIAFLFGSALLFAMHGATILAVTRFGGDRELEQITDRGTA WP_009021756_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGATILAVTRYGGDREIDQITDRGTA WP_010405242_1  181  WTAAFSIRYGNLFYNPFHCLSIVFLYGSTLLFAMHGATILAVGRYGGERELEHITDRGTA WP_018411947_1  181  WTAAFSIRYGNLFYNPFHMLSIAFLYGSTLLFAMHGATILAVSRFGGEREIEQIVDRGTA WP_021076706_1  181  WTNNFSITHGNLFYNPFHMLSIAFLYGSALLFAMHGATILAVSRYGGEREIEQIVDRGTA WP_008944427_1  181  WTAAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILAVGKYGGERELEQITDRGTA WP_017664717_1  181  WTNNFSLTYGNLFYNPFHALSIVFLYGSAVLFAMHGATILALGRYGGEREIEQITDRGTA WP_010184133_1  181  WTAAFSIRYGNLFYNPFHCLSIVFLYGSTLLFAMHGATILAVGRYGGERELEHITDRGTA WP_019012961_1  181  WTAAFSIRYGNLFYNPFHALSIVFLYGSTLLFAMHGATILAVSRYGGEREIEQTLDRGTA WP_020086078_1  181  WTNQLSLDYGNLFYNPFHMLSIAFLYGSALLFAMHGATVLAISRYGGDRELEQITDRGTA YP_004110280_1  181  WTNNFSIKYGNLYYNPFHCLSIAFLYGSALLFAMHGATILAVSRYGGEREIEQMLDRGTA YP_570871_1     181  WTNNFSIKYGNLYYNPFHCLSIAFLYGSALLFAMHGATILAVSRYGGEREIEQMLDRGTA YP_002362361_1  181  WTAAFSIRYGNLFYNPFHMLSIVFLYGSTLLFAMHAATILAVSRFGGEREIEQAVDRGTA WP_020503997_1  181  WTAAFSIAYGNLYYNPFHMLSIAFLYGSAVLFAMHGGTILAVSRYGGDREIDQITDRGTA BAA25552_1      181  WTVSFSLRYGNLYYDPFHMLSIVFLYGSVLLFAMHGATILAVSRFGGEREVEQTVDRGTA 197  YP_487598_1     181  WTNNFSIKYGNLYYNPFHCLSIAFLYGSALLFAMHGATILAVSRYGGEREIEQMLDRGTA YP_001925549_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGGTILATSRYGAEREIDQIVDRGTA WP_010162300_1  181  WTAAFSIRYGNLFYNPFHCLSIVFLYGSTLLFAMHGATILAVGRYGGEREIEHITDRGTA WP_006724825_1  181  WTAAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILAVGKYGGERELEQITDRGTA NP_946876_1     181  WTNNFSIKYGNLYYNPFHCLSIAFLYGSALLFAMHGATILAVSRYGGEREIEQMLDRGTA WP_007196596_1  179  WTAAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILAVTRYGGERELEQITDRGTA YP_002421731_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGGTILATSRYGAEREIDQIVDRGTA YP_002298302_1  181  WTAAFSIRYGNLFYNPFHALSIAFLYGSTLLFAMHGATVLATSRFGAEREVEQIVDRGTA            YP_002963963_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGGTILATSRYGAEREIDQIVDRGTA YP_003068997_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGGTILATSRYGAEREIDQIVDRGTA YP_001755478_1  181  WTAAFSLRYGNLFYNPFHMLSITFLYGSTLLFAMHGGTILACSRYGAEREIDQIVDRGTA YP_001640199_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGGTILATSRYGAEREIDQIVDRGTA WP_008189646_1  180  WTAAFSIRYGNLYYNPFHCLSIVFLYGSVLLFAMHGATILAVGKYGGERELEQITDRGTA WP_017486011_1  181  WTAAFSLRYGNLFYNPFHMLSIVFLYGSTLLFAMHGGTILATSRYGSEREIDQIVDRGTA WP_022723276_1  181  WTNNFSIKYGNLYYNPFHCLSIAFLYGSALLFAMHGATILAVSRYGGEREIEQMLDRGTA WP_022680670_1  180  WTAAFSIRYGNLFYNPFHALSIAFLYGSTLLFAMHGATILAVGRYGGEREIEQITDRGTA YP_680524_1     181  WTTAFSIRYGNLYYNPFHCLSIVFLYGSVLLFCMHGGTILAVTRYGGDRELEQIYDRGTA WP_020091995_1  181  WTAAFSLRYGNLFYNPFHMLSITFLYGSTLLFAMHGGTILACSRYGAEREIDQIVDRGTA P26279_1        181  WTTAFSIRYGNLYYNPFHCLSIVFLYGSVLLFCMHGGTILAVTRYGGDRELEQIYDRGTA WP_007563666_1  181  WTAAFSLRYGNLFYNPFHMLSITFLYGSTLLFAMHGGTILACSRYGADREIDQIVDRGTA YP_531181_1     180  WTNNFSIKYGNLYYNPFHCLSIAFLYGSALLFAMHGATILAVSRYGGEREIEQVVDRGTA WP_007421829_1  181  WTVAFSLRYGNLYYDPFHMLSIVFLYGSVLLFAMHGATILAVSRFGGEREVEQTVDRGTA YP_003819968_1  181  WTGAFSITYGNLFYNPFHCLSIVFLYGSTLLFAMHGATILATSRYGAEREVEQITDRGTG WP_007039476_1  181  WTAAFSIRYGNLYYNPFHMLSIAFLYGSALLFAMHGATILSVSRFGGDREIDQIVDRGTA CBM41421_1      181  WTNLFSLTYGNLFYNPFHALSIVFLYGSALLFAMHGATILAVSRLGGDRELEQIVDRGTA YP_004284102_1  181  WTVAFSLRYGNLYYDPFHMLSIVFLYGSVLLFAMHGATILAVSRFGGEREVEQTVDRGTA WP_010682616_1  181  WTAAFSLRYGNLFYNPFHMLSITFLYGSTLLFAMHGGTILACSRYGAEREIDQIVDRGTA WP_008383958_1  181  WTAAFSIRYGNLYYNPFHMLSIAFLYGSAVLFAMHGATVLAMSHYGGDREVEQIVDRGTA WP_017667956_1  180  WTAAFSIRYGNLFYNPFHALSIAFLYGSTLLFAMHGATILAVGRYGGEREIEQITDRGTA AGY14263_1      181  WTNNFSLVHGNLFYNPFHGLSIAFLYGSALLFAMHGATILAVSRFGGDRELEQIAD---- AEM00431_1      180  WTNLFSITHGNLFYNPFHALSIVFLYGSALLFAMHGATILAVSRFGGEREIEQIVDRGTA YP_006414082_1  181  WTAAFSIRYGNLYYNPFHMLSIAFLYGSALLFAMHGATILSVSRFGGDREIDQIVDRGTA CAQ16985_1      181  WTNLFSITYGNLFYNPFHALSIVFLYGSALLFAMHGATILAVSRFGGEREIEQIVDRGTA YP_003444494_1  181  WTAAFSIRYGNLYYNPFHMLSIAFLYGSALLFAMHGATILAVSRFGGDREIDQITDRGTA  198  * Accession nos. refer to numbers listed in NCBI’s protein database (http://www.ncbi.nlm.nih.gov/protein) **  Residues highlighted in black are conserved in 50% or more organisms. Residues highlighted in gray are not conserved, but are similar to a residue that is conserved in the same column (black letters). Residues in white are not conserved, nor are they similar to a residue that is conserved in the same column. Red column: position analogous to (M)214 in R. sphaeroides. Green: valine at position analogous to (M)214 in R. sphaeroides. blue: methionine at position analogous to (M)214 in R. sphaeroides.    199  APPENDIX 5: Light minus dark difference spectra of wild type and (M)L214G RCs before and after incubation with ubiquinone-10.   Absorbance spectra were not normalized, however the RC samples were kept at 42 µM concentration. Legend: WT, wild type RC; WT + Q, wild type RC after incubation with 1/10 volume of saturated ubiquinone-10 in ethanol; (M)L214G, (M)L214G mutant; (M)L214G + Q, (M)L214G mutant after incubation with 1/10 volume of saturated ubiquinone-10 in ethanol. 

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.24.1-0167255/manifest

Comment

Related Items