HOPPING ALONG THE WAY: GENETIC ANALYSIS OF PILI - MEDIATED CHARGE TRANSPORT IN GEOBACTER SULFURREDUCENS By Rebecca J. Steidl A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree o f Microbiology and Molecular Genetics - Doctor of Philosophy 2015 ABSTRACT HOPPING ALONG THE WAY: GENETIC ANALYSIS OF PILI - MEDIATED CHARGE TRANSPORT IN GEOBACTER SULFURREDUCENS By Rebecca J. Steidl Geobacter sulfurreducens is a dissimilatory iron - reduc ing bacterium that is able to utilize insoluble external electron acceptors such as Fe(III) oxides, radionuclides, and the anode of a microbial electrochemical cell (MEC ) . To accomplish this process it uses direct contact mechanisms involving a host of c - t ype cytochromes. It also pr oduces microbial nanowires that are necessary for efficient growth with all of the aforementioned insoluble electron acceptors. The question is then: A re these nanowires transferring electrons to the acceptors, or are they merely a scaffold for electron transporting cytochromes. Thus, I i nvestigated the contribution of electroact ive pili to electron transfer (ET) to exte rnal electron acceptors as well as the mechanism of electron transfer in the pili. In chapter 2 I demonstrate the necessity of nanowires for electron transfer through an anode biofilm by generating a mutant that displays only pili defects. Deleting gene encoding the pilus motor , PilB, produces a pili - deficient mutant with wild - type cytochrome expression that f orme d anode biofilms as thick ( ca. 10 µm) as the wild type yet with reduced electroactivity. Furthermore, the growth and electroactivity of thicker biofilms required the expression and conductivity of the pili. The results support a model in which the conducti ve pili form a nanopower grid that permeates the biofilms to wire the cells in all biofilm strata to the underlying electrode. The pili operate coordinately with cytochromes in the lower strata until the biofilm reaches a threshold thickness where the pili are required as electronic conduits and thus without this function the biofilm is unable to continue to grow . In chapter 3 I investigate the mechanism of pilus conductivity by generating amino acid replacements of residues predicted to be involved in ET in the pilus. Pili isolated from a Y27A mutant strain displayed a decreased ability to transfer electrons along the length of the pilus. This mutant, however, displayed no defects in biological external electron transfer assays. Electron transfer rates exp lained this phenotype, as the mutated pili were still able to transfer electrons faster than the acetate respiration rates. Two additional tyrosine residues, Y32 and Y57 were essential for MEC electroactivity, as were the negatively charged amino acids D53 and D54. Replacement of the three tyrosine residues with the aromatic amino acid phenylalanine, however, resu lted in a strain with no reduction in electroactivity. Thus, the presence of these aromatic and charged amino acids is required for optimal char ge transfer in the G. sulfurreducens pilus. In Chapter 4 I develop a system to determine the role of intramolecular ET of the pilin monomer. To this end, I collaborated with Dr. Castro - Forero of the Worden Lab to generate a method for the in vitro produc tion of soluble pilin monomers. These recombinant pilins were able to form conductive filaments in vitro and thus retain their ET capabilities. I then modified these recombinant pilins to attach to a gold electrode and used thes e self - assembled monolayers to study the role of intram olecular electron transfer. I generated several point mutations in the codons for aromatic and charged residues predicted to be involved in ET. Copyright by REBECCA J. STEIDL 2015 v I w ould like to dedicate my dissertation to my parents Robert and Barbara Steidl. They constantly pea ked my interest in science and the world around me , and were always supportive of my seemingly eternal academic career. My dad constantly showed me the wonder s of biology from teaching me about different species of plants to bringing me interesting bugs or seeds containing multiple embryos that he found while wor king outside . His doctorate and research background let me know from a young age that I wanted to st rive for the same. I know he would have loved to see me complete my doctorate. vi ACKOWLEDGEMENTS I would like to thank my parents Barb and Bob Steidl for their support through my long graduate career. You have always been supportive and helped me see the light at the end of the tunnel. I would also like to thank the Stern family, Laura, Travis and Ellie for providing me with reasons to leave the lab and go home (namely an adorable niece to play with) and sending random cards and care packages to make me smile. I would also like to thank Laura and Travis for the endless amount of help and support they have provided to my mom through first grief and later illness that I was not able to provide myself. I would like to thank my boyfriend and next doo r lab neighbor Lee Macomber and my lab mate and best friend, Mike Manzella. You were both always there for me when I needed it and made my time in here exponentially better . You were great science - related sou nding boards and also many a family dinner would not have been possible without my best sous chefs. Also in attendance at many of my dinners and often assisting in food prep were many of my other good friends I made while in grad school, including Nik McP herson, Ben and Sarah Koestler, and Darin Quach. You are the best friends ! I would like to also thank all present and past members of the Hungry Hungry Hippos Kickball team it was good to have a fun activity to organize with great people . I would like to thank all past and present members of the Reguera Lab. I would like to thank my undergraduate researchers that have both contributed to my work and allowed me to grow as a teacher: Carlos , Mike, and Alex. In particular, Mike and Alex who both assisted in the many EPS isolations and cytochrome gels. I would like to thank Allison and Bhushan for help with my electrochemical work and Krista and Sanela for all of their CP - AFM work. I would li ke to thank Dena for her moral support and always being there to bounce ideas off of. vii My committee members have also been integral to my graduate work and my success here: Dr. Bob Hausinger, Dr. Rob Britton, Dr. Mark Worden, and Dr. Cecilia Martinez - Gomez . You were very patient as I tried to schedule meetings and gave me needed perspective on my research, plans, and writing. I would also like to thank all of those in the Microscopy center for their help including Alicia, Melinda, and Abby. I would like to thank those that have collaborated with me on this work: Dr. Angelines Castro Forero and Dr. Gustavo Feliciano. Last but certainly not least , I would like to thank Dr. Gemma Reguera for the advice and support during my tenure here. It took a long time but you stuck with me through all of it and offered me support through all of the challenges. viii TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ...................... xi LIST OF FIGURES ................................ ................................ ................................ ................... xii KEY TO ABBREVIATIONS ................................ ................................ ................................ ...... xv CHAPTER 1. INTRODUCTION ................................ ................................ ................................ ... 1 The Geobacteraceae ................................ ................................ ................................ ........ 2 Extracellular electron transfer in Geobacter sulfurreducens ................................ .............. 3 Geobacter pili structure and function ................................ ................................ ................ 7 1.3.1 Overview of type IVa pili ................................ ................................ ............................ 7 1.3.2 The G. sulfurreducens pilin is divergent in structure and amino acid composition ..... 9 Metallic - like conductance in pili ................................ ................................ ....................... 10 Alternative methods of electron transfer in pili ................................ ................................ 11 1.5.1 Tunneling ................................ ................................ ................................ ................ 11 1.5.2 Hopping ................................ ................................ ................................ .................. 12 1.5.3 Protein structure and charge (electron or hole) hopping ................................ .......... 14 Dissertation outline ................................ ................................ ................................ ......... 16 1.6.1 Mechanistic stratification in electroactive biofilms of Geobacter sulfurreducens mediated by pilus nanowires ................................ ................................ ............................ 16 1.6.2 Site - directed mutagenesis reveals the role of aromatic amino acids and local electrostatics in pi lus conductivity and extracellular electron transfer ............................... 17 1.6.3 Intramolecular charge transport in pilus nanowires investigated in pilin - electrode interfaces . 18 CHAPTER 2. MECHANISTIC STRATIFICATION IN ELEC TROACTIVE BIOFILMS OF GEOBACTER SULFURREDUCENS MEDIATED BY PILUS NANOWIRES ............................ 19 Introduction ................................ ................................ ................................ ..................... 20 Materials and Methods ................................ ................................ ................................ ... 22 2.2.1 Bacterial strains and culture conditions. ................................ ................................ .. 22 2.2.2 DNA manipulations and mutant construction. ................................ .......................... 23 2.2.3 Microbial electrochemical cells (MECs). ................................ ................................ .. 25 2.2.4 Static biofilm assays on plastic surfaces . ................................ ................................ 2 6 2.2.5 Gene expression analyses by quantitative Real Time - PCR (qRT - PCR). ................. 26 2.2.6 Assays for c - type cytochrome content and profiling. ................................ ................ 27 2.2.7 Pili purification and conductivity measurements by CP - AFM. ................................ .. 28 Results ................................ ................................ ................................ ........................... 29 2.3.1 Inactivation of the PilB assembly motor does not affect c - type cytochromes expression. .. 29 2.3.2 Pili expression is required for optimal electrochemical act ivity of thin biofilms. ........ 31 2.3.3 The expression and conductivity of the pili are required to grow thick biofilms. ....... 33 Discussion ................................ ................................ ................................ ...................... 35 ix CHAPTER 3. SITE - DIRECTED MUTAGENESIS REVEAL S THE ROLE OF AROMATIC AMINO ACIDS AND LOCAL ELECTROSTATICS IN PILUS CONDUCTIVITY AND EXTRACELLULAR ELECTRON TRANSFER ................................ ................................ ................................ .......... 40 Introduction ................................ ................................ ................................ ..................... 41 Materials and Methods ................................ ................................ ................................ ... 45 3.2.1 Gr owth conditions ................................ ................................ ................................ ... 45 3.2.2 Mutant construction ................................ ................................ ................................ . 46 3.2.3 Pili purification and conductivity analyses. ................................ ............................... 46 3.2.4 Calculation of Fe(III) oxide respiratory rates. ................................ ........................... 48 3.2.5 Electrochemical activity of anode biofilms in MECs ................................ ................. 48 Results and Discussion ................................ ................................ ................................ .. 49 3.3.1 Modeling the structure of the Geobacter pilus at high resolution via molecular dynamics . ................................ ................................ ................................ ..................... 49 3.3.1.1 Potential ET pathways via aromatic residues ................................ .................. 50 3.3.1.2 Potential ET pathway via the amide backbone ................................ ................ 52 3.3.1.3 ................................ ....... 53 3.3.2 The role of tyrosine residues in biofilm electroactivity and Fe(III) oxide reduction .... 54 3.3.3 The effect of alanine replacement of Y27 on pilus conductivity ............................... 55 3.3.4 Effect of phenyla lanine replacements of the pilus tyrosines ................................ ..... 57 3.3.5 Effect of negatively - charged amino acids of the pilus ................................ .............. 58 Conclusions ................................ ................................ ................................ .................... 59 CHAPTER 4. INTRAMOLECULAR CHARGE TRANSPORT IN PILU S NANOWIRES INVESTIGATED IN PILIN - ELECTRODE INTERFACES ................................ ........................... 62 Introduction ................................ ................................ ................................ ..................... 63 Materials and Methods ................................ ................................ ................................ ... 66 4.2.1 Construction of Expression Strains ................................ ................................ ......... 66 4.2.1.1 QIAexpress TM system (His - tagged pilins) ................................ ......................... 66 4.2.1. 2 pMAL TM system (MBP - tagged pilin) ................................ ................................ . 66 4.2.1.3 IMPACT TM system (CBD - tagged pilins) ................................ ............................ 67 4.2.1.4 Mutagenesis of codons for pil in amino acids ................................ .................... 67 4.2.2 Expr ession of recombinant pilins ................................ ................................ ............. 68 4.2.3 Purification of recombinant pilins carrying a Chitin - binding domain protein tag ....... 69 4.2.4 Electrochemical charac terization of PilA 19 - A20C monolayers on gold electrodes .... 70 Results ................................ ................................ ................................ ........................... 71 4.3.1 Rationale for partially truncating the N - terminal hydrophobic pilin region ................. 71 4.3.2 Construction a nd heterologous expression of recombinant pilins ............................ 74 4.3.2.1 Recombinant production of pilins carrying an N - t His - Tag ............................... 74 4.3.2.2 Recombinant production of pilins carrying a MBP ................................ ............ 75 4.3.2.3 Recombinant production of pilins carrying a chitin - binding domain (CBD) ....... 77 4.3.3 Electrochemical characterization of recombinant PilA 19 ................................ ........... 79 4.3.3.1 Functionalization of PilA 19 with a cysteine tag and construction of pilin - electrode interfaces ................................ ................................ ................................ .......... 79 4.3.3.2 Cyclic voltammetry of pilin - electrode interfaces ................................ ............... 80 4.3.3.3 CP - AFM probing of pilin - electrode interfaces ................................ ................... 81 4.3.4 Intramolecular pilin pathway investigated in mutated pilin - electrode interfaces ....... 82 Conclusions ................................ ................................ ................................ .................... 83 CHAPTER 5. Conclusions and Future Directions ................................ ................................ . 86 Review o f project ................................ ................................ ................................ ............ 87 x 5.1.1 Review of mechanistic stratification in electroactive biofilms of Geobacter sulfurreducens mediated by pilus nanowires ................................ ................................ .... 87 5.1.2 Review of site - directed mutagenesis reveals a role of aro matic amino acids and local electrostatics in pilus conductivity and extracellular electron transfer ............................... 88 5.1.3 Review of intramolecular charge transport in pilus nanowires investigated in pilin - electrode interfaces ................................ ................................ ................................ ......... 89 Future directions ................................ ................................ ................................ ............. 90 5.2.1 Future directions for studying mechanistic stratification in electroactive biofilms of Geobacter sulfurreducens mediated by pilus nanowires ................................ .................. 90 5.2.2 Future directions for site - directed mutagenesis to reveal the role of aromatic amino acids and local electrostatics in pilus conductivity and extracellular electron transfer ....... 91 5.2.3 Future directions for examining intramolecular charge transport in pilus nanowires via pilin - electrode interfaces ................................ ................................ ................................ .. 93 APPENDICES ................................ ................................ ................................ ........................... 95 APPENDIX A MEC ADAPTATION IN PILB MUTANT STRAIN ................................ ............. 96 APPENDIX B TABLES ................................ ................................ ................................ ....... 101 APPENDIX C FIGURES ................................ ................................ ................................ ..... 112 REFERENCES ................................ ................................ ................................ ....................... 156 xi LIST OF TABLES Table 1 Strains and plasmids used in Chapter 2 ................................ ................................ ..... 102 Table 2 Primers used in Chapter 2 ................................ ................................ .......................... 103 Table 3 Bacterial strains and plasmids used in Chapter 4 ................................ ....................... 105 Table 4 Primers used in Chapter 3 ................................ ................................ .......................... 106 Table 5 Strains used in Chapter 4 ................................ ................................ ........................... 107 Table 6 Primers used in Chapter 4 ................................ ................................ .......................... 108 Table 7 Plasmids used in Chapter 4 ................................ ................................ ........................ 110 Table 8 Buffers used to purify recombinant pilins as fusion proteins with a Chitin - binding Domain ( CBD) using the IMPACT TM system using a chitin resin column ................................ .............. 111 xii LIST OF FIGURES Figure 1 Structure of the PAK pilin and geopilin ................................ ................................ ...... 113 Figure 2 Properties of the G. sulfurreducens pilin ................................ ................................ ... 114 Figure 3 - stacking configurations ................................ ................................ .......................... 115 Figure 4 Charge tra nsfer mechanisms ................................ ................................ .................... 116 Figure 5 Amide backbone charge hopping ................................ ................................ .............. 117 Figure 6 Absorption spectra of oxidize and reduced cell extracts ................................ ............ 118 Figure 7 Characterization of cytochrome content in pili - deficient mutant cultures .................... 119 Figure 8 Expression of biofilm related genes in pilus deficient mutants ................................ ... 120 Figure 9 Phenotypic characterization of pili - deficient mutant (pilB, pilA, and pilA - E5A) biofilms ................................ ................................ ................................ ................................ ............... 121 Figure 10 B iofilm electroactivity of selected mutants in MECs fed with 1 mM acetate ............. 122 Figure 11 Current production as a function of biofilm thickness in the WT and pilB strains ...... 123 Figure 12 Role of pilus conductivity in the growth and electroactivity of anode biofilms ........... 124 Figure 13 Model of mechanistic stratification of electroactive biofilms ................................ ..... 125 Figure 14 Surface m aps of MS - optimized GS pilus ................................ ................................ . 126 Figure 15 Potential charge transfer pathways identified in the MD model ................................ 127 Figure 16 Electron transfer paths predicted in the G. sulfurreducens pilus from the MD - refined pilus model ................................ ................................ ................................ .............................. 128 Figure 17 Surface exposure of Y32 and Y57 ................................ ................................ ........... 129 Figure 18 Amide electron hole hopping and phenylalanine dimer intermolecular transfer pathway ................................ ................................ ................................ ................................ .. 130 Figure 19 Biofi lm electroactivity in MECs fed with 3 mM acetate for Y32 and Y57 .................. 131 Figure 20 Biofilm electroactivity in MECs fed with 3 mM acetate for Y27 amino acid replacement mutants ................................ ................................ ................................ ................................ ... 132 xiii Figure 21 Transversal CP - AFM measurements of Y27 amino acid replacements ................... 133 Figure 22 CP - AFM transport measurements along WT and Y27A pili ................................ ..... 134 Figure 23 Electron transport rates per cell during growth of G. sulfurreducens with poorly crystalline Fe(III) oxides ................................ ................................ ................................ .......... 135 Figure 24 Biofilm electroactivity of phenylalanine vs. alanine replacement in MECs fed with 3 mM acetate ................................ ................................ ................................ ............................. 136 Figure 25 Comparison of the reduction of external electron acceptors by Asp2 and Tyr3 ........ 137 Figure 26 NMR - derived PilA structural model of the G. sulfurreducens PilA and predicted structure of recombinant PilA 19 pilins ................................ ................................ ....................... 138 Figure 27 AGGRESCAN and Kyte - Doolittle analyses ................................ ............................. 139 Figure 28 Expression and purification of recombinant pilins carrying a 6 His - tag ................... 140 Figure 29 Western blot analysis of periplasmic (left) and cytoplasmic (right) proteins fractions during the expression and purification of MBP - PilA 20 ................................ .............................. 141 Figure 30 Cleavage effi ciency of PilA 20 from its MBP tag by Factor Xa protease ..................... 142 Figure 31 Production of PilA 20 subunits after treatment with Factor Xa protease ..................... 143 Figure 32 Expression of the full - length pilin (PilA) and various truncated pilins (PilA n ), all in fusions with a CBD tag (CBD - PilA n ) ................................ ................................ ........................ 144 Figure 33 Presence of pilin monomers in elution from chitin column after cleavage from CBD tag determined by 16 - 20% Tricine SDS - PAGE ................................ ................................ ............. 145 Figure 34 Expression and purification of recombinant PilA 19 - A20C subunits ........................... 146 Figure 35 Effect of DTT on the electrochemical activity of PilA 19 - A20C pilins on a gold electrode ................................ ................................ ................................ ................................ ............... 147 Figure 36 Effect of solvent (acetonitrile) in the deposition of PilA 19 - A20C on a gold electrode . 148 Figure 37 Effect of length of deposition time of PilA 19 - A20C in acetonitrile on a gold electrode ................................ ................................ ................................ ................................ ............... 149 Figure 38 Effect of mixing and age of pilin sample on deposition of PilA 19 - A20C in acetonitrile ................................ ................................ ................................ ................................ ............... 150 Figure 39 Effect of concentration and temperature of pilin sample on deposition of PilA 19 - A20C in acetonitrile ................................ ................................ ................................ ........................... 151 Figure 40 CP - AFM of PilA 19 - A20C monolayers on gold electrodes ................................ ......... 152 Figure 41 Effect of amino acid substitutions on cyclic voltammogramms ................................ . 153 xiv Figure 42 Test of media component limitation ................................ ................................ ......... 154 Figure 43 Growth adaptation of pilB after extended incubation in MFC ................................ ... 155 xv KEY TO ABBREVIATIONS AEM anion exchange membrane Ag/AgCl silver/silver chloride reference electrode ATP adenosine triphosphate BCA bicincho ninic acid BLAST basic local alignment search tool BSA bovine serum albumin C - t C - terminal CBD chitin binding domain CD circular dichroism CEM cation exchange membrane CLSM c onfocal laser scanning microscopy CP - AFM conductive probe atomic force micr oscopy CV cyclic voltammetry DBAF fuel cell media DI deionized DLS dynamic light scattering DMSO dimethyl sulfoxide DTT dithiothreitol E° standard redox potential EDTA ethylenediaminetetraacetic acid EM electron microscopy EPS exopolysaccharide xvi E T electron transfer FWAF fresh water acetate fumarate Gm gentamycin GRAVY grand average of hydropathicity index HEPES 4 - (2 - hydroxyethyl)piperazine - 1 - ethanesulfonic acid HOPG highly oriented pyrolytic graphite HPLC high performance liquid chromatograph y I current IPTG isopropyl - 1 - thio - D - galactopyranoside LB lysogeny media MALDI matrix assisted laser desorption ionization MBP maltose binding protein MD molecular dynamics MEC microbial electrochemical cell MS mass spectrometry N - t N - terminal NA numerical aperture NBAF nutrient broth acetate fumarate NHE normal hydrogen electrode NMR nuclear magnetic resonance NTA nitrilotriacetic acid OD optical density OG octyl - D - glucopyranoside PBS phosphate - buffered saline PCET proton coupled electron tr ansfer xvii PCR polymerase chain reaction PSII photosystem II QCM quartz crystal microbalance R resistance RE reference electrode RNR ribonucleotide reductase RT real time SAM self - assembled monolayer SDS - PAGE sodium dodecyl sulfate polyacrylamide gel e lectrophoresis SEM scanning electron microscopy SHE standard hydrogen electrode Sp spectinomycin STM scanning tunneling microscopy TMBZ - tetramethylbenzidine TEM transmission electron microscopy TFP type IV pili UV - vis ultraviolet - visible spectroscopy V voltage WT wild - type YE yeast extract 1 CHAPTER 1. INTRODUCTION 2 The Geobacteraceae The Geobacteraceae are a diverse family of dissimilatory iron reducers in the - proteobacteria , thus sharing the ability to gain energy for growth from the reducti on of Fe(III) coupled to the complete oxidation of acetate , other organic compounds, and hydrogen gas 1 . Because o f this type of metabolism , members of the family are ubiquitous in environments where Fe(III) reduction is a significant process, particularly in subsurface environments. The genus Geobacter is perhaps the best studied and includes several model representa tives with sequenced genomes and genetic systems 2 5 . All Geobacter spp. are mesoph iles and prefer low salt and neutral pH 1 . Members within this genus are abundant in soils where Fe(III) reduction is an important ecosystem driver 1 and are often enriched on anode electrodes from sediment fuel cells, which har ness electricity from the oxidation of organic compounds by the anode - associated bacteria in sediments using a cathode positioned at the aerobic zones 6 . While Geobacter spp. are typically classified as strict anaerobes, some tolerate oxygen and can even grow with it as sole terminal electron acceptor for respiration under microaerophilic conditions 7,8 . This physiological trait allows them to survive and even boost their growth during oxygen intrusions in to the soils and sediments fo r brief periods of time. Electron transfer (ET) is not restricted to Fe(III) oxides and other chemical electron acceptors. Geobacter are also able to transfer electrons to syntrop h ic partners 9 and often account for a significant proportion of the microorganisms that form methanogenic aggregates 10 . In addition, many members of the Geobacteraceae can degrade a wide variety of organic substrates, including common pollutants such as hydrocarbons 11 and chlorinated contaminant s 12 , thus showing potential as bioremediation agents. Similarly, many members of the Geobacteraceae can reduce toxic metals such U(VI) 13 and the anode of a microbial fuel cell 14 , activities that show promise for the bioremediation of metal contaminants and bioenergy applications 13,15 17 . 3 Extracellular electron transfer in Geobacter sulfurreducens The insoluble nature of Fe(III) oxides, the natural electr on acceptor for the Geobacteraceae , has selected for a respiratory mechanism that allows the cells to establish direct electronic contact with the mineral. Such a strategy requires cells to first attach to the surface of the mineral and then transfer resp iratory electrons across the cell envelope to terminal reductases o f the outer surface, which can transfer electrons to the oxides. The most common iron reductases in dissimilatory iron reducers with a direct - contact mechanism are c - type cytochromes 18 . Geobacter sulfurreducens 19 , for example, one of the most well - studied bacteria in the Geob acteraceae , contains over 100 genes predicted to encode c - type cytochromes 3 . However, cytochrome conservation among Geobacter genomes is low. For instance, only 14 % of the cytochromes encoded in the genome (the equivalent to 2 % of all of the genes) of G. sulfurreducens are conserved across the Geobacter species 1 . In addition to being the first species within the Geobacteraceae to have a sequenced genome 3 , G . sulfurreducens was also the first member in the family to become genetically tractable 5 . Therefore, much of the research on extracellular electron transfer in this family has been performed in G. sulfurreducens . Genetic studies support the idea that the unprecedented number of c - type cytochromes in this bacterium allows it to utilize a wide range of electron acceptors in addition to Fe(III) oxides, such as Mn(IV) oxides, U(VI) and the anode of a microbial fuel cell 14,20 26 whereby the removal of a key cytochrome is rapi dly compensated by the expression of another or several more cytochromes 22,27 . Such redundancy allows Geobacteraceae to maintain different electron transfer pathways for the reduction of electron acceptors with different redox potentials 28 . This would be important for the utilization of the many different metal oxides and other electron acceptors available in the environment, which have different redox potentials 29 . 4 Because the pressure is high to directly contact the insoluble Fe(III) oxides in order to dischar ge respiratory electrons , Geobacter cells have evolved mechanism s to exten d their electroactive surface beyond the confines of the outer membrane. One such method is to produce an exopolysaccharide (EPS) matrix that anchors c - type cytochromes 30 . The matrix also allows the cell to build electroactive biofilms on electrodes, thereby harnessing the respiratory activities of the biofilm cells as an electrical current 30 . Another important component of the electroactive biofilm matrix of Geobacter are type IV pili 31,32 , extracellular protein appendages used by many bacteria for surface attachment, translocation, and to provide structural support within multilayered biofilms 33 . As in other bacteria, the pili of G. sulfurreducens play a structural role in the biofilm matrix, allowing cells to grow as multilayered communitie s on inert and electron - accepting surfaces 31,32 However, i n contrast to other bacterial pili, those of G. sulfurreducens are conductive 34 and have been proposed to form a nanopower grid that wire s the biofilms to an underlying electron - accepting surface such as Fe(III) oxides or an anode electrode poised at an oxidizing metabolic potential 31,32 . Furthermore, the pili are also required for the respiration of insoluble Fe(III) oxides 34 . These early findings relied heavily on genetic studies with a pilin - deficient mutant of G. sulfurreducen s , which lacked pili and was impaired in its ability to transfer electrons to a variety of external electron acceptors 13,31,34 . However, the mutant was later reported to also have defect s in c - type cytochromes of the outer membrane 13,15 , thus raising doubts about the role of the pili in electron transfer in planktonic cells and biofilms. Evidence that the pili do in fact function as electronic conduits was provided by studies that examined the mechanism that allows G. sulfurreducens cells to reduce the soluble uranyl cation 13 . These stu dies demonstrated the correspondence between piliation, but not c - type cytochromes , and the ability of the cells to reduce the U(VI) extracellularly 13 . Hence, t he pili are the primary site for uranium reduction 13 . By strongly binding the soluble uranyl cation and 5 reductively precipitating it as a mononuclear solid phase of U(IV) on the pilus fibers, the pili prevent the soluble uranyl ion from permeating and being reduced in the cell envelope , thus shield ing the cell from the toxic effects of uranium 13 . The role of pili in wiring electroactive biofilms remains, however, controversial. Cells within G. sulfurreducens biofilms transfer electrons to an underlying electrode at rates substantially higher than other species, even mixed cultures 35 . Genetic studies have identifie d only one cytochrome, the matrix - bound cytochrome OmcZ 30 , as required for biofilm electroactivity 20 . OmcZ is an outer membrane c - type cytochromes that is proteolytically dig ested into a small form, OmcZ S , for export to the biofilm matrix 20 . OmcZ S localizes preferentially to the biofilm layers closer to the electrode surface 36 , suggesting a role in electronically connecting the biofilm to the electrode. Other cytochromes of the biofilm matrix may participate in electro n transfer. However, genetic s tudies have failed to identify cytochromes other than OmcZ essential to ET in the anode biofilms, presumably because of the redundancy and compensatory adaptations often observed in cytochrome mutants 22,27 . E lectrochemical studies also demonstrated that the c - type cytochromes of the electroactive biofilm matrix can be reversibly oxidized or reduced depending on the anode potential 37,38 . In thin biofilms (1 - 2 cells thick) , all cytochromes are reduced at positive voltages whereas in thick biofilms ( ca. 20 µ m thick) only f ifty percent of cytochromes in the biofilm remain reduced 38 . Furthermore, a cytochrome redox gradient is established across thick biofilms, where by the concentration of electrons is higher in the upper layers and decreases progressively towards the electrode - attached biofilm 39 . Such an electron gradient is predicted to drive ET from the upper to the lower strata of the biofilm. However, it also suggests that the conduction of electrons via cytochromes becomes limiting as the biofilm grows progressively away from the electron - accepting anode. The low redox potential of the upper strata of th ick biofilms may begin as little as 10 µm from the oxidizing electrode based on source - drain 6 experiments 39 . Yet, despite this limitation, current production increases linearly with the biofilm thickness 31 and cells in the upper la yers of thick biofilms r emain metabolically active 40 and contribute to current production 31 . The pili could be the elusive electronic conduits that allow cells in the upper stratum to maintain optimal catalytic currents. Not only are they conductive and required for the reduction of Fe (III) oxides 34 and the uranyl cation 13 , they are also required for efficient current production in MECs 31 . Furthermore, the pili provide structural support to the biofilms 32 , and pili - deficient strains only form thin biofilms and produce low levels of current 31 . However , the pili - defective strains used in earlier studies ( the pilin - pilA mutant 34 and the pili regula tor mutant pilR 41 ) also have defects in the expression of outer - membrane cytochromes 13,41 . Hence, the observed phenotypes could have been caused by the pilus deficiency, the cytochrome defects, or both. The role of pili as electronic conduits has also be en challenged by the report that polyclonal antibodies raised against the outer membrane cytochrome OmcS bind pili in chemically - fixed preparations of piliated cells 42 . This has le d to the proposal that OmcS, and perhaps other cytochromes, bind to the pili and contribute to its conductivity 42 . Results from this study are, however, difficult to interpret because the pili preparations were chemically fixed and the polyc lonal antibodies could have bound non - specifical ly to the pili. Furthermore, when cell - associated p ili were deposited on graphite without a ny type chemical treatment and then probed by scanning tunneling microscopy, they showed conserved topographic features of type IV pili and electronic features consistent with hotspots of conductivity; however, none of the electronic signatures matched those typical of the heme groups of c - type cytochromes 43 . Hence , the contrib ution of the pil us protein matrix and the po tential of pilus - associated cyt o c hromes to facilitate electron transfer remains controversial . Recently, biofilms of the Aro - 5 strain of G. sulfurreducens , which carries alanine replacements in five of the six aromatic amino acids of the pilin (including all tyrosine residues) , 7 were shown to have reduced conductivity while still associating with the cytochrome OmcS on the pili 44 . As aromatic amino acids , in particular tyrosine residues, have been proposed to contribute to the pilus conductivity 45 , these results suggest that some or all of these amino acids ar e required for pilus conductivity and, consequently, that pili are necessary for biofilm electroactivity. However, potential effects of the extensive mutagenesis on the integrity of the pilus structure w ere never investigated 46 . Furthermore, defects in cytochrome production in the electroactive biofilms by the Aro - 5 strain, which have been reported for other pili - deficient mutants 13,15,41 , were not evaluated in this study 46 . Thus, the role of pili as electronic cond uit s to external electron acceptors remains controversial. Geobacter pili s tructure and f unction 1.3.1 Overview of type IVa pili Type IVa pili are thin ( 6 to 9 nm in diameter ) pilus filaments that are assembled on the cell envelope and can grow to several micro ns in length . Despite their small diameter, type IVa pili are very strong and can withstand elastic forces of up to 110 pN 47 . T hey are widespread in bacteria, particula rly in Proteobacteria , and serve vari ous functions such as surface motility, microcolony and biofilm formation, surface adhesion, immune evasion, cell signaling, transformation of DNA, and phage attachment 48 . Despite their multiplicity of functions, they are surprisingly simple at the bio chemical level, consisting of a single peptide subunit or pilin that polymerizes via hydrophobic interactions at the base of the fiber in a repeating , helical fashion . Hydrophobic interactions between adjacent subunits are mediated by the conserved N - termi nal (N - t) - helix of all type IVa pilins , a region that can be further divided in two domains: the highly conserved and hydrophobic - N , which spans approximately the first 25 amino acids - t, and the - C that contains a similar number of amino acids in the C - terminal (C - t) end of the - helix ( Figure 1 ) . In addition, most bacterial type IVa pilins 8 also contain a variable globular head of four anti - parallel - sheets connect ed to the C - t end of the - helix via an - loop , which mediates pilin - pilin interactions 49 . After the globular domain, there is also a highly variable region or D - region containing two conserved cysteine residues 50 . The globular domain and D - region confer on the pili much of their surface chemistry and biological functions such as cell - cell aggregation, adhesion, and surface motility and also stabilize the pilus filament through pilin - pilin interactions 48 . Another conserved feature of type IVa pilins is that they are synthesiz ed as a precursor or prepilin , which carries a short hydrophilic signal peptide at the N - t end. The last amino acid of this signal peptide is a conserved glycine and is recognized and required for efficient cleaving of the signal peptide by a dedicated pre pilin peptidase 51,52 . Also conserved are the phenylalanine at position 1 (F1) of the processed pilin, which is methylated by the signal peptide upon cleavage of the signal peptide 52 , and t he glutamic acid at position 5 (E5), which interact during assembly to align the incoming pilin at the base of the growing pilus fiber 51,52 . Also conserved are proteins of the type IVa apparatus required for pilin assembly such as PilC and ATPases that polymerize (PilB) and depolymerize (PilT) the pil ins to maintain optimal cycles of protrusion and retraction 53 . S tructural s tudies of type IVa pili and their pilin subunit have proven difficult due to the insoluble nature of both . The small diameter and smooth surface of the pili make the structural characterization of type IVa pili difficult. The pilins , on the other hand, are insoluble due to the - N region, making NMR and X - ray crystallography approaches challenging a s well. Despite these hurdles , full and truncated type IVa pilin structures have been crystallized and an NMR - derived structure for the p ilin of G. sulfurreducens in lipid micelles is available 49,54 65 . Structural information i s also available for two type Iv a pili : a computational model of the P. aeruginosa strain K ( PAK ) pil i (derived from X - ray fiber di ffraction data) 50 and the pseudoatomic model of Neisseria gonorrhea 9 structure and cryo - elect ron microscopy [ cryoEM ]) 66 . The most detailed experim ental structure is that of the GC pilus, which was generated b y fitt ing the crystal structure of a full - length GC pilin into a model of the GC pilus fiber reconstructed with 12.5 Å resolution by cryoEM 66 . The structure reveals a pilus with a diameter of approximately 60 Å with an inner hydrophobic core of 6 - accounts for 75 % of the total buried surface area of the pilus. Furthermore, the pilus has a symmetry of 3.6 pilins per turn with a 10.5 Å rise between subunits 66 . 1.3.2 The G. sulfurreducens pilin is divergent in structure and amino acid composition The type IVa pilin s of the family Geobacteraceae are shorter than other bacterial pilins and form an independent line of descent ; for this reason they are sometimes referred to as 34 . They contain the conserved of other type IV a pilins, but lack the globular domain and D - loop of other type IVa pilins , which are replaced with a short C - t random coil 45 . As shown in Figure 1 for the pilin of G. sulfurreducens , this large truncation makes the pilin structure primarily - helical. - has less sequence homology than the - N region required for pilus assembly 50 . It is in this 1 - C region of the pilin of G. sulfurreducens as well as in the C - t random coil where several aromatic amino acids conserved in the Geobacteraceae are located 34 ( Figure 2 ) The C - t random co il region is also highly flexible and mo bile with no fixed position, as observed in the NMR structure of the G. sulfurreducens pilin 58 ( Figure 2 ). Because the structural differences between the G. sulfurreducens pilin and oth er bacterial pilins are so profound, the predictive value of structural models for the PAK and GC pilins is limited. To address this limitation, my group collaborated with Drs. Artacho (University of Cambridge, U.K.) and Feliciano (University of Sao Paulo, Brazil) to generate a structural model of the pilin of G. sulfurreducens in solution via molecular dynamics (MD) simulations 45 . The electrostatic potential 10 of the pilin revealed in the simulations displayed a dipole along the length of the monomer with negative electrostatic region at the C - terminus and a positive region in the middle of the polypeptide confirming those calc u lated previously ( Figure 2 ). This dipole was not merely due to the structure of the pept ide as a poly - alanine peptide displayed no such effect 45 . The large dipole could dramatically increase the ET rate through the polypeptide similar to the effect of the natural dipole created by the N - terminus and C - terminus ends of helic al peptides 67,68 . The pilin model also predicted t hat the pilin monomer is conductive with a band gap similar to conventional semiconductors and charge transfer would be more favorable in one direction 45 . Evidence also suggests that type IVa pilins are post - translational modified and t hat such chemical modifications contribute to pili functions such as twitching motility, aggregation, attachment, and binding 51,62,69 73 . The role of post - translational modifications in G. sulfurreducens pilin are not known, except for an isolated report that suggested that one tyrosine of the pilin (Y32) could be modi fied with a glycerophosphate 74 as reported for other bacterial pilins 75 . Such post - translational modifications may be particularly relevant for a metal reducer such as G. sulfurreducens . For example, the S - layer of the bacterium Bacillus sphaericus JG - A12 is post - translational ly modified with phosphate groups for efficient binding of uranium 76 . Metallic - like conductance in pili As mentioned earl ier, the G. sulfurreducens pilus is conductive 34 and required for G. sulfurreducens to transfer electrons to a variety of external electron acceptors 13,31,34 . However, the mechanism by which this filament is conductive remains controversial. The temperature dependence of piliated anode biofilms and crude preparations of pili dried on electrodes has been interpreted as the pili behaving like metalli c conductors 77 . However, this interpretation has been challenged based on limitations of the electrochemical approach used for the conductivity 11 measurements and the fact that the crude preparations of pili contained other proteins, including c - type cytochromes 78 . The metallic model of pilus conductance re lies heavily on the assumption that aromatic residues of the pilus are clustered sufficiently so as to maintain inter - aromatic distances ( ca. 3 Å ) and configurations (sandwich type) that promote - stacking 79 ( Figure 3 ). Such configurations are unstable and rarely seen in proteins due to their rigidity 80 . Furthermore, it is unlikely that dynamic appendages such as pili can stably maintain these aromatic stacking g eometries because even subtle dealignments or increases in inter - aromatic distances can have dramatic effects on charge mobility 81,82 . Furthermore, homology models of the G. sulfurreducens pilus reveal inter - aromatic residues too large (15 - 21 Å ) 83 - stacking and metallic conductance . However, more refined structural models are necessary to test the model of metallic conductance. Alternative methods of electron transfer in pili In addition to the proposed metallic - like conductance there are two other well characterized mechanisms for ET in proteins: electron tunneling and hopping ( Figure 4 ). 1.5.1 Tunneling One well - established mechanism of ET in proteins is via electron tunneling 84 . This mechanism is based on quantum mechanics rather than classical mechanics and it involves the direct transfer of an electron from donor to acceptor through the barrier ( Figure 4 ). This process creates the first defi ning factor of electron tunneling: it is relatively temperature independent 85 , unlike classical thermodynamics. Because of the quantum - mechanical na ture of electron tunneling, ET rates decrease exponentially with distance 84 . Hence, the ET rate constant for tunneling is an exponential function dependent on the decay constant and the distance between the donor and acceptor 84 . Because of this, ET over 14 Å 86 to 20 Å 84 becomes too slow to be biologically relevant. 12 Tunneling has been primarily studied by Gray and colleagues using flash - quench techniques that measured ET rates in ruthenium - modified metalloproteins. In these studies, the electrons tunnel from the metal cofactor (e.g., Cu in azurin) to a Ru moeity attached to histidine residues of the protei ns positioned at defined distances from the metal cofactor 84 . The results of the ir studies demonstrated the exponential distance dependence and temperature independence of the ET rates, which were in the order of milliseconds a t distances of about 20 Å at 170 K 84 . Electron tunneling has been demonstrated in electrochemical studies of short helical peptides, however at lengths greater than 20 Å evidence does not support a single step tunneling mechani sm 87 89 . As the pilins themselves are longer ( ca. 80 Å ) than both predicted and demonstrated single - step tunneling mechanism s for ET, it is highly unlikely that tunneling regimes are the p rimary mechanism of ET through the pilin or pilus. 1.5.2 Hopping In contrast to the quantum mechanical electron tunneling mechanism , electron hopping occurs by the classical theory of ET . As such, the electron is transferred via redox acti ve groups that function ( Figure 4 ) . Because in the hopping model electrons do in fact reside in the relay residues, activation energies must be overcome for ET to occur leading to a strong temperature dependence. A s the electron m akes shorter hops from redox active site to site , the rates of ET no longer show a strong distance dependence and the rate constant is instead a linear function of the distance between the acceptor and donor 90 . One amino acid commonly involved in electron hopping in proteins is tryptophan which is well - characterized in ET in several proteins such as photolyase 91 and ribonucleotide reductase (RNR) 92 . However as the pilin monomer contains no tryptophan residues, they ca nnot be involved in ET in the G. sulfurreducens pilus. Tyrosine residues have also been implicated in electron hopping in several well - studied proteins , such as RNR and within the photosystem II 13 (PSII) complex. In RNR an electron is transferred more than 3 5 Å along an established pathway: 92 . This is one of the longest distances for ET recorded in a protein. The space between W48 and Y731 is 25 Å , with Y356 the only redox active residue between 92 . This means that a jump of more than 12.5 Å must be necessary. In PSII, on the other hand, a tyrosine residue transfers an electron to the next carrier, P 680 , over distances estimated to range from 8.3 93 to 13.8 94 Å. Thus, even though many hopping steps are required to enable long distance transfer, the individual steps can be greater than 12 Å in length themselves. Because r emoval of an electron from a tyrosine requires an oxidation potential (1.46 V vs NHE) 92 much higher than those than operate in biological systems , e lectron hopping via tyrosine s in proteins occurs by a pr oton coupled electron transfer (PCET) reaction 92 . Acidic residues, histidine, and cysteine have been documented to act as proton acceptors for tyrosine s in PCET reactions 92,95,96 . The removal of a proton from the tyrosine by these amino acids lowers the oxidation potential of the tyrosine residue and facilitates ET. In addition to work characterizing protein systems, Cordes et. al developed a system to study ET rates in synthesized peptides 97 . This system was used to investigate the relay capabilities of aroma tic amino acids 97 . The studies demonstrated that the presence of redox - active peptide side chains increase s ch arge transfer rates along these peptides 97 . They also showed that redox - active amino acids such as tyrosine an d redox - active aromatic derivatives 98 . The aromatic amino acid phenylalanine has a predicted redox potential (2.2 V vs. NHE) 99 much higher than any other aromatic residue and too high to allow it to function as a relay amino acid in an ET pathway in vivo ; however, in several cytochrome c proteins studie d, phenylalanine residues are surprisingly essential for optimal ET rates 100,101 . E lect ronic structure calculations in bacterial NADH:ubiquione oxidoreductase have also identified c onserved phenylalanine residues as essential to ET 102 . One phenylalanine in particular, F328 , was 14 involved in all the fast paths of ET ide ntified . Peptide systems have also been used to investigate the role of phenylalanine residues as stepping stones in ET 103 . Despite having an oxidation potential higher tha n that of its final electron acceptor, a phenylalanine residue in the middle of the peptide still resulted in a high ET efficiency. Surprisingly, this efficiency was slightly higher than a similar peptide carrying a redox - active methoxy - substituted relay i nstead of a phenylalanine, suggesting that, despite its high oxidation potential, the phenylalanine was a ble to act as a stepping stone. 1.5.3 Protein structure and charge (electron or hole) hopping In vitro studies have demonstrated that peptides, particularly helical ones, allow electrons to hop through the peptide backbone. Interestingly, electrons hop up to 120 Å ; a distance significantly longer than the G. sulfurreducens pilin 68,88 . Hopping down the length of a peptide is theorized to occur via the amide groups of the peptide backbone, 88 with the amide taking on a positive charge. Ab initio calculations of single amino acids injected with an additional positive charge support this model, as 88 % of the introduced positive c harge resides in the amide region 90 . Thus, hopping by this method is theorized to occur by electron hole hopping where the electron would first hop to the acceptor R - group and then the cation radical would hop along the amide groups to the electron donor 88 ( Figure 4 ). Therefor e , the amide region of a peptide may play an important role in charge transfer. In addition to the amide region, electron - rich bonds may also act as stepping stones and contribute to hopping. When the ab initio calculations were repeated on a serine residu e modified with an alkene - rich side group, the amide region held only 68 % of the positive charge injected and 22 % of the charge resided in the electron - rich side group 90 . Electrochemi cal studies confirmed the contribution of these electron - rich groups to hopping while also demonstrating the negative role that backbone rigidity plays on charge transfer 90 . Hence, elec tron - rich groups not typically thought to be involved in charge transfer could in fact participate in charge hopping and could help explain 15 perplexing observations about phenylalanine residues acting as stepping stones for hole hopping 103 . It has also been proposed that other non - traditional sources may act as stepping stones for charge transfer including the formation of lone pair ( lp ) 104 , lone pair lone pair (lp lp ) three elec tron bonds 105 ) 106 , and the C - - helices 107 . As the aromatic residues are essential to the conductivity of the pili, the interactions mentioned earlier in that two aromatic rings must come ca. 3 Å from each other to facilitate ET . However, by the hole hopping mechanism, the formation of a hole causes the two rings to be brought closer together followed by the formation of a delocalized three electron bond between the rings. Calculations predict that subsequent separation of the rings will promote the electron hole relay to another grou p along the peptide. This contrasts with - interactions for metallic - like conductance, which require a near static ring position and distance across the pilus. Thus, the advantage of hole hopping is that even transient aromatic interactions and thus limited bonding could facilitate the move ment of the hole relay 106 . transition to this state. The rate of charge transfer via the backbone hopping method has been shown to increase with the natural dipole moment of a helical pep tide 67,108 110 , from positive to negative. - helices is also predicted to increase ET through these peptides ( Figure 5 ). Short peptides unable to form hydrogen bonds exhibit a decrease d rate constant with a dditional length that is consistent with one - step tunneling. However, as longer peptide lengths permit hydrogen bonding, a transition is observed whereby a hopping mechanism takes over 111 . Similar relationships between the hydrogen bond number and rate cons tants of ET have been demonstrated in other peptide systems 112 114 . As mentioned earlier, 16 the flexibility of molecules also affects ET 90 and, as such, the natural movement and flexibility of organic structures such as DNA and peptides is predicted to increase their charge transfer rates 115,116 . As stated earlier, the G. sulfurreducens pilin is predominantly an - helical structure with large dipole 45 . Given that - helices are structures that promote charge transfer and that the rate of charge transfer is affected by the dipole of the peptide, the structure of the pilin is consistent with a peptide environment evolved for intramolecular charge transport by the hopp ing mechanism. Furthermore, once the pilins assemble, the aromatic amino acids of the pili could cluster sufficiently to create additional pathways for charge hopping through the pilus. Hence, inter - and intramolecular pathways likely mediate charge hoppin g in the pilus. Because of its complexity, it is important to dissect the contribution of both. Dissertation o utline 1.6.1 Mec hanistic stratification in electroactive biofi lms of Geobacter sulfurreducens mediated by pilus nanowires Chapter 2 describes genetic st udies to elucidate the role of conductive pili in electroactive bio films formed on the anode electrode of MECs fed with acetate. The effect of several pili - inactivatin g mutations on biofilm formation and conductivity and epistatic effects on c - type cytochr omes production were studied. A deletion mutant (pilB), which cannot produce the PilB ATPase that polymerizes pilins, had no detectable changes in cytochrome production or localization, and provided the elusive genetic tool needed to assess the contributi on of pili to the growth and electrochemical activity of anode biofilms. A strain that produces pili with reduced conductivity was also analyzed. The results supported a model of mechanistic stratification in anode biofilms mediated by the conductive pili, whereby pili and cytochromes 17 work coordinately to transfer electrons to the underlying electrode until the biofilm reaches a threshold thickness ( ca. 10 µm) that limits the efficiency of the cytochrome pathway but not of the pili. All of the experiments reported in this chapter were conducted by me, except for the conductivity measurements with a conductive probe atomic force microscopy (CP - AFM), which were performed by a post - doctoral member of our lab, Dr. Sanela Lampa - Pastirk. 1.6.2 Site - directed mutagenesi s reveals the role of aromatic amino acids and local electrostatics in pilus conductivity and extracellular electron transfer Chapter 3 describes the mutational analysis of key residues in the pilus predicted to be involved in ET along the pilus. A homolog y model for the G. sulfurreducens pilus was refined by MD simulations by Dr. Feliciano and then used to predict potential ET pathways in the pilus. R esidues implicated in a hopping pathway of ET (Y27, Y32, and Y5 7) and several residues that contribute to a natural dipole in the pilus (D53, D54, and R28). Alanine replacement mutations were generated in the gene encoding the pilin to substitute these residues and study their effect in pilus conductivity in vivo (ele ctrochemical activity of biofilms in MECs and growth in Fe(III) oxide cultures). Some of the variant pili were then purified and their transversal and axial conductivities were measured with a conductive probe atomic force microscope ( CP - AFM ) . The experim ental data were used to calculate the ET rates of individual pilus fibers and that of piliated cells respiring Fe(III) oxides or anode electrodes. The results support the predictions of the MD model, which indicated that the clustering and configuration of the aromatic residues is critical for optimal multistep hopping. They also revealed redundant pathways for ET in the pilus and potential paths for the transfer of electrons from the pilus to external electron acceptors such as Fe(III) oxides and the urany l cation. 18 All of the experiments reported in this chapter were conducted by me, except for th e conductivity measurements by CP - AFM , which were performed by, Dr. Sanela Lampa - Pastirk. Assistance from Dr. Allison Speers setting up the Fe(III) oxides cultures is acknowledged. 1.6.3 Intramolecular charge transport in pilus nanowires investigated in pilin - electrode interfaces Chapter 4 describes the heterologous expression of truncated pilin monomers of G . sulfurreducens and their deposition as pilin monolayers on gol d electrodes to isolate and study the intramolecular ET pathway of the pilus . Protein expression constructs of truncated pilins were generated using the following systems: QIAexpressionist, pMAL , and IMPACT. The expression and purification conditions were then optimized for the mass - production of recombinant pilins . One of them, PilA 19 , which was purified at high yields, was modified with an N - termina l cysteine to attach it to gold electrodes as a confluent monolayer . The conductivity of the pilin monolayer was demonstrated by cyclic voltammetry (CV). Furthermore, CV was also performed on the mutated pilin monolayers , which carried alanine replacements in residues predicted to be involved in pilin conductivity (Y27, Y32, Y57, D53, D54, and R28) . The results suggest that intramolecular ET is not influenced by these amino acids. Potential limitations of the interfaces to unmask the mutant defects are discussed. All of the experiments reported in this chapter were conducted by me, except for the optimization of the expression and purification of the recombinant pilins, which was performed by my collaborator Dr. Castro Forero (a graduate student in the lab of Dr. R. Mark Worden , Department of Chemical Engineering ). She used the selected recombinant pilin to assemb le conductive pili in vitro . I am also grateful to Dr. Bhushan Awate for his invaluable help in the early phase of the electrochemical experiments. 19 CHAPTER 2. MECHANISTIC STRATIFICATION IN ELECTROACTIVE BIOFILMS OF GEOBACTER SULFURREDUCENS MEDIATED BY PILUS NANOWIR ES 20 Introduction The ability of Geobacter bacteria to completely oxidize organic compounds to CO 2 with an electrode poised at a metabolically oxidizing potential shows promise for the conversion of organic wastes and renewable biomass into electricity, hy drogen gas , and/or liquid fuels in microbial electrochemical cell s ( MEC s) 16,17,117 . Energy recoveries in these devices depend in a great manner on the electroactivity of the electrode - associated biofilms 118 yet the mechanism of conductance is not fully understood. Genetic studies in the model representative Geobacter sulfurreducens have been instrumental to identify biofilm components required for biofilm formation and electroactivity. Among the iden tified genes are those encoding the Xap exopolysaccharide (EPS), which are required for the development of multilayered biofilms and for anchoring redox - active proteins, mainly c - type cytochromes 30 . Although the genome of G. sulfurreducens contains over 100 open reading frames annotated as c - type cytochromes, many of which are essential for the reduction of extracellular electron acceptors 21,22,119 , genetic studies have identified only one cytochrome, the matrix - associated OmcZ, as essential for biofilm electroactivity 120 . Other cytochromes may be required, but their genetic identification is challenging because of compensatory effects observed in some cytochrome mutants 22,27 . The matrix - associated c - type cytochromes can be reversibly oxidized or reduced depending on the anode potential 37,38 , consistent with their role as electron carriers in the biofilms. However, redox potentials decrease in b iofilm layers positioned ca. 10 µm or further away from the electrode, and a redox gradient is established whereby more electrons are concentrated in the upper, electron acceptor - limited biofilm stratum and decrease progressively towards the electrode - atta ched layers 39 . The redox gradient provides the driving force for electron transport across the biofilms in a manner analogous to how electrons diffuse through redox polymers by hopping among immobilized redox cofactors 121 . Consistent with this process , while it is possible to reduce all of the biofilm cytochromes of thin (< 10 µm) biofilms when the 21 potential is set at a positive voltage, only half are reduced in thicker ( ca. 20 µm) biofilms 38 . Yet, despite this limitation, G. sulfurreducens can grow electroactive biofilms that are tens of micrometers from the electrode surface while generating current proportionally to the biofilm thickness 31 . Furthermore, the cells in the biofilm layers farthest from the anode surface are metabolically active and continue to oxidize acetate while contributing to curre nt production 40 . This result suggests that electron carriers other than cytochromes operate in electroactive biofilms, particularly in the upper stratum of thick biofilms. Biofilm formation in G. sulfurreducens also requires the expression of conductive protein filaments termed pili 31 . The pili permeate the biofilm matrix and provide structural support for the growth of the multilayered community 32 . Yet their conductive properties 34 s uggest they could also form a nanopower grid across the biofilms to electronically connect the biofilm cells to the underlying electrode 31 . The pili have also been proposed to anchor cytochromes, inasmuch as polyclonal antibodies raised against the outer membrane c - type cytochromes OmcS hybridize to antigens along chemically - fixed, cell - associated pilus fibers 42 . Because the pil i are part of a biofilm matrix with abundant cytochromes, they may also function as electronic conduits between the biofilm cells and the matrix - associated cytochromes, similarly to how planktonic cells use the pili to electronically connect with extracell ular electron acceptors such as Fe(III) oxides 34 and uranium 13 . Much of the difficulty in assessing the contribution of the pili to biofilm electroactivity stems from the fact that p ili - deficient mutants reported thus far also have defects in cytochromes required for extracellular electron transfer 13,15,41 . Deleting the gene encoding the pilus subunit (th e PilA pilin), for example, prevents pili formation and decreases current production by biofilms 31 . However, the mutation also results in defects in outer m embrane cytochromes 13 and prevents the expression of OmcZ in the biofilm matrix 15 5 aromatic amino acids with alanines resulted in a strain (Aro - 5) that formed thick anode 22 biofilms yet had reduced conductivity 44 . The study did not investigate potential pleitrophic effects of the Aro - 5 mutation in cytochrome expression 46 . Furthermore, although Aro - 5 pili sheared from the cells had reduced conductivity 44 , the shearing procedure also releases c - type cytochromes 77 , wh ich could contribute to the measured conductivity 78 . Yet the studies raise the interesting possibility that the conductive propertie s of the pili influence biofilm conductance. In this study, I addressed current limitations of genetic studies with pili - deficient mutants and investigated the effect of several pili - inactivating mutations in the expression of cytochromes. One of the mutat ions, a deletion in the gene encoding the pilin polymerization motor, PilB, inactivated pili expression without affecting cytochrome expression . Hence, I used this mutant to investigate the role of pili in the growth and electro chemical activity of G. sulf urreducens biofilms. The genetic studies indicate that the pili are required for optimal growth and current production by the biofilm cells. Electron transfer in thin ( ca. 10 µm) biofilms required the expression of both the pili and the matrix - associated c ytochrome OmcZ, suggesting that both componets function as charge carriers. Yet to grow thicker biofilms and sustain linear increases of current the biofilm cells require the expression and conductive properties of the pili. Presence of conductive pili con fers on Geobacter cells a metabolic advantage over other bacteria relying solely on cytochromes or diffusible electron carriers, as it maximizes energy generation through extracellular electron transfer even in cells positioned at tens of micromete r distan ces from the electrode. Materials and Method s 2.2.1 B acterial strains and culture conditions. The bacterial strains used in this study are described in (Table 1) . The wild - type (WT) strain, G. sulfurreducens strain PCA, was kindly provided by Daniel Bond (Univ ersity of Minnesota) and was used to construct three knock - out mutants: pilB (carrying a deletion in pilB , GSU1491), pilA 23 (carrying a deletion in the pilin gene pilA , GSU1496), and omcZ (carrying a deletion in omcZ , GSU2076). When indicated, the pilB mutat ion was complemented in trans by expressing a wild - type copy of pilB from the pRG5 plasmid (pilB+ strain). The WT strain was also used to construct the pilA - E5A mutant (carrying a single alanine replacement of the glutamic 5 residue in the mature pilin Pil three tyrosine residues Y27, Y32, and Y57). The WT and mutant strains were routinely cultured anaerobically in NBAF, FWAF or DBAF media, which are NB medium 5 , modified freshwater FW medium 13 , and DB medium 51 supplemented with 15 mM (FWAF and NBAF) or 20 mM (DBAF) acetate as the electron donor and 40 mM fumarate as the electron acceptor. Na 2 SeO 4 (1 M) was also added to NBAF and DBAF media to stimulate growth, as reported elsewhere 122 . 2.2.2 D NA manipulations and mutant construction. Deletions of the pilB , pilA , and omcZ genes were constructed using the cre - lox system 52 with the primers listed in Table 2 . The general procedure included the PCR - amplification of the gentamycin (Gm) resistance cassette ( aaaC1 ) flanked by loxP sites (Gm - loxP ) from plasmid pCM351 52 using primer set RS21 - RS22 and of the upstream/downstream target chromosomal regions using primer sets RS1 - RS2/RS3 - RS4 ( pilB ), RS5 - RS6/RS7 - RS8 ( pilA ) , and RS9 - RS10/RS11 - RS12 ( omcZ ). The upstream region of the target gene, the Gm - loxP cassette, and the downstream region of each gene were then fused in that order by overlap extension PCR using the corresponding external forward and reverse primers and th e Herculase II Fusion DNA Polymerase (Agilent Technologies). PCR conditions for this last amplification step were: 2 min of denaturation at 95°C; 35 cycles of 20 s at 95°C, 25 s at 54°C, and 2 min at 72°C; and a final 3 - min extension at 72°C. The PCR produ cts were then separated on an agarose gel, purified using the Zymoclean TM Gel DNA Recovery Kit (Zymo Research), and cloned into the pCR2.1 plasmid ( Table 1 ) using the TOPO® TA Cloning kit (Life Technologies) for sequence 24 confirmat ion. The cloned fragments were then PCR - amplified with the external primers and the linear fragment, once purified from an agarose gel, was electroporated into electrocompetent cells of G. sulfurreducens by following a previously published procedure 5 . Selection of recombinant strains was performed on NBAF plates supplem ented with 5 µg/ml of Gm . When indicated, the Gm cassette carried by the pilB mutant was excised from its chromosomal location by expressing the Cre recombinase fr om plasmid pCM158 123 an d selecting for marker was confirmed by PCR and the resulting mutant ( pilB ) was transferred twice in NBAF medium without kanamycin and then plated on NBAF with and without kanamycin to confirm the loss of the plasmid. The pilB strain was used to construct the double pi lB gspE, and pilB mshE mutants (Table 1) using the general PCR procedure described above but with primer sets RS9 - RS12, RS13 - RS16, and RS17 - RS20 ( Table 2 ), respectively. I constructed the pilA - E5A and Tyr3 mutants by introducing one (E5A) or three (Y27A Y32A Y57A) alanine replacements, respectively, in targeted codons in the pilA gene (GSU1496). The primers used to construct th ese mutants are listed in Table 2 . The general procedure was to PCR - amplify the pilA gene and the pilA downstream region using primer sets RS23 - RS24 and RS25 - RS26, respectively, as well as the spectinomycin (Sp) resistance cassett e from plasmid pRG5 124 using primer set R S27 - RS28. The three fragments were then fused by overlap extension PCR with the external RS23 and RS26 primers to generate a 1684 - bp DNA construct ( pilA - Sp) containing the pilA gene, the Sp resistance cassette ( aadA ), and the pilA downstrea m region. The co nstruct was gel purified before cloning it into plasmid pCR®2.1 - TOPO® TA vector (Invitrogen TM ) and introducing targeted nucleotide substitutions in the pilA gene using the QuikChange Lightning Site - Directed Mutagenesis kit (Agilent Technologies). The E5A s ubstitution was generated with primers RS29 and RS30. Tyr3 was generated by sequentially introducing Y27A first (RS31 - RS32), then Y32A (RS33 - RS34) and, lastly, Y57A 25 (RS35 - RS36). The mutated fragments were confirmed by sequencing before PCR - amplification us ing the external primers, gel purification, and electroporation into electrocompetent cells of G. sulfurreducens . Se lection of recombinant strains w as performed on NBAF plates supplemented with 75 µg/ml of Sp. 2.2.3 Microbial electrochemical cells ( MEC s). The growth and electro chemical activity of the WT and mutant biofilms were assayed in H - type MECs equipped with anode and cathode graphite rod electrodes and a 3 M Ag/AgCl reference electrode. The MECs were set up, inoculated with cell suspensions, and operate d with a poised anode electrode (0.24 V versus reference electrode) as described previously 122 . T he electron donor in the anode chambers was always acetate, provided in 1, 2, or 3 mM concentrations, as indicated. Supernatant samples were periodically removed from the anode medium broth, filtered (0.4 5 µm), and analyzed by high - performance liquid chromatography (HPLC), as described elsewhere 54 , to monitor ele ctron donor removal. Current production was recorded with a VSP potentiostat (BioLogic) . The ability of the cells to grow from the oxidation of acetate coupled to the reduction of the anode electrode was inferred from the rates of linear current increase ( mA/day) before maximum current was reached and the deceleration phase in current production was initiated . The exponential phase of current production was also fitted statistically (R 2 > 0.98) to an exponential curve and the exponent of the resulting formu la was used to estimate the generation times of the anode biofilm cells. At the end of the MEC experiment, when current had decreased to < 0.1 mA, the anode electrodes were removed from the chamber and the live and dead biofilm cells were differentially st ained in green and red, respectively, with the SYTO 9 and propidium iodide dyes of the BacLight TM viability kit (Invitrogen). The anode electrodes with the stained biofilms were then immersed gently in a Lab - Tek coverglass chamber (Nunc) filled with 3 ml of phosphate 26 buffer ed saline and examined using a FluoView FV1000 inverted microscope system (Olympus, Center Valley, PA) equipped with an Olympus UPLFLN 40X oil immersion objective (numerical aperture, 1.30). SYTO 9 was excited at 488 nm and propidium iod ide was excited at 543 nm. Vertical 2D - images of the biofilms were collected every 1 µm from approximately 10 random fields (1,024 by 1,024 pixels, 0.31 µm/pixel) per electrode, using a minimum of two biological MEC replicates. The biofilm thickness was th en manually analyzed by averaging thickness measurements of 5 representative areas per field. When indicated, the COMSTAT software 126 was used to estimate the biofilm biomass based on the fluorescence emitted from the live cells, as described previously 122 . 2.2.4 Static biofilm assays on plastic surfaces . When indicated, the biofilm phenotype of th e WT and mutant strains was also investigated on plastic surfaces using a soluble electron acceptor (fumarate) essentially as described elsewhere 15 . Briefly, the strains were grown in FW medium with 30 mM acetate and 40 mM fumarate (FWAF) to mid - exponential phase and inoculated to a final OD 600 of 0.02 into 6 - well, polystyrene, tissue culture treated plate (Costar , Corning ® Life Sciences). After 48 h of incubation at 30°C, the culture broth was discarded and the biofilms were stained for 30 min with 15 solution of crystal violet. After staining, the biofilms were washed with double distilled H 2 O and dried overnight before resolubilizing the biofilm - associated crystal violet with 33% (v/v) acetic acid. The optical density of the solution was read at 580 nm to estimate the biofilm biomass. 2.2.5 Gene expression analyses by quantitative Real Time - PCR (qRT - PCR). The expression of key components of the biofilm matrix (pili, EPS matrix, and matrix - associated c - type cytochromes ) was st udied by measuring the transcript levels of selected genes by qRT - PCR. The gene targets were the pilin - encoding pilA (GSU1496) and its 27 downstream gene (GSU1497) in the pilA operon 34 , the gene encoding the ATP - depend ent EPS exporter , xapD (GSU1501) 30 , and omcZ (GSU2078), which encodes for the precursor of the matrix - associated c - type cytochromes OmcZ S 20 . The constitutive gene rpoD (GSU30 89) was used as a control. WT, pilB, pilA, and pilA - E5A biofilms were grown for 48 h in the wells of 48 - well polystyrene plates containing 600 resuspending the biofilm cells in 50% (v/v) of ice - cold methanol to stop transcription. The cell suspension s were centrifuged to isolate the cells as pellet s and t rizol reagent (Invitrogen) was used to extr act their RNA. T reatment of the RNA with RNase - free DNase (Promega) and reverse transcription (RT) with random primers (Promega) using the Super Script ® III Reverse Transcriptase (Invitrogen) were carried out following the manufacturer s recommendations. q RT - PCR was performed with the rEVAluation qPCR Master Mix (Syzygy), as recommended by the manufacturer, using the primers listed in Table 2 . The comparative C T method 127 was used to calculate the relative expression of each gene using rpoD constitutive expression as an T value or C T ( target ) C T ( rpoD ) ) for each strain, and the average of the T T T . The relative fold change of expression fo r each target gene versus the rpoD internal control for each mutant strain versus the WT was then calculated with the formula 2 - . 2.2.6 Assays for c - type cytochrome content and profiling. The total (cells and biofilm matrix) c - type cytochrome content of bio films of the WT, pilB, pilA, and pilA - E5A strains was estimated by redox difference spectroscopy . The biofilms were first grown for 48 h in 96 - well plates, as described above for the biofilm assays on plastic surfaces, before discarding the culture broth a nd freezing the biofilms by storing the plates at - 80 ° C for a minimum of 24 h. T he plates were thawed at room temperature for a minimum of 15 min before suspending the biofilms in FW medium at room temperature. For each strain, I pooled together biofilm s amples from 96 wells and adjusted the biofilm suspension to an OD 600 28 of 0.2 with FW medium . Sodium dodecyl sulfate (SDS) (final concentration of 0.1% w/v) and fresh 1 mM dithionite w ere then added to the suspension to lyse the cells and as a reducing agent , respectively. All of the steps were performed anaerobically inside a glove bag (Coy Laboratories). The ultraviolet - visible ( UV - Vis ) spectrum from 350 - 650 nm of the reduced biofilm cell extract was then collected before and after oxidation with 2 mM ferri cyanide using a Shimadzu UV - 2401 spectrophotometer . The difference between the reduced and oxidized and pili - deficient mutants (pilB, pilA, and pilA - E5A) was u sed to estimate the overall biofilm c - type cytochromes content (cells and biofilm matrix) 56 . Spectra were also collected from cell extracts obtained from planktonic cells harvested from exponential (OD 600 . 0.4 - 0.5) cultures grown in FWAF at 30 o C. The total heme content of the cells was analyzed using the alkaline pyridine hemochrome method as described previously 129 . P rote ins in the biofilm matrix in the WT and pili - deficient strains w ere also isolated using the procedure described by Rollefson et al. 30 and modified by Cologgi et al. 15 . The matrix - associate proteins were separated electrophoretically by SDS - PAGE as previously described 30 except that samples were loaded onto 12% Mini - Protean TGX gels (Bio - Rad) and Novex Sharp molecular weight markers (Invitrogen) were used as standards. Heme - containing protein ba nds were stained with - tetramethylbenzidine (TMBZ), as previously described 130 . Replicate gels were run and stained with Coomassie Blue to ensure that the total protein content per lane was comparable. When indicated, SDS - PAGE and heme - staining were also used to profile all heme - containing proteins within planktonic cells grown in FWAF to mid - exponential phase , as reported elsewhere 130 . 2.2.7 Pili purification and conductivity measurements by CP - AFM. Pili from WT or the Tyr3 mutant strain of G. sulfurreducens were purified as SDS - insoluble fractions by preparative electrophoresis, as previously described 13 except that all 29 buffers contained 1 mM ethylenediaminetetraacetic acid (EDTA) and all drying steps were carried out under a constant flow of filter - sterilized (0.22 µm) N 2 gas. The pili were deposited on the surface of freshly cleaved highly oriented pyrolytic graphite (HOPG) for 30 min, then blotted dry, before probing their transversal conductivity by Conductive Probe - Atomic Force Microscopy (CP - AFM). The sample s were scanned with the AFM tip in tapping mode to image the pili and to identify individual fibers. The CP - AFM tip was then positioned at different points along each pilus filament to measure its transversal conductivity while applying a bias voltage within the ± 1 V range (3 nN force, 1 Hz rate). Two to three current - voltage ( I - V ) curves were collected from each of at least three positions along each pilus filament and four to five pilus fibers were probed for each strain to account for technical and biological replication . During the measurements , the tip was periodically moved to the HOPG surface adjacent to the pili to control tip quality. T he resistance ( R ) values were calculated f rom the inverse of the linear portion of each I - V curve and averaged for each strain ( WT an d Tyr3 ). Results 2.3.1 I nactivation of the PilB assembly motor does not affect c - type cytochromes expression. The assembly of type IVa pili on the bacterial inner membrane is mediated by a protein apparatus involving three interacting subcomplexes: the pilus subcomplex (major pilin and minor pilins, if present), the motor subcomplex (containing several proteins, including the PilB and PilT ATPases that power pilus polymerization and retraction, respectively), and the alignment subcomplex (which properly aligns the assembled pilus through the outer membrane PilQF secretin complex) 131 . As mutations that render the pilus and motor subcomplexes inoperative also prevent pili expression 131 , I constructed mutants of G. sulfurreducens carrying deletions in the genes encoding the PilA pilin subunit (pilA mutant) and the PilB ATPase (pilB mutant), 30 respectively. Additionally, I constructed a pili - defici ent mutant (pilA - E5A) carrying an alanine substitution in the conserved E5 amino acid required for proper alignment and assembly of pilins 52,66 . Two of the pili - inactivating mutations, pilA and p ilA - E5A, had pleiotropic effects in cytochrome expression. Differential redox spectroscopy of reduced - minus - oxidized UV - VIS spectra ( Figure 6 ) revealed, for example, an increase d expression of c - type cytochromes in the pilA and pi lA - E 5A biofilms compared to the WT biofilm ( Figure 9 A ). Similar increases were observed in planktonic cells o f the pilA and pilA - E5A strains ( Figure 7 ), ruling out any influence of the physiological stat e of the cells (planktonic or biofilm) in the phenotype. Some of the pleiotropic effects observed in these mutants were transcriptional in nature. For example, transcripts levels for the gene encoding the outer membrane c - type cytochromes OmcB were up - regu lated in the pilA and pilA - E5A mutants ( Figure 8 ). On the other hand, other pleiotropic effects were post - transcriptional. Transcripts for the matrix - associated cytochrome OmcZ were, for example, similar in the WT and the pilA and pilA - E5A mutants ( Figure 8 ) . Yet the small (~30 - kDa), processed OmcZ S cytochrome that is release d in to the biofilm matrix 30 and is required for optimal biofilm electroactivity 15,119 was absent in the matrix of pilA and pilA - E5A biofilms ( Figure 9 ). Interestingly , OmcZ S was d etected in heme - stained preparations of planktonic cells ( Figure 7 ) , suggesting that the OmcZ S defect of the pilA and pilA - E5A mutant biofilms was related to the unique physiology of cells living within a surface - attached communit y. The xapD gene, which encodes an ATP - dependent exporter required for the synthesis of the biofilm EPS matrix that anchors c - type cytochromes 30 , was also transcriptio nally up - regulated in the pilA and pilA - E5A mutants ( Figure 8 ). These mutants also formed dense biofilms on plastic su rfaces under static conditions ( Figure 9 ) , a phenotype that ha s been linked to the ove rproduction of the biofilm EPS 132 . 31 In contrast with the pleiotropic nature of the pilA and pilA - E5A mutations, i nactivation of the PilB motor to prevent pilin assembly had no effect on the over all c - type cytochromes content of the biofilms ( Figure 9 ). Moreover, the mutation did not affect the transcription of the two genes of the pilin operon ( pilA and GSU1497) or the genes encoding the outer membrane c - type cytochromes omcB and omcZ ( Figure 8 ). The expression of OmcZ S in the pilB biof ilm matrix was also unaffected ( Figure 9 ). The synthesis of the biofilm EPS was not affected in the pilB mutant biofilms either, as xapD transcript levels were s imilar to those in WT biofilms ( Figure 8 ) . The pili deficiency of the pilB mutant also prevented the formati on of dense biofilms on plastic surfaces but the biofilm defect was rescued in the genetically - com plemented pilB+ strain, which restored pili production through the expression of the pilB gene in trans from a medium - copy plasmid ( Figure 9 ). Hence, the pilB strain provides the elusive genetic tool needed to assess the contribut ion of Geobacter conductive pili in the growth and electroactivity of anode biofilms. 2.3.2 Pili expression is required for optimal electrochemical activity of t hin biofilms. I investigated the ability of the pilB mutant to generate current in a n MEC containing 1 mM acetate in reference to the WT strain. Under these conditions, the WT biofilms grew reproducibly to a thickness ( ca. 10 µm) that is not predicted to limit electron transfer to the electrode surface via c - type cytochromes 37 39,133 135 . As a control, I also constructed a null omcZ mutant (omcZ), which cannot produce the matrix - associated cytochrome OmcZ S required for electron transport to the underlying electrode 119 . Maximum current in the pilA and pilA - E5A mutant biofilms, which lack both pili and OmcZ S ( Figure 10 ) , was low (0.08 ± 0.05 and 0.14 ± 0.08 mA, respectively, for duplicate MECs) and required prolonged incubations to oxidize all the acetate in the anode chamber ( Figure 10 ) . The phenotype was similar to the omcZ control biofilms, which also produced low levels of current (0.11 ± 0.07 mA for duplicate MECs) and req uired lengthy incubation times ( Figure 10 ). The defect of the pilA and pilA - E5A mutants in the 32 MECs cannot be attributed to a reduced number of cells growing on the electrode, because the biofilms eventually reached WT thickness ( ca. 13.5 ± 3.5 and 11.6 ± 1.9 µm respectively, for duplicate MECs) , they just grew slower . Hence, the results indicate that the inabi lity of the pilA and pilA - E5A mutant biofilms to produce pili and the matrix - associated OmcZ S c - type cytochromes limited their electroactivity. The pilB mutant, which has WT levels of c - type cytochromes and expresses OmcZ S in the biofilm matrix ( Figure 9 ) grew biofilms on the anode electrode with WT thickness (9.6 ± 1.4 µm for triplicate MECs). However, the maximum current harnessed from the anode biofilms was approximately half of that measured in the WT biofilms ( Figure 10 ) . Such low levels of current are still significantly higher than those measured in the other pili - deficient mutant strains pilA and pilA - E5A, which, in addition to the pili deficiency, failed to express OmcZ S in the biofilm matrix ( Figure 9 ). The partial restoration of biofilm electroactivity in pilB did not result from functional complementation of the pilB mutation by functionally homologous genes, because deletion of the two pilB homologues of G. sulfurreduc ens ( mshE and gspE , encoding putative ATPase motors of type II secretion systems) generated double mutants that were phenotypically indistinguishab le from the pilB single mutant ( Figure 10 inset) . Hence, the pilB mutant phenotype is consistent with reductions in biofilm electroactivity caused by the inability of the mutant cells to assemble the pili. It is interesting to note that the pili defect of the pilB mut ant could not be complemented by the co - inoculation of omcZ with pilB m utant cells ( Figure 10 inset ). I did observe a delay in electrode colonization in the co - culture - driven MECs, which was similar to the delay experienced by single mutants that failed to express OmcZ S ( omcZ , pilA and pilA - E5A) ( Figure 10 ). However, once the electrode was colonized, the biofilms produced current similarly to the pilB single mutant. Hence, the results suggest that biofilm cells need to express both pili and OmcZ S for optimal biofilm growth an d electroactivity. 33 I n addition to a 2 - fold reduction in current maxima, the rates of current increase , which correlate well with the rates of exponential growth of the biofilm cells on the anode 136 , measured in the MECs driven by the pilB mutant were red uced in half (0.61 ± 0.15 mA/d) compared to the WT biofilms (1.45 ± 0.06 mA/d). Coulombic efficiencies in the pilB - driven MECs (96.6 ± 3.1) were similar to the WT (94.6 ± 4.1), indicating that, on average, the pilB biofilms converted the same amount of ace tate to electricity as the WT biofilms, they just did so at slower rates. The pilB defects cannot be attributed to a reduction in the number of cells actively contributing to electron transfer or structural variations of the biofilms either. Confocal micro graphs of WT and pilB biofilms collected at the end of the experiment, when all of the acetate had been depleted, and stained with fluorescent viability dyes showed predominantly live cells and similar biofilm structure ( Figure 10 ) . Furthermore, the biofilm biomass estimated from the fluorescence emitted by the biofilm cells was similar in both strains (10.05 ± 1.00 in WT and 9.06 ± 1.46 m 3 / m 2 in pilB). The pilB phenotypes are therefore consistent with reductions in the respirato ry rate of individual cells, which is expected in a mutant with defects in ex tracellular electron transfer. 2.3.3 The expression and conductivity of the p ili are required to grow thick biofilms. The genetic and MEC data presented thus far indicate that op tima l electroactivity of thin ( ca. 10 - µm ) biofilms of G. sulfurreducens requires both the matrix - associated c - type cytochromes OmcZ S and the pili. As the biofilms grow thicker, cytochromes become progressively reduced and unable to accept electrons from cells in the distal layers. To test if the pili could be t he charge carriers in this distal biofilm stratum , I grew the pilB mutant in MECs fed with 2 and 3 mM acetate, which are concentrations of electron donor that support the growth of thicker ( ca. 15 and 20 µm, respectively) WT biofilms ( Figure 11 ) . In contrast to the linear increases in biofilm thickness observed in the WT - driven MECs as a function of acetate concentration, th e pilB biofilms remained thin ( ca. 10 µm) u nder all the c onditions tested ( Figure 34 11 ). Furthermore, whereas the rates of current increase during the linear phase of current production and current maxima increased proportionally to acetate availability in the WT MECs, they we re unaffecte d in the pilB MECs ( Figure 11 ). To rule out a str uctural role for the pili, where their absence could have prevented the pilB biofilms from growing beyond 10 µm, I constructed a mutant strain (Tyr3) that produced pili carrying al residues are predicted by ab initio calculations to contribute greatly to the unique electronic structure and low electron ban d gap of the pilin without interfering with pi lin assembly 45 . Furthermore, the structure of the pilus fiber revealed in molecular dynamics simulations supports a model of multistep hopping involving tyrosines 137 . Supporting this model , the Tyr3 mutant produced pili a t WT levels ( Figure 12 ) and formed biofilms as dense as the WT on plastic surfaces ( Figure 12 ) which are conditions that only require pili expression to provide structural support to grow multilayered bio films 32 . However , pili purified from the Tyr3 mutant were poorly conductive, having more than 5 - fold greater resistance to the passage of electrons than the WT pili ( Figure 12 ) . The reduced conductivity of the Tyr3 pili also reduced the electroactivity of the Tyr3 anode biofilms in MECs fed with 3 mM acetate and prevented the growth of the biofilms beyond the threshold 10 µm thickness, as observed in the pil i - deficient strain pilB ( Figure 11 ) . Hence, the conductive properties of the pili, rather than their expression to provide structural support, are required to build thick (> 10 m) biofilms and maintain optimal biofilm electroacti vity in the distal biofilm layers where the cytochrome pathway is inoperative. Hence, the results support the notion that the pili function as electronic conduits in the upper biofilm strat a of thick biofilms as well . 35 Discussion The studies described he rein provide the elusive genetic evidence supporting the role of Geobacter pili as electronic conduits in electroactive biofilms , but also reveal novel insights into how Geobacter cells coordinately export c - type cytochromes and assemble pili in the biofil m matrix to establish an electronic network that maximizes long - range electron transfer. Part of the difficulty in interpreting earlier genetic studies of pili function 31,34,77 stemmed from the pleiotropic effects that mutations in pilin expression have in c - type cytochromes required for optimal extracellular electron transfer 13,15 . Supporting these findings, I show that mutations in the pilin gene (pilA and pilA - E5A) have numerous pleiotropic effects in cytochrome expression and that such effects are likely exerted thro ugh diverse regulatory mechanisms, both transcriptional and post - transcriptional. Furthermore, as observed in other bacteria 138,139 , the pilin gene mutations were epistatic to EPS synthesis , at least at the transcriptional level . By contrast, inactivation of the gene encoding PilB, the ATPase that powers pilin assembly, prevented pili formation wit hout disrupting the expression of other components of the biofilm matrix ( c - type cytochromes and EPS). The lack of multiple phenotypes in the pilB mutant contrasted with those observed in the pilA or pilA - E5A strains and adds to the growing body of evidenc e indicating that type IV pili play multiple roles in bacteria. In Pseudomonas aeruginosa , for example, the PilA pilin 140 , but not the pilin assembly motor PilB 141 , is essential for efficient protein secretion. This is because PilA, in addition to functioning as the structural subunit of the pilus fiber, also interacts with components of the genera l secretion pathway to promote the export of proteins across the outer membrane 140 . As a result, mutations that inactivate PilA in this bacterium also affect the secretion of numerous proteins and lead to mutant strains with multiple phenotypes. By contrast, the function of PilB is restricted to pilus biogenesis, and strains of P. aeruginosa carryi ng inactivating mutations in pilB do not display export defects 141 . Interestingly, the PilA pilin protein of G. sulfurreducens is translated as both short and long prepilin isoforms 142 . Interactions 36 between the two isoforms promote the processing of the short pilin and its assembly to form the conductive pilus fiber but also influence the se cretion and localization of c - type cytochromes, such as OmcZ, to the outer - membrane 142 . Hence, pilin - inactivating mutations such as pilA and pilA - expression and localization. However, as in P. aeruginosa 141 , the PilB motor of G. sulfurreducens is only involved in pilus formation. Hence, inactivating pilB prevents pilin assembly and has no effect in the secretion and processing of other biofilm com ponents such matrix - associated c - type cytochromes . The clean phenotype of the pilB mutant provided the elusive genetic tool needed to investigate the role of pili as biofilm electronic conduits and to study the interactions between pili and cytochrome elec tron carriers in electroactive biofilms. By controlling the amount of electron donor added to the anode chamber of the MECs (1, 2, or 3 mM acetate), I restricted the growth of the anode WT biofilms to ca. as able to study the contrib ution of the pili to the biofilm electroactivity in multilayered communities where all of the biofilm cytochromes are predicted to function as electron carriers (10 - reduced in the distal biofilm layers (15 - and 20 - . The electroactivity of thin ( ca. - fold reductions measured in the rates of current production and the maximum current by pilB biofilms grown in MECs fed with 1 mM acetate ( Figure 9 ) . Such defects cannot be attributed to decreases in the number of biofilm cells contributing to electron transfer, because the thickness and the number of viable cells in the pilB biofilms were comparable to the WT. Hence, even in thin biofilms, where the efficiency of the cytochrome pathway is not distance - limited, pili expression is required for optimal electroactivity. Furthermore, biofilms permeated by the poorly conductive Tyr3 pili remaine d thin and were phenotypically indistinguishable from the pili - deficient pilB biofilms, producing current at rates half of those m easured in WT biofilms as well 37 ( Figure 12 ). Hence, even with a structural network of pili, the elect roactivity of Tyr3 biofilms was is restricted to providing structural support needed for cytochrome organization 42 . Rather, the genetic evidence indi cated that the pili and cytochrome electronic pathways may work coordinately in thin biofilms. Our results also add to the growing evidence that OmcZ S , the processed and predominant extracellular form of the outer membrane c - type cytochromes OmcZ 20 , is essential for optimal electron transfer across anode biofilms. Pili - deficient strains (pilA and pilA - E5A) with OmcZ S defects had a more pronounced defect in current production than the pilB strain ( Figure 10 ), whose EPS matrix contains Omc Z s ( Figure 9 ). Furthermore, the defect of the pilA and pilA - E5A strains was similar to the omcZ strain, which carries a deletion in the omcZ gene that prevents it from expressing the cytochrome. Given the preferential localization of OmcZ S in the biofilm stratum closer to the anode electrode 36 and the dramatic effect of inactivation of the omcZ gene on the growth and electroactivity of the anode biofilm 119 , it is likely that OmcZ S is the primary electron carrier of the cytochrome pathway in the regions of the bi ofilm matrix closer to the electrode . While it was possible to grow thicker ( ca. 15 and 20 µm) WT biofilms, preventing the assembly of the pilins (pilB mutant) or reducing the conductivity of the assembled pili (Tyr3 mutant) effectively prevented the biof ilms from growing more than 10 µm away from the electrode ( Figure 12 ) . Hence, the expression and conductive properties of the pili are required to maintain optimal electrical connectivity as the biofilms grow in thickness and elec tron transfer via the cytochrome pathway becomes limited 38,39 . The fact that cytochromes in the upper biofilm strata remain reduced even when a positive vo ltage is applied to the biofilm 38 , suggests that they do not have a primary role as electron carriers in these regions. Paradoxically, these are also the regions where electron donor (acetate) availability is highest 40 . Studies show that G. 38 sulfurreducens ca n continue to metabolize acetate in the absence of an electron acceptor by overexpressing its extracytoplasmic c - type cytochromes and using them as a capacitor to store up to 10 6 electrons per cell 143 . In support of this, the expression of the outer membrane c - type cytochromes OmcB is transcriptionally up - regulated under conditions of electron acceptor limitation 144 and levels of OmcB prot ein incre ase in biofilm cells located more than 10 µm from the electrode 54 . The capacitor role of cytochromes in these upper biofilm regions could allow cells to continue to metabolize acetate and generate a proton motive force for energy generation. The pilus apparatus is an chored in the cell envelope of G ram - negative bacteria; hence, it could potentially accept electrons from the extracytoplasmic cytochromes, promoting their discharge from the cell to the biofilm matrix. Furthermo re, pili can grow several µ m in length and intertwine to form a comp lex web of nanowires that could electronically connect the strata of the biofilm distal and proximal to the electrode , effectively bypassing the limitation of the cyt ochrome pathway in the electron acceptor - limited regions. This pili network may also be required as cytochrome electron acceptors in the distal strata of a biofilm; as the observed electron gra dient leads these cytochromes to constantly be in a reduce d state, making diffusion based mechanisms too slow to support growth Taken together, the result s support a model of mechanism - based stratification in electroactive biofilm s. The pili electronic net work is essential for charge transport in all the biofilm strata, working coordinately with the matrix - associated cytochrome carriers to transfer electrons in the stratum closest to the electrode. It has been previously proposed that electron transfer acr oss electroactive biofilms may require the expression of both cytochromes and pili to promote short - distance, electron transfer reactions 146 . For example, electron transfer out of a pilus (i.e. pili - to - pili or pili - to - electron acceptor) may be facilitated by cytochromes bound to the pilus fiber s 42 . Of note, no additional cytochromes are currently known to be necessary for 3 9 efficient current production, and it is unknown what role, if any, t he cytochromes play in the distal strata. As the biofilm thickness increases, a pH gradient is established where by the pH is lower in the regions closer to the anode 147 . The accu mulation of protons in the regions of the biofilms proximal to the electrode could limit the efficie ncy of the pili pathway locally if proton - coupled electron transfer (PCET) reactions influence pilus conductivity, as pr eviously shown for other proteins relying on tyrosines for electron transfer 148 and proposed for the Geobacter pili 137 . In PCET, tyrosine residues simultaneously transfer a proton and an electron, thus lowering the systems 92 . Thus, the efficiency of the pili pathway and its electronic interactions with cytochromes may be influenced by the local pH. Nutrient gradients also form in thick biofilms, which can influence the cell physiology, the express ion of biofilm components, and the biofilm electroactivity 37,134 . Such regulatory networks may be responsive to more than one biofilm parameter, such as pH, nutrient availability, and redox potential, allowing cells to control the composition of the biofilm electron carriers to maximize the rate s of electron transfer. Biofilms of oxygen - respiring bacteria are, for example, responsive to redox potentials and use redox cues from their microenvironment to alter the biofilm structure, maximize O 2 diffusion, and maintain redox homeostasis 149 . Similar mechanisms could control the expression of cytochromes and p ili to facilitate their coordinated interactions and redox homeostasis in G. sulfurreducens biofilms. Thus, understanding the multiple functions that pili play in G. sulfurreducens and their regulation may prove instrumental to improve the performance of G eobacter - driven electrochemical systems for applications in bioenergy. 40 CHAPTER 3. SITE - DIRECTED MUTAGENESIS REVEALS THE ROLE OF AROMATIC AMINO ACIDS AND LOCAL ELECTROSTATICS IN PILUS CONDUCTIVITY AND EXTRACELLULAR ELECTRON TRANSFER Parts of this chapter were publis hed previously and have been reprinted with permission. Copyright Royal Society of Chemistry, Feliciano, G. T., Steidl, R. J. & Reguera, G. Structural and functional insights into the conductive pili of Geobacter sulfurreducens revealed in molecular dyn amics simulations. Phys. Chem. Chem. Phys. 17, 22217 22226 (2015). Reguera, G., R. Steidl, Microbial nanowires. U.S. Patent Application Serial No. 13/221,495, filed August 30, 2011,claiming priority to U.S. Provisional Patent Application Serial No. 61/378 ,188, filed on August 30, 2010. 41 Introduction In Chapter 2 I demonstrated that the conductive pili of Geobacter sulfurreducens are required for the efficient electron trans fer (ET) through anode biofilms in a microbial electrochemical cell ( MEC ) . Furtherm tyrosines (Tyr3 mutant) reduced the pilus conductivity and electrochemical activity of the anode biofilms. In this chapter, I focused on gaining insights into the role of tyrosines in these functions and, more general ly , about the role of pili as electronic conduits in the biofilms. It has been propo sed that electron transfer across anode biofilms has metallic features and these properties are due to the intrinsic metallic - like conductance of the pili 77 . The metallic model of pilus conductance relies heavily on the assumption that aromatic amino acid residues of the pilin cluster together in the pilus fiber promoting - stacking and metallic regimes 77 . As I describe in this chapter, this type of stacking requires very short inter - aromatic distances and specific (sandwich type) aromatic di mer configurations. Such geometric configuration s are rare in proteins compared to the parallel - displaced or T - shaped ring conformations , which are prevalent in proteins due to their superior stability 150 . Charge transfer in proteins is only known to proce ed via two mechanisms: tunneling and hopping 85,151 . In tunn eling, electrons travel through the medium (e.g., the proteins matrix) from donor to acceptor without residing in relay stations. As a result, tunneling rates are inversely and exponentially proportional to the distance the electron travels and become too s low to be biologically relevant at distances greater than 20 Å 84,86 . Electron hopping, on the other hand, is the sequential transfer of charges in various steps using intermediates such as redox active amino ac ids and cofactors, which act as relay stations 151 . As a result, this mechanism of ET is faster than tunneling over longer distances (ca. 25 Å ) and is only weakly dependent o n distance 151 . 42 As the pili are often several micrometers long, tunneling alone cannot explain why the G. sulfurreducens pilus transport charges. Long - range charge hopping , on the other hand, is possible if amino acids with oxidizable side chains are clustered sufficiently in the pilus so as to maintain side chain distances ( ca. 25 Å ) that permit successive short and fast electron transfer steps 152 . The hopping of electrons in proteins is mediated by redox - active amino acids such as tryptophan, tyrosine, and ET pathways 92,153,154 . The G. sulfurreducens pilin does not contain any try ptophans, cysteines, or methionines but does have three tyrosines ( Figure 2 ). Multistep hopping via tyrosines has been demonstrated in some well - studied protein systems such as ribonucleotide reductase 96 and photosystem II 155 . In these systems, electron transfer via tyrosines is coupled to the transfer - charged residue (thus, it is a p roton - coupled electron transfer). The transfer of a proton decreases the reduction potential of the tyrosine and facilitates electron transfer at the low potentials that operate in biological systems 92 . The pilin also has three pheny alanines ( Figure 2 ). Phenylalanines have too high a redox potential to be oxidized during multistep hopping. However, the presence of phenylalanine residues in several cytochrome c proteins i ncreases the ET rates compared to those without phenylalanines 100,101 . Electronic structure calculations also predicted the co nserved phenylalanine residues of bacterial NADH:ubiquione oxidoreductase to be involved in electron transfer 102 . One phenylalanine in particular, F32 8, was involved in all fast paths of electron transfer calculated in the analyses. Although the exact mechanism of ET mediated by phenylalanines in vivo is not fully understood, studies with small peptides have demonstrated that phenylalanines act as a n electron in the ET pathway and are required for optimal ET rate s 103 . 43 In addition to specific amino acids in proteins acting as sites of electron hopping , peptide studies have demonstrated that charge transfer via hopping can happen along t he amide groups of the peptide backbone particularly in helical peptides 156 . Charge hopping by the amide groups is theorized to involve hole - hopping as opposed t o electron hopping, meaning the t by the lack of an electron is filled by an electron from another amide group. In these peptide systems, electrons hopped along lengths si gnificantly longer than the G. sulfurreducens pilin peptide 68,88 . The rate of charge transfer via this backbone hopping method is increase d by the natural dipole moment of the helical peptide 67,108,109 , from positive N - terminus to negative C - terminus. The natural dipole of G . sulfurreducens pilin is accentuated due to the placement of acidic and basic residues ( Figure 2 ). - helices ( Figure 5 ) is also thought to increase ET through the peptide 112,113 . Furthermore, the movement and flexibility of organic structures, such as DNA and peptides, is predicted to increase polymer charge tran sfer rates 115,116,157,158 . As a first step in understanding how electrons move through the pilus and what amino acids could be implicated, we first sought to construct a high resolution structural model of the pilus fiber. The model would not only identify residues that may be involved in ET, but could also predict the distances and orientations of these amino acid residues . Thus it can both predict the potential for those amino acids to part icipate in multistep hopping and discern whether the residues are align ed and cluster ed to promote metallic regimes. As with other insoluble peptide assemblies such as amyloid fibers, atomic resolution structures of pili are difficult to attain because of the difficulty in preparing suitable crystals or applying liquid - state NMR. To date, the only published pseudoatomic model of a type IV pilus is that of N. gonorrhea (gonococcus or GC) 66 . This model was derived from cryo - EM data of the pilus fibers with a 12.5 Å resolution, which was used as a template to superimpose the crystal structure of the GC pili n. Similar approaches were used to construct three different homology models of the G. sulfurreducens 44 pilus fiber 58,79,83 . However, there are many discrepancies in the predictions inferred from the se models , which likely reflect differences in the structural assumptions used to construct them. The first homology model , for example, superimposed a computational homology model of the G. sulfurreducens pilin or PilA onto the GC pilus template and reported the clustering of aromatic amino acids of the pilin within a 15 - 21 Å range 83 . These inter - aromatic residue dis tances are still optimal for charge hopping, but are too large to promote the - stacking required for metallic conductance. A second model also used the GC pilus template, but inserted an atomic resolution NMR structure of the G. sulfurreducens pilin 58 . This approach generated a structural model in which the aromatic residue s clustered as 15 Å - wide bands along the pili, which were separated by aromatic ring - free regions. The model fails to explain how charges would move from one aromatic ring - dense band to another. A third homology model used the G. sulfurreducens NMR pilin model , but superimpo se d the structure into computationally generated P. aeruginosa or PAK pilus fiber template, which was inferred from diffraction data for the pilus 50 . This study reported a path of aromatic residues 3 - 4 Å apart from each other, which are inter - aromatic residue distances that could support metallic conducta nce. However, the aromatic ring residues are not face - centered (sandwich conformation), wh ich is required for metallic regimes to operate. Furthermore, some of the reported 3 - 4 Å distances involved the peptide backbone rather than the aromatic ring itself. Other deficiencies were noted in this homology model, particularly the fact that the phen ylalanine F1 is bonded to the F24 of an adjacent pilin in the lower portion of the pilus fiber. Inter - aromatic residue distances in the upper portion of the fiber, where the F1 - F24 bond is not present, exceed 3 - 4 Å . Key to constructing a reliable homology model is to minimize the number of structural assumptions by using experimentally validated structures for both the pilin and the pilus fiber used as template. To date, the only validated structures are the NMR - derived structural model of the pilin resolve d in lipid micelle s 58 and the GC pilus structure derived from cryo - EM data 66 . 45 Our collaborator Dr. Gustavo Feliciano used molecular dynamics (MD) to optimize the NMR pilin structure in solution and superimpose it into the GC pilus template. This allowed him to generate a homology model of the G. sulfurreducens pilus wi th minimal structural assumptions. He then improve d the accuracy of the homology model by refin ing it to higher resol ution in MD simulations 137 . My role in this project was to test the model predictions by introducing targeted amino acid re placements in residues of the pilin predicted to have a direct role in electron transfer (tyrosines) and those influencing the electrostatic environment around the tyrosines (charged amino acids), which can affect the redox potential, configuration and dis tances of the aromatic residues and, indirectly, the rates of electron transfer. l also performed independent analyses of the model to interpret the phenotypes of some pilin mutants and develop a model for ET along the pilus and to external electron accept ors such as insoluble Fe(III) oxides or the soluble uranyl cation. When indicated, I worked with other members of the Reguera lab (Dr. Sanela Lampa - Pastirk and Krista Cosert) to measure the conductivity of purified pi li carrying selected mutations. Materia ls and Methods 3.2.1 Growth conditions The bacterial strains used in this study are described in Table 3 . The wild - type (WT) strain, G. sulfurreducens strain PCA, was kindly provided by Daniel Bond (University of Minnesota) and was us ed to construct the amino acid replacement mutants. To distinguish this strain from the laboratory strain routinely used by other members of my lab, I designated it strain DB (for Daniel Bond). The WT and mutant strains were routinely cultured anaerobicall y in NBAF, FWAF or DBAF media, which are NB medium 5 , modified freshwater ( FW ) medium 13 , and DB medium 122 supplemented with 15 mM (FWAF and NBAF) or 20 mM (DBAF) acetate as the electron donor 46 and 40 mM fumarate as the electron acceptor. Na 2 S eO 4 (1 M) was also added to NBAF and DBAF media to stimulate growth, as reported elsewhere 122 . 3.2.2 M utant construction I constructed the amino acid replacement mutants by introducing alanine or phenylalanine replacements in targeted codons in the pilA gene (GSU1496). The primers used to construct these mu tants are listed in Table 4 . Amino a c id replacements were generated by introducing targeted nucleotide substitutions in the pilA gene using the QuikChange Lightning Site - Directed Mutagenesis kit (Agilent Technologies) and the a ppropriate primers (and pCR2.1 - pilASpec , as described in Chapte r 2 . The mutated fragments were confirmed by sequencing before PCR - amplification using the external primers, gel purification, and electroporation into electrocompetent cells of G. sulfurreduce ns . Selection of recombinant strains was performed on NBAF plates supplemented with 75 µg/ml of spectinomycin. 3.2.3 Pili purification and conductivity analyses. Pili from the WT or the Y27A mutant strains of G. sulfurreducens were purified to homogeneity as p reviously described 13 , except that all the buffers used during the purification contained 1 mM ethylenediaminetetraacetic acid (EDTA) and all drying steps were carried out with a constant flow of filter - sterilized N 2 gas, r ather than in a Speed Vac, to prevent contamination. Samples were stored on ice for s hort - term use or flash frozen in liquid nitrogen and stored at - 80°C for long - term use. The transversal conductivity of purified pili deposited on graphite was measured wi th a conductive probe - atomic force microscope (CP - AFM), as described in Chapter 2. When indicated, charge transport along pili purified from the WT and a Y27A mutant was also measured by Dr. Sanela Lampa - Pastirk. For these experiments, pili were deposited on patterned gold electrodes nanofabricated onto a silicon chip by Dr. Lampa - Pastirk, as described 47 elsewhere 159 . After pili deposition, the electrodes were stored in an air - tight container under a flow of N 2 gas for 1 h and immediately probed for axial conductivity. CP - AFM was performed with Ti/Ir - coated silicon cantilevers havi ng a nominal spring constant 2 N/m (ASYELEC - 01, Asylum Research) and using a Cypher scanning probe microscope (Asylum Research). Pilus fibers lying across the gold - SiO 2 interface were first identified in non - contact amplitude modulation (tapping or AC) ima ging mode and then probed with the conductive tip (3 nN force, 1 Hz rate) at different points along the pilus fiber starting at regions further apart from the electrode to those closest. T he region of the pilus in direct physical contact with gold (designa ted as 0 nm) was also probed as well as p ositive (bare gold) and negative ( SiO 2 substrate ) controls. The resistance ( R ) at each position along the pilus fibers was calculated from the inverse slope of the regression line that fit the portion of the I - V cur ve exhibiting the most significant ohmic dependence of current on voltage (usually within the ± 0.5 - 0.7 V range). When indicated, I - V curves were smoothed using the 75 point Savitzky - Golay smooth function prior to fitting using the IgorPro software (WaveMe trics). Several pili purified from 4 WT or 2 Y27A independent cultures were used to collect I - V curves for several regions per pilus fiber. Each pilus region was probed at several bias voltages (often 0.1, 0.5 - 0.7 and 0.9 - 1 V) to account for technical rep lication and the average resistance, R , for each pilus region was calculated from inverse slope of the linear portion of the I - V curves. The resistance values were then plotted as a function of the distance (pilus length from the point of probing to the go ld edge) using the Microsoft Excel software and the best fit was selected based on the trendline correlation coefficient (R 2 ). The equation obtained for the best fit in the lower linear fit range (ca. 600 n m and below) was used to calculate the average res istance ( R , - µm long pilus and this value was used to estimate the current ( I , in amps) along the pilus at an applied voltage ( V ). As one 48 amp represents the flow of 1 coulomb of electrical charge per sec ond, I calculated the electron transport rates by multiplying the current value ( I ) along a 1 - µm long pilus by the number of electrons in one coulomb (6.2415 × 10 18 ). The average resistance value of a 1 - µm long pilus was also used to calculate the resisti vity ( where L is the nanowire length (in cm) and A is the cross - sectional nanowire area (calculated in cm 2 using the 2 nm height measured by AFM for a cylindrical pilus nanowire. 3.2.4 Calculation of Fe(III) oxi de respiratory rates . The con ductivity of the pili was interpreted in the context of the respiratory rates of the cells expressing pili during the reduction of Fe(III) oxides. I used the previously reported 34 rates of Fe(II) accumulation from the reduction of poorly crystalline Fe(III) oxide reduction and cell growth (measured as number of cells from cultures doubling every 15 h) to infer the amount of Fe(II) (in mol) solubilized per cell. The moles of Fe(II) per ce ll were then used to estimate the electron transport rate per cell by multiplying the moles of Fe(II) per cell by 6.0221413x10 23 - electron reaction). The electrons exported to the Fe(III) o xid es per cell during the exponential phas e of growth are shown . From the linear fit of this plot, I estim ated transport rates of 9 x 10 6 electrons per cell per sec . 3.2.5 Electrochemical activity of anode biofilms in MECs The growth of cells and electroactivity of the biofilms from WT and mutant cells were assayed as in Chapter 2. HPLC and CSLM were also performed as in Chapter 2. I estimated the respiratory rates of anode biofilms as electrons per cell by dividing the current maxima of 49 WT anode biofilms grown i n MECs fed with 1 mM acetate by the biofilm biomass estimated from confocal micrographs using the COMSTAT software 126 as described in Chapter 2. Electrons transferred per cell per sec were calculated as for the cells grown with Fe(III) oxides. The surface area of the electrode (1235431 µ m 2 ) and previ ously reported cell dimensions 19 were used to determine the approximate number of cells in the anode biofilm. Results and Discussion 3.3.1 M odeling the structure of the Geobacter pilus at high resolution via molecular dynamics There is currently a lot of debate in the field about the mechanism that enables the Geobacter pili to transfer electrons. Homology models take structur al assumptions that lead to various interpretations and added to the controversy. Our approach has been to work synergistically with Dr. Gustavo Feliciano at the University of Sao Paulo in Brazil to couple theoretical and experimental studies that provide novel insights into the correlation between structure and function in this complex system. To this end, Dr. Feliciano constructed a high resolution structural model of the G. sulfurreducens pilus via molecular dynamics ( Figure 14 ) . My goal was to test the model predictions experimentally by constructing mutants with targeted amino acid replacements in amino acids predicted by the model to influence electron transfer through the pilus. As reported for other type IV pili 6 , the MD model revealed the tight packing of pilins via h ydrophobic interactions among neighboring - helices , which formed a central fiber core ca. 35 Å in diameter. The flexible C - t random coil s from each pilin protruded at a 40 ° angle from the 47 Å . Such diameters are within the ranges (2 - 5 nm) estim ated by atomic force microscopy 34,43 and scanning tunneling microscopy 43 . T he rise of adjacent pilin s in the fiber ( ca. 10.5 Å) was also similar to that reported 50 for other bacterial pili 66 and aligned residues F1 of one pilin with the E5 residue of the next. I demonstrated these interactions are required for in vivo pilin assembly in G. sulfurreducens in Chapter 2 . Positively and negatively charged amino aci ds on one pilin and those of four neighboring pilins also interacted, forming salt bridges (D53 with K30 and D54 with R28) that have been proposed to strengthen pilin - pilin contacts. 66 The salt bridges also maintain the bend - region, where a proline residue (P22) increases the flexibility o f the peptide , and allow for exposure of residues of the fiber core to the solvent for increased flexibility and dynamics ( Figure 14 ). The model also revealed the clustering of aromatic residues, which formed a continuous helical region of high aromatic ring density along the length of the pilus model ( Figure 14 ). This result is consistent with the idea that these residues are directly involved in electron transfer. 3.3.1.1 Potential ET pathways via aromatic resid ues I analyzed the MD modeling data in order to identify possible ET paths involving aromatic residue contacts ( Figure 15 ). Several ET steps are possible but two provide the shortest routes for electrons to move ( Figure 16 ): 1) I identified a path for electrons involving two amino acids of the fiber core (Y27 and F24). The path has both intra and intermolecular dimers of these amino acids, with inter - aromatic ring distances averaging 4.2 ± 2.5 Å and 14.4 ± 2. 4 Å , respectively. Thus, according to this model replacement of Y27 with an alanine should reduce ET rates significantly because the distance between the two closest F24 residues is too large (19.7 ± 3.1 Å ). 2) A second path is also possible involving four a mino acids of the fiber core (F24, Y27, Y32, and F51) and the only aromatic residue (Y57) of the C - t random coil. Inter - aromatic distances between F24 and Y27 average 4.2 ± 2.5 Å . In addition to the F24 51 to Y27 dimer, electrons can also move from F24 to Y32 ( 10.4 ± 2.5 Å ). The same F24 can also pass electrons to a Y57 from the neighboring pilin, which is located 14.2 ± 3.1 Å apart. Y27 and Y32 can also pass electrons to Y57 inasmuch inter - aromatic distances between dimers are 12.6 ± 2.5 Å and 17.6 ± 2.7 Å , re spectively. From Y57 electrons can move to F51 intramolecularly (9.1 ± 2.7 Å ) and from F51 electrons move to the F24 of the neighboring pilin ( 7.3 ± 2.5 Å ) , repeating the path to Y27, Y32, and Y57. Distances of residues to the Y57 in particular may vary gre atly in the pilus due to the highly flexible nature of the C - t random coil ( Figure 2 ). Inter - aromatic residue distances in both pathways are within the ranges reported for multi - step hopping pathways in other biological systems. I n photosystem II, for example, tyrosine Y Z transfer s electrons to P 680, w hich is positioned 13.8 Å apart 94 . The first path involves amino acids (F24 a nd Y27) that are buried in the fiber core. Thus, the path is unlikely to participate in the discharge of electrons to external electron acceptors such as Fe(III) oxides and uranium. By contrast, the second path includes amino acids (Y32 and Y57) with suffi cient exposure to transfer electrons externally. Tyrosine 32 is located near a putative heme - binding motif (S25, R28, V29, K30, A31, S37, R41, L47, A50) 83 and could t ransfer electrons to c - type cytochromes . Such reactions are particularly relevant for the pili in the matrix of anode biofilms, which contains abundant c - type cytochromes required for ET to the underlying electrode. Y32 could bind the matrix - associated cyt ochromes, as it is located on the external surface of the pilus, and discharge electrons to them ( Figure 17 ). Y57 is also located on the external surface of the pilus. Y57 appears to be integral to pilus electron transfer by our model unless ET occurs in the core between Y27 and F24 ca. ca. 14.4 ± 2.4 Å ), but it may also play a role similar to Y32. Y57 is also surrounded ( ca . 3 - 10 Å radius) by negatively charge d residues (D39, D53, D54, E60) as well as the C - terminal carboxyl group of the terminal serine S61 52 residue. The negative charges surrounding Y57 and exposure on the pilus surface could promote the binding of positively charged electron acceptors (e.g., Fe(III) oxides and the uranyl [U( IV)] cation). In particular D39, the carboxyl group of S61 and Y57 could form a cage of oxygen atoms all ca. 2 - 8 Å from each other ( Figure 15 ). Studies of the human apotransferrin 160 show the uranyl ion bound by carboxylic acid groups from two amino acids and an oxygen from a tyrosine, with oxygen - to - uranium bond lengths of 1.8 2.5 Å. Given that the uranyl cation is ca . 2 Å in diameter, bond lengths in this range could be achieved in the pocket containing Y57. Consistent with this hypothesis, an atomic model derived from the L III - edge Extended X - ray Absorption Spectroscopy (EXAFS) spectra of uranium bound to Geobacter pili revealed a U(IV) atom coo rdinated by two bidentate carboxyl ligands like those found in the carboxyl groups of acidic amin o acids 13 as well as a monodentate carbon ligand. Thus, the most exposed tyrosine,Y57, may be the terminal relay amino acid in the ET pathway to electron acceptors such as the uranyl cation, which are likely bound with grea t affinity by nearby negatively - charged ligands. This mechanism allows the pili to bind and reduce U(VI) extracellularly, preventing the toxic uranium from permeating the cell envelope and killing the cell 13 . 3.3.1.2 Potential ET p athway via the amide backbone - helix, the polypeptide backbone follows a helical path of 3.6 amino acid residues per turn of the helix , which aligns the backbone of amino and carbonyl groups and promotes hydrogen bonding ( Figure 5 - helices has been theorized to increase ET rates 112,113 . Hence, the possibility exists that ET through the pilus also involves a path through the amide backbone. A hybrid pathway containing a path through the amide backbone and short distance hopping between peptides was identified in the pilus structure ( Figure 18 ). Charge hopping is possible between the F51 of one pilin subunit and the F24 of another, which are positioned 7.3 Å (± 2.5) apart. Given the fle xible nature of both - 53 helices and pili, it is possible that these residues could periodically be drawn within 3 - 4 Å of each other to create a three electron bond 106 . The possibility of this shortening process is incr eased by the prediction that an adjacent hole draws the rings close to each ot her. The transient nature of this close distance would then cause the hole relay to continue 106 . The electron hole would then hop from F51 along the amino acid backbone via the amide groups until it reaches the next F24, an d the path will repeat itself. 3.3.1.3 Contribution of the pilin - helix , which is accentuated by the unique distribution of charged amino acids, has been proposed to promote ET along the pilus 45 . The accentuated charge dipole created by the presence of these charged amino acids ( Figure 2 ) could potentially contribute to charge hopping along the amide backbone 67,108 110 . Negatively charged amino acid residues can also act as proto electron transfer (PCET) 92 . As mentioned in Chapters 1 and 2, the oxidation potential of tyrosine is greatly reduced if it is first deprotonated, allowing it to mediate char ge hopping at the low potentials that operate in biological systems. Not surprisingly, PCET is the major mechanism of ET in many well studied protein systems 92 . Consistent with a PCET mechanism, the helical band of high aromatic density revealed in the MD model of the G. sulfurreducens pilus is interspersed with the regions of the pili of highest negative potential. Furthermore, the tyrosine fluorescence emission spectrum of isolated G. sulfurreducens pili demonstrated that most of the ty rosines were in the deprotonated form (tyrosinate) 159 . Hence, the theoretic al and experimental evidence supports the notion that acidic residues near tyrosine residues could facilitate PCET. As tyrosine residues have been more thoroughly studied in electron transfer through proteins, I focused on investigating the effects of tyro sine removal on ET first. 54 3.3.2 The role of tyrosine residues in biofilm electroactivity and Fe(III) oxide reduction In chapter 2, I demonstrated that Tyr3, a mutant that carries alanine substitutions in the stance to the passage of electrons. Furthermore, Tyr3 biofilms grown in MECs had a reduced electrochemical activity and were phenotypically indistinguishable from a pilB mutant, which does not produce pili. To gain further insights into the role of tyrosin es in ET through the pili, I constructed single alanine effect of the mutations in the electroactivity of anode biofilms in MECs fed with 3 mM acetate. Strains carry ing the Y57A pili or Y32A mutations had defects in current production similar to the Tyr3 mutant ( Figure 19 Figure 12 ). However, no defect was observed for the strain producing Y27A pili ( Figure 20 ) which is buried in the pilus core, Y32 and Y57 are exposed on the pilus surface. Hence, the exposed tyrosines are more likely to transfer electrons to external electron acceptors in the biofilm matrix such as c - type cytochromes, other p ili, and other cells. Y32, in particular, forms a heme - binding motif with other amino acids of the pilus surface 83 , which could facilitate ET from the pili to c - type cytochromes . Y57, on the other hand, is in the flexible C - t random coil surrounded by negatively - charged amino acids and could bind ligands of the biofilm matrix via electrostatic interactions. Similarly, I observed no significant differences in Fe(III) o x ide cultures, with generation times for all strains tested averaging 2.7 ± 0.5 days. These results indicate that ET pathways relying on Y27 as a keystone residue such as ET through the core via Y27 - > F24 - >Y27 hopping cannot be the sole pathway for ET. The l ack of phenotype in the Y27A mutant strain indicates that pathways such as Y27 - > F24 - >Y27 hopping, where Y27 is key to the formation of intermolecular and intramolecular (highlighted in gray) are not essential. Therefore the pathways such as the one that in cludes the tyrosine on the C - t random coil (Y57) are operational for long distance charge transfer. The Y27 is not essential to this pathway due to the ability of F24 to 55 transfer electrons to Y57 (at a slightly farther average distance away). Therefore, th e lack of a phenotype in the Y27A mutant strain does not dispute this model. Interestingly, there were no intermediate phenotypes for current production in the MECs. As current is direct ly proportional to biofilm t hickness 31 and it was found in Chapter 2 that strains that were defective only in pilus producti on were unable to grow beyond 10 µm, this make sense. If pili of a certain conductivity are required to grow beyond this threshold thickness, then the current will increase proportionally with biofilm thickness as they do although intermediate ph enotypes may still be possible. 3.3.3 The effect of alanine replacement of Y27 on pilus conductivity To test the cor relation between pilus conductivity, current production in the MEC , and the predictions made by our model, I purified the Y27A pili and Dr. Sanela Lampa - Pastrik measured their transversal conductivity with a CP - AFM. As a control, the conductivity of the pu rified Y27F pili was also measured. As predicted by the model, the Y27A mutations had the mo st significant de creases in pilus conductivity, and genetically complementing the mutation with a phenyalanine in the Y27F culture partially restored the pilus cond uctivity to WT levels ( Figure 21 ). The results thus support our earlier conclusion that Y27 is an important relay amino pping pathway. Furthermore, these findings indicate that the benzene ring rath er than the hydroxyl group of the tyrosine is critical for optimal rates of ET. It is important to note, however, that the Y27A mutation could have caused a conformational change in the structure of the pilus such that the ET pathway was interrupted. Howev er, the tight packing of pilins via hydrophobic interactions and salt bridges revealed in the MD model makes this interpretation less plausible. To further investigate the effect of the Y27A in pilus conductivity, Dr. Sanela Lampa - Pastirk also measured cha rge transport along the pilus fibers in reference to the WT pili ( Figure 22 ). For these experiments, the pili were deposited on the edge of a gold electrode 56 nanofabricated on an insulating SiO 2 substrate and the conductive tip of the CP - AFM was used to measure charge transport between the tip and the gold electrode through the pilus fiber while applying a bias voltage. Charge transport along the WT pili was measured at distances of up to pilus resistance with the distance of probing ( Figure 22 ). Significantly , the resistance of the Y27A sample was greater at all the distances probed. In analyzing the data, I identified two different correlations between distance and resistance in both the WT and Y27A pili ( Figure 21 ). While the average resistance of the Y27A pili was 4.5 - fold greater than the WT for distance s of 500 nm or less, they were 5.5 - fold greater beyond this threshold distance. F or example, the average resistance of a 1 µm long WT pilus was ca. 430 M or 720 M depending on whether the slope of the first (less than 500 nm) or second (more than 500 nm) portions of the plot were considered. Similarly, the average resistance for a 1 µm long Y27A pilus was ca. 1.99 M or 3.96 M , respectively. Even when using the greatest resistance values, I calculated a pilus resistivity of 0.23 .cm and 1.25 .cm for the WT and Y27A pili, respectively. The calculation of these resistivities assum ed a pilus diameter of 2 nm estimated or the pilus height AFM. As the AFM tip can flatten the sample during probing and underestimate the pilus height, I also calculated the resistivity of the pilus using the 47 Å diameter estimated for the preferred MD pi lus conformer, which has the C - t random coils at a 40 ° angle from the pilus core 137 . This diameter also matches the diameter of the pilus estimated by scanning tunneling microscopy ( ca. 5 nm) 43 . The resistivity of a pilus with a 47 Å diameter was 1.25 .cm for the WT and 6.88 .cm for Y27A pili. All of these values are within the lowest ranges report ed for moderately doped nanowires 161 . I also calculated the average electron transport rate for the WT and Y27A pili using the lower and upper ranges of resistance described above. For the WT pilus, I estimated electron transport rates betwe en 9 x 10 8 to 1 x 10 9 electrons per sec at a potential of 100 mV. For the 57 Y27A pilus, I estimated electron transport rates between 1.5 x 10 8 to 3 x 10 8 electrons per second at the same potential. These rates of electron transfer are much higher than the ce llular rate of respiration of Fe(III) oxides, which I estimated to be ca . 9 x 10 6 electrons per cell per sec ( Figure 23 ). This result helps explain why there was no significant difference between the WT and Y27A during the reducti on of Fe(III) oxides ( Figure 20 ). I also calculated a cellular rate of respiration of anode electrodes of ca. 2 - 3 x 10 8 electrons per cell per sec using the maximum current and biofilm biomass measured experimentally in MECs fed w ith 1 mM acetate (Chapter 2). This rate is within the ranges calculated for electron transport rates for both the WT and Y27A pili. Hence, as observed for the Fe(III) o xide cultures, the Y27A variant does not reduce charge transport through the pili suffic iently to limit optimal current production in MECs. The increased resistance along the pili maybe be due to the closer proximity of Y27 to Y57 (the next aromatic in the ET chain). It may also be due to the loss of a bond slowing the rate of electron h ole transfer. This is supported to the close proximity of Y27 to F24. F24 is part of the phenylalanine dimer predicted to be used for intermolecular pilin transfer in the case of amide bond hopping ( Figure 18 Amide electron hole hopping and phenylalanine dimer intermolecular transfer pathway ). Y27 is also six - helical hydroge n bonds away from the other half of this dimer, F51 and could therefore affect transfer between these two aromatic residues. 3.3.4 Effect of phenylalanine replacements of the pilus tyrosines Though the role of phenyalanines in protein ET is not fully understood, phenylalanine residues have been reported to influence and even mediate charge transfer in proteins and peptides 103 . Hence, I investigated the effect of tyrosines on the electrochemical activity of anode biofilms in MECs. I therefore constructed a mutant carrying Y27F, Y32F, and Y57F substitutions in the pili (Tyr3 - F) and compared its growth and electrochemi cal activity in MECs to the Tyr3 mutant, which carries alanine 58 replacements in the same tyrosines. Unlike the Tyr3 mutant, which has a defect in current production comparable to a non - piliated strain pilB (Chapter 2), the Tyr3 - F mutant had no defect in cur rent production ( Figure 24 ). This suggests that the phenylalanine residues are able to participate in ET in the G. sulfurreducens pilus. Additionally, the phenotype of the Tyr3 - F mutant also rules out a role for post translational modifications of the tyrosines (which target the group) in ET through the pilus. 3.3.5 Effect of negatively - charged amino acids of the pilus tyrosines and cou ld serve as proton acceptors for PCET. In addition, the two aspartic acid r esidues are located in the C - t end of the a - helix and contribute to the steep dipole of the pilin, which is predicted to promote ET 45 . Hence, I took a genetic app roach to investigate the role of these residues in charge transfer through the pilus. I constructed a mutant carrying alanine substitutions in both D53 and D54 (Asp2 mutant) and compared its ability to produce current in a MEC to the Tyr3 mutant. As shown in Figure 25 , the Asp2 culture produced current at rates (0.52 ± 0.11 mA/day) four - fold lower than the WT cells and also had two - fold decreases in current maxima (0.68 ± 0.05 mA) . The Asp2 defect was similar to the Tyr3 mutant, in dicating that both mutants have a reduced ability to discharge respiratory electrons in anode biofilms. Furthermore, the Asp2 cells initially colonized the electrode like the WT culture , but they required more time (~12 h) to begin the exponential phase of biofilm growth and current generation ( Figure 25 , inset). Hence, the in vivo studies support the model predictions that the local electrostatic environment around the aromatic residue contacts and the surface properties of the pi lus influence the rates of ele ctron transfer through the pili. G eneration times for the Asp2 during growth with Fe(III) oxides were also two - fold greater in the Asp2 cultures compared to the WT controls , though the defect was not as pronounced as the Tyr3 mutant , but 59 supplementing the Asp2 cultures with the metal chelator NTA to alleviate the need of the cells to transfer electrons via pili 34 chemically rescued the growth defect ( Figure 25 ). It is unlikely that the only role of the aspartic acid residues is to deprotonate tyrosines for PCET b ecause the Tyr3 - F mutant had no defects in MECs ( Figure 24 ). The MD model of the pilus does predict that an Asp2 mutation red uces the flexibility of the pilus because D53 and D54 form salt bridges with R28 and K30 with adjacent pilin monomers, and in the mutant new salt bridges form that make the pilus more rigid 137 . This effect could decrease the ability of the pilus to bind and transfer electrons to extracellular electron acceptors of the biofilm matrix such as c - type cytochromes . Furthermore, although the Asp2 pilus fiber maintained the same number of aromatic residue contacts as the WT pilus, the type of a roma tic residues involved in the contacts and their configuration changed and average inter - aromatic residue distances increased 137 . Thus, the Asp2 phenotypes in MECs and Fe(III) oxide cultures indicate that the local electrostatic environment around the aromatic residue contacts influence s the aromatic ring configuration so as to promote efficient ET. Conclusions At the beginning of my graduate work, it was not known if the pili of G. sulfurreducens could transport electrons through the length of the pilus . My collaborative work with Dr. Sanela Lampa - Pastirk demonstrated that the pili are protein nanowires a nd utilize specific amino acids of the pilin ( such as Y27) for long - range charge transport. Although this residue was not required for optim al electrochemical activity of anode biofilms, the other two tyrosines (Y32 and Y57) were required . These tyrosines, unlike Y27, are exposed on the pilus surface and may serve as terminal relay amino acids of the pilus pathway for the reduction of redox ac tive components of the biofilm matrix such as c - type cytochromes . 60 Insights into the amino acids that mediate ET in the pilus also resulted from a collaborative effort with Dr. Feliciano, who constructed a high resolution model of the G. sulfurreducens pil us using molecular dynamics simulations 137 . The packing, alignment, and geometric configuration of aromatic residues in the MD model are consistent with a mechanism for charge transport dominated by multistep hopping - sta cking required for metallic conductance. I tested the model predictions about the role of tyrosines in the multistep hopping pathways by constructing a Ty r3 mutant and demonstrated its inability to discharge respiratory electrons to Fe(III) oxides or produ ce current in MECs optimally. However, the tyrosines could be replaced with phenylalanines without any defects in pili - functions, highlighting the importance of the electron rich - bonds found in these aromatic side chains in ET through the pilus. importance of the local electrostatic environment around the aromatic contacts to align the aromatic resi dues in configurations that promote optimal ET . My studies also highlight ed the role of tyrosines expose d on the pilus surface (Y57 and Y32) ( Figure 17 ) on ET in vivo . Y57 is of particular interest in transversal ET as it is not o nly on the external face of the pilus , but is also in close proximity to several negatively charged amino acids (D39, D53, D54, E60) and the C - terminal carboxyl group of S61. The negatively charged residues position carboxyl ligands in close proximity to Y 57, which could promote the binding of positively charge d electron acceptors such as Fe (III) oxides and the uranyl cation ( Figure 17 ). Experimental evidence exists in support of this model: the atomic environment of the pili - bound uranium modeled from the uranium L III - edge Extended X - ray Absorption Spectroscopy (EXAFS) spectra is that of a reduced U(IV) atom coo rdinated by two bidentate carboxyl ligands like those found in the carboxyl groups of acidic amino acids 13 . Furthermore, other proteins utilized carboxyl groups to coordinate uranium 160,162 as well as tyrosine residues 160 . 61 While the results presented in this chapter provide novel mechanistic understanding of ET though the G. sulfurreducens pilus, many questions still remain. A l anine replacements of a ll three tyrosine residues investigated in this study displayed negative ET phenotype s . This supports electron hopping via redox - active stepping stones; however, it does not dismiss the possible role of amide hopping to ET. Also, whi le the charged amino acid residues tested are re quired for both WT levels of Fe (III) oxide reduction and current production, this result could be indirect by the generation of the pilin dipole, as important structural components, by promoting binding to ca tionic electron acceptors, or as proton acceptors. Finally, the contribution of PCET to the pilus conductivity cannot be dismissed. 62 CHAPTER 4. INTRAMOLECULAR CHARGE TRANSPORT IN PILUS NANOWIRES INVESTIGATED IN PILIN - ELECTRODE INTERFACES 63 Introduction The ability of Geobacter sulfurreducens to transfer electrons to external electron acceptors such as metals, radionuclides, or electrodes in microbial electrochemical cells (MECs) has applications in a variety of fields including bioenergy, bioremediation, and nanotec hnology 118,163 . At the most fundamental level, these application s depend on the ability of the cells to transfer electrons to external electron acceptors using type IVa pili (T4P) 13,32,34 . The T4P of G. sulfurreducens are only 2 - 5 nanometers in diameter yet several micrometers in length 34 . In addition, the pili are conductive 3 4 , a property that is not mediated by bound metal ions or organic cofactors but, rather, depends on the ability of the protein fiber to conduct electrons (Dr. Sanela Lampa - Pastirk, unpublished data). These findings indicate that the conductive nature of t he pili is due to the amino acids within the pilin subunits. Due to their conductive nature and geometry, pili have the potential to be used as nanowires in bioelectronic applications 34,159,164 . Critical to the biotechnological applications of Geobacter pili is the elucidat ion of their mechanism of conductivity. In Chapters 2 and 3 I presented evidence supporting the role of tyrosines of the pili in charge transport and the contribution of the local electrostatic environment around the aromatic residues in pilus conductivity and biological functions. Furthermore, replacement of the two tyrosines with highest exposure on the pilus surface was sufficient to reduce the electrochemical activity of G. sulfurreducens anode biofilms ( Figure 19 ). This result highlighted the importance of interactions between the pili and other redox - active components of the biofilm matrix such as c - type cytochromes in biofilm conductance. I also demonstrated that the loss of Y27, the tyrosine involved in the formation of an i ntramolecular contact in the pilin assembly 137 ( Figure 22 ). These findings support the multistep hopping mechanism via tyrosines predi cted in molecular dynamics (MD) simulations of t he G. sulfurreducens pilus in solution 137 . In this model, 64 the assembly of the pilins clusters the aromatic residues (phenylalanines and tyrosines) of neighboring subunits with inter - aromatic d istances and dimer configurations optimal for multistep hopping. The aromatic contacts were between aromatic residues of the same or adjacent pilins, consistent with a mechanism that integrates intra - and inter - molecular pathways for electron transfer thro ugh the pili 137 . Evidence for intramolecular charge transport also comes from ab initio studies of the G. sulfurreducens pilin, PilA 45 . The G. sulfurreducens pilin is significantly truncated compared to the typical type IVa pilin , lacking the conserved large globular head domain of other bacterial pilins and reducing the Geobacter pilin to a small (61 - aa) and predominantly - helical peptide ( Figure 1 ). Helical structures accelerate electron transfer reactions due to the hydrogen bonding and natural dipole moment inherent in the structure 67,112,113,165 , even at lengths significantly longe r than the G. sulfurreducens pilin 68 . As such, the predominantly helical structure of the G. sulfurreducens pilin is predicted to create a peptide environment optimal for electron transfer 45 . In addition, the dipoles of the peptide bonds align in - helix and polarize the pilin ( Figure 2 ). Such dipole moments have been previously found to enhance electron transfer in helical peptides 67 and are predicted to lower the electron band gap of the G. sulfurreducens pilin 45 . Furthermore, specific a mino acids of the pilin en hance the natural dipole and contribute to the electrostatic effects that reduce the electronic band gap of the pilin 45 . Among the amino acid acids predicted to contribute to the unique electronic structure of the pilin are s everal charg ed and aromatic residues. Hence, the results indicate that the divergent structure and amino acid composition of the pilin favor electron transfer (ET) reactions . Furthermore, they also suggest that intramolecular charge transfer is possible through the pi lin . As the pilin is 61 amino acids long and therefore too long for electron tunneling to occur, it is likely that this charge transfer - helical peptides 68,88 . 65 Given the multimeric nature of T4P, it is challenging to determine the contribution of the predicted int ramolecular conductivity to the pilus conductivity. To address this limitation, I used recombinant technologies to mass - produce a truncated PilA peptide that could be assembled in vitro into fully functional protein nanowires, in a collaborative effort bet ween myself and Dr. Angelines Castro - Forero, a student in the laboratory of Dr. R. Mark Worden (Chemical Engineering Department, Michigan State University). As a team we sought to mass - produce protein nanowires for biotechnological applications. Individual ly, I aimed to develop hybrid pilin - intramolecular charge transport. To this end, I constructed genetic systems for the large - scale recombinant production - helical structure and amino acids critical to electron transport. My collaborator Dr. Castro - Forero screened some of the recombinant systems, and developed protocols for the purification of pilin mo nomers at high yields and their in vitro self - assembly 166 . I then functionalized one of the truncated peptides (PilA 19 ) with an amino - terminal (N - t) cysteine residue (PilA 19 - A20C) to chemically attach the pilins to gold electrodes and electrochemically interrogate the intramolecular pathway of electron transfer in pilin - electrode interfaces. I also constructed PilA 19 - A20C derivatives carrying ala nine replacements in the Y27, Y32, and Y57 residues in order to characterize the role that these amino acids play in pilus conductivity. Elucidating the method of pilin ET will not only enable us to understand protein - dependent long distance ET in organis ms like Geobacter , it will allow us to manipulate this system for producing wires with differing electronic capabilities. 66 Materials and Methods 4.2.1 Construction of Expression Strains 4.2.1.1 QIA express TM system (His - tagged pilins) Primers were design to generate both the full length protein and various truncation variants to determine the minimum number of hydrophobic amino acids on the pilin N - terminus required to enable efficient expression and purification of the protein ( Table 6 ). The QIA express TM expression system (Qiagen) was used to generate N - terminal hexapolyhistidine - tagged (6 His - Tag) truncations of the PilA peptide to facilitate purification. PCR amplification of purified G. sulfurreducens genomic DNA (gDNA) was performed (denatura tion time of 5 m in at 95 °C, 30 cycles of 30 s at 95 °C, 20 s at 50.7 °C, 1 min at 72 °C followed by an extension time of 5 min at 72 °C) with 100 ng of gDNA and Platinum Taq DNA Polymerase High Fidelity (Invitrogen). PCR products were gel purified using the Zymoclean TM Gel DNA Recovery Kit (Zymo Research). The products were then ligated into the pQE - 30UA (Table) expression vector and transformed into E. coli M15[pREP4] cells ( Table 5 ) . 4.2.1.2 pMAL TM system (MBP - tagged pilin) The pMAL TM protein fusion and purification system (New England Biolabs Inc.) was used to generate N - t maltose binding protein (MBP) fusions of the PilA peptide. PCR amplification of purified G. sulfurreducens genomic DNA was performed as described above with the use of Herculase II Fusion DNA Polymerase (Agilent Technologies) using primers ( Table 6 ). PCR products were gel purified using the Zymoclean TM Gel DNA Recovery Kit (Zymo Research). Plasmids were digested with HindIII and XmnI (NEB) a nd the PCR products with HindIII. The digested plasmids and PCR products were purified and ligated using T4 DNA ligase (New 67 Table 5 ) . For protein expression experiments the recombinant plasmids were transformed in to K12 TB1 cells ( Table 5 ). 4.2.1.3 IMPACT TM system (CBD - tagged pilins) The IMPACT TM - CN protein expression system was used to generate N - t and C - t fusion proteins containing the PilA peptide (full length) or truncated derivatives (PilA 10 , PilA 19 , PilA 20 , and PilA 22 ) and a chitin binding domain (CBD) - intein fusion (New England Biolabs Inc. Incorporated). PCR amplification of purified G. sulfurreducens genomic DNA was performed as described above with the use of Herculase II Fusion DNA Polym erase (Agilent Technologies) using primers ( Table 6 ). PCR products were gel purified using the Zymoclean TM Gel DNA Recovery Kit (Zymo Research). The pTYBII plasmid and PCR products were subsequently digested with SapI and PstI (Li fe Technologies). The digested plasmid and PCR products were purified, ligated using were then transformed into Rosetta TM 2 (DE3) pLysS cells ( Table 5 ) for protein expression experi ments. 4.2.1.4 Mutagenesis of codons for pilin amino acids All amino acid replace ment s were made with the DNA encoding the 19 - amino acid truncation variant . I first constructed a PilA 19 derivative carrying a cysteine tag (PilA 19 - A20C) for utilization in gold elec trode experiments. Then, several PilA 19 derivatives carrying either: amino acid acid residues (D53 and D54), and R28 were generated in this strain. All mutati ons were introduced using the QuikChange Lightning Site - Directed Mutagenesis kit (Agilent Technologies). To generate the tr iple tyrosine variant (Tyr3) after the first mutation of the DNA was confirmed, the process was repeated two times, each time with a different primer set. The 68 double aspartic acid variant (Asp2) was generated with a single mutagenesis run with a primer containing both mutations. Primers carrying the desired point mutations ( Table 6 ) were used to PCR - amplify the pTYB11:: pilA 19 plasmid ( Table 6 ) (denaturation time of 2 min at 95 °C, 18 cycles of 20 s at 95 °C, 10 s at 60 °C, 6 min 30 s at 68 °C followed by an extension time of 5 min at 68 °C). Methylated DNA was then digested with DpnI (A gilent Technologies) and the plasmids were transformed into XL10 - Gold cells ( Table 5 ) following the manufacturer recommendations. Plasmids were isolated from ampicillin resistant colonies using the ZR Plasmid Miniprep TM - Classic kit (Zymo Research) and sequenced to c onfirm the mutation s . Confirmed plasmids were transformed into Rosetta TM 2 (DE3) pLysS cells ( Table 5 ) for protein expression experiments. 4.2.2 E xpression of recombinant pilins The expression eff iciency in E. coli of recombinant pilins tagged with a 6xHis region was first assessed in small - scale (50 ml) cultures grown in lysogeny broth (LB) medium. Antibiotics protein expression was induced with 1 mM isopropyl - 1 - thio - D - galactopyranoside (IPTG) once the cultures reached an OD 600 between 0.5 and 0.7. Cells were harvested 5 h post - induction by centrifugation (4,000 x g for 20 m in at 4 °C), suspended in 20 mM Tris - HCl pH 7, and sonicated on ice 3 times (duty cycle 30%, output 3, for a total of 10 pulses) using a Branson sonifier 450. Supernatant fluids were recovered by centrifugation and 4x Tris - glycine loading buffer was added. T he pellets were suspended in Tris - glycine loading buffer (62.5 mM Tris - HCl, pH 6.8, 25 - mercaptoethanol). Approximately 10 - Glycine acrylamide gel (Bio - Rad Laboratories, Inc.) which was run at 100 V for 70 m in . Pr oteins were transferred to a Hybond - ECL nitrocellulose membrane (Amersham Biosciences) using a Mini 69 Trans - Blot cell ® (Biorad) over 1.5 h at 170 m A. The membrane was blocked wi th 5 % (w/v) non - fat dry milk in Tris - buffered saline ( 20 mM Tris, pH 7.4, 0.9% N aCl ) for 3 h. The membrane was - - His tag (Qiagen); 1:100 or 1:2,000 dilution, respectively) o vernight. The membrane was incubated with the secondary antibody (anti - rabbit IgG - AP conjugated (Sigma) 1:2,000 dilution). Visualization was performed at room temperature using nitro - blue tetrazolium and 5 - bromo - 4 - chloro - 3' - indoly l phosphate (NBT/BCIP) solution (Sigma - Aldrich). Expression studies of MBP - and CBP - tagged pilins were performed by collaborator Dr. Angelines Castro - Forero and have be en described in detail elsewhere 166 . 4.2.3 Purification of recombinant pilins carrying a Chitin - binding domain protein tag ystem was selected for the large - scale production of pilins. The recombinant strains were grown to mid - expotential phase (OD 600 of 0.4 0.5) in one liter of LB medium at 37 ° C with shaking (250 rpm) . Protein production was induced with 0.5 mM IPTG at 16 °C overnight. The cells were harvested by centrifugation and the cell pellets were frozen at - 80 °C for at least one day. The cell pellets were suspended in 30 ml of lysis buffer ( Table 8 ) and lysed by sonication ( see above) i n an ic e bath . The soluble fraction was recovered by centrifugation at 110,000 x g for 30 m in at 4 °C and pass ed through an affinity column containing ca. 30 ml of chitin beads (New England Biolabs Inc.) previously washed with 4 column volumes of column buffer ( Table 8 ) and equilibrated for 30 min. Non - specifically bound proteins were removed from the column with 4 column volumes of wash buffers 1 and then 2 ( Table 8 ), containing increasing concentrations of NaCl . Recombinant pilins, which bound the chitin resin in the column beads, were subsequently removed from the column after self - cleaving the peptides from the chitin - binding domain with 50 mM dithiothreitol ( DTT ) dissolved in elution buffer ( Table 8 ). On - column cleavage was allowed to proceed at room temperature for 24 h. Cleaved recombinant pilins were eluted from the column with elution buffer ( containing 5 70 mM DTT for the cysteine - modified pilins) and fractions containing protein we re identified by their absorbance at 280 nm and pooled. The buffer solution containing recombinant pilins was exchanged after passing it through a Waters Sep - Pak 3 cc C18 vac cartridge. The C18 cartridge was first wet with 1.5 column volumes of acetonitri le and then equilibrated with 1.7 column volumes of ddH 2 O. Eluted pilin samples were then loaded onto the column and washed with ddH 2 O (6x column volumes). Protein was eluted in 1.7 column volumes of acetonitrile and stored in a 5 ml serum vial to prevent oxygen contamination. The same procedure was used for solutions containing recombinant pilins functionalized with an N - terminal cysteine amino acid , but inside an anaerobic glove bag to prevent oxidation of the cysteine tag. For dialysis experiments recomb inant pilins were dialyzed overnight two times in anaerobic elution buffer (without DTT) at 4 °C. The purified recombinant pilins were stored on ice , or flash - frozen in liquid N 2 and stored at - 80 °C for long - term storage. 4.2.4 Electrochemical characterization of PilA 19 - A20C monolayers on gold electrodes Gold electrodes ( LGA Thin Films, Santa Clara, CA) were cut into small pieces and cleaned with piranha solution (7 parts by volume concentrated sulfuric acid to 3 parts by volum e 30% aqueous hydrogen peroxide) a . Each piece was then washed copiously with ddH 2 O, dried with N 2 and placed in a separate plastic scintillation vial. A 1:2 dilution of the recombinant PilA 19 - A20C pilin suspended in acetonitrile or elution buffer ( Table 8 ) was a dded to each vial with an electrode and incubated for various periods of time ( 48 h ended up the standard deposition time to ensure coverage) at room temperature inside an anaerobic chamber containing a H 2 :CO 2 : N 2 a Caution: piranha solution is an extremely strong oxidant and potentially explosive; it must be handled with extreme care. 71 (7:10:83) atmosphere. The pilin - electrode i nterfaces were washed with 100% ethanol and incubated with 1 mM 1 - undecanethiol (in ethanol) for 48 h at room temperature in the anaerobic chamber. When indicated, controls of electrodes covered with 1 - undecanethiol were generated by incubation of a clean gold electrode with a 1 mM 1 - undecanethiol solution for 48 h at 25 °C. 1 - undecanethiol was also used to cover any gaps in pilin coverage on the electrode. Gold electrode controls were incubated in the solvents without recombinant pilins or 1 - undecanthiol for use as a positive control. The thickness es of the pilin monolayer that formed on the electrode was measured with a M - 44 rotating analyzer ellipsometer (J.A. Woollam Co., Inc., Lincoln, NE) controlled by WVASE32 software. The incident angle was set at 7 5 o using 44 wavelengths of light between 414.0 and 736.1 nm. The r efractive index (n) and extinction coefficient (k) were assumed to be n = 1.5 and k = 0 , respectively. Thickness measurements were averaged from measurements of 3 random spots taken from eac h of five electrodes . Before the electrochemical experiments, the pilin - electrode interfaces were rinsed with ethanol and dr ied under N 2 . C yclic voltamme try measurements used a conventional three - electrode cell consisting of the pilin - functionalize d workin g electrode, a platinum auxiliary electrode, and a 3M silver/silver chloride (Ag/Ag Cl) reference electrode . Measurements were performed in a 100 mM phosphate buffer, pH 7.0 containing 100 mM NaCl and 5 mM potassium ferricyanide (K 3 [Fe(CN) 6 ]). The electroch emical experiments were performed using a CHI660 potentiostat (CH Instruments, Austin, TX). Results 4.3.1 Rationale for partially truncating the N - terminal hydrophobic pilin region The heterologous expression of the G. sulfurreducens pilins in a recombinant host such as Escherichia coli is necessary for their mass - production, yet is challenging due to the 72 hydrophobicity of the pilin peptides. Expression in the native host is possible because the pilin protein is expressed as precursor containing an N - t leade r peptide containing a large number of charged amino acids. The polar leader peptide of type IV pilins effectively decreases the by a dedicated signal peptidas e, which also methylates the phenylalanine F1 of the processed pilin to prepare it for assembly 50 . Once assembled, the pilin subunits of T4P are held together predominantly by interactions within the hydrophobic - helical region, which spans more than 50 amino acids at the N - terminus 66 . Such hydrophobic interactions can promote the aggregation of recombinant proteins inside the expression host and the formation of inclusion bodies. The latter make purification of the target protein challenging and time intensive whi ch is undesirable in an industrial setting. Furthermore, at the high levels of expression required for our project, proteins containing hydrophobic regions often insert into host membranes, killing the recombinant host cell and reducing expression yields. One way to address the challenges associated with the recombinant expression of hydrophobic pilins is to truncate the most N - terminal (N - t) amino a cids. These amino acids include F1 and E5, which, as I showed in Chapter 2, are required for the recognition and assembly of the pilins in the native host , but they are not required for the in vitro assembly. N - t truncations thus make the pilin less hydrophobic, thereby enabling both their recombinant expression and purification. Truncation of 35 amino acids of t he N - t region of the pilin of Pseudomonas aeruginosa w as used, for example, to produce recombinant PA pilins at high yields and assemble them into fibers in the presence of a hydrophobic matrix 57 . These studies thus demonstrated that it is possible to assemble pilins carrying trunca tions of more than half of the - helix region. This is because the truncated pilins retain charged amino acids involv ed in the formation of salt bridges, which align the pilins in the proper orientation, and strengthen the hydrophobic interactions between adjacent helices 57 . Such truncations are not desirable in the 73 Geobacter pilin because several aromatic amino acids predicted to be critical to electron transport (F24, Y27, and Y32) and the formation of salt bridges (R28 and K30) are located in this N - t region 137 . Hence, ideally, the truncations of the pilin of G. sulfurreducens had to be of 23 amino acids or less. The truncation design involved first an analysis of the hydrophobic regions and aggregation potential of the mature PilA peptide (i. e., the pilin without its signal peptide) using the AGGRESCAN, GRAVY, and a Kyte Doolittle plot servers. AGGRESCAN 167 uses an - propensity value (a positive value indicates a propensity for aggregation). This approach identified two hot spots of aggregation in the PilA peptide spanning re sidues 1 - 22 and 25 - 31 ( Figure 27 ). The average aggregation - propensity value for the protein remains positive until the truncation of the first 10 amino acids. To corroborate the AGGRESCAN analysis, I used GRAVY 168 to analyze the average hydrophobicity of the PilA protein. GRAVY calculates the average hydropathy value of the amino acids in a protein, with positive values indicating overall hydrophobicity. The analysis of the PilA peptide using this approach indicated that a truncation of 11 amino acids within the N - t region is necessary before the GRAVY value becomes negative. By contrast, a Kyte Doolittle plot 168 , which calculates the localized hydrophobicity along the length of the protein, shows the portion of the peptide containing the first 21 amino acids is considered hydrophobic ( Figure 27 ). All together, the computational predictions indicated that a truncation of at least 10 20 amino acids is necessary to solubilize the pilin peptide and facilitate its recombinant production. T runcations within this range also retain the amino acids predicted to serve as relay stations for multistep hopping pathways through the pilus 137 . Furthermore, the truncations do not remove charged amino acids of the pilin, which are requir ed to form salt - bridges between neighboring pilins and influence the local electrostatic 74 environment around the aromatics that is critical to maintain proper inter - aromatic distances and configurations for electron transfer 137 . 4.3.2 Constructio n and heterologous expression of recombinant pilins 4.3.2.1 Recombinant production of pilins carrying an N - t His - Tag I first explored the possibility of heterologously expressing truncated pilins fused to an N - t hexa polyhistidine tag (6 His - Tag) using the QIA expr ess TM expression system . The 6 His - Tag increases the solubility of the recombinant protein and capitalizes on the metal binding affinity of the histidine residues for nickel and cobalt to purify them in a metal - containing matrix. Furthermore, the small siz e of the 6x - His - tag has been reported to not interfere with protein stability or folding. As the expression of large quantities of foreign proteins, especially those that are hydrophobic, can often be cytotoxic, the M15[pREP4] ( Table 5 ) host strain was used for expression. This host strain permits high levels of expression for the recombinant proteins when induced, but permits the culture to grow normally up to high cell densities before induction due to strong repression of the he terologous gene. Therefore, in case of toxicity, the culture is already at densities sufficient for large amounts of protein production prior to induction. Full - length pilin as well as sequential 5 - amino acid truncations (up to 20) were successfully fused to the 6xHis tag. Despite the LacI repression of pilin transcription expected before IPTG induction, each trial expression study showed protein expression in both induced and un - induced cultures ( Figure 28 ). Furthermore, Western b lots revealed multiple protein bands in all of th e samples His6x antibody ( Figure 28 ). This result suggested that the recombinant peptides were aggregating. Consistent with this proposal , the majority of the protein bands were identif ied in the insoluble fraction, regardless of truncation length. The results indicated that even with substantial truncations (up to 20 amino acids) to decrease the 75 hydrophobicity of the pilin, larger tags are needed to increase the solubility of the peptid es to levels such as that recombinant expression yields are high and aggregation is minimized. 4.3.2.2 Recombinant production of pilins carrying a MBP I chose the pMAL TM protein fusion and purification system because it allowed me to fuse the pilin to a large glo bular protein, the maltose binding protein (MBP), which is expected to hydrophobic proteins, including other bacterial pilins 61 . Additionally, the MBP tag can be cleaved by the protease Factor Xa at a specific recognition site (I - E/D - G - R) introduced before the pilin. Thus, clea vage removes the tag in its entirety and does not leave non - native amino acids in the recombinant protein. For these experiments, I first cloned the pilA 20 pilin gene into the XmnI restriction site of the plasmid, which is located directly after the region encoding a Factor Xa recognition sequence that separates the MBP region from the target protein and allows for the cleavage of the full tag with the Factor Xa protease. My collaborator, Dr. Angelines Castro Forero, tested the expression system and demonst rated its suitability to express the PilA 20 - MBP fusion protein and its purification using an amylose affinity column 166 . Furthermore, she was a ble to recover the PilA 20 peptide upon protease cleavage ( Figure 31 ). Another advantage of this system is that it allows for the periplasmic expression of the recombinant fusion. Periplasmic expression of recombinant proteins has been shown to provide many benefits including increased solubility, decreased production of inclusion bodies, increased protein stability, and a decreased cellular protein contamination 169 . The recombinant fusions can be purified by the use of an amylose resin, which binds MBP, and eluted by the addition of maltose to the co lumn. Hence, I constructed both cytoplasmic and periplasmic expression systems for PilA 20 and cloned the plasmids into E. coli K12 TB1 (Table 4.1). As shown in ( Figure 29 ) and ( Figure 30 ), the MBP - PilA 20 fusion protein were successfully expressed and purified by amylose affinity purification by Dr. Castro Forero using the 76 periplasmic and cytoplasmic recombinant systems, respectively. Expression yields were lower in the periplasmic system compared to that o bserved in the cytoplasmic system. Furthermore, MALDI - TOF analysis of the periplasmically - expressed fusion protein indicated that the mass of the fusion protein was 43,031.9 Da . This value is significantly lower than the predicted molecular weight of the M BP - PilA 20 protein (46,902.6 Da), suggesting that proteolytic cleavage occurred. By contrast, a protein with a mass (47,042.6 Da) similar to that predicted for the MBP - PilA 20 protein expressed with the cytoplasmic system (47,033.9 Da) was detected by SDS - PA GE, Western blot and MALDI - TOF analyses (Figure 4.6). However, a small band was also observed in Western blots of the cytoplasmically - expressed fusion proteins, whose mass, as estimated by MALDI - TOF analysis, was 43,671 Da ( Figure 29 ). Hence, as in the periplasmic system, proteolytic cleavage likely occurred that truncated the fusion protein. Despite the non - specific proteolysis observed in the cytoplasmic system, Dr. Angelines was able to cleave the MBP tag at 23°C after a 48 h i ncubation of the MBP - PilA 20 fusion protein(s) ( Figure 30 ). MALDI - TOF analyses revealed the presence of two peptides in the protease - cleaved samples: one at 4,527.4 Da and the other at 3,563.3 Da. The larger peptide matches well t he expected fragment size for the PilA 20 (4,523.9 Da). The smaller peptide, on the other hand, is consistent with a PilA 20 peptide that has lost 8 amino acids at the N - t ( Figure 31 ) hence, a PilA 28 peptide, which has a predicted m olecular weight of 3,560.8 Da). This result suggests that the Factor Xa protease cleaved after the arginine in the recognition site as well as 20 peptide. This was confirm ed by peptide sequencing of the first 5 amino acids, which produced the sequence V - AYN from the N - terminal end of the cleaved peptide (the lysine in position 2 was not found). Thus, Factor Xa successfully cleaved PilA 20 from the MBP domain but also produce d a PilA 28 peptide, which removes amino acids of the pilin critical for electron transfer. Though the two peptides could be separated by HPLC and other techniques, such approaches are time - 77 consuming, costly, and reduce the yields of the recombinant pilins. Hence, as an alternative approach, I developed a third pilin expression system as described below. 4.3.2.3 Recombinant production of pilins carrying a chitin - binding domain (CBD) Similar to the MBP fusion system described above, the Impact TM system utilizes a ch itin binding domain (CBD) tag. This protein tag, like MBP, increases the solubility of the recombinant protein and allows f or its selective purification using a column packed with a proprietary chitin resin. An advantage of the Impact TM system is that it i ncorporates an intein in the fusion protein for self - cleavage of the target protein from the CBD tag. Self - excision is induced in the presenc e of a reducing agent such as DTT and is highly specific to the intein sequence, thereby minimizing the chance for non - specific cleavage at other sites. Self - cleavage with reducing agents also decreases the cost of purification of the recombinant pilins, a factor to consider for applications in industrial settings. In addition, this method produces a recombinant protei n free of non - native amino acids due to the utilization of the SapI enzyme which results in a start codon directly behind the cleavage site. Thus, I used the Impact system to construct CBD fusion proteins of the PilA peptide (full length) and several trunc ated pilins (PilA 10 , PilA 19 , PilA 20 , and PilA 22 ). The genetic constructs were cloned in to plasmid pTYB11 (Table 4.2) and transformed into Rosetta 2(DE3)pLysS E. coli cells (Table 4.1). The Rosetta 2 (DE3) cells contain a plasmid, pRARE, that encodes tRNAs for 7 codons rarely used in E. coli : AUA, AGG, AGA, CUA, CCC, GGA, and CGG. This strain background is therefore suitable for the efficient expression of foreign proteins. Rosetta 2(DE3)pLysS also contains a plasmid, pLysS, which encodes a T7 lysozyme to de crease the un - induced expression of the fusion construct which is under the control of the T7 promoter. Dr. Castro - Forero tested the efficiency of the system by comparing the protein profiles of cell extracts from induced and un - induced cultures b y SDS - PA GE gels of proteins to demonstrate the expression of the fusion protein for all the recombinant pilins ( Figure 32 ). 78 Except for the full - length CBD - PilA fusion protein, which was not detected in the induced cultures, all of the tru ncated pilins fused to the CBD were expressed upon induction ( Figure 32 ). Cultures expressing the CBD - PilA fusion protein also grew to a cell density half that of cultures producing fusions of CBD and truncated pilins. This findin g suggests that, even with a large protein tag to increase its solubility, expression of the full - length pilin was toxic to the cells. As such, expression of the full - length pilin was not pursued further. All of the fusion proteins of CBD and truncated pi lins were purifi ed by affinity chromatography using a chitin resin column and the recombinant pilins were subsequently cleaved on the column with the addition of DTT to yield pure recombinant pilins. The cleaved peptides that eluted from the column were ex amined by T ris - tricine SDS - PAGE, which is better suited to the resolution of small molecular weight peptides and their separation from SDS micelles 170 ( Figure 33 ). Hig hest yields for recombinant pilins were obtained when cleavage was performed at 23 °C compared to 4 °C ( Figure 33 ). Yields of the 19 - and 22 - residue truncation s (PilA 19 and PilA 22 ) were highest, with nearly 100 % of the recombinan t pilins being recovered after only 24 h of incubation at 23 °C in the presence of DTT. By contrast, the 10 - residue truncation (PilA 10 ) was not eluted after 24 h at 23 °C ( Figure 33 ) and it took 72 h of cleavage to recover ca. 50% of the truncated pilin from the column. Such prolon ged incubation with DTT also le d to the clogging of the chitin column, suggesting that once the PilA 10 peptide was cleaved from the CBD tag it aggregated inside the column, preventing its efficient elutio n and recovery. Recovery of PilA 10 improved when cleavage was performed at 4 °C but additional bands were also observed ( Figure 33 ), indicative of non - specific cleavage. Hence, although computational predictions of the aggregation factor and total hydrophobicity of the PilA 10 suggested it was a suitable truncation for efficient purification, the experimental evidence suggests otherwise. The 20 - residue truncation (PilA 20 ) could not be recovered at any temperature ( Figure 33 ). Analysis of the amino acid sequence of the PilA 20 , which could not be 79 cleaved from the CBD domain, revealed an isoleucine in the most N - t position of this pilin, a residue that the manufacturer of the expression system (New England Biol abs Inc.) reports to inhibit DTT - induced cleavage. 4.3.3 Electrochemical characterization of recombinant PilA 19 The results presented above indicated that both PilA 19 and PilA 22 pilins can be efficiently expressed as fusion proteins with a CBD tag and recover ed as soluble pilin monomers in the presence of DTT ( Figure 34 ). As the PilA 19 pilin is the one with the shortest truncation and therefore carries more of the native amino acids of the PilA pilin, we chose it for further studies. My collaborator, Dr. Angelines Castro - Forero also demonstrated that PilA 19 retains the - helical structure of PilA using circular dichroism 166 , which is required for assembly and conductivity of the pili n assembly 45 . Furthermore, she demonstrated that this truncated pilin can be assembled in vitro into pilus fibers that are otherwise indistinguishable from the native fibers by both transmission electron microscopy (TEM) and atomic force microscopy (AFM) 166 . Furthermore, scanning tunneling microscopy (STM) demonstrated that PilA 19 fibers are conductive and have topographic and electronic features similar to those observed in the native pili 18 . Hence, I used PilA 19 as a model pilin to investigate the intramolecular charge transport of the pili. 4.3.3.1 Functionalization of PilA 19 with a cysteine tag and construction of pilin - electrode interfaces Addition of a cysteine residue to a protein enables its attachment to gold surfaces via interaction with the sulfhydryl group 171 . Therefore, an amino acid replacement was introduced into a PilA 19 construct, whereby the codon for the N - t alanine (A20) was mutated to encode a cysteine (A20C), thus generating a PilA 19 peptide carrying an N - t cysteine (PilA 19 - A20C). My collaborator, Dr. Castro - Forero, confirmed that the functionalized pilin can be expressed as a 80 fusion protein with a CBD in E. coli , as described above, and purified from a chitin column after inducing self - cleavage with DTT ( Figure 34 ). Furthermore, MALDI TOF MS analysis confirmed the presence of a 4,632.3 Da peptide, similar to the predicted molecular weight of 4,627.1 Da. Furthermore, she also confirmed that the PilA 19 - A20C monomers can be assembled in vitro into pilus fibers 166 . I used the PilA 19 - A20C monomers to develop protocols for their deposition as a monolayer on gold electrodes. The pilin monolayer formed on gold electrodes had an average thickness of 37.11 ( ± 7.89 Å) as measured by ellipsometry. As the predicted length of the PilA 19 - A20C p ilin is approximately 55 Å, the thickness of the peptide monolayer is consistent with a peptide tilt angle of 42° with respect to the gold support, which is similar to the values reported for - helical peptides attached to gold electrodes 172,173 . 4.3.3.2 Cyclic voltammetry of pilin - electrode interfaces The ability of the pilin (PilA 19 - A20C) monolayer to transfer electrons between a soluble mediator (ferricyanide) and the underlying electrode was investigated by cyclic voltammetry (CV) ( Figure 35 ). For these experiments, I used pilin samples directly eluted from the chitin columns , which contained DTT and pilin samples dialyzed to remove the DTT. This allowed me to assess if the DTT in the solution affected the formation of sulfhydryl bonds between the pilin and the electrode, which could potentially reduce the monolayer coverage a nd leave areas of gold exposed to the solution and electrochemically interacting with the ferricyanide. As controls, I also obtained cyclic voltammograms for bare gold electrodes (positive control) and gold electrodes coated with an insulating layer of an undecanethiol self - assembled monolayer (SAM), which insulates the electrode (negative control) ( Figure 35 ). Cyclic voltammograms for electrodes with PilA 19 - A20C were similar to those with bare gold, consistent with a conductive peptide monolayer. However, incubation of the PilA 19 - A20C monomers and the electrode in the presence of the reducing agent DTT, prevented the attachment of the pilins to the gold 81 electrode. This w as demonstrated by the decrease of current den sity when 1 - undecanthiol was layered on top of the pilin monolayer. This decrease was comparable to levels of a gold electrodes coated with 1 - undecanethiol ( Figure 35 ). This result demonstrates that a complete pili n monolayer was not achieved. Less of an effect upon 1 - undecanethiol addition was observed when the recombinant pilins were dialyzed two times anaerobically to remove the DTT ( Figure 35 ). Furthermore, a buffer exchange into aceto nitrile resulted in a CV curve that was of similar height and shape to one with 1 - undecanethiol added, demonstrating a fully established SAM ( Figure 36 ). However, the current density of the pilin monolayer formed on the gold elect rode did not decrease after adding the insulating 1 - undecanethiol, indicating that the monolayer covered the electrode in its entirety. Hence, the results suggest that the PilA 19 - A20C peptide is conductive; consistent with computational predictions of an i ntramolecular electron transfer pathway in G. sulfurreducens nanowires 45,137 . In all cases, pilin monolayer formation was completed after 48 h ( Figure 37 ). Initial tests showed a decrease in current with longer incubation times, most likely due t o oxygen leaking into the pre ssure vials over time, so further experiments were performed in an anaerobic chamber. The e ffect of mixing the PilA 19 - A20C solution during deposition on the electrodes was also investigated and found to have no effect ( Figure 38 ). Cyclic voltammograms of 30 - day old pilin - electrode interfaces were also similar to those formed after a fresh monolayer deposition ( Figure 38 ). Similarly, lowering the concentration of the pilins in s olution or depositing the pilins at lower (4 °C) temperatures did not affect the current response ( Figure 39 ). 4.3.3.3 CP - AFM probing of pilin - electrode interfaces As a complementary approach, I used conductive probe atomic force microsc opy (CP - AFM) to directly measure the conductivity of the pilin monolayer formed on gold electrodes. Results were variable , but applying a high force allowed the conductive tip of the AFM to penetrate the pilin monolayer and directly transfer electrons to a nd from the gold electrode. At 82 very low pressures the pilin monolayer prevented the transfer of electrons, indicating that the monolayer was insulting the tip from the electrode. However, applying forces between these 2 extremes resulted in a current resp onse from the pilin molayer, and resulted in sigmoidal current - voltage ( IV ) plots typical of semi - conductor materials ( Figure 40 ). The ability of the recombinant pilins to bind to gold electrodes shows promise for the study of pilin conductivity as well as serving as the starting point for the gen eration of new bi oelectronics. 4.3.4 Intramolecular pilin pathway investigated in mutated pilin - electrode interfaces In order to investigate the roles of certain amino acid residues on the intramolecular conductivity of t he pilin, I generated mutants encoding CBP fusions of PilA 19 - A20C carrying alanine replacements in amino acids that molecular modeling 45 as well as my previous studies implicated in the conductivity of the pilin. As mentioned previously, the pilin has a unique structure with a strong dipole across the length of the polypeptide. This charge differential contributes to the low electronic band gap estimated for the pilin 45 , increasing its conductivity. Also, aromatic residues located in the charged regions were implicated previously 137 (and in Chapters 2 and 3) in electron transfer. Thus, I constructed PilA 19 - A20C derivatives carrying ee tyrosines (Tyr3) and the two aspartic acid resid ues that influence the aromatic residue configuration and electron transfer through the pili (Asp2) 137 . Additionally, I also constructed an R28A derivative, to investigate the role of posit ive charges in intramolecular electron transfer. I expressed and purified the Tyr3, Asp2, and R28A PilA 19 - A20C pilins, as described for the PilA 19 - A20C peptide, and used them to form monolayers on a gold electrode. Undecanethiol was used to cover and insul ate areas of the electrode that could have remained uncovered. Interestingly, cyclic voltammograms were similar for all the monolayers ( Figure 41 ), indicating that all samples cycle electrons between the ferricyanide and the under lying gold electrode at 83 similar rates. This finding suggests that intramolecular electron transfer is not influenced by the tyrosines or the electrostatics contributed by the charged amino acids mutated in the Asp2 and R28A mutants. This does not rule out that intramolecular electron hopping operates in t he pilins, because two aromatic residues (phenylanines) are still present in the pilins. Furthermore, dip ole. Hence, more aggressive replacements may be needed to see an effect in electron transfer using this approach. Additionally, charge transfer via hopping may be occurring across the amide groups of the peptide backbone particularly in helical peptides 156 and without contributions from specific amino acids, which only play a role once the pilins assemble and intermolecular electron tran sfer is necessary. Charge hopping has been demonstrated in - helical peptides significantly longer than the G. sulfurreducens pilin peptide 68,88 . The rate of c ha rge transfer may increase beyond that observed with the natural dipole moment of the helical peptide 67,108,109 , due to the accentuated dipole in the geopilin ( Figure 2 ). The hydrogen bonding as well as the natural flexibility of - helices is also predicted to increase charge transfer rates 112,113,115,116,157,158 . Finally, it is important to note that the dipolar distribution of charges in the pilin, along with the flexible nature of a relatively long - helix that a lso contains a kink caused by a proline at position 22 49,55 , could have prevented the formation of a monolayer dense enough so as to prevent the penetration of the soluble ferricyanide and interaction s with the underlying gold electrode. Conclusions A soluble, truncated pilin derived from the PilA peptide of G. sulfurreducens pilin and termed PilA 19 was successfully expressed in E. coli as a fusion protein with a chitin - binding domain using the Impact TM system. This expression system also enabled the purification of the 84 fusion protein in a chitin column and the cleavage and purification of PilA 19 . In collaboration with Dr. Castro - Forero, we used this pilin to assemble conductive pili in vitro . I also constructed a PilA 19 derivative carrying a cysteine at the N - terminus and used it to deposit the PilA 19 as a monolayer on a gold electrode. The pilin monolayer was confluent and prevented the smaller, nonconductive 1 - undecanethiol from attaching to the gol d electrode. Due to the lack of a redox - active center in the pilin itself I was unable to measure current transferred directly through the pilin. Nevertheless, I demonstrated electron transfer between ferricyanide and the electrode mediated by the pilin m onolayer. Though promising, more studies are needed to demonstrate that the PilA 19 monolayer is impermeable to solutes, such as ferricyanide, and to validate the CV approach as a method to measure p ilin - mediated ET. Electrochemical techniques such as imped ance spectroscopy could be used to test for the presence of pinholes or defects in the monolayers 174 . Similar approaches have been used to demonstrate, for example, that alkanet hiol SAMs, such as the 1 - undecanethiol preparation utilized in this study, limit access of molecules in solution to the electrode 174 . As the pilins are significantly larger and have a more complex structure than an alkanethiol, and are also charged, it is possible that pilin monolayers cannot be formed that are dense enough to exclude ferricyanide. If so, alternative approaches, such as the CP - AFM approach I used, may be needed t o investigate the conductivity of the pilin monolayer and to assess the contribution of amino acid replacements to intramolecular ET . Finally, the possibility of using PilA 19 to produce conductive fibers in vitro shows promise as a tool to investigate the contribution of specific amino acids of the pilin to charge transport. These experiments, coupled to CP - AFM measurements of intramolecular charge transport through the pilin monolayers could help dissect the electron transfer pathways that allow the pili to function as prot ein nanowires. ET - helical peptides has been directly measured by - helical peptides were similarly deposited on a gold 85 electrode 87,172,173,175 - helical peptide then interacts with the tip of the conductive probe of a scanning tunneling microscope or current - sensing AFM to measure electron transfer through the peptide. These studies have shown that - helical peptides are often able to transfer electrons and that the transfer is favored in one direction 175,176 . This ET is thought to involv e the transfer of electrons directly through the peptide backbone, and is facilitated by the secondary structure and the dipole created by the N - terminal amine and C - terminal carboxylic acid 69 . The use of these pilin monolayers in CV experiments may als o be possible by directly binding a redox active ferrocene molecule to the peptide 68,88 . This would all ow us to determine the conductivity of the pilin itself, and thus its intramolecular charge transfer capabilities. Since I am now able to produce a uniform monolayer with the PilA 19 - A20C pilins, the packing should be tight enough such that no ferrocene uni ts will directly interact with the gold electrode. As a result, the only CV signal produced would be generated from the electrons moving from the ferrocene to the gol d through the pilin backbone. 86 CHAPTER 5. Conclusions and Future Directions 87 Review of project The elucidation of both the role of the conductive G. sulfurreducens pilus nanowires and the method of electron transfer (ET) are of great importance to the study of Geobacteraceae and their roles in dissimilatory metal reduction, performance of microbial elec trochemical cell s , and bioremediation of toxic metals and organic pollutants. Furthermore, the identification of conductive pili in these organisms provides a new paradigm in protein ET and may prove instrumental to discover other protein - mediated mechanis ms for long distance ET in other organisms in the emerging field of electromicrobiology. From a practical point of view, studies such as those I presented in this dissertation provide the foundation for applications of protein nanowires in nanotechnology, sensor design, and bioremediation, for improving the performance of microbial electrochemical cell s such as microbial electrolysis cells (MECs), and for developing more efficient schemes for the environmental restoration of environments impacted with metal contaminants. 5.1.1 Review of mechanistic stratification in electroactive biofilms of Geobacter sulfurreducens mediated by pilus nanowires In Chapter 2, I took a genetic approach to investigate the role of pilus nanowires in electroactive biofilms using the m odel representative bacterium G. sulfurreducens . I constructed several pilus - deficient mutants and identified one (pilB, which carried a deletion in the PilB ATPase that energizes pilin assembly) with no apparent defects in c - type cytochrome production or localization. This mutant provided the elusive genetic tool that prevented other studies from conclusively linking electrochemical signatures of anode biofilms of G. sulfurreducens to the pilus conductivity. The pilB mutant had, for example, a ca. 50 % red uction in current production in thin biofilms and was unable to grow biofilms thicker than ca. 10 µ m. The defect was the direct result of a reduced ability of the cells to efficiently transfer electrons 88 across the biofilms because a mutant that produced pi li with increased resistance to the passage of electrons (Tyr3) had the same phenotype. I also demonstrated that thick anode biofilms are stratified in terms of their mechanism of ET , with the bottom stratum ( ca. 10 µ m thick) requiring the coordinated acti vity of pili and matrix - associated c - type cytochromes such as OmcZ, and the conductive pili being essentially for growth and electrochemical activity of biofilm cells growing b eyond this threshold thickness. 5.1.2 Review of site - directed mutagenesis reveals a r ole of aromatic amino acids and local electrostatics in pilus conductivity and extracellular electron transfer I was able to identify amino acid residues of the pilin monomer that were essential for pilus conductivity in vivo and in vitro . Substitution of a single tyrosine residue, Y27, resulted in a 4.5 to 5.5 increase in resistance to the passage of electrons along the length of the pilus. This change in residue , however, did not affect the ability of Y27A mutant cells to produce current in acetate - fed ME Cs or reduce Fe(III) oxides. Hence, while Y27 is necessary for optimal charge transport in vitro , the defect is not rate - limiting in vivo , where the rates of acetate oxidation limit growth. However, subtitution of each of the remaining tyrosine residues (Y 32 and Y57) in the Y32A and Y57A variants reduced the performance of the anode biofilms in acetate - fed MECs to levels comparable to the Tyr3 mutant. The exposure of these two tyrosines on the pilus surface suggests that they play a critical role in ET to t he matrix - associated c - type cytochromes . I also demonstrated that the phenotypic defects of the Tyr3 mutant in anode biofilms can be rescued result suggests th at phenylalanine residues can also transfer electrons between the pili and matrix - associated c - type cytochromes . Furthermore, the result suggest that post - translational oduction in MECs. By contrast, a double aspartic acid mutant (D53A D54A or Asp2) had a severe defect in ET to both Fe(III) oxides and across anode biofilms. The Asp2 phenotype thus validates 89 predictions from a molecular dynamics (MD) model of the pilus gen erated by our collaborator, Dr. Feliciano, which indicated that the local electrostatic environment around the aromatic ring contacts in the Asp2 pilus influence s the packing an d configuration of the aromatic residues and, in turn, ET. 5.1.3 Review of intramol ecular charge transport in pilus nanowires investigated in pilin - electrode interfaces Analysis of the MD model of the pilus revealed intermolecular and intramolecular interactions between aromatic residues of the pilins and also identified pathways for int 137 . Fur thermore, the steep dipolar lan dscape of the pilin has been proposed to promote intramolecular ET as well 45 . Thus, in chapter 4, I tested the hypothesis that the pili n, by itself, conducts electrons. To test this hypothesis, I constructed pilin monolayers attached to gold electrodes and used cyclic voltammetry to investigate if the pilins could cycle electrons between the underlying gold electrode and ferricyanide in t he medium. To do this, I first constructed truncations of the pilins and tested several expression systems to identify one that enabled its production and purification in the quantities needed for my experiments. This part of the work was part of a collabo ration with Dr. Angelines Castro Forero, who used the truncated pilins to assemble protein nanowires in vitro 166 . Together, we selected a const ruct and expression system that allowed us to produce a recombinant PilA 19 pilin with high yields and assembled a set of these into conductive PilA 19 pili. I then constructed a PilA 19 derivative carrying a cysteine at the N - terminus (PilA 19 - A20C) for deposi tion on a gold electrode. In Chapter 4, I demonstrated that it is possible to deposit a confluent pilin monolayer on gold electrodes. Using insulating self - assembled monolayers (SAMs) of undecanethiol as controls, I demonstrated that the pilin monolayer fu nctioned as a mediator for electrons between ferricyanide and the underlying electrode surface. I did not observe any change in electron cycling using pilin monolayers 90 carrying amino acid replacements in the three tyrosines (Tyr3 mutant), the two D53 and D 54 residues (Asp2 mutant), or the positively - charged amino acid R28 (R28A mutant). The results thus indicate that the pilins have paths for intramolecular ET , but these pathways do not involve amino acids that are needed for ET once the pilins assemble to form the pilus fiber. This finding suggests that aromatic residues need to cluster in the pilin assembly to create paths for ET and charged amino acids influen ce the rates of ET via aromatic residues because they are position ed in the assembly close enough to the aromatic rings to provide the local electrostatic environment required for optimal ET. The results also indicate that charge hopping within the peptide backbone is the predominant m echanism for intramolecular ET. F uture directions 5.2.1 Future direction s for studying mechanistic stratification in electroactive biofilms of Geobacter sulfurreducens mediated by pilus nanowires In Appendix 1 I showed that it is possible to increase the thickness of pilB biofilms growing on anode electrodes after successive t ransfers of the anode biofilms in to fresh medium. Furthermore, I also documented occasional increases in current production in this strain. This result suggests that compensatory mechanisms exist to overcome the pilus deficiency though they are not as effi cient as those provided by the conductive pili. Future work could, for example, investigate if the type and/or quantity of c - type cytochromes in the biofil m matrix have changed so as to promote ET in thick biofilms despite the lack of pili. Alternatively, pili - like filaments may have been assembled that provide structural support to the biofilms, allowing them to grow in thickness and, perhaps, conduct electrons. Such compensatory mechanisms may be similar to those reported for a pilin - deficient strain, whi ch appears to produce pili - like filaments at some low frequency 177 and can be adaptively evolved to grow in Fe(III) oxides by secreting a small c - type cytochromes that functions as an electron shuttle 178 . Finally, it is still not 91 clear how pili and c - type cytochromes coordinate their activities in the b ottom stratum of the biofilm or what is the exact role of cytochromes in the outer layers of the biofilm ; these questions should be further investigated. Electrochemical characterization of the pilB mutant and other mutants I constructed and characterized will prove instrumental to answer ing these out standing questions. 5.2.2 Future directions for site - directed mutagenesis to reveal the role of aromatic amino acids and local electrostatics in pilus conductivity and extracellular electron transfer In Chapter 3, I demonstrated that defects in pilus conductivity do not necessarily translate to changes in biological phenotypes associated with pilus function. Still, many questions remain unresolved. For example, why Y27 is not critical to current production in MECs bu t Y32 and Y57 are. The conductivity of the Y32A and Y57A mutant pili has never been examined and could help explain the observed phenoty pes. Y27A pili on the other hand , were defective in charge transport but performed like the WT in the MECs. However, the re may exist a minimum charge transfer rate that allows the pili to grow thick, electroactive biofilms that can only be achieved when certain amino acids are present in the pilus ET pathway. Increasing the sensitivity of the CP - AFM measurements to unmask p henotypes at the low voltages that operate in the cells may also prove instrumental to understand their role in ET. Furthermore, their role in ET to c - type cytochromes and other pili also needs to be evaluated. Future work will also need to evaluate how t hese mutations affect the binding of metals such as oxidized iron and uranium. The MD model predicts that the tyrosines exposed on the pilus surface will be the ones mediating the final step in ET to external electron acceptors. Furthermore, negatively - cha rged ligands in the C - t random coils of the pilins could form a metal them close enough to the exposed aromatic residues for efficient ET. Finally, the role of post - translational modifications of t he pilin warrants attention. Tyrosine residues are post - translationally modified in other bacterial pilins 92 and influence the binding properties of the pili 70,73,179 . I showed that phenyalanine replacements - F mutant) do not affect the elec trochemical activity of the anode biofilms. Hence, post - translational modifications of the pilin, if they exist, are not critical to pili functions in the electroactive matrix. However, post - translational modifications could influence the binding of metals such as iron and uranium. If so, Tyr3 - F pili are expected to have defects in metal binding and perhaps also in their reduction. Lastly, it is important to construct atomic or pseudoatomic structural models of the pili based on experimental validated struc tures or, in lieu of this, to further refine the MD pilus model to provide information about the electronic structure of the pili. An electronic structural model will allow one to make further predictions about the roles of specific residues in ET, which c ould be tested experimentally by in vitro and in vivo assays. The MD model that I experimentally tested in Chapter 3 suggests that small perturbations in protein structure, such as those resulting from the Asp2 mutation, influenced the configuration of the aromatics and ET. Phenylalanine residues are common replacements for tyrosine or tryptophan residues to minimize structural changes in proteins 180,181 . However, such substitutions also removed the hydroxyl group in tyrosines, which could have affected the structure of the pilin and the number and configurations of the contacts. Hence, it will be important to introduce the Tyr3 - F mutation in the MD model and investigate how the pilus structure is affected. The contribution of PCET to pilus ET also needs to be studied. While the Tyr3 - F had no observable phenotype in the MECs, the pilus conductivity may have been affected, as we observed in the Y27A mutant. If so, the pH dependence of the conductivity of the Tyr3 and Tyr3 - F pili in vivo and in vitro could elucidate whether the tyro sines are involved in PCET. In addition, chemical complementation of the mutant anode biofilms in MECs or in Fe(III) oxide cultures could be attempted. Chemical complementation of a histidine proton acceptor mutant with imidazole has been demonstrated prev iously 95 . Changing the buffering capacity of the medium 93 or directly measuring the pH in a biofilm would also help us to investigate the role of PCET 182 . Bulk conductivity measurements on the pili, such as CV, can be performed in solution, and therefore, direct effects on chemical complementation or buffering capaci ty o n conductivity can be measured. 5.2.3 Future directions for examining intramolecular charge transport in pilus nanowires via pilin - electrode interfaces In Chapter 4, I demonstrated that it is possible to deposit confluent pilin monolayers of the PilA 19 - A20 C peptide that prevent the small 1 - undecanethiol molecule from attaching and insulating the electrode. Furthermore, thickness measurements consistently indicated that the pilins are packed at a 40 ° tilt in the monolayer. Hence, these nanostructured interf aces show promise as a tool to investigate intramolecular ET through the pilin peptide. Future directions could focus on improving the sensitivity of the assays so as to unmask potential conductivity changes in the peptide due to single amino acid replacem ents. One way to do this is to rule out that the soluble ferricyanide is not leaking through the pilin monolayer and artifactually enhancing the cyclic voltammograms. Similar to other small conductive peptides, the pilin does not produce its own redox peak s, and therefore, a direct measurement of the conductivity of the pilin is not possible. For this reason, my assays investigated the conductivity of the pilin monolayer through its ability to cycle electrons between ferricyanide and the gold electrode. How - terminus to sensitively measure ET across the pilin monolayer. In this interface, cyclic voltammograms would reflect direct ET from ferrocene to the electrode through the pilins only. This syste m could also be adapted to investigate the binding affinity of the pilins for metals such as iron and uranium and to test the effect of ta rgeted amino acid replacements o n ET to specific metals. Such interfaces could be the foundation of larger interfaces for the remediation and/or reclamation of metals and for the development of metal sensors. 94 Direct conductivity measurements of the pilin monolayers with CP - AFM also show promise as a tool to investigate intramolecular ET via pilins. AFM - based approaches o r other methods also need to be developed to determine how tightly packed the pilins are, as the packing of peptides in a monolayer is predicted to affect charge transfer 183,184 . Lastly, the possibility of assembling the same recombinant pilins as conductive pili can be used as a complementary tool. The conductivity of the synthetic pili produced with recombinant pilins integrates intermolecular and intramolecular ET pathways, whereas the pilin - electrode interfaces isolate the intramolecular path only. Furthermore, synthetic pili do not contain any po st - translational modifications, where as native pili (i.e., purified from G. sulfurreducens ) do. Hence, the three systems are complementary and provide the necessary tools to dissect the complex pilus system and potential ET pathways. 95 AP PENDICES 96 APPENDIX A MEC ADAPTATION IN PILB MUTANT STRAIN 97 Abstract The effect of nutrient availability and/or accumulation of toxic metabolic products was tested in pilB - driven m icrobial electroche mic al cells ( MECs) to rule out any influence in the mutant phenotype. I also performed adaptive evolution experiments to investigate the stability of the pilB mutant phenotype. The results suggest that the pilB mutant biofilms can grow to WT thicknesses after repeated transfers of the anode biofilms to MECs with fresh medium and even increase their current production, though never to the levels observed in the WT MECs. Introduction In Chapter 2 I demonstrated that the pili - deficiency of the pilB mutant preve nted it from growing biofilms beyond a threshold thickness ( ca. 10 µm) and also reduced the rates of current increase and maximum current. Here, I describe experiments that tested whether growth was limited by a component of the media (i.e., nutrient avail ability and/or accumulation of toxic pilB mutant to adapt to and overcome the biof ilm growth limitation in MECs. Materials and Methods T esting media components in MECs MEC exp eriments were performed as in Chapter 2 and contained 1 mM acetate. The WT strain of G. sulf u rreducens was grown on the anode electrode of the MECs until maximum current was reached and current began to decrease (deceleration phase), different media compon ents were added to determine if their availability could restore maximum levels of current production. I tested acetate, vitamins, minerals, and salts. 98 Sequential t r an sfers of anode biofilms in MECs Anode biofilms of the WT, pilB, and omcZ strains wer e grown in MECs with 3 mM acetate. Once all of the acetate was depleted and current had decreased to < 0.1 mA, the MECs were taken into an anaerobic chamber (COY Labs) to rinse the anode chamber with ddH 2 O, and replenish it with fresh media containing 3 mM acetate. Results To determine if a component of the media was limiting in the pilB MECs, acetate, mineral mix, vitamins, or media salts were added to the MEC when current began to decrease (about 32 h) ( F igure 42 ). Addition of the vitamin mix had no effect on current, but supplementing the anode medium with the electron donor ( 3 mM acetate) restored current production to the original maximum current levels. Current production plateaued at this level for a length of time similar to that of control pilB MECs fed 3 mM acetate from the beginning ( F igure 42 ) Once current production declined, I supplemented the anode medium with a second dose of 3 mM acetate but the deceleration phase continued. Media suppleme ntation with mineral mix, salts, or vitamins had no effect either, suggesting that other factors limited the growth and electroactivity of the biofilms. To determine if the production of a waste product was participating in the pilB phenotype, the medium was replaced after the culture had consumed 3 mM acetate. With the media replacement, a slight increase in current was seen for the pilB strain; however with fresh media and the biofilm previously formed on the anode, the current still plateaued at a value considerably lower compared to the WT culture (which produce 2.3 mA of current). It has been reported that pili - deficient strains of G. sulfurreducens , which cannot reduce Fe(III) oxides optimally, can bypass the limitation by adaptively evolving alternat ive mechanisms 178 . Thus, I investigated whether the pilB strain was able to adapt under selective pressure to form thick biofilms and produce current in MECs. To do this I sequentially replenish 99 the anode medium with fresh medium up to 10 times. Interestingly, both t he WT and the pilB biofilms grew to more than 50 m thicknesses after 10 sequential media transfers. Hence, pilB biofilms can bypass the pilus deficiency and grow thicker biofilms under selective pressure to couple the oxidation of acetate to current production. As for the effect on current production, I observed two different phenotypes. Current in some of the pilB - driven MECs remained low (less than 1 mA) throughout the experiment, but increased above 1 mA in some of the MECs after multiple medium replacements ( Figure 43 ). Such increases are still low compared to those observed in the WT strain, which reached ca. 5 mA of maximum current, and much higher than controls with the omcZ mutant ( Figure 43 ). Yet they suggest that there are compensatory mechanis ms that allow the cell to bypass the pili deficiency. If, as reported 178 , a soluble cytochrome can be produced to bypass the pilus deficiency and shuttle electrons between the cell and insoluble electron acceptors, pilB cells could grow in anode biofilms by dischargi ng respiratory electrons to the shuttle. However, it is unlikely that this cytochrome can effectively shuttle electrons between the cells and the electrode, explaining the inability of the pilB mutant t o produce current at WT levels. Conclusions There ar e several reasons why G. sulfurreducens biofilms might stop producing current in an MEC (e.g., depletion of a required nutrient and buildup of a toxic byproduct). To determine if limitations in current production was due to acetate limitation or the buildu p of waste productions, MEC experiments were performed in which the media in the MEC was replaced after the culture had stopped producing current. Initially, addition of acetate was sufficient to restore current production, but not to provide any increase in current, however, current production could not be rescued by addition of any media component. This result suggested that there was a buildup of a toxic byproduct that upon removal allowed the restoration of current production. Replacing the anode medi um with fresh medium once acetate had been 100 depleted, allowed us to adaptively grow thicker ( ca. 50 m) biofilms of both the WT and pilB strains after 10 sequential media transfers. Current increased over 1 mA in some of the pilB - driven MECs but never to WT levels, supporting our model ( C hapter 2) that the pili are required for optimal electron transfer in anode biofilms. The mechanism by which pilB mutants were able to grow thick biofilms without changing current production has yet to be determined. 101 APPENDIX B TABLES 102 Table 1 Strains and plasmids used in Chapter 2 Bacterial strain or plasmid Relevant genotype and properties a Source or reference(s) Geobacter sulfurreducens WT Wild type strain PCA 19 pilA pilA :: aaaC1 , Gm r This study pilB pilB :: aaaC1 , Gm r This study omcZ omcZ :: aaaC1 , Gm r This study pilA - E5A pilA E5A :: aadA , Spec r This study pilB pilB :: loxP This study pilB + pilB complemented with pRG5 - pilB This study pilB gspE pilB strain carrying the gspE: : aaaC1 mutation, Gm r This study pilB mshE pilB strain carrying the mshE: : aaaC1 mutation, Gm r This st udy Tyr3 pilA Y27,32,57A:: aadA , Spec r This study Plasmids pCM351 Amp r , Tet r , Gm r , ColE1 ori , oriT 123 pCM158 Km r , trfA , oriT , oriV , ColE1 ori , cre 123 pCR2.1 - TOPO Amp r , Km r , ColE1 ori Invitrogen pRG5 Shuttle vector for G. sulfurreducens Spc r , P taclac 124 pRG5 - pilB pRG5 with G. sulfurreducens pilB This study 103 Table 2 Primers used in Chapter 2 Primer - Amplificat ion Use RS1 GATCTGGTCGGATACAACACC pilB upstream pilB RS2 TTATGCGGCCGCCATATGCATCTGCTAGCCTGCATAGACTCTCC pilB upstream pilB RS3 GTGTTAACCGGTCATATGCAGTGGCTGACGACTAAACAAAATGCC pilB downstream pilB RS4 CTCTTGTGAGGATGCAGGTAC pilB downstream pilB RS5 ATGGACCT CGAAGCCTACCT pilA upstream pilA RS6 TTATGCGGCCGCCATATGCATCTGAAGCATAAGTG pilA upstream pilA RS7 GTGTTAACCGGTCATATGCACGCCCGAAAGTTAA pilA downstream pilA RS8 TCCATGCATCATTTTCGATG pilA downstream pilA RS9 ATATGGCGTTTACCGCAGAG omcZ upstream omcZ RS10 TTAT GCGGCCGCCATATGCAGCTCCGAAGAAAGTCAAACG omcZ upstream omcZ RS11 GTGTTAACCGGTCATATGCAGATGCGCCAATCAGTACCTT omcZ downstream omcZ RS12 CACAGCCAGGTACCATCTGA omcZ downstream omcZ RS13 TTTCTCAGCAATCCATCGAG gspE upstream pilB gspE RS14 TTATGCGGCCGCCATATGCAATCTGT TCCATGTCGTGTGC gspE upstream pilB gspE RS15 GTGTTAACCGGTCATATGCAGTATGCCGACCTTCCGGTAT gspE downstream pilB gspE RS16 AGTGATTTTGCTCCGAATGG gspE downstream pilB gspE RS17 TTTCGGCCATGTACTCCTTT mshE upstream pilB mshE RS18 TTATGCGGCCGCCATATGCATCCTTGACGATGC TTTCCAT mshE upstream pilB mshE RS19 GTGTTAACCGGTCATATGCACTCCTCCCTGCACGGTTGA mshE downstream pilB mshE RS20 CGACATCTTGTTCTCGTGGA mshE downstream pilB mshE RS21 TGCATATGGCGGCCGCATAA aaaC1 loxP Excible Gm marker RS22 TGCATATGACCGGTTAACAC aaaC1 loxP Excib le Gm marker RS23 GTGGTGAAGGGGTAGGTTGA pilA upstream pilA::aadA RS24 GTTAGGCGTCATCCTGTGCTTAACTTTCGGGCGGATAGG pilA upstream pilA::aadA RS25 CAAGCCGACGCCGCTTCTTGATTAAATACATACTGGAGG pilA downstream pilA - Sp RS26 GCGACTTCCACTCGGTACC pilA downstream pilA - Sp RS27 GCACAGGATGACGCCTAAC aadA Sp marker RS28 GAAGCGGCGTCGGCTTG aadA Sp marker RS29 GGTTTCACCCTTATC G C G CTGCTGATCGTCGTT pilAE5A::aadA pilA - E5A RS30 AACGACGATCAGCAG C G C GATAAGGGTGAAACC pilAE5A::aadA pilA - E5A RS31 TCCGCAGTTCTCGGCG GC T CGTGTCAAGGCGTAC pilAY27A ::aadA Tyr3 104 Table RS32 GTACGCCTTGACACG A GC CGCCGAGAACTGCGGA pilAY27A::aadA Tyr3 RS33 GTATCGTGTCAAGGCG GC C AACAGCGCGGCGTC pilAY32A::aadA Tyr3 RS34 GACGCCGCGCTGTT G GC CGCCTTGACACGATAC pilAY32A::aadA Tyr3 RS35 CCGCATTTGCTGATGATCAAACC GC T CCGCCCGAAAG TTAA pilAY57A::aadA Tyr3 RS36 TTAACTTTCGGGCGG A GC GGTTTGATCATCAGCAAATGCGG pilAY57A::aadA Tyr3 RS37 CCAACACAAGCAGCAAAAAG pilA qPCR RS38 GCAGCGAGAATACCGATGAT pilA qPCR RS39 ATGGGTGGCAAGGACTTTAC GSU1497 qPCR RS40 ACACCCGGTTACCAGAAGAG GSU1497 qPCR RS41 GTC CAACAAAGGGAACTGCT xapD qPCR RS42 CCTCCGCAGAGAGGTAATCA xapD qPCR RS43 CACGAGCCTGACACTCACTC omcZ qPCR RS44 AAGGTTGCTGACCTTGTTGG omcZ qPCR RS45 GACACGGTCAACCAGAACAA omcB qPCR RS46 GGTCCCAGTTTACGACAGGA omcB qPCR RS47 AGTTCTCGACGTACGCCACT rpoD qPCR RS48 TCAGCTTGTTGATGGTCTCG rpoD qPCR 105 Table 3 Bacterial strains and plasmids used in Chapter 4 Bacterial strain or plasmid Relevant genotype and properties a Source or reference(s) Geobacter sulfurreducens WT Wild type strain PCA 19 Tyr3 pilA Y27,32,57A:: aadA , Spec r This study Y27A pilA Y27A:: aadA , Spec r This study Y27F pilA Y27F:: aadA , Spec r This study Y32A pilA Y32A:: aadA , Spec r This study Y57A pilA Y57A:: aadA , Spec r This study Tyr3 - F pilA Y27,32,57F:: aadA , Spec r This study Asp2 pilA D53,54A:: aadA , Spec r This study Plasmids pCR2.1 - TOPO Amp r , Km r , ColE1 ori Invitrogen pCR2.1 - pilASpec Amp r , Km r , ColE1 ori pilA::aadA Spec r Chapter 2 pCR2.1 - Tyr3 Amp r , Km r , ColE1 ori pilA Y27,32,57A:: aadA , Spec r Chapter 2 pCR2.1 - Y27A Amp r , Km r , ColE1 ori pilAY27A::aadA Spec r This study pCR2.1 - Y27F Amp r , Km r , ColE1 ori pilAY27F::aadA Spec r This study pCR 2.1 - Y32A Amp r , Km r , ColE1 ori pilAY32A::aadA Spec r This study pCR2.1 - Y57A Amp r , Km r , ColE1 ori pilAY57A::aadA Spec r This study pCR2.1 - Tyr3 - F Amp r , Km r , ColE1 ori pilA Y27,32,57F:: aadA , Spec r This study pCR2.1 - D5354A Amp r , Km r , ColE1 ori pilAD53,54A::aadA Spec r This study pRG5 Shuttle vector for G. sulfurreducens Spc r , P taclac 124 106 Table 4 Primers used in Chapter 3 Primer - Amplification Use RS31 TCCGCAGTTCTCGGCG GC T CGTGTCAAGGCGTAC pilAY27A::aadA Y27A RS32 GTACGCCTTGACACG A GC CGCCGAGAACTGCGGA pilAY27A::aadA Y27A RS33 GTATCGTGTCAAGGCG GC C AACAGCGCGGCGTC pilAY32A::aadA Y32A RS34 GACGCCGCGCTGTT G GC CGCCTTGACACGATAC pilAY32A::aadA Y32A RS35 CCGCATTTGCTGATGATCAAACC GC T CCGCCCGAAAGTTAA pilAY57A::aadA Y57A RS36 TTAACTTTCGGGCGG A GC GGTTTGATCATCAGCAAATGCGG pilAY57A::aadA Y57A RS49 TCCGCAGTTCTCGGCG T T T CGTGTCAAGGCGTAC pilAY27F::aadA Tyr 3 - F, Y27F RS50 GTACGCCTTGACACG A A A CGCCGAGAACTGCGGA pilAY27::aadA Tyr3 - F, Y27F RS51 GTATCGTGTCAAGGCG T T C AACAGCGCGGCGTC pilAY32F::aadA Tyr3 - F RS52 GACGCCGCGCTGTT G A A CGCCTTGACACGATAC pilAY32F::aadA Tyr3 - F RS53 CCGCATTTGCTGATGATCAAACC T T T CCGCCCGAAAGTTAA pilAY5 7F::aadA Tyr3 - F RS54 TTAACTTTCGGGCGG A A A GGTTTGATCATCAGCAAATGCGG pilAY57F::aadA Tyr3 - F RS55 GAGTCCGCATTTGCT G C TG C T CAAACCTATCCGCCC pilAD53,54A::aadA D5354A RS56 GGGCGGATAGGTTTG A G CA G C AGCAAATGCGGACTC pilAD53,54A::aadA D5354A 107 Table 5 Strains used in Chapter 4 E. coli Strain Genotype Reference M15[pREP4] NaI s , Str s , Rif s , Thi - ,Lac - , Ara + , Gal + , Mtl - , F - , RecA - , Uvr + ,Lon + , Kan R Qiagen® K12 TB1 ara (lac - proAB) [ 80dlac (lacZ)M15] rpsL (Str R ) thi hsdR New England Biolabs In c.® Rosetta TM 2 (DE3)pLysS ompT hsdS B (r B - m B - ) gal dcm (DE3) pLysSRARE2 (Cam R ) Novagen® - argF)U169 recA1 endA1 hsdR17(rk - , mk+) phoA supE44 thi - Life Technologies Co. ® XL10 - Gold - hsdSMR - mr r)173 endA1 supE44 thi - 1 recA1 gyrA96 relA1 lac Hte [F´ proAB Agilent Technologies® 108 Table 6 Primers used in Chapter 4 Mutated codons are shown in bold whereas mutated nucleotides are u nderlined. Restriction sites are represented in italics. Plasmid Sequence Restriction Site All pQE - 30::pilA ACTTTCGGGCGGATAGGTTT - pQE - 30::pilA TTCACCCTTATCGAGCTGCT - pQE - 30::pilA 5 CTGCTGATCGTCGTTGCGAT - pQE - 30::pilA 10 GCGATCATCGGTATTCTCGC - pQE - 30::p ilA 15 CTCGCTGCAATTGCGATTCC - pQE - 30::pilA 20 ATTCCGCAGTTCTCGGCGTA - pMAL - c2x::pilA / pMAL - p2x::pilA CCC GAAGCCGTTC TTCACCCTTATCGAGCTGCT CCC AAGCTT TTAACTTTCGGGCGGATAGGT XmnI HindIII pMAL - c2x::pilA 20 / pMAL - p2x::pilA 20 CCC GAAGCCGTTC ATTCCGCAGTTCTCGGCGTA XmnI CCC AAGCTT TTAACTTTCGGGCGGATAGGT HindIII pTYB11::pilA GGTGGTT GCTCTTCC AACTTCACCCTTATCGAGCTGCT SapI GGTGGT CTGCAG TCATTAACTTTCGGGCGGATAGGT PstI pTYB11::pilA 10 GGTGGT CTGCAG TCATTAACTTTCGGGCGGATAGGT PstI GGTGGTT GCTCTTC CAACGCGATCATCGGTATTCTCGC SapI pTYB11::pi lA 19 GGTGGTT GCTCTTC CAACGCGATTCCGCAGTTCTCGGC SapI GGTGGT CTGCAG TCATTAACTTTCGGGCGGATAGGT PstI pTYB11::pilA 22 GGTGGTT GCTCTTC CAACCAGTTCTCGGCGTATCGTGT SapI GGTGGT CTGCAG TCATTAACTTTCGGGCGGATAGGT PstI pTYB11::pilA 19 Y27 A TCCGCAGTTCTCGGCG GC T CGTGTCAAGGCGTAC - GTACGCCTTGACACGAGCCGCCGAGAACTGCGGA - pTYB11::pilA 19 Y32A GTATCGTGTCAAGGCG GC C AACAGCGCGGCGTC - GACGCCGCGCTGTTGGCCGCCTTGACACGATAC - pTYB11::pilA 19 Y57A CCGCATTTGCTGATGATCAAACC GC T CCGCCCGAAAGTTAA - TTAACTTTCGGGCGGAGCGGTTTGATCATCAGCAAATGCGG - pTYB11::pilA 1 9 R28A CCGCAGTTCTCG GC G TATGCTGTCAAGGCG CGCCTTGACAGCATACGCCGAGAACTGCGG - - pTYB11::pilA 19 D53,54A GAGTCCGCATTTGCT G C TG C T CAAACCTATCCGCCC - GGGCGGATAGGTTTGAGCAGCAGCAAATGCGGACTC - 109 Table 6 pTYB11::pilA 19 A20 C GGTTGTTGTACAG AA C TGCATTCCGCAGTTCTCGG - CCGAGAACTGCGGAATGCAGTTCTGTACAACAACC - 110 Table 7 Plasmids used in Chapter 4 Amino acid position based on the mature pilA gene sequence (i.e., after processing of its leader peptide) . Plasmid Genotype Reference pQE - 30 UA P T5 His 6 - tag A mp r Qiagen® pQE - 30::pilA pQE - 30 plus pilA Amp r This study pQE - 30::pilA 5 pQE - 30 plus - Amp r This study pQE - 30::pilA 10 pQE - 30 plus - Amp r This study pQE - 30::pilA 15 pQE - 30 plus - Amp r This study pQE - 30::pilA 20 pQE - 30 plus - Amp r This study pMAL - c2x P tac MBD - tag malE signal sequence Amp r New England Biol abs Inc.® pMAL - p2x P tac MBD - tag Amp r New England Biolabs Inc.® pMAL - c2x::pilA 20 pMAL - c2x plus - inserted into XmnI - HindIII Amp r This study pMAL - p2x::pilA 20 pMAL - p2x plus - inserted into XmnI - HindIII Amp r This study pTYB11 P T7 Sce VMA intein CBD - tag A mp r New England Biolabs Inc.® pTYB11::pilA pTYB11 plus pilA inserted into NdeI - SapI Amp r This study pTYB11::pilA 10 pTYB11 plus - inserted into NdeI - SapI Amp r This study pTYB11::pilA 19 pTYB11 plus - inserted into NdeI - Sa pI Amp r This study pTYB11::pilA 20 pTYB11 plus - inserted into NdeI - SapI Amp r This study pTYB11::pilA 21 pTYB11 plus - inserted into NdeI - SapI Amp r This study pTYB11::pilA 19 A20C pTYB11::pilA 19 pilA A20A a Amp r This study pTYB11::pilA 19 A2 0C Y27A pTYB11::pilA 19 pilA A20CY27A a Amp r This study pTYB11::pilA 19 A20C Y32A pTYB11::pilA 19 pilA A20CY32A a Amp r This study pTYB11::pilA 19 A20C Y57A pTYB11::pilA 19 A20C Y27,32,57 A pTYB11::pilA 19 A20C R28 A pTYB11::pilA 19 A20C D5354 A pTYB11::pilA 19 pilA A20CY57A a Am p r pTYB11::pilA 19 pilA A20CY27,32,57A a Amp r pTYB11::pilA 19 pilA A20CR28A a Amp r pTYB11::pilA 19 pilA A20CD53,54A a Amp r This study This study This study This study 111 Table 8 Buffers used to purify recombinant pilins as fusion pr oteins with a Chitin - binding Domain (CBD) using the IMPACT TM system using a chitin resin column Buffer Composition Column Buffer 20 mM Tris - HCl pH 7.4, 100 mM NaCl, 1 mM EDTA Lysis Buffer 20 mM Tris - HCl pH 7.4, 100 mM NaCl, 1 mM EDTA, 2% CHAPS Wash Buf fer 1 20 mM Tris - HCl pH 7.4, 600 mM 1 M NaCl, 1 mM EDTA Wash Buffer 2 20 mM Tris - HCl pH 7.4, 1 M NaCl, 1 mM EDTA Cleavage Buffer 20 mM Tris - HCl pH 9.0, 100 mM NaCl 50 mM DTT Elution Buffer 20 mM Tris - HCl pH 9.0, 100 mM NaCl 112 APPEND IX C FIGURES 113 Figure 1 Structure of the PAK pilin and geopili n Crystal structure of the P. aeruginosa strain K pilin 49 with the - loop shown in red and the D region shown in blue. Representative NMR structure of the G. sulfurreducens pilin 58 with the random coil region sh own in red 114 Figure 2 Properties of the G. sulfurreducens pilin Representative NMR structure of the G. sulfurreducens pilin 58 showing: A) aromatic residues (phenylalanine in green and tyrosine in yellow) B) charged residues (negative residues in red and positive residues in blue) C) electrostatic potential calculated 185 for the pilin (positive potential in blue and negative potential in red). D) All NMR structures collected of the G. sulfurreducens pilin 58 showing the flexibility of the C - t random coil. Electros tatic potentials were computed for the Geobacter sulfurreducens NMR atomic structure 58 using Swiss - Pbd Viewer 185 . 115 Figure 3 - stacking configurations Images representing the three - stacking configurations. 116 Figure 4 Charge transfer mechanisms Depiction of tunneling (A) and hopping (B). Relay states are represented by 1, 2, 3, and 4. E i s the barrier for movement into the relay state. (C) Representation of the possible mechanisms for charge transfer in an - helix (electron tunneling, electron hopping, and electron hole hopping through the amide groups of the peptide backbone). All black single headed arrows represent the movement of electrons. White single headed arrows represent the movement of electron holes. 117 Figure 5 Amide backbone charge hopping An image of the pilin - helical backbone (carbon and hyd rogen in gray, oxygen in red, and nitrogen in blue). Hydrogen bonds are shown as dashed lines and the amide electron hole hopping pathway designated with arrows. 118 Figure 6 Absorption spectra of oxidize and reduced cell extr acts Absorption spectra of WT (a), pilB (b), pilA (c), and pilA - E5A (d) dithionite - reduced (solid) and then ferricyanide oxidized (dashed) and biofilm cell extracts, which were used to estimate the overall biofilm cytochrome content s. 119 Figure 7 C haracterization of cytochrome content in pili - deficient mutant cultures (a) Cytochrome content (total cellular heme, measured with the pyridine hemochrome method) of planktonic cells of the pili - deficient strains pilB, pilA, and p ilA - E5A relative to the WT. Significant changes relative to the WT values are indicated with stars (* p <0.05) (b) Heme - stained proteins in whole - cell extracts of planktonic cells of WT (lane 1), pilB (lane 2), pilA, (lane 3), and omcZ (lane 4), which was used as negative control lane for the ~ 30 - kDa OmcZ S band (arrow). 120 Figure 8 Expression of biofilm related genes in pilus deficient mutants Expression of genes in the pilin operon ( pilA and GSU1497), EPS synthesis ( xapD ), and outer membrane c - type cytochromes ( omcZ and omcB ) in 48 - h old biofilms formed by the pili - deficient strains pilB (black), pilA (gray) and pilA - E5A (white) compared to WT. The constitutive gene rpoD was used as internal control. Significant changes rela tive to the pilB values are indicated (* p <0.05; ** p <0.005) . 121 Figure 9 Phenotypic characterization of pili - deficient mutant (pilB, pilA, and pilA - E5A) biofilms (a) Relative c - type cytochromes content of the pili - de ficient mutant strains in reference to WT cells (average and standard error of two biological replicates for each). (b) Biofilm biomass on plastic surfaces estimated from the OD 580 of the biofilm - associated crystal violet solubilized with acetic acid (show n are averages and standard deviation of 8 biofilm replicates of the WT, pili - deficient mutants, and the genetic complemented pilB+ strain). (c) Heme - containing protein bands isolated from the biofilm matrix of the WT and pili - deficient strains. Lanes were loaded with 20 µg of protein. Numbers at left are relative molecular masses of protein standards in kDa. 122 Figure 10 B iofilm electroactivity of selected mutants in MECs fed with 1 mM acetate (a) Current generation by WT, pili - deficient strains (pilB, pilA, and pilA - E5A), and omcZ mutant in MECs fed with 1 mM acetate. Inset shows current generation (Y axis, in mA) over time (X axis, in days) by double mutants pilB mshE (solid gray), pilB gspE (dashed gray), and an equal mix ture of pilB and omcZ cells (blue). (b and c) Confocal micrographs showing top and side views of biofilms of the WT (b) and pilB (c) strains stained with the BacLight viability kit (green, live; red, dead). Scale bar, 20 µm. 123 Figure 11 Current production as a function of biofilm thickness in the WT and pilB strains Anode biofilm thickness (a), maximum current (b), and maximum rates of current production (c) in MECs fed with 1, 2 and 3 mM acetate driven by the WT (solid sym bols) or the pilB mutant (open symbols) strains. 124 Figure 12 Role of pilus conductivity in the growth and electroactivity of anode biofilms (a) AFM topographic images of pili purified from the WT and Tyr3 strains and depos ited on a HOPG surface. (b) Biomass of 48 - h old biofilms of the pilB and Tyr3 mutants relative to WT (shown are averages and standard deviation of 8 biofilm replicates for each strain). (c). Current - voltage ( I - V ) plots obtained after probing the transversa l conductivity of individual WT (black) and Try3 (blue) pili d eposited on the HOPG surface as shown in (a). Inset shows the average resistance values obtained from the I - V plots and standard deviation. (d) Current production by WT, pilB, and Tyr3 biofilms grown in MECs fed with 3 mM acetate. 125 Figure 13 Model of mechanistic stratification of electroactive biofilms Cytochromes (circles) and pili (crisscrossed lines) permeate the biofilm matrix and contribute to electron tra nsfer in the layers of biofilm stratum proximal to the electrode . The efficiency of the cytochrome pathway is limited in the distal region of the biofilm (> 10 m away from the electrode), resulting in the accumulation of reduced cytochromes (indicated by the solid circles) and leaving the pili network required as an electronic conduit. 126 Figure 14 Surface maps of MS - optimized GS pilus A) The position of one pilin in the optimized pilus structure, (orange), B) the expos ed core amino acids (green), and C) the aromatic residue isodensity. Reprinted with permission 137 . 127 Figure 15 Potential charge transfer pathways identified in the MD model Relationship of aromatic residues i n the G. sulfurreducens MD pilus model. Average distances between the amino acids across the 19 subunit assembly are indicated (arrows). 128 Figure 16 Electron transfer paths predicted in the G. sulfurreducens pilus from the MD - refined pilus model The fiber core only path involves the electron hopping between two residues buried in the fiber core (F24 and Y27). The fiber core + C - t coil path involves both buried (F24 and Y27) and exposed aromatic residues (Y32 and F51) on fibe r core as well as the C - t random coil (Y57). Y27 is in gray to indicate that it is an alternate path for ET. For each path, intramolecular transfers ar e enclosed in boxes. 129 Figure 17 Surface exposure of Y32 and Y57 (a) MD - m odel of the G. sulfurreducens pilus fiber showing the surface exposure of tyrosine residues Y32 (bronze) and Y57 (yellow). (b) Close - up of Y57 interacting with an uranium atom (orange) through its hydroxyl group and coordinations by the carboxyl groups of D39 and the terminal S61 residues. (c) Atomic model showing the coordination of U(IV) with pili ligands (oxygen, red; U(IV), gray; carbon, black) showing two bidentate and one monodentate carboxyl linkages. Reproduced with permission from 13 . 130 Figure 18 Amide electron hole hopping and phenylalanine dimer intermolecular transfer pathway Direction of electron hole hopping indicated by arrows. 131 Figure 19 Biof ilm electroactivity in MECs fed with 3 mM acetate for Y32 and Y57 Current generation by WT (black), Tyr3 (red), Y57A (green), and Y32A (blue) MECs fed with 3 mM acetate. 132 Figure 20 Biofilm electroactivity in MECs fed with 3 mM acetate for Y27 amino acid replacement mutants Current generation by WT (black), Y27F (pink), and Y27A (green) in MECs fed with 3 mM acetate. 133 Figure 21 Transversal CP - AFM measurements of Y27 amino acid replacements Average resistance values obtained from the I - V plots and standard deviation. 134 Figure 22 CP - AFM transport measurements along WT and Y27A pili Correlations of resistance values and distance from the gold electrode were gene rated from measurements performed on WT (black) and Y27A mutant (green ) pili, respectively. Expanded view of WT resistance values (inset). 135 Figure 23 Electron transport rates per cell during growth of G. sulfurreducens with poorly crystalline Fe(III) oxides Calculated from the rates of Fe(II) production and cell growth reported previously 34 . 136 Figure 24 Biofilm electroactivity of phenylalanine vs. ala nine replacement in MECs fed with 3 mM acetate Current generation by WT (black), Tyr3 - F (blue), and Tyr3 (red) in MECs fed with 3 mM acetate. 137 Figure 25 Comparison of the reduction of external electron acceptors by Asp2 an d Tyr3 a) Current generation by WT (black), Asp2 (red), and Tyr3 (green) in MECs fed with 1 mM acetate. b) Fe(III) oxide reduction f or WT - SPEC, Asp2, and Tyr3 with (white) and without (back) the addition of 4 mM NTA. 138 Figure 26 NMR - derived PilA structural model of the G. sulfurreducens PilA and predicted structure of recombinant PilA 19 pilins Structure of a) PilA, b) PilA 19 and c) PilA 19 - A20C . All residues mutated for this study are labelled. 139 Figure 27 A GGRESCAN and Kyte - Doolittle analyses The Kyte - Doolittle hydropathy plot (black) was determined using a 3 amino acid window for the the PilA pilin precursor, which contains a hydrophilic N - t leader peptide. The AGGRESCAN plot for the PilA precursor is also shown (red). The straight lines indicate the aggregation (red) and hydrophobicity (black) cut - offs (at 0 and 1.5, respectively) used for the analyses. 140 Figure 28 Expression and purification of recombinant pilins carrying a 6 His - tag Western blot analysis using a) anti - PilA and b) anti - His antibodies. For gel A the soluble fraction of the untruncated PilA and the 10, 15, and 20 amino acid truncations were analyzed before ( - ) and after (+) induction. For gel B the soluble (S) and insoluble (I) fractions were analyzed before ( - ) and after (+) induction. 141 Figure 29 Western blot analysis of p eriplasmic (left) and cytoplasmic (right) proteins fractions during the expression and purification of MBP - PilA 20 Proteins in c rude extract s (A) before ( - ) and after (+) induction with IPTG, soluble protein fraction (B), and elution fraction from amylose column (C) were separated in a 1 2% glycine SDS - PAGE and blotted to a membrane for hyb ridization with anti - MBP antibodies . The migration of the periplasmic and cytoplasmic MBP - PilA 20 fusion protein is indicated with arrows . Reproduced from 166 with permission from Dr. Angelines Castro - Forero . 142 Figure 30 Cleavage efficiency of PilA 20 from its MBP tag by Factor Xa protease Elution fractions from an amylose column were incubated with Factor Xa prot ease at 4 o C and 2 3 ° C for 24 h (left) and 48 h (right) . Proteins in the s amples were analyzed by 7.5% glycine SDS - PAGE. Shown are proteins in g els from control s incubated in buffer without protease (A), and samples incubated with 0.5 g/mL (B) or 1.0 g/mL (C) of protease . Reproduced from 166 with permission from Dr. Angelines Castro - Forero . 143 Figure 31 Production of PilA 20 subunits after t reatment with Factor Xa protease MALDI - TOF MS analysis of a) sample untreated with protease used as a control, and b) sample treated with protease. MS results indicate two peptides are produced during cleavage. The peptide with MW of 4,527.4 Da corresponds to the PilA 20 subunit. c) N - terminal sequencing of the peptide with MW of 3,563.3 Da. Sequencing indicated that non - specific cleavage of the pilin subunit occurred. Reproduced from 166 with permission from Dr. Angelines Castro - Forero . 144 Figure 32 Expression of the full - length pilin (PilA) and various truncat ed pilins (PilA n ), all in fusions with a CBD tag (CBD - PilA n ) Expression was determined by analyzing cell pellet samples before (A - ) and after (A + ) induction using a 12% glycine SDS - PAGE. Modified from 166 with permission from Dr. Angelines Castro - Forero . 145 Figure 33 Presence of pilin monomers in elution from chitin column after cleavage from CBD tag determined by 16 - 20% Tricine SDS - PAG E Reproduced from 166 with permission from Dr. Angelines Castro - Forero . 146 Figure 34 Expression and purif ication of recombinant PilA 19 - A20C subunits a) 12% SDS - PAGE and Western blot analysis of cells before (A - ) and after (A + ) induction. Anti - chitin serum was used for the Western blot. b) MALDI - TOF mass spectrometry and c) 10 - 20% tricine gel of elution fracti on after purification . Reproduced from 166 with permission from Dr. Angelines Castro - Forero . 147 Figure 35 E ffect o f DTT on the electrochemical activity of PilA 19 - A20C pilin s on a gold electrode Shown in blue are c yclic voltammograms of gold electrodes incubated with PilA 19 - A20C to form a pilin monolayer in the presence (left) or absence (right) of DTT. Controls with pilin monolayers treated with 1 - undecanethiol SAM (red), bare gold (black) , and electrodes insulated with a 1 - undecanethiol SAM ( gray ) are also shown . Data were recorded at room temperature in 100 mM sodium phosphate buffer at pH 7.0 containing 100 mM NaC l, and 5 mM K 3 [Fe(CN) 6 ] at a potential scan rate of 100 mV s - 1 . 148 Figure 36 Effect of solvent ( acetonitrile ) in the d eposition of PilA 19 - A20C on a gold electrode Shown are c yclic voltammograms of gold electrodes incubated w ith either PilA 19 - A20C that has been buffer exchanged into acetonitrile (blue) or dialyzed into elution buffer (red) to form a pilin monolayer. Controls bare gold (black) , and electrodes insulated with a 1 - undecanethiol SAM ( gray ) are also shown . Data were recorded at room temperature in 100 mM sodium phosphate buffer at pH 7.0 containing100 mM NaCl, and 5 mM K 3 [Fe(CN) 6 ] at a potential scan rate of 100 mV s - 1 . 149 Figure 37 Effect of length of d eposition time of PilA 19 - A20C in acetonitrile on a gold electrode Recombinant pilins were incubated at room temperature in an anerobic glove bag for differing times; 0 h (gray, solid), 12 h (light blue, dash), 24 h (light blue, solid), 48 h (dark blue, dashes), and 120 h (dark blue, soli d). Electrodes were then washed with 100 % ethanol and incubated for 48 h with 1 mM 1 - undecanethiol. Data were recorded at room temperature in 100 mM sodium phosphate buffer at pH 7.0 containing 100 mM NaCl, and 5 mM K 3 [Fe(CN) 6 ] at a potential scan rate of 100 mV s - 1 . 150 Figure 38 Effect of mixing and age of pilin sample on deposition of PilA 19 - A20C in acetonitrile (a) Effect of mixing the PilA 19 - A20C solution: electrodes were rocked (blue) or stationary (black) and were depo sited for 48 h. (b) A freshly purified pilin sample (blue) and an old pilin sample (stored for 30 days at 4 °C, black) were deposited on gold electrodes for 48 h. The pilin - electrode interfaces were then washed with 100 % ethanol and incubated for 48 h wi th 1 mM 1 - undecanethiol to cover any exposed electrode areas. Current density was recorded at room temperature in 100 mM sodium phosphate buffer at pH 7.0 containing100 mM NaCl, and 5 mM K 3 [Fe(CN) 6 ] at a potential scan rate of 100 mV s - 1 . 151 Figure 39 Effect of concentration and temperature of pilin sample on deposition of PilA 19 - A20C in acetonitrile Pilins were deposited at a) the original concentration (blue) or a 1:10 dilution (black) or b) at 25 °C (black) or 4 °C for 48 h. Electrodes were then washed with 100 % ethanol and incubated for 48 h with 1 mM 1 - undecanethiol. Data was recorded at room temperature in 100 mM sodium phosphate buffer at pH 7.0 containing100 mM NaCl, and 5 mM K 3 [Fe(CN) 6 ] at a potential scan rate of 10 0 mV s - 1 . 152 Figure 40 CP - AFM of PilA 19 - A20C monolayers on gold electrodes Sample c urrent - voltage ( I - V ) plots obtained after probing the transversal conductivity of a PilA 19 - A20C SAM coated gold electrode (black) and bare go ld (yellow) at a force of 6 nN. 153 Figure 41 Effect of amino acid substitutions on cyclic voltammogramms Cyclic voltammograms of PilA 19 - A20C variants deposited for 48 h: Tyr3 (green), Asp2 (red), R28A (blue) compared to wil d - type (black). Electrodes were then washed with 100 % ethanol and incubated for 48 h with 1 mM 1 - undecanethiol. Data were recorded at room temperature in 100 mM sodium phosphate buffer at pH 7.0 containing100 mM NaCl, and 5 mM K 3 [Fe(CN) 6 ] at a potential s can rate of 100 mV s - 1 . 154 F igure 42 Test of media component limitation After growth with 4 mM acetate current decreases regardless of the addition of more electron donor (acetate) or any other media components. 155 Figure 43 Growth adaptation of pilB after extended incubation in MFC Current production (a) and CSLM micrographs of WT (b) and pilB (c) of MFCs after 10 media replacements (WT and pilB). CSLM micrographs are stained with the BacLig ht viability kit (green, live; red, dead). Scale bar, 20 µ m. 156 REFERENCES 157 REFERENCES 1. Röling, W. F. M. in The Prokaryotes 157 172 (2014). doi:10.1007/978 - 3 - 642 - 39044 - 9_381. 2. Aklujkar, M. et al. The genome sequence of Geobacter metallireducens : features of metabolism, physiology and regulation common and dissimilar to Geobacter sulfurreducens . BMC Microbiol. 9, 109 (2009). 3. Methé, B. A. et al. Genome of Geobacter sulfurreducens : Metal reduction in subsurface environments. Science 302, 1967 1969 (2003). 4. Tremblay, P. L., Aklujkar, M., Leang, C., Nevin, K. P. & Lovley, D. A genetic system for Geobacter metallireducens : Role of the flagellin and pilin in the reduction of Fe(III) oxide. Environ. Microbiol. Rep. 4, 82 88 (2012). 5. Coppi, M. V, Leang, C., Sandler, S. J. & Lovley, D. R. Development of a genetic system for Geobacter sulfurreducens . Appl. Environ. Microbiol. 67, 3180 3187 (2001). 6. Lovley, D. R. et al. Geobacter . The microbe applications. Adv. Microb. Physiol. 59, 1 100 (2011). 7. Lin, W. C., Coppi, M. V. & Lovley, D. R. Geobacter sulfurreducens can grow with oxygen as a terminal electron acceptor. Appl. Environ. Microbiol. 70, 25 25 2528 (2004). 8. Aklujkar, M. et al. The genome of Geobacter bemidjiensis , exemplar for the subsurface clade of Geobacter species that predominate in Fe(III) - reducing subsurface environments. BMC Genomics 11, 490 (2010). 9. Summers, Z. M. et al. Direct e xchange of electrons within aggregates of an evolved syntrophic coculture of anaerobic bacteria. Science 330, 1413 1415 (2010). 10. Morita, M. et al. Potential for direct interspecies electron transfer in methanogenic wastewater digester aggregates. MBio 2 , (2011). 11. Botton, S., Van Harmelen, M., Braster, M., Parsons, J. R. & Röling, W. F. M. Dominance of Geobacteraceae in BTX - degrading enrichments from an iron - reducing aquifer. FEMS Microbiol. Ecol. 62, 118 130 (2007). 12. Yan, J., Ritalahti, K. M., Wagn er, D. D. & Löffler, F. E. Unexpected specificity of interspecies cobamide transfer from Geobacter spp. to organohalide - respiring Dehalococcoides mccartyi strains. Appl. Environ. Microbiol. 78, 6630 6636 (2012). 13. Cologgi, D. L., Lampa - Pastirk, S., Speer s, A. M., Kelly, S. D. & Reguera, G. Extracellular reduction of uranium via Geobacter conductive pili as a protective cellular mechanism. Proc. Natl. Acad. Sci. U. S. A. 108, 15248 15252 (2011). 158 14. Bond, D. R. & Lovley, D. R. Electricity production by Geo bacter sulfurreducens attached to electrodes. Appl. Environ. Microbiol. 69, 1548 1555 (2003). 15. Cologgi, D. L., Speers, A. M., Bullard, B. A., Kelly, S. D. & Reguera, G. Enhanced uranium immobilization and reduction by Geobacter sulfurreducens biofilms. Appl. Environ. Microbiol. 80, 6638 46 (2014). 16. Speers, A. M. & Reguera, G. Consolidated bioprocessing of AFEX - pretreated corn stover to ethanol and hydrogen in a microbial electrolysis cell. Environ. Sci. Technol. 46, 7875 7881 (2012). 17. Speers, A. M. , Young, J. M. & Reguera, G. Fermentation of glycerol into ethanol in a microbial electrolysis cell driven by a customized consortium. Environ. Sci. Technol. 48, 6350 6358 (2014). 18. Weber, K. A., Achenbach, L. A. & Coates, J. D. Microorganisms pumping ir on: Anaerobic microbial iron oxidation and reduction. Nat. Rev. Microbiol. 4, 752 764 (2006). 19. Caccavo, F. et al. Geobacter sulfurreducens sp. nov., a hydrogen - and acetate - oxidizing dissimilatory metal - reducing microorganism. Appl. Environ. Microbiol. 60, 3752 3759 (1994). 20. Inoue, K. et al. Purification and characterization of OmcZ, an outer - surface, octaheme c - type cytochrome essential for optimal current production by Geobacter sulfurreducens . Appl. Environ. Microbiol. 76, 3999 4007 (2010). 21. Lea ng, C., Coppi, M. V & Lovley, D. R. OmcB, a c - type polyheme cytochrome, involved in Fe(III) reduction in Geobacter sulfurreducens . J. Bacteriol. 185, 2096 103 (2003). 22. Mehta, T., Coppi, M. V., Childers, S. E. & Lovley, D. R. Outer membrane c - type cytoch romes required for Fe(III) and Mn(IV) oxide reduction in Geobacter sulfurreducens . Appl. Environ. Microbiol. 71, 8634 8641 (2005). 23. Shelobolina, E. S. et al. Importance of c - type cytochromes for U(VI) reduction by Geobacter sulfurreducens . BMC Microbiol . 7, 16 (2007). 24. Qian, X. et al. Biochemical characterization of purified OmcS, a c - type cytochrome required for insoluble Fe(III) reduction in Geobacter sulfurreducens . Biochim. Biophys. Acta - Bioenerg. 1807, 404 412 (2011). 25. Butler, J. E., Kaufman n, F., Coppi, M. V., Núñez, C. & Lovley, D. R. MacA, a diheme c - type cytochrome involved in Fe(III) reduction by Geobacter sulfurreducens . J. Bacteriol. 186, 4042 4045 (2004). 26. Kim, B. C. & Lovley, D. R. Investigation of direct vs. indirect involvement of the c - type cytochrome MacA in Fe(III) reduction by Geobacter sulfurreducens . FEMS Microbiol. Lett. 286, 39 44 (2008). 159 27. Holmes, D. E. et al. Microarray and genetic analysis of electron transfer to electrodes in Geobacter sulfurreducens . Environ. Micro biol. 8, 1805 1815 (2006). 28. Levar, C. E., Chan, C. H., Mehta - Kolte, M. G. & Bond, D. R. An inner membrane cytochrome required only for reduction of high redox potential extracellular electron acceptors. MBio 5, e02034 (2014). 29. Schwertmann, U. & Corne ll, R. M. Iron oxides in the laboratory. (Wiley - VCH Verlag GmbH, 2000). doi:10.1002/9783527613229 30. Rollefson, J. B., Stephen, C. S., Tien, M. & Bond, D. R. Identification of an extracellular polysaccharide network essential for cytochrome anchoring and biofilm formation in Geobacter sulfurreducens . J. Bacteriol. 193, 1023 1033 (2011). 31. Reguera, G. et al. Biofilm and nanowire production leads to increased current in Geobacter sulfurreducens fuel cells. Appl. Environ. Microbiol. 72, 7345 7348 (2006). 32. Reguera, G., Pollina, R. B., Nicoll, J. S. & Lovley, D. R. Possible nonconductive role of Geobacter sulfurreducens pilus nanowires in biofilm formation. J. Bacteriol. 189, 2125 2127 (2007). 33. ty are necessary for Pseudomonas aeruginosa biofilm development. Mol. Microbiol. 30, 295 304 (1998). 34. Reguera, G. et al. Extracellular electron transfer via microbial nanowires. Nature 435, 1098 1101 (2005). 35. Torres, C. I. et al. A kinetic perspectiv e on extracellular electron transfer by anode - respiring bacteria. FEMS Microbiol. Rev. 34, 3 17 (2010). 36. Inoue, K. et al. Specific localization of the c - type cytochrome OmcZ at the anode surface in current - producing biofilms of Geobacter sulfurreducens . Environ. Microbiol. Rep. 3, 211 217 (2011). 37. Liu, Y., Kim, H., Franklin, R. R. & Bond, D. R. Linking spectral and electrochemical analysis to monitor c - type cytochrome redox status in living Geobacter sulfurreducens biofilms. ChemPhysChem 12, 2235 2241 (2011). 38. Liu, Y. & Bond, D. R. Long - distance electron transfer by G. sulfurreducens biofilms results in accumulation of reduced c - type cytochromes. ChemSusChem 5, 1047 1053 (2012). 39. Snider, R. M., Strycharz - Glaven, S. M., Tsoi, S. D., Erickson, J. S . & Tender, L. M. Long - range electron transport in Geobacter sulfurreducens biofilms is redox gradient - driven. Proc. Natl. Acad. Sci. 109, 15467 15472 (2012). 40. Renslow, R. et al. Metabolic spatial variability in electrode - respiring Geobacter sulfurreduc ens biofilms. Energy Environ. Sci. 6, 1827 1836 (2013). 160 41. Juárez, K. et al. PilR, a transcriptional regulator for pilin and other genes required for Fe(III) reduction in Geobacter sulfurreducens . J. Mol. Microbiol. Biotechnol. 16, 146 158 (2009). 42. Lea ng, C., Qian, X., Mester, T. & Lovley, D. R. Alignment of the c - type cytochrome OmcS along pili of Geobacter sulfurreducens . Appl. Environ. Microbiol. 76, 4080 4084 (2010). 43. Veazey, J. P., Reguera, G. & Tessmer, S. H. Electronic properties of conductive pili of the metal - reducing bacterium Geobacter sulfurreducens probed by scanning tunneling microscopy. Phys. Rev. E - Stat. Nonlinear, Soft Matter Phys. 84, (2011). 44. Vargas, M. et al. Aromatic amino acids required for pili conductivity and long - range e xtracellular electron transport in Geobacter sulfurreducens . MBio 4, (2013). 45. Feliciano, G. T. et al. Molecular and electronic structure of the peptide subunit of Geobacter sulfurreducens conductive pili from first principles. J. Phys. Chem. A 116, 8023 8030 (2012). 46. Boesen, T., Nielsen, L. P. & Nielsen, P. Molecular dissection of bacterial nanowires transport mechanism. MBio 4, (2013). 47. Maier, B., Potter, L., So, M., Seifert, H. S. & Sheetz, M. P. Single pilus motor forces exceed 100 pN. Proc. Nat l. Acad. Sci. USA 99, 16012 16017 (2002). 48. Craig, L. & Li, J. Type IV pili: paradoxes in form and function. Curr. Opin. Struct. Biol. 18, 267 277 (2008). 49. Craig, L. et al. Type IV pilin structure and assembly: X - ray and EM analyses of Vibrio cholerae toxin - coregulated pilus and Pseudomonas aeruginosa PAK pilin. Mol. Cell 11, 1139 1150 (2003). 50. Craig, L., Pique, M. E. & Tainer, J. A. Type IV pilus structure and bacterial pathogenicity. Nat. Rev. Microbiol. 2, 363 378 (2004). 51. Aas, F. E. et al. Su bstitutions in the N - terminal alpha helical spine of Neisseria gonorrhoeae pilin affect Type IV pilus assembly, dynamics and associated functions. Mol. Microbiol. 63, 69 85 (2007). 52. Strom, M. S. & Lory, S. Amino acid substitutions in pilin of Pseudomona s aeruginosa . Effect on leader peptide cleavage, amino - terminal methylation, and pilus assembly. J. Biol. Chem. 266, 1656 1664 (1991). 53. Burrows, L. L. Twitching motility: Type IV pili in action. Annu. Rev. Microbiol. (2012). doi:10.1146/annurev - micro - 09 2611 - 150055 54. Hartung, S. et al. Ultrahigh resolution and full - length pilin structures with insights for filament assembly, pathogenic functions, and vaccine potential. J. Biol. Chem. 286, 44254 44265 (2011). 161 55. Parge, H. E. et al. Biochemical purificat ion and crystallographic characterization of the fiber - forming protein pilin from Neisseria gonorrhoeae . J. Biol. Chem. 265, 2278 2285 (1990). 56. Hazes, B., Sastry, P. A., Hayakawa, K., Read, R. J. & Irvin, R. T. Crystal structure of Pseudomonas aeruginos a PAK pilin suggests a main - chain - dominated mode of receptor binding. J. Mol. Biol. (2000). doi:10.1006/jmbi.2000.3801 57. Audette, G. F., Irvin, R. T. & Hazes, B. Purification, crystallization and preliminary diffraction studies of the Pseudomonas aerugin osa strain K122 - 4 monomeric pilin. Acta Crystallogr. - Sect. D Biol. Crystallogr. 59, 1665 1667 (2003). 58. Reardon, P. N. & Mueller, K. T. Structure of the type IVa major pilin from the electrically conductive bacterial nanowires of Geobacter sulfurreduce ns . J. Biol. Chem. 288, 29260 29266 (2013). 59. Keizer, D. W. et al. Structure of a pilin monomer from Pseudomonas aeruginosa : Implications for the assembly of pili. J. Biol. Chem. 276, 24186 24193 (2001). 60. Nguyen, Y., Jackson, S. G., Aidoo, F., Junop, M. & Burrows, L. L. Structural characterization of novel Pseudomonas aeruginosa type IV pilins. J. Mol. Biol. 395, 491 503 (2010). 61. Pirbadian, S. et al. Shewanella oneidensis MR - 1 nanowires are outer membrane and periplasmic extensions of the extracellu lar electron transport components. Proc. Natl. Acad. Sci. 111, 1 6 (2014). 62. Parge, H. E. et al. Structure of the fibre - forming protein pilin at 2.6 Å resolution. Nature 378, 32 38 (1995). 63. Gorgel, M. et al. High - resolution structure of a type IV pili n from the metal - reducing bacterium Shewanella oneidensis . BMC Struct. Biol. 15, 4 (2015). 64. Piepenbrink, K. H. et al. Structure of Clostridium difficile PilJ exhibits unprecedented divergence from known type iv pilins. J. Biol. Chem. 289, 4334 4345 (201 4). 65. Forest, K. T., Dunham, S. A., Koomey, M. & Tainer, J. A. Crystallographic structure reveals phosphorylated pilin from Neisseria : Phosphoserine sites modify type IV pilus surface chemistry and fibre morphology. Mol. Microbiol. 31, 743 752 (1999). 66 . Craig, L. et al. Type IV pilus structure by cryo - electron microscopy and crystallography: Implications for pilus assembly and functions. Mol. Cell 23, 651 62 (2006). 67. Lauz, M., Eckhardt, S., Fromm, K. M. & Giese, B. The influence of dipole moments on the mechanism of electron transfer through helical peptides. Phys. Chem. Chem. Phys. 14, 13785 8 (2012). 162 68. Arikuma, Y., Nakayama, H., Morita, T. & Kimura, S. Ultra - long - range electron transfer through a self - assembled monolayer on gold composed of 120 Å - helices. Langmuir 27, 1530 1535 (2011). 69. Gastric, P. pilO , a gene required for glycosylation of Pseudomonas aeruginosa 1244 pilin. Microbiology 141, 1247 1254 (1995). 70. Virji, M. Post - translational modifications of meningococcal pili. Identific ation of common substituents: Glycans and alpha - glycerophosphate -- a review. Gene 192, 141 7 (1997). 71. Warren, M. J., Roddam, L. F., Power, P. M., Terry, T. D. & Jennings, M. P. Analysis of the role of pglI in pilin glycosylation of Neisseria meningitidis . FEMS Immunol. Med. Microbiol. 41, 43 50 (2004). 72. Smedley, J. G. et al. Influence of pilin glycosylation on Pseudomonas aeruginosa 1244 pilus function. Infect. Immun. 73, 7922 7931 (2005). 73. Chamot - Rooke, J. et al. Posttranslational modification of p ili upon cell contact triggers N. meningitidis dissemination. Science 331, 778 782 (2011). 74. Richter, L. V. Mutation al analysis of geopilin function in Geobacter sulfurreducens . Access (Doctoral dissertation). (2011). Retrieved from ProQuest Dissertation s and Theses. (Accession Order No. [ 3465075] ) . 75. Stimson, E. et al. Discovery of a novel protein modification: Alpha - glycerophosphate is a substituent of meningococcal pilin. Biochem. J. 316, 29 33 (1996). 76. Merroun, M. L. et al. Complexation of uraniu m by cells and S - layer sheets of Bacillus sphaericus JG - A12. Appl. Environ. Microbiol. 71, 5532 5543 (2005). 77. Malvankar, N. S. et al. Tunable metallic - like conductivity in microbial nanowire networks. Nat. Nanotechnol. 6, 573 579 (2011). 78. Strycharz - G and D. R. Lovley, Energy Environ. Sci., 2012, 5, DOI: 10.1039/c2ee02613a. Energy Environ. Sci. 5, 625 0 (2012). 79. Malvankar, N. S. et al. Structural basis for metallic - like conductivity in microbial nanowires. MBio 6, e00084 15 (2015). 80. - Chem. Sci. 3, 2191 2201 (2012). 81. Solomon, G. C., Herrmann, C., Vura - Weis, J., Wasielewski, M. R. & Ratner, M. a. The - stacked systems. J. Am. Chem. Soc. 132, 7887 7889 (2010). 163 82. Vura - Weis, J., Ratner, M. A. & Wasielewski, M. R. Geometry and electronic coupling in perylenediimid e stacks: Mapping structure - charge transport relationships. J. Am. Chem. Soc. 132, 1738 1739 (2010). 83. Bonanni, P. S., Massazza, D. & Busalmen, J. P. Stepping stones in the electron transport from cells to electrodes in Geobacter sulfurreducens biofilm s. Phys. Chem. Chem. Phys. 15, 10300 10306 (2013). 84. Gray, H. B. & Winkler, J. R. Electron tunneling through proteins. Q. Rev. Biophys. 36, 341 372 (2003). 85. Gray, H. B. & Winkler, J. R. Electron flow through proteins. Chem. Phys. Lett. 483, 1 9 (2009) . 86. Page, C. C., Moser, C. C., Chen, X. & Dutton, P. L. Natural engineering principles of electron tunnelling in biological oxidation - reduction. Nature 402, 47 52 (1999). 87. Sek, S., Misicka, A., Swiatek, K. & Maicka, E. Conductance of alpha - helical pep tides trapped within molecular junctions. J. Phys. Chem. B 110, 19671 19677 (2006). 88. Arikuma, Y., Nakayama, H., Morita, T. & Kimura, S. Electron hopping over 100 Å along an Angew. Chemie - Int. Ed. 49, 1800 1804 (2010). 89. Brooksby, P. A., Anderson, K. H., Downard, A. J. & Abell, A. D. Voltammetric and - peptide monolayers on gold. J. Phys. Chem. C 115, 7516 7 526 (2011). 90. Horsley, J. R., Yu, J., Moore, K. E., Shapter, J. G. & Abell, A. D. Unraveling the interplay of backbone rigidity and electron rich side - chains on electron transfer in peptides: the realization of tunable molecular wires. J. Am. Chem. Soc. 136, 12479 12488 (2014). 91. Zhong, D. Electron transfer mechanisms of DNA repair by photolyase. Annu. Rev. Phys. Chem. 66, 691 715 (2015). 92. Reece, S. Y. & Nocera, D. G. Proton - coupled electron transfer in biology: Results from synergistic studies in na tural and model systems. Annu. Rev. Biochem. 78, 673 699 (2009). 93. Migliore, A., Polizzi, N. F., Therien, M. J. & Beratan, D. N. Biochemistry and theory of proton - coupled electron transfer. Chem. Rev. 114, 3381 3465 (2014). 94. Ferreira, K. N., Iverson, T. M., Maghlaoui, K., Barber, J. & Iwata, S. Architecture of the photosynthetic oxygen - evolving center. Science 303, 1831 1838 (2004). 95. Hays, A. M. A., Vassiliev, I. R., Golbeck, J. H. & Debus, R. J. Role of D1 - His190 in proton - coupled electron transfer reactions in photosystem II: A chemical complementation study. Biochemistry 37, 11352 11365 (1998). 164 96. Stubbe, J., Nocera, D. G., Yee, C. S. & Chang, M. C. Y. Radical initiation in the class I ribonucleotide reductase: Long - range proton - coupled electron transfer? Chem. Rev. 103, 2167 2201 (2003). 97. Cordes, M. et al. Development of a model system for the study of long distance electron transfer in peptides. Adv. Synth. Catal. 350, 1053 1062 (2008). 98. Cordes, M. et al. Influence of amino acid side chain s on long - distance electron transfer in Angew. Chemie - Int. Ed. 47, 3461 3463 (2008). 99. Dileesh, S. & Gopidas, K. R. Photophysical and electron transfer studies of a stable carbocation. Chem. Phys. Lett. 330, 397 402 (2000). 100. Liang, N., Pielak, G. J., Mauk, A. G., Smith, M. & Hoffman, B. M. Yeast cytochrome c with phenylalanine or tyrosine at position 87 transfers electrons to (zinc cytochrome c peroxidase)+ at a rate ten thousand times that of the se rine - 87 or glycine - 87 variants. Proc. Natl. Acad. Sci. U. S. A. 84, 1249 1252 (1987). 101. Porter, T. D. Mutagenesis at a highly conserved phenylalanine in cytochrome P450 2E1 affects heme incorporation and catalytic activity. Biochemistry 33, 5942 5946 (1 994). 102. Wittekindt, C., Schwarz, M., Friedrich, T. & Koslowski, T. Aromatic amino acids as stepping stones in charge transfer in respiratory complex I: An unusual mechanism deduced from atomistic theory and bioinformatics. J. Am. Chem. Soc. 131, 8134 81 40 (2009). 103. Cordes, M. How d o amino acids transport electrons through peptides? (Doctoral dissertation). (2008). Retrieved from http://edoc.unibas.ch/754/1/DissB_8284.pdf. 104. Chen, X. et al. Aromatic residues regulating electron relay ability of S - co ntaining amino - electron bonds in proteins. J. Phys. Chem. C 116, 19682 19688 (2012). 105. Chen, X. et al. Relay stations for electron hole migration in peptides: Possibility for formation of three - electron bonds along peptide chains. J. Phys. Chem. B 112, 14302 14311 (2008). 106. Sun, W. et al. - electron bonds. J. Phys. Chem. C 117, 18325 18333 (2013). 107. Chen, X. et al. - Helix c - terminus acting as a relay to mediate long - range hole migration in proteins. J. Phys. Chem. Lett. 1, 1637 1641 (2010). 108. - helices. Int. J. Quantum Chem. 79, 120 124 (2000). 165 109. Gao, J. et al. Electron transfer in peptides: The influence of charged amino acids. Angew. Chemie - Int. Ed. 50, 1926 1930 (2011). 110. Nakayama, H., Morita, T. & Kimura, S. Dipole effects on molecular and electronic structures in a novel conjugate of oligo(phenyleneethynylene) and helic al peptide. Phys. Chem. Chem. Phys. 11, 3967 3976 (2009). 111. Yu, J., Horsley, J. R. & Abell, A. D. The influence of secondary structure on electron transfer in peptides. Aust. J. Chem. 66, 848 851 (2013). 112. Polo, F., Antonello, S., Formaggio, F., Toni olo, C. & Maran, F. Evidence against the hopping mechanism as an important electron transfer pathway for conformationally constrained oligopeptides. J. Am. Chem. Soc. 127, 492 493 (2005). 113. Dey, S. K. et al. Study of electron transfer in ferrocene - label ed collagen - like peptides. Langmuir 23, 6475 6477 (2007). 114. Yu, J. et al. Electro - peptides attached to vertically aligned carbon nanotube arrays: A mechanistic transition. Chem. Commun. 48, 1132 1134 (2012). 115. Arikuma, Y., Takeda, K., Morita, T., Ohmae, M. & Kimura, S. Linker effects on monolayer formation and lo ng - range electron transfer in helical peptide monolayers. J. Phys. Chem. B 113, 6256 6266 (2009). 116. Hatcher, E., Balaeff, A., Keinan, S., Venkatramani, R. & Beratan, D. N. PNA versus DNA: Effects of structural fluctuations on electronic structure and ho le - transport mechanisms. J. Am. Chem. Soc. 130, 11752 11761 (2008). 117. Sleutels, T. H. J. A., Ter Heijne, A., Buisman, C. J. N. & Hamelers, H. V. M. Bioelectrochemical systems: An outlook for practical applications. ChemSusChem 5, 1012 1019 (2012). 118. Borole, A. P. et al. Electroactive biofilms: Current status and future research needs. Energy Environ. Sci. 4, 4813 4834 (2011). 119. Nevin, K. P. et al. Anode biofilm transcriptomics reveals outer surface components essential for high density current prod uction in Geobacter sulfurreducens fuel cells. PLoS One 4, e5628 (2009). 120. Richter, H. et al. Cyclic voltammetry of biofilms of wild type and mutant Geobacter sulfurreducens on fuel cell anodes indicates possible roles of OmcB, OmcZ, type IV pili, and p rotons in extracellular electron transfer. Energy Environ. Sci. 2, 506 516 (2009). 121. Dalton, E. F. et al. Charge transport in electroactive polymers consisting of fixed molecular redox sites. Chem. Phys. 141, 143 157 (1990). 122. Speers, A. M. & Reguera , G. Electron donors supporting growth and electroactivity of Geobacter sulfurreducens anode biofilms. Appl. Environ. Microbiol. 78, 437 444 (2012). 166 123. Marx, C. J. & Lidstrom, M. E. Broad - host - range cre - lox system for antibiotic marker recycling in Gram - negative bacteria. Biotechniques 33, 1062 1067 (2002). 124. Kim, B. C. et al. OmcF, a putative c - type monoheme outer membrane cytochrome required for the expression of other outer membrane cytochromes in Geobacter sulfurreducens . J. Bacteriol. 187, 4505 45 13 (2005). 125. McKinlay, J. B., Zeikus, J. G. & Vieille, C. Insights into Actinobacillus succinogenes fermentative metabolism in a chemically defined growth medium. Appl. Environ. Microbiol. 71, 6651 6656 (2005). 126. Heydorn, a et al. Quantification of b iofilm structures by the novel computer program COMSTAT. Microbiology 146 ( Pt 1, 2395 2407 (2000). 127. Schmittgen, T. D. & Livak, K. J. Analyzing real - time PCR data by the comparative C(T) method. Nat. Protoc. 3, 1101 1108 (2008). 128. Senko, J. M. & Sto lz, J. F. Evidence for iron - dependent nitrate respiration in the dissimilatory iron - reducing bacterium Geobacter metallireducens . Appl. Environ. Microbiol. 67, 3750 3752 (2001). 129. Takaichi, S. & Morita, S. Procedures and conditions for application of th e pyridine hemochrome method to photosynthetically grown cells of Rhodopseudomonas sphaeroides . J. Biochem. 89, 1513 1519 (1981). 130. Thomas, P. E., Ryan, D. & Levin, W. An improved staining procedure for the detection of the peroxidase activity of cytoch rome P - 450 on sodium dodecyl sulfate polyacrylamide gels. Anal. Biochem. 75, 168 176 (1976). 131. Burrows, L. L. Pseudomonas aeruginosa twitching motility: Type IV pili in action. Annu. Rev. Microbiol. 66, 493 520 (2012). 132. Hentzer, M. et al. Alginate o verproduction affects Pseudomonas aeruginosa biofilm structure and function. J. Bacteriol. 183, 5395 5401 (2001). 133. Strycharz - Glaven, S. M., Snider, R. M., Guiseppi - Elie, A. & Tender, L. M. On the electrical conductivity of microbial nanowires and biofi lms. Energy Environ. Sci. 4, 4366 4379 (2011). 134. Babauta, J. T., Nguyen, H. D., Harrington, T. D., Renslow, R. & Beyenal, H. pH, redox potential and local biofilm potential microenvironments within Geobacter sulfurreducens biofilms and their roles in el ectron transfer. Biotechnol. Bioeng. 109, 2651 2662 (2012). 135. Robuschi, L. et al. Spectroscopic slicing to reveal internal redox gradients in electricity - producing biofilms. Angew. Chem. Int. Ed. Engl. 52, 925 928 (2013). 167 136. Marsili, E., Sun, J. & Bon d, D. R. Voltammetry and growth physiology of Geobacter sulfurreducens biofilms as a function of growth stage and imposed electrode potential. Electroanalysis 22, 865 874 (2010). 137. Feliciano, G. T., Steidl, R. J. & Reguera, G. Structural and functional insights into the conductive pili of Geobacter sulfurreducens revealed in molecular dynamics simulations. Phys. Chem. Chem. Phys. 17, 22217 22226 (2015). 138. Yang, Z., Lux, R., Hu, W., Hu, C. & Shi, W. PilA localization affects extracellular polysaccharid e production and fruiting body formation in Myxococcus xanthus . Mol. Microbiol. 76, 1500 1513 (2010). 139. Withers, R. T., Damron, H. F., Yin, Y. & Yu, H. D. Truncation of type IV pilin induces mucoidy in Pseudomonas aeruginosa strain PAO579. Microbiologyo pen 2, 459 470 (2013). 140. Lu, H. M., Motley, S. T. & Lory, S. Interactions of the components of the general secretion pathway: Role of Pseudomonas aeruginosa type IV pilin subunits in complex formation and extracellular protein secretion. Mol. Microbiol. 25, 247 259 (1997). 141. Strom, M. S., Nunn, D. & Lory, S. Multiple roles of the pilus biogenesis protein PilD: involvement of PilD in excretion of enzymes from Pseudomonas aeruginosa . J. Bacteriol. 173, 1175 1180 (1991). 142. Richter, L. V, Sandler, S. J . & Weis, R. M. Two isoforms of Geobacter sulfurreducens PilA have distinct roles in pilus biogenesis, cytochrome localization, extracellular electron transfer, and biofilm formation. J. Bacteriol. 194, 2551 2563 (2012). 143. Esteve - Núñez, A., Sosnik, J., Visconti, P. & Lovley, D. R. Fluorescent properties of c - type cytochromes reveal their potential role as an extracytoplasmic electron sink in Geobacter sulfurreducens . Environ. Microbiol. 10, 497 505 (2008). 144. Chin, K., Esteve - Núñez, A., Leang, C. & Lov ley, D. R. Direct correlation between rates of anaerobic respiration and levels of mRNA for key respiratory genes in Geobacter sulfurreducens . Appl. Environ. Microbiol. 70, 5183 5189 (2004). 145. Stephen, C. S., LaBelle, E. V, Brantley, S. L. & Bond, D. R. Abundance of the multiheme c - type cytochrome OmcB increases in outer biofilm layers of electrode - grown Geobacter sulfurreducens . PLoS One 9, e104336 (2014). 146. Bonanni, P. S., Bradley, D. F., Schrott, G. D. & Busalmen, J. P. Limitations for current prod uction in Geobacter sulfurreducens biofilms. ChemSusChem 6, 711 720 (2013). 147. Franks, A. E. et al. Novel strategy for three - dimensional real - time imaging of microbial fuel cell communities: monitoring the inhibitory effects of proton accumulation within the anode biofilm. Energy Environ. Sci. 2, 113 (2009). 168 148. Carra, C., Iordanova, N. & Hammes - Schiffer, S. Proton - coupled electron transfer in a model for tyrosine oxidation in photosystem II. J. Am. Chem. Soc. 125, 10429 10436 (2003). 149. Okegbe, C., Pr ice - Whelan, A. & Dietrich, L. E. P. Redox - driven regulation of microbial community morphogenesis. Curr. Opin. Microbiol. 18, 39 45 (2014). 150. - Stacking interactions. Alive and well in p roteins. J. Biol. Chem. 273, 15458 15463 (1998). 151. Cordes, M. & Giese, B. Electron transfer in peptides and proteins. Chem. Soc. Rev. 38, 892 901 (2009). 152. Giese, B., Graber, M. & Cordes, M. Electron transfer in peptides and proteins. Curr. Opin. Che m. Biol. 12, 755 759 (2008). 153. Shih, C. et al. Tryptophan - accelerated electron flow through proteins. Science 320, 1760 1762 (2008). 154. Wang, M., Gao, J., Müller, P. & Giese, B. Electron transfer in peptides with cysteine and methionine as relay amino acids. Angew. Chemie - Int. Ed. 48, 4232 4234 (2009). 155. Nugent, J. H. Oxygenic photosynthesis. Electron transfer in photosystem I and photosystem II. Eur. J. Biochem. 237, 519 531 (1996). 156. Yu, J., Huang, D. M., Shapter, J. G. & Abell, A. D. Electro chemical and computational - peptides. J. Phys. Chem. C 116, 26608 26617 (2012). 157. Okamoto, S., Morita, T. & Kimura, S. Electron transfer through a self - assembled monolayer of a double - helix pe ptide with linking the terminals by ferrocene. Langmuir 25, 3297 3304 (2009). 158. Yu, J., Horsley, J. R., Moore, K. E., Shapter, J. G. & Abell, A. D. The effect of a macrocyclic constraint on electron transfer in helical peptides: a step towards tunable m olecular wires. Chem. Commun. (Camb). 50, 1652 4 (2014). 159. Reguera, G., Tessmer, S., Veazey, J. & Lampa - Pastirk, S. Microbial nanowires and products rel ated thereto. (2012 ). U.S. Patent Application Serial No. 13/221,495, filed August 30, 2011, claiming priority to U.S. Provisional Patent Application Serial No. 61/378,240 filed on August 30, 2010. 160. Benavides - Garcia, M. G. & Balasubramanian, K. Structural insights into the binding of uranyl with human serum protein apotransferrin structure and spectra of protein - uranyl interactions. Chem. Res. Toxicol. 22, 1613 1621 (2009). 161. Yu, J. - Y., Chung, S. - W. & Heath, J. R. Silicon nanowires: Preparation, device fabrication, and transport properties. J. Phys. Chem. B 104, 11864 11870 (2000). 169 162. Barkleit, A., Moll, H. & Bernhard, G. Complexation of uranium(VI) with peptidoglycan. Dalton Trans. 5379 5385 (2009). doi:10.1039/b818702a 163. Reguera, G. Electron transfer at the cell - uranium interface in Geobacter spp . Biochem. Soc. Trans. 40, 1227 32 (2012). 164. R eguera, G. & Steidl, R. Microbial nanowires. (2014). U.S. Patent Application Serial No. 13/221,495, filed August 30, 2011,claiming priority to U.S. Provisional Patent Application Serial No. 61/378,188, filed on August 30, 2010. 165. Shin, Y. G. K., Newton, M. D. & Isied, S. S. Distance dependence of electron transfer across peptides with different secondary structures: The role of peptide energetics and electronic coupling. J. Am. Chem. Soc. 125, 3722 3732 (2003). 166. Castro Forero, A. Properties and appli cations of self - assembled biomolecules in nanostructured biomimetic inte rfaces ( Doctoral dissertation). (2011). Retrieved from ProQuest Dissertations and Theses. (Accession Order No. [ 3489777] ) . 167. Conchillo - Solé, O. et al. AGGRESCAN: A server for the pr BMC Bioinformatics 8, 65 (2007). 168. Kyte, J. & Doolittle, R. F. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105 132 (1982). 169. Rosano, G. L. & Ceccarelli, E. A. Recombinant protein expression in Escherichia coli : Advances and challenges. Front. Microbiol. 5, 177 202 (2014). 170. Schägger, H. Tricine - SDS - PAGE. Nat. Protoc. 1, 16 22 (2006). 171. Sasaki, Y. C. et al. Two - dimensional arrangem ent of a functional protein by cysteine - gold interaction: enzyme activity and characterization of a protein monolayer on a gold substrate. Biophys. J. 72, 1842 1848 (1997). 172. Sek, S., Swiatek, K. & Misicka, A. Electrical behavior of molecular junctions incorporating alpha - helical peptide. J. Phys. Chem. B 109, 23121 23124 (2005). 173. - helical peptide monolayers: Importance of interchain coupling. Langmuir 28, 17287 17294 (2012). 174. Agonafer, D. D., Chainani, E., Oruc, M. E., Lee, K. S. & Shannon, M. A. Study of insulating properties of alkanethiol self - assembled monolayers formed under prolonged incubation using electrochemical impedance spectroscopy. J. Nanotechnol. Eng. Med. 3, 031006 (2013). 175. Sek, S., Tolak, A., Misicka, A., Palys, B. & Bilewic z, R. Asymmetry of electron transmission through monolayers of helical polyalanine adsorbed on gold surfaces. J. Phys. Chem. B 109, 18433 18438 (2005). 170 176. Uji, H., Morita, T. & Kimura, S. Molecular direction dependence of single - molecule conductance of a helical peptide in molecular junction. Phys. Chem. Chem. Phys. 15, 757 760 (2013). 177. Klimes, A. et al. Production of pilus - like filaments in Geobacter sulfurreducens in the absence of the type IV pilin protein PilA. FEMS Microbiol. Lett. 310, 62 68 (20 10). 178. Smith, J. A. et al. Going wireless: Fe(III) oxide reduction without pili by Geobacter sulfurreducens strain JS - 1. Appl. Environ. Microbiol. 80, 4331 4340 (2014). 179. Aas, F. E. et al. Neisseria gonorrhoeae type IV pili undergo multisite, hierarc hical modifications with phosphoethanolamine and phosphocholine requiring an enzyme structurally related to lipopolysaccharide phosphoethanolamine transferases. J. Biol. Chem. 281, 27712 27723 (2006). 180. Chang, M. C. Y., Yee, C. S., Stubbe, J. & Nocera, D. G. Turning on ribonucleotide reductase by light - initiated amino acid radical generation. Proc. Natl. Acad. Sci. U. S. A. 101, 6882 6887 (2004). 181. Lukacs, A., Eker, A. P. M., Byrdin, M., Brettel, K. & Vos, M. H. Electron hopping through the 15 Å tripl e tryptophan molecular wire in DNA photolyase occurs within 30 ps. J. Am. Chem. Soc. 130, 14394 14395 (2008). 182. Irebo, T., Reece, S. Y., Sjodin, M., Nocera, D. G. & Hammarstrom, L. Proton - coupled electron transfer of tyrosine oxidation: Buffer dependenc e and parallel mechanisms. J. Am. Chem. Soc 129, 15462 15464 (2007). 183. Arikuma, Y., Nakayama, H., Morita, T. & Kimura, S. Ultra - long - range electron transfer through a self - assembled monolayer on gold composed of 120 - Å - - helices. Langmuir 27, 1530 1 535 (2011). 184. Arikuma, Y., Takeda, K., Morita, T., Ohmae, M. & Kimura, S. Linker effects on monolayer formation and long - range electron transfer in helical peptide monolayers. J. Phys. Chem. B 113, 6256 6266 (2009). 185. Guex, N. & Peitsch, M. C. SWISS - MODEL and the Swiss - PdbViewer: An environment for comparative protein modeling. Electrophoresis 18, 2714 2723 (1997).