<%BANNER%>

Metabolic Engineering of Escherichia coli ATCC 8739 for Production of Bioelectricity

Permanent Link: http://ufdc.ufl.edu/UFE0041113/00001

Material Information

Title: Metabolic Engineering of Escherichia coli ATCC 8739 for Production of Bioelectricity
Physical Description: 1 online resource (143 p.)
Language: english
Creator: Moore, Jonathan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: atp, escherichia, metabolic, microbial, redox, tricarboxylic
Microbiology and Cell Science -- Dissertations, Academic -- UF
Genre: Microbiology and Cell Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Escherichia coli ATCC 8739 was genetically engineered to increase electric current and Coulombic yield from glucose in a microbial fuel cell (MFC). Initial testing of strains was done using aerobic batch cultures. Deletion of genes for the ATP synthase stator stalk (atpFH) resulted in a 44% increase in rate of glucose utilization. With the higher rate of glucose catabolism, NADH reoxidation by the electron transport system (ETS) became limiting, despite increased levels of ETS proteins. The change in redox ratio led to acetate production as part of an overflow metabolism. To increase the rate of NADH oxidation and to minimize acetate production, ackA (encoding acetate kinase) was replaced by naoX from Streptococcus mutans, encoding a cytoplasmic, water-forming NADH oxidase. The arcA gene was deleted to remove redox sensitive controls. These modifications enabled the complete oxidation of glucose in aerobic cultures. In a poised potential MFC with a chemical mediator, the maximum current produced by the engineered strain JC93 was 19% higher than the wild type, with a Coulombic efficiency (percent theoretical yield of electrons) of 76% (versus 49% for wild type). The arcA deletion did not significantly affect current or Coulombic efficiency. However, deletion of arcA in an naoX^+ background increased flux through the tricarboxylic acid (TCA) cycle during aerobic glucose metabolism. Genes (omcA and mtrCAB) encoding c-type cytochromes of the Shewanella oneidensis MR-1 extracellular ETS were integrated into the chromosome of the engineered strain. Heterologous expression of these genes was investigated to allow mediatorless MFC operation and to establish the minimal sufficient set for extracellular electron transfer. Current production by the omcA-mtrCAB integration strains was low and further study is required. Successful optimization of the MFC biocatalyst is likely to shift the rate limiting steps back to the design of the MFC. An enhanced anode was designed to increase active surface area and decrease concentration overpotentials. The design incorporated single-wall carbon nanotubes and mixed-length carbon fibers and resulted in up to 3-fold higher average MFC current output. The practical adoption of MFC technology requires higher power densities, which can be achieved by combining MFC design optimization with biocatalyst improvement.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jonathan Moore.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Ingram, Lonnie O.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041113:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041113/00001

Material Information

Title: Metabolic Engineering of Escherichia coli ATCC 8739 for Production of Bioelectricity
Physical Description: 1 online resource (143 p.)
Language: english
Creator: Moore, Jonathan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: atp, escherichia, metabolic, microbial, redox, tricarboxylic
Microbiology and Cell Science -- Dissertations, Academic -- UF
Genre: Microbiology and Cell Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Escherichia coli ATCC 8739 was genetically engineered to increase electric current and Coulombic yield from glucose in a microbial fuel cell (MFC). Initial testing of strains was done using aerobic batch cultures. Deletion of genes for the ATP synthase stator stalk (atpFH) resulted in a 44% increase in rate of glucose utilization. With the higher rate of glucose catabolism, NADH reoxidation by the electron transport system (ETS) became limiting, despite increased levels of ETS proteins. The change in redox ratio led to acetate production as part of an overflow metabolism. To increase the rate of NADH oxidation and to minimize acetate production, ackA (encoding acetate kinase) was replaced by naoX from Streptococcus mutans, encoding a cytoplasmic, water-forming NADH oxidase. The arcA gene was deleted to remove redox sensitive controls. These modifications enabled the complete oxidation of glucose in aerobic cultures. In a poised potential MFC with a chemical mediator, the maximum current produced by the engineered strain JC93 was 19% higher than the wild type, with a Coulombic efficiency (percent theoretical yield of electrons) of 76% (versus 49% for wild type). The arcA deletion did not significantly affect current or Coulombic efficiency. However, deletion of arcA in an naoX^+ background increased flux through the tricarboxylic acid (TCA) cycle during aerobic glucose metabolism. Genes (omcA and mtrCAB) encoding c-type cytochromes of the Shewanella oneidensis MR-1 extracellular ETS were integrated into the chromosome of the engineered strain. Heterologous expression of these genes was investigated to allow mediatorless MFC operation and to establish the minimal sufficient set for extracellular electron transfer. Current production by the omcA-mtrCAB integration strains was low and further study is required. Successful optimization of the MFC biocatalyst is likely to shift the rate limiting steps back to the design of the MFC. An enhanced anode was designed to increase active surface area and decrease concentration overpotentials. The design incorporated single-wall carbon nanotubes and mixed-length carbon fibers and resulted in up to 3-fold higher average MFC current output. The practical adoption of MFC technology requires higher power densities, which can be achieved by combining MFC design optimization with biocatalyst improvement.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jonathan Moore.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Ingram, Lonnie O.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041113:00001


This item has the following downloads:


Full Text

PAGE 1

1 METABOLIC ENGINEERING OF ESCHERICHIA COLI ATCC 8739 FOR PRODUCTION OF BIOELECTRICITY By JONATHAN C. MOORE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREME NTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Jonathan C. Moore

PAGE 3

3 To Negarre and Murv Moore

PAGE 4

4 ACKNOWLEDGMENTS I thank my family for their encouragement and support and Dr. Lonnie Ingram and the other members of my Supervisory Committee for their guidance. I also thank members of the Ingram Lab, past and present, for their helpful discussions. I am indebted to Dr. Xuan Wang for his helpful comments and suggestion during prep aration of this manuscript. I thank Dr. Andrew Rinzler and Dr. Zhuangchun Wu for providing the single walled carbon nanotubes and the micro scale carbon fibers for our study and for their collaboration on microbial fuel cell electrode optimization. I am gr ateful for the assistance of Ms. Marjorie Chow and of Dr. Matthew Humbard with the proteomic comparisons. I thank Dr. Donald Court of the National Cancer Institute for the gift of plasmid pEL04 and Dr. Linda Thony Meyer for the gift of pEC86. I thank Dr. R obert H. Fillingame of the University of Washington for the kind gift of antisera.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 8 LIST OF FIGURES ................................ ................................ ................................ ......................... 9 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................................ ... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 16 Bioelectricity ................................ ................................ ................................ ........................... 16 Microbial Fuel Cells (MFCs) ................................ ................................ .......................... 17 History of MFC Research ................................ ................................ ................................ 19 MFC Design and Applications ................................ ................................ ........................ 19 Dissimilatory Meta l Reducing Bacteria (DMRB) ................................ ................................ .. 26 Limitations of DMRB for Use in MFC ................................ ................................ ........... 26 Shewanella oneidensis MR 1 ................................ ................................ .......................... 28 S. oneidensis MR 1 central metabolism ................................ ................................ ... 28 S. oneidensis extracellular electron transfer ................................ ............................. 29 Esch erichia coli ................................ ................................ ................................ ...................... 29 E. coli Central Metabolism ................................ ................................ .............................. 30 ATP/ADP ratio ................................ ................................ ................................ ......... 31 C ellular redox balance ................................ ................................ .............................. 31 Electron Transport ................................ ................................ ................................ ........... 33 Engineered E. coli as a Clean Background Model for Dissimilatory Metal Reduction and as a Biocatalyst for Increased MFC Power Output ................................ ...................... 34 2 PHYSIOLOGY AND METABOLISM OF AN E COLI ATCC 8739 STRAIN LACKING THE STATOR STALK OF THE ATP SYNTHASE ................................ .......... 39 Introduction ................................ ................................ ................................ ............................. 39 Materials and Methods ................................ ................................ ................................ ........... 41 Growth of Cultures and Media ................................ ................................ ........................ 41 Anaerobic Batch Fermentation ................................ ................................ ........................ 41 Construction of ATP Synthase Gene Deletions ................................ .............................. 41 Dye Reduction Assay ................................ ................................ ................................ ...... 43 Enzyme Assays ................................ ................................ ................................ ................ 43 SDS Electrophoresis and Immunoblotting ................................ ................................ ...... 44 Analyses ................................ ................................ ................................ .......................... 45 Results ................................ ................................ ................................ ................................ ..... 45

PAGE 6

6 Construction of JC25 and JC27 Through Targeted Deletions in atp Operon ................. 4 5 Comparison of Aerobic Growth Rates and Yields ................................ .......................... 46 Product Formation and Specific Rates of Glucose Consumption ................................ ... 47 Methylene Blue Reduction Rates ................................ ................................ .................... 47 Anaerobic vs. Aerobic Growth Rates and Metabolism ................................ ................... 48 NAD H Oxidase and NADH Dehydrogenase Activities ................................ .................. 49 Cytochrome Oxidase Activity ................................ ................................ ......................... 50 Discussion ................................ ................................ ................................ ............................... 50 3 DEVELOPMENT OF GENETIC TOOLS FOR STREAMLINING CHROMOSOMAL MODIFICATION OF E COLI ................................ ................................ .............................. 62 Introduction ................................ ................................ ................................ ............................. 62 Construction of Plasmids ................................ ................................ ................................ ........ 66 Construction of pLOI4162 Containing a cat sacB Cassette for Markerless Gene Deletions ................................ ................................ ................................ ...................... 66 Construction of pLOI4151 for Removal of FRT Sites ................................ .................... 67 pLOI4162 based Chromosomal Modification Procedure ................................ ...................... 67 Applications of pLOI4 162 Method ................................ ................................ ........................ 69 Applications of pLOI4151 ................................ ................................ ................................ ...... 69 4 METABOLIC ENGINEERING OF E COLI ATCC 8739 FOR INCREASED CURRENT AND COULOM BIC EFFICIENCY IN AN ELECTROCHEMICAL CELL .... 75 Introduction ................................ ................................ ................................ ............................. 75 Materials and Methods ................................ ................................ ................................ ........... 77 Growth Conditions and Media ................................ ................................ ........................ 77 Genetic Methods ................................ ................................ ................................ .............. 77 Electrochemical Analytical Methods ................................ ................................ .............. 79 NADH Oxidation Assay ................................ ................................ ................................ .. 80 Results ................................ ................................ ................................ ................................ ..... 81 Elimination of Competition for NADH ................................ ................................ ........... 81 Deletion of ackA ................................ ................................ ................................ .............. 82 Heterologous Expression of naoX Sm ................................ ................................ ................ 83 Aerobic Batch T ransfers in Pyruvate for Increased T ricarboxylic Acid (T CA ) Cycle Flux ................................ ................................ ................................ .............................. 85 Introduction of lpd101 for Lower Sensitivity of P yruvate Dehydrogenas e (P DH ) K etoglutarate Dehy d r ogenase ( K GDH ) Complexes to NADH ..................... 85 Deletion of arcA ................................ ................................ ................................ .............. 86 Comparison of Contribution of Individual Genetic Modifications to Complete Aerobic Glucose Oxidation ................................ ................................ .......................... 86 Current Production and Coulombic Efficiencies Using Engineered Strains in Poised Potential Electrochemical Cell ................................ ................................ ..................... 87 Discussion ................................ ................................ ................................ ............................... 87

PAGE 7

7 5 HETEROLOGOUS EXPRESSION OF GENES ENCODING EXTRACYTOPLASMIC CYTOCHROMES FROM DISSIMILATORY METAL REDUCING BACTERIA IN E COLI FOR EXTRACELLULAR ELECTRON TRANSFER ................................ ................................ ................................ .......................... 103 Introduction ................................ ................................ ................................ ........................... 103 Materials and Methods ................................ ................................ ................................ ......... 104 Growth Conditions and Media ................................ ................................ ...................... 104 Integration of Heterologous Genes and Replacement of gsp Promoters ....................... 105 SDS PAGE and Heme Staining ................................ ................................ .................... 106 Electroc hemical Analytical Methods ................................ ................................ ............ 107 Results and Discussion ................................ ................................ ................................ ......... 107 Chromosomal Integration of omcA So and mtrCAB So and Replacement of gsp Promoters in JC85 ................................ ................................ ................................ ...... 107 Production of Current by JC131 ................................ ................................ .................... 108 6 OPTIMIZATION OF ELECTRON TRANSFER TO THE ANODE IN A MICROBIAL FUEL CE LL USING SINGLE WALL CARBON NANOTUBES (SWNT) AND MIXED LENGTH CARBON FIBERS (CF) ................................ ................................ ....... 116 Introduction ................................ ................................ ................................ ........................... 116 Materials and Methods ................................ ................................ ................................ ......... 117 Materials, Instrumentation and MFC Assembly ................................ ........................... 117 Media and Growth of Cultures ................................ ................................ ...................... 118 SWNT MFC Anolyte Preparation and Inoculation ................................ ....................... 119 CF MFC Preparation ................................ ................................ ................................ ..... 119 Results ................................ ................................ ................................ ................................ ... 120 SWNT MFC Current Production ................................ ................................ ................... 120 CF MFC Current Production ................................ ................................ ......................... 120 Discussion ................................ ................................ ................................ ............................. 121 7 CONCLUSIONS AND FUTURE DIRECTIONS ................................ ............................... 127 REFERENCES ................................ ................................ ................................ ............................ 130 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 143

PAGE 8

8 LIST OF TABLES Table page 2 1 Strains, plasmids, and PCR primers ................................ ................................ ................... 53 2 2 Growth and glucose consumption of atp deletion strains ................................ .................. 54 2 3 Respiratory chain activities ................................ ................................ ................................ 55 3 1 Plasmids and PCR primers ................................ ................................ ................................ 71 4 1 Strains, plasmids, and PCR primers ................................ ................................ ................... 90 4 2 Coulombic efficiencies and maximum current from engineered strains in glucose fed bulk electrolysis cell (BEC) testing ................................ ................................ ................... 92 5 1 Strains, plasmids, and PCR primers ................................ ................................ ................. 110

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Example of a simple tw o chambered microbial fuel cel l ................................ .................. 36 1 2 Simple model of extracellular electron transfer in Shewanella oneidensis ...................... 37 1 3 Simplified view of central metabolism of aerobically cultured Esc herichia coli ATCC 8739 ................................ ................................ ................................ ........................ 38 2 1 Genetic organization and gene products of the atpIBEFHAGDC operon ........................ 56 2 2 Immunoblot of membrane fractions ................................ ................................ ................... 57 2 3 (A) Growth of aerobic cultures in glucose mineral salts medium. Product formation and glucose consumption by (B) ATCC 8739 (C) JC25 and (D) JC27 ............................ 58 2 4 Net electron transfer from glucose as determined by methylene blue (MB) reduction by aerobically grown whole cells ................................ ................................ ...................... 59 2 5 Growth rates of aerobic all y and anaerobic ally grown cultures ................................ ......... 60 2 6 Circuit model s for aerobically grown strains with glucose as carbon source .................... 61 3 1 A simple method for making chromo somal deletions or replacements ............................. 72 3 2 Plasmid map of pLOI4162 ................................ ................................ ................................ 73 3 3 Plasmid map of pLOI4151 ................................ ................................ ................................ 74 4 1 Photograph (A) and expanded schematic (B) of the bulk electrolysis cell used in this study for electrochemical measurements ................................ ................................ ........... 93 4 2 Summary of strain constructio n ................................ ................................ ......................... 94 4 3 Decreased pyruvate accumulation with overexpression of naoX from Streptococc us mutans ................................ ................................ ................................ ................................ 95 4 4 Partial restoration, through naoX Sm integration and pyruvate transfers of JC6 2 growth rate after deletion of ackA ................................ ................................ ................................ 96 4 5 Cytoplasmic NADH oxidase activity from aerobically grown strains with different naoX Sm integrations ................................ ................................ ................................ ............ 97 4 6 Aerobic batch culture transfers of JC68 a nd JC72 in pyruvate mineral salts medium for s horter generation time s ................................ ................................ ............................... 98

PAGE 10

10 4 7 Growth curves of aerobic glucose mineral salts medium cultures of isogenic set of strains with restoration of each chromoso mal modification present in JC8 5 .................... 99 4 8 Glucose utilization and product formation of aerobic cultures of isogenic set of strains with restoration of each chromosomal modification present in JC85. ................. 100 4 9 Current production over time with glucos e fe d cells in bulk electrolysis cell ................ 101 4 10 Simplified metabol ic overview of JC93 ................................ ................................ .......... 102 5 1 Arrangement of gsp (gene ral secretory pathway ) genes in representative strai ns ........... 112 5 2 A model for extracellular electron transfer by an engineered E. coli stra in producing periplasmic and outer me mbrane cyto chromes from Shewanella oneidensis MR 1 ....... 113 5 3 Heme stained crude memb rane fr actions of engineered strains ................................ ...... 114 5 4 Electrochemical cell (EC) anolytes (post run) from the first EC trial with JC131/pEC86 (A) and JC85/pEC86 (B) ................................ ................................ .......... 115 6 1 Initial hour (mixing) average current with and without addition of single walled ca rbon nanotubes (SWNT) ................................ ................................ .............................. 122 6 2 Average current with and without addition of single walled carbon nanotubes (SWNT) ................................ ................................ ................................ ............................ 123 6 3 Initial hour (post mixing) average current with and without addition of single wall ed carbon nanotubes (SWNT) ................................ ................................ .............................. 124 6 4 Average current with and without addition of micro sca le carbon fibers (CF). .............. 125 6 5 Second trial average current with and without addition of micro scale carbon fibers (CF) ................................ ................................ ................................ ................................ .. 126

PAGE 11

11 LIST OF ABBREVIATION S KGDH alpha ketoglutarate dehydrogenase ACK acetate kinase ADP adenosine diphosphate AMP adenosine monophosphate ArcA P phosphorylated response regulator of redox sensitive two component system ATP adenosine triphosphate BCA bicinchoninic acid BCIP bromo chloroindolyl phosphate BEC bulk electrolysis cell cAMP CRP cyclic AMP catabolism repressor protein C E Coulombic efficiency CF carbon fiber CoA coenzyme A DCM dry cell mass DMRB dissimilatory metal reducing bacteria DMSO dimethyl sulfoxide d NADH deamino n icotinamide adenine dinucleotide (reduced) EC electrochemical cell EMP Embden Meyerhoff Pathway ETS electron transport system FNR transcriptional regulator named for role in fumarate and nitrate reduction FRT FLP recombinase recognition target HGA homogent isic acid H NS histone like nucleoid structuring protein

PAGE 12

12 HPLC high pressure liquid chromatography KCN potassium cyanide MB methylene blue MFC microbial fuel cell MOPS 4 morpholinopropanesulphonic acid MtrA periplasmic cytochrome c MtrB outer membrane prote in associated with OmcA and MtrC MtrC outer membrane cytochrome c MV membrane vesicle NADH/NAD + nicotinamide adenine dinucleotide (reduced/oxidized) NapC cytoplasmic ( cytochrome c ) subunit of periplasmic nitrate reductase NBSM New Brunswick Scientific MOPS mineral salts medium NBT nitro blue tetrazolium NDH NADH dehydrogenase NOX NADH oxidase OM outer membrane OmcA outer membrane cytochrome c PCR polymerase chain reaction PDH pyruvate dehydrogenase PMF proton motive force PTA phosphotransacetylase PTFE poly tetrafluoroethylene PVDF polyvinylidene fluoride RVC reticulated vitreous carbon SDS PAGE sodium dodecyl sulf ate polyacrylamide gel electrophoresis

PAGE 13

13 SS stator stalk of the F 1 F O ATP synthase T2SS type II secretion system TCA tricarboxylic acid cycle TMAO trimethylamine N oxide TMPD N,N tetramethyl phenylenediamine v redMB specific rate of methylene blue reduction

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy METABOLIC ENGINEERING OF ESCHERICHIA COLI ATCC 8739 FOR PRODUCTION OF BIOELECTRICITY By Jonathan Moore December 2009 Chair: Lonnie O. Ingram Major: Microbiology and Cell Science Escherichia coli ATCC 8739 was genetically engineered to increase electri c current and Coulombic yield from glucose in a microbial fuel cell (MFC). Initial testing of strains was done using aerobic batch cultures. Deletion of genes for the ATP synthase stator stalk ( atpFH ) resulted in a 44% increase in rate of glucose utilizati on. With the higher rate of glucose catabolism, NADH reoxidation by the electron transport system (ETS) became limiting, despite increased levels of ETS proteins. The change in redox ratio led to acetate production as part of an overflow metabolism. To inc rease the rate of NADH oxidation and to minimize acetate production, ackA (encoding acetate kinase) was replaced by naoX from Streptococcus mutans encoding a cytoplasmic, water forming NADH oxidase. The arcA gene was deleted to remove redox sensitive cont rols. These modifications enabled the complete oxidation of glucose in aerobic cultures. In a poised potential MFC with a chemical mediator, the maximum current produced by the e ngineered strain JC93 was 19% higher than the wild type, with a Coulombic effi ciency (percent theoretical yield of electrons) of 76% (versus 49% for wild type). T he arcA deletion did not significantly affect current or Coulombic efficiency. However, deletion of arcA in an naoX Sm + background increased flux through the tricarboxylic a cid (TCA) cycle during aerobic gl ucose metabolism

PAGE 15

15 Genes ( omcA and mtrCAB ) encoding c type cytochromes of the Shewanella oneidensis MR 1 extracellular ETS were integrated into the chromosome of the engineered strain. Heterologous expression of these gene s was investigated to allow mediatorless MFC operation and to establish the minimal sufficient set for extracellular electron transfer. Current production by the omcA mtrCAB integration strains was low and further study is required. Successful optimizatio n of the MFC biocatalyst is likely to shift the rate limiting steps back to the design of the MFC. An enhanced anode was designed to increase active surface area and decrease concentration overpotentials. The design incorporated single wall carbon nanotube s and mixed length carbon fibers and resulted in up to 3 fold higher average MFC current output. The practical adoption of MFC technology requires higher power densities, which can be achieved by combining MFC design optimization with biocatalyst improveme nt.

PAGE 16

16 CHAPTER 1 INTRODUCTION Bioelectricity Electrical energy production from renewable feedstocks and organic wastes has great potential as a component of a comprehensive strategy to offset the use of fossil fuels. Microorganisms are capable of transfor ming the stored chemical energy in organic materials such as cellulosic biomass and animal wastes to useful electrical energy. The most direct and theoretically efficient strategy to harvest microbially derived electrical energy is the use of a microbial f uel cell (MFC). Other strategies include microbially catalyzed production of liquid biofuels, methane or hydrogen. All of these biofuels require additional steps to convert the stored energy to electricity, thus lowering the overall efficiency of the proce ss. Alternately, the stored energy in biofuels may be used to do work through combustion. A thorough lifecycle analysis of biofuel production and combustion for transportation shows this process to be less efficient than direct combustion of biomass to sup ply an electrical infrastructure (18) Most abiotic transformations of biomass to electricity involve combustion and other significant process losses. Even the promising route to bioelectricity from gasification combined with solid oxide fuel cells (105) while avoiding the inefficiency of combustion, requires significant energy inputs. In the context of a global energy economy that is shifting towards increased efficiency and sustainability, the preferred process is one that is able to convert solar energy and waste streams to electrical energy most directly. Most of the effort in MFC research has been focused upon MFC design optim ization, extracellular electron transfer and the microbial ecology of consortia colonizing the electrodes of wastewater fed MFCs. Substantial gains in MFC electrical power output have been made (1, 23, 141) However, few studies have reported attempts to complement these approaches with genetic

PAGE 17

17 engineering of the MFC biocatalysts. No natural environments have been studied that are similar to that of an MFC. An ideal MFC biocatalyst is capable of rapidly and completely oxidizing any fuel and of efficiently transferring the liberated electrons to an electrode. While certain organisms have evolved to do some of these things well, no known ecological niche appear s to select for all of the desired attributes. The greatest strength of genetic engineering is the ability to modify biocatalysts for tasks for which there has been no natural selection. Production of bioelectricity can only achieve its potential through m etabolic optimization of the biocatalysts, in conjunction with MFC design efforts. Microbial Fuel Cells Microbial fuel cells employ microorganisms to convert stored chemical energy into useful electrical energy. The electrons liberated from the oxidation o f organic material through the metabolic activity of these microbes are transferred to an anode in an MFC. The electrons flow through an external circuit across a load (such as a resistor, a motor, or a light) to the cathode, where they combine with an oxi dant in the balancing half cell reaction (Figure 1 1). Protons from the anodic reaction are conducted through the electrolyte and across a proton exchange membrane (or salt bridge, etc.) to the cathode, where they are reunited with the electrons in the cat hode oxidation reaction. Figure 1 1 depicts a simple, two chambered MFC using glucose as an electron donor. If the glucose is completely oxidized to CO 2 by the microbial biocatalyst in the anode chamber, the anode half cell reaction would be: C 6 H 12 O 6 + 6H 2 2 + 24H + +24e (1 1) If oxygen was used as the oxidant in the cathode half cell reaction (potassium ferricyanide is shown in the figure and is frequently used in laboratory experiments because of improved electrode kinetics) : 24e + 24H + + 6 O 2 2 O (1 2)

PAGE 18

18 The overall reaction in the MFC is the same as the one for the complete catabolism of glucose by a respiring cell: C 6 H 12 O 6 + 6O 2 2 + 6H 2 O (1 3) Figure 1 1 shows a mediator (a diffusible electron carrier) carrying electr ons from the bacterial inner membrane electron transport chain to the anode. MFCs reroute electrons from microbial metabolism through an external circuit to capture a portion of the stored energy in the fuel as electricity. In heterotrophic organisms, the oxidation of organic material is coupled to storage of the released energy in high group transfer potential compounds such as adenosine triphosphate (ATP). These compounds are subsequently used in cellular biosynthetic processes. This energy transformation involves the stepwise transfer of electrons up a reduction potential gradient to a terminal electron acceptor. This terminal electron acceptor can be an endogenous metabolite, as in fermentation, or an exogenous acceptor, in respiration. In an MFC, electr ons can be transferred from the respiratory chain to the anode (as the terminal electron acceptor), or fermentation products may be oxidized by the anode. When electrons are captured from the respiratory chain, the transfer can be direct or indirect (via a soluble mediator). The soluble mediators involved in the indirect electron transfer can be endogenously produced or exogenous, supplied compounds. Since the discovery of microbes that are capable of electron transfer to an electrode (electrode reducers) without requiring added mediators (59) most MFC studies have used mediatorless MFCs to avoid the expense and toxicity issues associated with the use of added mediators. Much of the research has focused on MFC design and the mechanisms of electron transfer. However, low power density is the greatest challenge for the widespread adoption of MFCs as a means of

PAGE 19

19 sustainable energy transformation (24, 36) The largest remaining gains in MFC power density are to be made through studying and e ngineering of microbial metabolism (88) to increase the rate of electron transfer (current) as well as the conversion efficiency from the fuel. History of MFC R esearch M. C. Potter is generally recognized as publishing the first study of the microbially catalyzed production of electricity from organic material in 1911 (96) In that study, he measured the electrical potent ial developed from the fermentation of sugars by Saccharomyces cerevisiae and by E. coli preliminary experiments provided a proof of concept, few subsequent MFC studies were publishe d for over half a century. In the 1960s, several MFC studies were published including an article describing the operation of a methane oxidizing bacterial fuel cell employing Psuedomonas methanica at the anode (122) In 1993, Allen and Bennetto studied redox dye mediated MFCs using various bacterial strains and conditions (3) Kim et al (59) revolutionized the field by describing a microbial fuel cell which used Shewanella putrefaciens (later changed to S. oneidensis ) MR 1 to oxidize lactate with transfer of electrons to the electrode without the need for any added mediator. Many othe r microbes were subsequently found to be capable of this extracellular electron transfer. These organisms have been termed exoelectrogens (73) There is extensive but not complete overlap of the exoelec trogen group with the dissimilatory metal reducers, microorganisms capable of energy coupling electron transfer to metal oxides and oxyhydroxides. MFC Design and Applications All fuel cells share the same fundamental components: an anode, to which electron s from the fuel are transferred, a cathode, where the oxidation reaction occurs, and an electrolyte, which conducts positively charged ions (cations) from the anode to the cathode for the completion of

PAGE 20

20 the circuit. Electrons flow through the external circu it across the load to the cathode, but do not flow through an electrically insulating electrolyte polymer or membrane. MFCs are generally categorized as one or two chambered. Biological fuel cells have aqueous anode chambers, whereas other fuel cells gener ally have solid electrolytes, sandwiched between the two electrodes. The liquid in the anode chamber (anolyte) of an MFC contains the nutrients required by the electron providing microbes. It is kept anaerobic to prevent direct electron transfer to oxygen, which would decrease MFC power output and efficiency. Furthermore, many of the microbes used in MFCs are obligate anaerobes. The fuel (electron donor) is added to the anolyte, which may be stirred and operated in batch mode, or operated in flowthrough mod e. Anaerobic conditions (and sometimes mixing) can be maintained by flowing gases such as argon or nitrogen/carbon dioxide into the anode chamber. A diffusible electron carrier (mediator) can be added to the anolyte, if necessary. A two chambered MFC (Figu re 1 1) has separate cathode chamber that is generally filled with an aqueous catholyte. The catholyte contains the oxidant. The oxidant can be oxygen, or it can be a surrogate such as the potassium ferricyanide (Figure 1 1), frequently used because of its favorable electrode kinetics. The two compartments and their contents may be separated by an ion permeable membrane, such as the proton permeable Nafion from DuPont, or simply by a salt bridge. Single chambered MFCs have an anode chamber and a direct air cathode. They commonly use a membrane electrode assembly, similar to those used in other types of fuel cells (68) The proton exchange membrane is laminated onto the cathode, which is exposed to the air outside t he assembly. Single chambered MFCs can also be membraneless. Membraneless MFCs are not separated from the anolyte by a membrane or a salt bridge. This design is made possible by the

PAGE 21

21 unique properties of MFCs that use microbes that form a biofilm on the ano de and do not require the addition of a mediator. Thus, electrons are primarily transferred to the anode and do not significantly decrease MFC efficiency by traveling to the cathode through the anolyte. MFC design has focused upon maximizing Coulombic eff iciency and power density. Coulombic efficiency (C E Eq. 1 4) is the percentage of total electrons in the fuel (theoretical Coulombic yield) that are recovered by the MFC (experimental Coulombic yield). C E = Coulombic yield (experimental)/ Coulombic yield (theoretical) (1 4) The theoretical Coulombic yield from 1 mol glucose is 24 mol electrons (or 2.32 10 6 C, based 5) F = 96,485 C mol 1 (1 5) Experimental Coulombic yield is calculated by measuring current produc ed over time by the MFC with a known quantity of fuel added. Integration of the area under the peak gives the total charge transferred to the electrode, which can be divided by the theoretical value to give the C E The C E of an MFC is never 100%, due to bi ological and non biological losses. Microbial biosynthesis and cellular maintenance reactions can require reduced cofactors and therefore decrease MFC Coulombic yields. Any incomplete microbial fuel catabolism resulting in fermentation or other side produc ts further decreases Coulombic yield. In addition, any diffusion of competing electron acceptors into the anode chamber (oxygen, nitrate, etc.) from the cathode chamber or from outside the MFC will result in a lower C E Therefore it is important to conside r both the metabolic pathways used by the microbial biocatalyst and the MFC design when attempting to optimize electron recovery. Total energy recovery by an MFC is ultimately more informative than C E because C E neglects the potential difference between anode and cathode. The total MFC energy recovery is

PAGE 22

22 the electrical energy produced as a percentage of the total energy content of the fuel (heat of combustion). MFC power integrated over time gives the MFC energy recovery, just as MFC current integrated ov er time provides the Coulombic yield. Power is the product of the potential difference (MFC voltage) and the current produced. As the primary focus of our study was on improvement of current production and of Coulombic yield through metabolic engineering o f the microbial biocatalyst, a poised potential electrochemical cell was used for most of the experiments. With this type of system, MFC voltage and power output (and by extension, total energy recovered) are not considered. As mentioned above, MFC power d ensity is a critical factor for useful application of MFC technology. Power densities of MFCs have frequently been reported as a function of electrode surface area (W/m 2 ). Many studies have reported current densities (mA/m 2 ). The lack of convention has mad e direct comparison of MFC designs difficult. Some MFC researchers have agreed upon volumetric power (W/m 3 ) as the most useful measure of MFC power density (21, 71) Volumetric power takes into account most of the relevant factors involved in MFC operation, but notably ignores C E and operational longevity. These are very important parameters, as the y constitute the primary advantages that MFC offer over enzymatic fuel cells, for example (along with lower cost). Enzymatic fuel cells (133) can achie ve higher power densities than existing MFCs, but are subject to fouling from products and are not capable of self repair and enzyme replenishment. Furthe rmore, enzymatic fuel cells canno t recover much of the energ y present in more complex fuels. For example, a glucose oxidase fuel cell (57) recovers only two electrons per glucose (maximum theoretical C E is 8.3%). While these additional factors must be considered, volumetric power is an important M FC operational metric by which to compare different MFCs for most applications.

PAGE 23

23 MFC power output is dependent upon the voltage (potential difference) across the load. The open circuit voltage is not simply the difference between the reduction potentials fo r the two half cell reactions. For example, in the glucose fed MFC with an oxygen cathode, the difference between the reduction potentials of the O 2 /H 2 O (Eq. 1 2) and glucose/CO 2 (Eq. 1 1) couples is approximately 1.2 V (0.805 V ( 0.414 V), both vs. norm al hydrogen electrode (116) ). However, the actual MFC voltage would be lower. First, the anode chamber redox potential is ultimately determined not by the fuel, but by the potential of the immediate electron donor to the anode. That is the mediator, or perhaps an outer membrane cytochrome in a mediatorless MFC. Next, there are several reasons for voltage losses (ohmic losses and overpotentials) in an MFC (21, 24) Polarization curves are routinel y used to evaluate MFC performance. Since voltage is inversely proportional to resistance, the external resistance can be changed to vary the voltage. The current is plotted as a function of voltage. A power curve can then be plotted based upon the the pol arization curve (and the relationship between power, voltage and current, Eq. 1 6). P = V I (1 6) These types of curves provide both the maximum power production and the effect of the MFC internal resistance. Current varies in a nonlinear manne r with voltage because of current dependent overpotentials. The relationship would be linear if there were no internal resistance (Eq. 1 7). I = V/ R (1 7) However, the total resistance is the sum of external and internal resistance and it is t he total resistance that must be used in the calculations. Internal resistance is dependent upon ohmic (conductive) losses as well as anodic and cathodic overpotentials Types of overpotentials include activation overpotentials (due to electron transfer k inetics), losses due to bacterial

PAGE 24

24 metabolism, and mass transfer losses (concentration overpotentials) due to substrate limitation of biocatalysis or at the electrode. Part of our study focused on overcoming anodic overpotentials due to metabolism and due t o activation overpotentials using metabolic engineering and improvement of electron transfer kinetics. Activation overpotentials are greatest at lower current, while concentration overpotentials become a factor when current increases (72, 138) Our study also investigated optimization of anode design to minimize ohmic losses and concentration overpotentials. There are many potential applications for MFC technology. MFCs offer unique benefits for niche appli cations such as power for remote sensing equipment, power and waste recycling for space travel, small scale distributed power generation in areas lacking infrastructure, and for use as biosensors. The larger scale applications that are currently being deve loped include MFCs for processing of wastewater and of food processing wastes. MFCs have been developed for use in providing electricity to low power sensing and communications equipment in remote locations. Benthic unattended generators (BUGs) are exampl es of sediment MFCs (74, 115) BUGs have an anode, buried in seafloo r sediment, that harvests electrons from the metabolic processes of the natural microbial ecosystem and couples the anodic half reaction to the oxidation of a cathode located in the oxygenated seawater above. One could also imagine MFCs powering telemetry equipment in a forest or jungle, using tree sap or other readily and continuously available fuel to allow long term, low cost, maintenance free operation. Exotic niche applications such as MFCs for space travel (83) and in foraging robotic systems (132) may be somewhat limited in scale, but these research efforts can help fund further MFC research for more widespread practical usage. Projects such as the gastrobot research effo rt

PAGE 25

25 (131) and the slugbot, an autonomous (carnivorous) pest contol sytem (58) both powered by MFCs, generated headlines and awareness of the possibilities for MFC technology. Another MFC application that is att racting attention is the use of very simple, inexpensive MFCs (9) in third world countries for small scale distributed power. Several non profit organizations are distributing plans and teaching the people in these areas to make these MFCs from cheap and locally available materials. Initial MFC applications that benefit from economies of scale include the treatment of industrial, agricultural, municipal and domestic sewage wastes. The organic material content of these wastes is currently release d directly into the environment, or pretreated before release. These processes involve significant costs, both financial and environmental. As an example, treatment of sewage wastewater (generally by aerobic bacteria in staged mixing tanks) in the United S tates costs over $25 billion annually (71) The energy in the organic wastes is not currently recovered. Removal of organic material via large scale flowthrough MFCs could offset energy (and therefore operational) costs through the recovery of this stored energy as electricity Progress is being made in MFC scale up design and operation. Recently, a 1 m 3 pilot MFC by and is operated by a group from the Advanced Water Management Centre at the University of Queensland, in cooperation with the brewery. A collaboration between the Center for Environmental Sciences and Engineering at the University of Connecticut and a private environmental engineering company has received funding to build and operate pilot scale wastewater treatment MFCs in New York. These types of projects mark a new phase for MFC research, which is still quite young compared with most other energy harvesting technologies.

PAGE 26

26 Dissimilatory Metal Reducing Bacteria The number of MFC studies increased dramatically after the discovery that some dissimilatory metal reducing bacteria were able to reduce an MFC anode (59) Some mi croorganisms are capable of transferring electrons to insoluble metal oxides and oxyhydroxides in an energy coupling manner (43, 75, 85) The use of extracellular metals as terminal electron acceptors to support growth is known as dissimilatory metal reduction. Bacteria that can respire on these metals are known as dissimilatory metal reducing bacteria (DMRB). Many DMRB can transfer electro ns to an MFC electrode (70) While som e archaea are capable of dissimilatory metal reduction (56) we will focus on dissimilatory metal reduction by eubacteria. Limitations of DMRB for Use in MFC Although DMRB possess desirable attributes for MFC application in that most are capable of reducing an electrode without requiring addition of an electron carrying mediator, these organisms have significant limitations. The DMRB can be present as isolates in MFCs, or more commonly, as members of a consortium in the anode containing chamber. DMRB such as Shewanella or Geobacter have a very limited substrate range and are incapable of using sugars as electron donors. MFCs using Shewanella are inefficient because they can only partially oxidize the fuel ( primarily lactate is oxidized to acetate) Other DMRB have very low metabolic rates and grow slowly. One example of a DMRB that is capable of completely oxidizing sugars and transferring electrons to an electrode with a high Coulombic yield is Rhodoferax ferrireducens (22) Despite its ability to efficiently transform the chemical energy of sugars into electrical energy and to transfer electrons directly without the need for a mediator, the doubling time for this organism i s more than a week and sugars are catabolized very slowly (22) This low catabolic rate translates to low current and power production. DMRB generally live in an

PAGE 27

27 environment of low substrate availability and the electron donors present are often inorganic compounds or breakdown products from the catabolism of more complex organic substances by other organisms. Therefore, high metabolic rates and fast growth would not seem to offer a survival advantage, nor would complete oxidation of complex carbohydrates or other organic compounds It is possible that these traits could be introduced and improved through genetic engineering and directed evolution. However, the necessary genetic tools have not been readily available for long and in many cases they are still lacking for DMRB. Furthermore, many of these organisms have additional nutrient requirements and are obligate anaerobes, both of which make culturing and genetic manipulation more d ifficult. The use of consortia in mediatorless fuel cells has been the most common approach since the first report of direct electron transfer to an electrode (59) The anodic chamber in these studies generally contains some type of organic wastewater with an ill defined chemical composition and microbial flora already present in the source. A stable, electrical current producing consortium is then select ed for through prolonged MFC or poised potential bioelectrochemical cell operation. The optimization of MFC design and operation in these studies has led to considerable increases in power output and densities. The challenges associated with these black bo x MFCs are that their complex interactions and metabolic networks are difficult to study and lessons learned from researching one system may not be applicable to another. A poor understanding of the underlying metabolic processes might make stable and repr oducible MFC performance difficult and certainly complicates efforts to increase microbial current production via metabolic engineering. MFCs employing isolated organisms may limit power output over the short term, but the ability to study simpler systems has been integral to the understanding of fundamental principles. Isolated DMRB such as Shewanella oneidensis are

PAGE 28

28 being used as model systems in extracellular electron transfer r esearch. This research can provide the basis for the combination of efficient, mediatorless electron transfer with increased substrate range and current production in a single organism, via a genetic engineering approach. Shewanella oneidensis MR 1 S. oneidensis strain MR 1 is a model organism for the study of dissimilatory metal re duction. Its genome has been sequenced and many studies have focused on its system for transfer of electrons to insoluble extracellular acceptors. S.oneidensis is a Gram negative, facultative anaerobe belonging to the Gammaproteobacteria. S. oneidensis was first isolated from Lake Oneida (New York) sediment (126) Shewanella species, while primarily aquatic, may be found in many environments, from soil and sediment, to spoiled fish, to Antarctic seawater (15) While S. oneidensis does not appear to use simple or complex sugars and incompletely oxidizes growth substrates (64) i t is capable of transferring electrons to many soluble and insoluble acceptors, including electrodes, without the need for exogenous mediators (43, 78, 128) S. oneide nsis MR 1 central metabolism S. oneidensis MR 1 contains most of the genes necessary for a complete Embden Meyerhoff Pathway (EMP), but is missing phosphofructokinase (64) It contains the genes encoding the complete pentose phosphate and Entner Doudoroff pathways. However, S. oneidensis MR 1 does not use hexose or pentose sugars. One and three carbon compounds serve (primarily) as carbon sources for growth. S. oneidensis M R 1 has a complete tricarboxylic acid cycle (TCA) under some conditions, but flux is limited and it generally does not completely oxidize carbon sources to CO 2 S. oneidensis MR 1 is a facultative anaerobe and is capable of aerobic respiration. It is not k nown to be fermentative, but has most of the fermentative pathway genes. Terminal electron

PAGE 29

29 acceptors that S. oneidensis MR 1 can use in the absence of oxygen include: dimethyl sulfoxide (DMSO), trimethyl amine N oxide (TMAO), sulfate, elemental sulfur, nitr ate, nitrite, various metal oxides and oxyhydroxides, and carbon electrodes (78, 126) S. oneidensis extracell ular electron transfer The mechanism of extracellular electron transfer used by S. oneidensis has not been fully elucidated. An electron transport chain linking cytoplasmic membrane embedded cytochromes to soluble periplasmic and outer membrane cytochromes has been shown (Figure1 2) (108) along with evidence of conductive pili, called nanowires (42) The outer membrane decaheme, c type cytochromes OmcA and MtrC are terminal reductases in the S. oneidensis MR 1 extracellular electron transfer system (26) OmcA and MtrC form a complex with MtrB in the outer membrane (102) MtrB is n ot a cytochrome and no direct role in electron transfer has been reported. MtrA is a periplasmic decaheme c type cytochrome (94) that associates with MtrB and transfers electrons to the terminal reductases. All of these proteins, along with the inner membrane bound tetraheme c type cytochrome CymA, have been reproted to be necessary for extracellular electron transfer (108) S. oneidensis MR 1 has 46 c type cytochromes (108) making it difficult to determine which are sufficient for insoluble metal (or electrode) reduction. Indirect electron transfer to insoluble acceptors was also reported, but u ntil recently, the endogenous mediators used in this process were unidentified. It has now been shown that flavins are used by S. oneidensis as soluble electron carriers (78, 128) Escherichia coli E. coli is a Gram negative, heterotrophic, facultative anaerobe belonging to the Gammaproteobacteria It is the best studied prokaryote, with minimal nutrient requirements and a short generation time. Exce llent genetic tools are available and several complete E. coli strain genome sequences have been published and well annotated. These characteristics all contribute

PAGE 30

30 to making E. coli an ideal organism for basic and applied research projects. E. coli has bee n used extensively by all types of researchers as a molecular biology tool, by microbiologists as a model system, and by the biotechnology industry for production of everything from pharmaceuticals to fuels. E. coli Central Metabolism The primary central m etabolic pathways of E. coli that are involved in aerobi c glucose catabolism (Figure 1 3 ) consist of glycolysis (Embden Meyerhof Parnas pathway), intermediate (pyruvate) metabolism, and the tricarboxylic acid cycle (TCA cycle, also called the Krebs or citr ic acid cycle). E. coli has a wide substrate range and is capable, under aerobic conditions, of completely oxidizing a variety of organic compounds including: sugars (hexoses and pentoses), sugar alcohols, organic acids, fats, and proteins. E. coli can sy nthesize all amino acids, nucleic acids and cofactors needed for its growth and metabolism, and can be cultured in a mineral salts medium. It can respire anaerobically and can use nitrate, nitrite, DMSO, TMAO, or fumarate as terminal electron acceptors. Ho wever, E. coli is not generally capable of mediatorless extracellular electron transfer. There have been reports of strains of E. coli that have been serially transferred in MFCs to select for mediatorless extracellular electron transfer (97, 129) but further characterization of these strains has not been reported, n or have the initial reports been independently confirmed. In the absence of exogenous electron acceptors, E. coli metabolizes sugars via mixed acid fermentation. Primary fermentat ion products are lactate, acet ate, formate and ethanol, accompanied by succin ate and pyruvate as side products. Under these conditions, the TCA cycle is not complete and flux to succinate a ketoglutarate is via reductive and oxidative branches (respectively) of the split pathway. In addition, flux through the pyruvate dehydrogenase ( PDH ) complex, also important in aerobic catabolism, is greatly

PAGE 31

31 diminished. The control of flux through the se metabolic pathways is primarily exerted by the cellular ATP/ADP and NADH/NAD + ratios. ATP/ADP ratio The ratio of ATP to ADP has been shown to be largely responsible for control of flux through glycolysis. By overexpressing the genes of the F 1 sector (cy toplasmic ATPase portion) of the F 1 F O ATP synthase, Koebmann et al determined that greater than 75% of glycolytic flux control is exerted by ATP/ADP (63) Therefore overexpression of genes encoding glycolytic enzymes had little effect upon g lycolytic flux. Other groups have reported that futile metabolic cycles (gratuitously consuming energy) increase the rate of glucose catabolism as well (67, 91) Disruption of the atp operon (encoding the subunits of the F 1 F O ATP synthase) eliminates oxidative phosphorylation, decreasing the ATP/ADP ratio and increasing the rate of glycolysis (52) The ATP/ADP ratio affects the rate of glucose catabolism through substrate limitation and by allosteric control of key glycolytic enzymes (20) Manipulation of the ATP/ADP ratio by the abovementioned means has been successfully used to increase production rates of compounds such as acetate (20) and pyruvate (19) in E. coli and glutamic acid in Corynebacterium glutamicum (5) Cellular redox balance The cellular NADH/NAD + ratio is another critical factor in determining metabolic flux and routing in E. coli In the context of the complete and rapid oxidation of glucose, this ratio must be kept low. Increased glycolytic flux without a similar increase in the rate of NADH reoxidation will result in the incomplete catabolism of the glucose. In the absence of oxygen or other exogenous electron acceptors, respiration is not possible and the N ADH/NAD + ratio is elevated (109) Fermentative pathways permit the reoxidation of the NADH via reduction of intracellular metabolites such a s pyruvate and acetyl CoA, but limit MFC yield and stable operation. Even in

PAGE 32

32 the presence of an exogenous electron acceptor, the respiratory chain can become rate limiting, leading to phenomena such as acetate overflow metabolism (76, 124) High glycolytic flux results in a high NA DH production rate. If NADH canno t be reoxidized rapidly enough by the resp iratory chain, NADH can limit flux through the TCA cycle and divert carbon to acetate via the phosphotransacetylase acetate kinase ( PTA ACK ) pathway. This shunting to acetate avoids the extra reducing equivalents that would otherwise be produced via the TC A cycle and helps to maintain redox balance. NADH exerts its negative control of TCA cycle flux primarily through transcriptional regulation, through allosteric inhibition of key enzymes, and through its inhibition of dihydrolipamide dehydrogenase (the E3 component of both PDH and KGDH). Overcoming the problem of rate and yield limitation due to redox imbalance requires eithe r increasing the rate of NADH oxidation, or removing the negative controls of NADH over PDH and the TCA cycle. Both approaches have resulted in some success. Vemuri et al have combined these approaches to address the problem of acetate overflow metabolism (124) using an engineered arcA deletion strain expressing a gene from Streptococcus pneumoniae encoding a water forming NADH dehydrogenase. Veit et a l replaced the sdhC promoter region to eliminate repression of gltA (encoding citrate synthase) and of the sdhCDAB sucABCD operon (123) The intergenic region between gltA and sdhC contains several ArcA P binding sites (28) as well as binding sites for cAMP CRP (cyclic AMP bound catabolite repressor protein) and FNR. FNR and ArcA P are redox sensitive (ArcA is the response regulator of the ArcAB system and is phosphorylated by ArcB under reducing conditions) transcriptional dual regulators (77) Kim et al characterized an lpd (encoding dihydrolipoamide dehydrogenase) mutant whose PDH activity is less sensitive to NADH inhibition (62) Each of these strategies for increasing flux through intermediate metabolism and the TCA cycle has been shown to be beneficial and several are incorporated into

PAGE 33

33 our study. However, the more global approach to solving this problem of increasing flux by directly addressing the reoxidat ion of NADH is simpler and may be more effective than trying to remove individual bottlenecks. This work will therefore focus upon maintaining a low NADH/NAD + ratio to enable higher electron yields from glucose while increasing the catabolic rate by keepin g the ATP/ADP ratio low. Electron Transport Unlike the DMRB, E. coli is incapable of extracellular electron transfer in the absence of exogenous electron acceptors. The electron transport chain is confined primarily to the inner membrane, with a few peripl asmic proteins that function in nitrate and nitrite reduction. There are no outer membrane cytochromes in E. coli and therefore no conduit for the electrons from glucose metabolism to cross the outer membrane. Nor are there any confirmed reports of native E. coli producing diffusible electron carriers (such as the flavins of Shewanella species) for extracellular electron transfer. The electron transport chain of E. coli can contain several different dehydrogenases and terminal oxidases, connected by the qu inone pool (17) Our study primarily discusses the components normally involved in aerobic (and nitrate) respiration, with NADH as the primary electron donor. It is important to note, however, the presence (under the appropriate conditions) of dehydrogenases that accept electrons from: glucose, lactate, pyruvate, succinate, formate hydrogen, and glycerol 3 phosphate. Also DMSO, TMAO, nitrate, nitrite and fumarate reductases can accept electrons from the quinone pool. With all of its dehydrogenases and terminal reductases, along with three types of quinones (ubiquinone, menaquinone, and dimethylmenaquinone) there are a num ber of possible combinations and therefore many possible electron transport chains. In reality, the cellular redox potential and energy state determine the path of the electrons. The greater the difference in

PAGE 34

34 potential between the electron donor and the te rminal electron acceptor, the more energy that is available from the overall reaction and the more proton translocating coupling steps that are involved. Electron donor and acceptor pairs with small differences in potential, such as formate and DMSO, are n ot energy conserving transfer pathways, whereas NADH and oxygen (aerobic respiration) provide the greatest potential difference that can result in the most proton translocation for energy conservation. Therefore when oxygen is available, the cell can deri ve more energy from using the aerobic respiratory chain than from respiration with any other acceptors. Engineered E. coli as a Clean Background Model for Dissimilatory Metal Reduction and as a Biocatalyst for Increased MFC Power Output Despite significant ly improved models of extracellular electron transport systems in bacteria such as Shewanella and Geobacter species, much is still unknown about these systems. In particular, it has not been easy to determine which components are sufficient for the reducti on of extracellular acceptors. This is primarily due to the large number of periplasmic and outer membrane cytochromes that are present in these DMRB. There is a complicated and redundant set of electron transfer pathways, with some leading to specific ter minal electron acceptors and some pathways that can be used for reduction of many different types of acceptors. E. coli offers a clean background in which to assemble the minimal set of components to allow reduction of extracellular acceptors. The extracel lular electron transport chain of S. oneidensis MR 1 is the best studied system and Pitts et al have already demonstrated that the periplasmic MtrA, heterologously produced in E. coli can accept electrons from the native NapC (94) Further, the outer membrane cytochromes OmcA and MtrC from S. oneidensis have been produced in E. coli and can localize to the o uter face of the OM, permitting their reduction and oxidation by

PAGE 35

35 chemical agents (34) However, the chain has not yet been completed and the minimal set of pro teins (and other carriers) has not been determined. The establishment of mediatorless extracellular electron transport in E. coli could also be useful for production of bioelectricity in an MFC. Combining the broad substrate range, ease of genetic manipula tion and high productivity of E. coli with an extracellular electron transport system would provide an excellent biocatalyst for robust and versatile MFCs. Such a strain could be readily engineered and adapt ed to provide high power output and high Coulombi c efficiency

PAGE 36

36 Figure 1 1. Example of a simple two chambered microbial fuel cell. The current (I) is shown counter to the flow of electrons (e ) by convention. Potassium ferricyanide is used as the oxidant in the cathode chamber in this example. A rgon gas is used to maintain anaerobic anode chamber. Light rectangles represent microbial biocatalyst. Filled and open circles, oxidized and reduced diffusible mediator dye (respectively)

PAGE 37

37 Figure 1 2. Simple model of extracellular electron transfer in Shewanella oneidensis Light blue arrows represent electron transfer. X red/ox reduced/oxidized dehydrogenase substrate; DH, f ormate (or other cytoplasmic membrane associated) dehydrogenase; Q, quinone pool; CymA, cytoplasmic membrane associated c type cy tochrome; MtrA, periplasmic c type cytochrome, OmcA and MtrC, outer membrane associated c type cytochromes; MtrB, protein of unknown function (complexes with OmcA and MtrC). Electron transfer to MtrA may be directly from CymA or via an (unknown) intermedia te. Some putative components of system omitted

PAGE 38

38 Figure 1 3. Simplified view of central metabolism of aerobically cultured Escherichia coli ATCC 8739. Glyoxylate shunt (grayed, dotted lines) is not active and is omitted. Major fermentation pathwa ys are not active and are shown in gray, with dotted lines. Inset shows simplified oxidative phosphorylation with the electron transport system (ETS) driving the net translocation of protons into the periplasmic space (Out) and their return to the cytoplas m (In) through the F 1 F O ATP synthase. Reducing equivalents are shown as electrons (e ) and ATP is shown as ~P for simplicity. P i inorganic phosphate

PAGE 39

39 CHAPTER 2 PHYSIOLOGY AND METAB OLISM OF AN E COLI ATCC 8739 STRAIN LAC KING THE STATOR STALK OF THE ATP SYNTHASE Introduction The F 1 F O ATP synthase of Escherichia coli is a well studied molecular motor which couples the proton potential generated by electron transport through the respiratory chain with the phosphorylation of ADP. The complex is co mposed of two sectors, a cytoplasmic membrane bound F O sector and a catalytic F 1 sector. The F O sector is made up of an a subunit, a ring of c 1 sector. The structure of the F 1 F O ATP synthase has been reviewed by others (41, 130) Functionally, the F O se ctor contains the channel (between the a and c subunits) through which protons are translocated across the membrane, causing the rotation of the c subunit r otor (16, 87) (112) which 1 sector to the a subunit of the F O sector rives the phosphorylation of ADP within the stator of the F 1 sector. The process is reversible, with ATP hydrolysis capable of pumping protons out of the cytoplasm (29) Thus the F 1 F O ATP synthase is able to tightly control the cytoplasmic pH over a wide range of conditions (140) in addition to its more commonly discussed role in oxidative phosphorylation. The eight genes encoding the structural subunits of t he E. coli F 1 F O ATP synthase are part of a nine gene operon ( atpIBEFHAGDC ), where atpI does not code for a structural subunit of the F 1 F O ATP synthase, but may have a regulatory function (37) or might assist in assembly of the complex (90) The other genes encode, respectively (with subunit stoichiometry): a 1 c 10 b 2 1 3 1 3 1 (Figure 2 1).

PAGE 40

40 Flux through glycolysis is increased through overexpression of F 1 genes (63) or by disruption of the atp operon (20, 52, 86) The first approach lowers the cellular ATP/ADP ratio as a result of gr atuitous cytoplasmic F 1 ATPase activity. Control of glycolytic flux is largely dependent upon this ratio (63) The lower ATP/ADP ratio in strains with deletions in the atp operon leads to an increase in glycolytic flux via activation of phosp hofructokinase by ADP and AMP (32) AMP activation of pyruvate kinase (121) and inhibition of the gluconeogenic fructose bisphosphatase by AMP (7) The increase in ADP availability for phosphoglycerate kinase and pyruvate kinase is also stimulatory. This leads to an in creased pyruvate pool and acetate overflow metabolism, since downstream steps become rate limiting (27) The primary reason for this rate limitation seems to be the inability of the respiratory chain to reoxidize the NADH produced by glucose catabolism (19) This respiratory limitat ion is partially alleviated because increased pyruvate levels inactivate the pyruvate dehydrogenase repressor (PdhR). PdhR has been found to negatively regulate transcription of ndh and cyoABCDE (encoding cytochrome bo 3 ) (89) Mutation of the atp operon by insertion or deletion also lowers the ATP/ADP ratio by eliminating oxidative phosphorylation. Deletion of the atpFH genes physically uncouples the F 1 and F O sectors through the elimination of the second stalk and can further lower the ratio through a combination of the aboveme ntioned effects (20) Other atp operon deletion strains are partially complemented with the F 1 genes expressed from a plasmid (6 3) Previous work in our laboratory used atpFH deletion strains to produce acetate and pyruvate with high rates and yields (19, 20) These strains have hi gh flux from glucose to acetate or to pyruvate w hen oxidative routes to acetate are eliminated.

PAGE 41

41 Here I have made a detailed investigation of the physiology and metabolism of an atpFH deletion strain Respiratory and overall electron transfer routes were examined by measurement of cytoplasmic membrane associated NADH oxidation and of dye reduction activities. Relative transmembr ane proton conductance and membrane potentials were evaluated. I underscore the important role of the F 1 F O ATP synthase in pH homeostasis and in metabolic flux distribution for cellular redox balance. Materials and Methods Growth of Cultures and Media The strains used in this study are derivatives of E. coli ATCC 8739 (Table 1 1 ). Unless otherwise specified, cells were grown aerobically in NBS mineral salts (20) containing MOPS (4 morpholinopropanesulphonic acid, 0.1 M, pH 7.4) and glucose (50 mM glucose), using baffled flasks (50 ml broth in 250 ml flasks, 250 rpm, 37 C ). LB and LB agar plates were used only for plasmid and strain construct ion. Glucose was included (20 g/liter) for strains that contained deletions in the atp operon. Antibiotics were included as appropriate (kanamycin, 50 m g /liter; ampicillin, 100 m g /liter; chloramphenicol, 40 m g /liter). Arabinose (20 g /liter) was used to ind uce expression of Red recombinase genes (strain construction). Anaerobic Batch Fermentation Fermentations were done in 500 ml vessels (250 ml working volume, 37C, 150 rpm) under argon. Cells were grown in NBSM mineral salts with glucose (50 mM glucose). Samples were removed for the measurement of pH, optical density (cell ma ss), and fermentation products ( by high pressure liquid chromatography, HPLC). Construction of ATP Synthase Gene Deletions E. coli ATCC 8739 was used as the parental strain for all mo difications. PCR primers, plasmids and strains used in this work are listed in Table 1 1 Standard methods (103) were used

PAGE 42

42 in PCR amplification, plasmid construction, and transformati ons. Deletions were made by Red recombinase (30) catalyzed integration of an FRT (FLP recognition target) flanked cassette containing a selective antibiotic mark er. This cassette was excised by a FLP recombinase catalyzed intramolecular recombination (95) leaving a single FRT containing sequence artifact. FLP recombinase also enabled the removal of longer chromosomal sequences between similarly oriented FRT sites from two individual deletions. A port ion of atpD was deleted to produce JC10. Primers JM atpC C Kpn I and JM atpD N Nde I were used to amplify part of atpC through part of atpD from ATCC 8739 genomic DNA by PCR. The product was cloned into pCR2.1 TOPO (Invitrogen, Carlsbad, CA) to produce pLOI4135. The Kpn I/ Nde I atpCD multiple cloning site of pFLAG CTC to produce pLOI4123. The Eco RI/ Sma I FRT kan FRT fragment from pLOI2511 was ligated into the Stu I/ Eco RI digested inside out PCR product ( J M atpD Stu I/ JM atpD Eco RI primers, pLOI4123 template) The resulting plasmid (pLOI4125) was digested with Kpn I/ Nde I and used a template for PCR with JM atpC C Kpn I and JM atpD NNde I primers. Construction of JC07 was by Red recombinase facilitated, double cross over homologous recombination with the PCR product and ATCC 8739 transformed with pKD46 (30) The successful integrants were selected for chloramphenicol resista nce. The FRT flanked chloramphenicol marker was removed via the FLP recombinase method previously described (79, 95) to produce JC10. Replacement of atpFH with FRT cat FRT in ATCC 8739 and JC10, followed by FLP recombin ation, resulted (respectively) in strains JC27 ( atpFH ) and JC25 ( atpFH AGD ). JC25 resulted from an FLP recombinase catalyzed excision of sequence between integration sites. JM atpFH down Kpn I and JM atpFH up Nde I primers were used for PCR cloning of atpFH

PAGE 43

43 (pLOI 4139) and for amplification of the fragment for integration (from pLOI4134). Primers JM atpFH Stu I and JM atpFH Eco RI were used for inside out PCR. The Stu I/ Eco RI FRT cat FRT fragment from pLOI4131 was ligated into the Stu I/ Eco RI digested inside out PCR prod uct (for pLOI4134). Dye Reduction Assay Electron transfer rates were determined during glucose catabolism using methylene blue ( MB, colorless in reduced form) as the terminal electron acceptor (8, 84) Specific rates were measured using aerobically grown cultures (NBSM, 10 g /liter glucose) harvested in early exponential growth (approximately 0.33 gDCM/liter). Equal volumes (1.5 ml each) of culture and m edium containing carboxymethylcellulose (10 g/liter) and MB (50 M) were mixed in a stoppered cuvett e Carboxymethylcellulose was included to increase viscosity and minimize mixing of the oxidized surface. Change in A 609nm was monitored at 37 C a nd used to calculate the specific rate of dye reduction. Enzyme Assays Cells were lysed and fractionated for the determination of membrane associated NADH dehydrogenase, NADH oxidase, and TMPD ( tetramethyl phenylenediamine) oxidase activities. Aerobic cul tures (50 ml) were harvested (0.33 g DCM /liter) by centrifugation and stored overnight at 20 C. Pellets were thawed on ice, washed with 10 ml ice cold TM buffer (50 mM Tris HCl, 10 mM MgSO 4 pH 7.5) and suspended to a cell density of 0.35 g DCM /liter in ice cold TM buffer containing 10 g /ml DNase I. Cells were disrupted by passage though a French pressure cell at 14,000 psi. Unbroken cells and debris were removed by two centrifugation steps (10 min each 8,000g, 4 C) Supernatant v olumes were adjusted to 1 0 ml with TM buffer and membrane pellets prepared by ultracentrifugation (1 h, 165,000g, 4 C). Membrane pellets were homogenized (vortex mixing and Teflon homogenizer) in 1 ml of TM

PAGE 44

44 buffer. Both the supernatants (cytoplasmic fractions) and membrane fracti ons ( membrane vesicles, MVs) were held on ice and used immediately for enzyme assays. Enzyme activities were measured at room temperature (22 C). Values reported are from triplicate determinations of two independent samples. Activities are expressed as nmo l min 1 mg protein 1 Total membrane bound NADH oxidation activity was determined spectrophotometrically (340 nm) using the molar extinction coefficient of NADH (6220 M 1 cm 1 ). Reac tions were initiated by adding MVs ( 30 g protein) to a 1 ml assay mixtur e containing TM buffer and 0.25 mM NADH (110) Deamino NADH (d NADH) oxidation was used to measure the NADH oxidase activity from the proton translocat ing NADH dehydrogenase (NDH 1) in a similar manner. Note that deamino NADH is a substrate for NDH 1, but not for NDH 2 (81) Membrane bound NADH dehydrogenase activity was measured spectrophotometrically (110) using potassium ferri cyanide as the electron acceptor. KCN was included in the reaction mix to inhibit cytochrome oxidase activity. NADH (0.5 mM, final) was used as substrate to measure combined NDH activities. The deamino NADH (0.5 mM, final) was used to measure NDH 1. Reacti ons were initiated by addition of g protein). Cytochrome oxidase activity w as measured as the oxidation of TMPD (53) at 609 nm usin g the molar extinction coefficient for oxidized TMPD (12,200 M 1 cm 1 ). Reac tions were initiated by adding g protein). SDS Electrophoresis and Immunoblotting otein) were incubated for 1 h at room temperature in reducing SDS sample buffer (46) and resolved by e lectrophoresis using 12% polyacrylamide slabs (200 V for 35 min) with Tris glycine running buffer (65) Protein was transferred to a polyvinylidene fluoride ( PVDF Hybond) membrane (44) and blocked with 5% (w/v) nonfat powdered milk. Primary rabbit antise ra against subunits a and c (provided by Dr. Robert Fillingame) were diluted

PAGE 45

45 (1:5,000 and 1:10,000, respectively), combined and incubated with the PVDF membrane (1 h at ambient temperature). Alkaline phosphatase conjugated goat anti rabbit secondary antibo dy was diluted (1:10,000 in TBS Tween20) and incubated with the PVDF membrane (40 min at ambient temperature). The washed membrane was then developed as previously described (44) with bromochloroindolyl phosp h ate/nitro blue tetrazolium (BCIP/NBT) as a subst rate, rinsed in deionized water and dried for imaging. Analyses Glucose and organic acids concentrations were measured by HPLC using an HP 1090 Series II equipped with refractive index and UV(210 nm) detectors, a Bio Rad HPX 87H column and with a 4 mM H 2 SO 4 mobile phase (20) Cell mass was estimated by measuring optical density at 550 nm using a Bausch & Lomb Spectronic 70 spectrophotometer (1 OD 550 is equivalent to 0.33 gDCM/liter). Protein concentration was determined using the bicinchoninic acid method (BCA Protein Assay Kit, Thermo Scientific, Rockford, IL) with bovine serum albumin as a protein standard Results Construction of JC25 and J C27 Through Targeted D eletions in atp O peron Two isogenic strains were constructed using ATCC 8739 by making targeted deletions in the atp operon (Table 2 1, Figure 2 1). Both of the deletion strains lacked the genes encoding the 1 F O ATP synthase (33 bp of atpF encoding the first 11 amino acids of the b subunit, remained in JC25 and JC27). In addition, JC25 lacked the F 1 sector genes ( atpC remained). These strains provided a clean genetic background and a means to compare the metabolic and physiological effects of removal of each of these major F 1 F O ATP synthase structural elemen ts, in vivo

PAGE 46

46 The membrane proteins from aerobically grown ATCC 8739 and JC27 were electrophoretically separated by SDS PAGE and subjected to immunoblotting to determine whether or not subunits a and c were present in the membrane of JC27 (Figure 2 2). Su bunit c was present in the membrane fraction of JC27, but subunit a was absent. The band that migrated just above subunit a was due to nonspecific labeling, as confirmed by its presence in the membrane fraction of an F O strain (Figure 2 2, lane 3). This r esult is consistent with previous reports of incomplete F O assembly in the absence of the b subunit (46) In particular, JC27 lacks codons for residues that have been found to contact the a subunit (82) These results show that JC27 is functionally F O and should not allow F O sector mediated proton transduction across the inner membrane. The F O transmembrane proton channel has been reported to be at the interface betwe en a and c subunits (113) While the JC25 membrane fraction was not analyzed, it likely contained only the c subunits of the F O sector, since it (like JC27) was missing the b subunit. The only genotypic difference between JC27 and JC25 is the presence of the genes for the subunits of the F 1 (ATPase) sector in the former. Therefore, both stains were expected to lack oxidative phosphorylation and any phenotypic difference should have res ulted from gratuitous ATP hydrolysis in JC27 (20) Comparison of Aerobic Growth Rates and Yields ell densities and biomass yields from glucose (Y g ) of the strains ATCC 8739, JC25, and JC27 grown aerobically in a mineral salts medium, were compared in order to characterize the effects of each deletion in the atp operon (Table 2 2). The growth rate and Y g of JC27 were approximately 30% and 50% lower, respectively, than ATCC 8739. Growth rates and cell yields of the JC25 and JC27 strains were similar.

PAGE 47

47 Product Formation and Specific Rates of Glucose Consumption Both strains with deletions in the atp op eron had approximately 44% higher specific rates of glucose consumption ( J gluc ) vs. the wild type, when the cultures were grown aerobically (Table 2 2). The higher glucose consumption rates were consistent with those reported for other atp mutants (2 0, 52, 86) There have been few studies of the product profiles of E. coli strains with targeted atp operon deletion s grown aerobically in glucose mineral salts medium (20, 86) However, the accumulation of acetate by the modified strains (Figure 2 3) was not unexpected, given the increased glycolytic flux and the well established phenomenon of overflow metabolism to acetate (123, 124) Furthermore, others have used engineered atp strains to produce high titers of acetate and pyruvate from glucose (20, 135) JC25 produced 22.3 mM acetat e from 50 mM glucose by 11 h after which the acetate levels decreased slightly to 14.3 mM by 24 h JC27 accumulated 21.1 mM acetat e by 11 h which decreased to 15.5 mM by 24 h In contrast, the wild type strain transiently produced a small amoun t of acetate (3.2 mM) by 8 h w hich was undetectable by 9 h JC25 and JC27 transiently produced a small amount of pyruvate and trace amounts of lacta te. Most of the acetate produced in the mutant strains was not subsequently catabolized by the cells, presumably due to energetic constraints. Both of the primary acetate assimilation pathways require one ATP per acetate for conversion to acetyl CoA. In th e absence of oxidative phosphorylation, there is no net energetic benefit in catabolism of acetate, because a maximum of one ATP per acetyl CoA can be gained via the TCA cycle. Methylene Blue Reduction Rates The rates of net electron transfer from glucos e catabolism in aerobically grown cultures were evaluated on the basis of the specific rates of reduction of MB (v redMB ). MB is a membrane permeable, water soluble phenothiazine redox dye with a standard biological midpoint potential

PAGE 48

48 of +0.011V. It has bee n used as an indicator of bacterial metabolism in the food industry (2) and to quantitate viable E. coli cells on the basis of total reducing power (8) The dye ch anges from blue in its oxidized form (absorbance max at 609 nm) to colorless when reduced, in a two electron transfer. The oxidative phosphorylation deficient strains showed higher v redMB values than the wild type strain (Figure 2 4). Strain JC27 ( atpFH ) had a v redMB that was over six fold higher than that of the wild type (100.3 vs. 16.3 nmol min 1 mgDCM 1 respectively). JC25 ( atpFHAGD ) lacked the F 1 sector and exhibited a v redMB that was five fold higher (82.0 nmol min 1 mgDCM 1 ) than ATCC 8739 These results suggest that elimination of oxidative phosphorylation is responsible for most of the increase in v redMB observed for the strains with deletions in the atp operon. The slightly higher rate of MB reduction shown by JC27 (vs. JC25) may have b een due to the effect of gratuitous ATP hydrolysis. While the glucose dependent reduction of MB allowed the measurement of the rate of production of reducing equivalents, it did not show which pathways were used for reoxidation of reduced cofactors in thes e strains. Anaerobic vs. Aerobic Growth Rates and Metabolism Strains lacking oxidative phosphorylation were expected to grow similarly in the presence or absence of oxygen. To test this hypothesis, ATCC8739 (wild type) and JC27 ( atpFH ) were grown anaerob ically in glucose minimal medium and their growth rates compared with the aerobic cultures (Figure 2 5). As mentioned above, the absence of oxidative phosphorylation led to the lower growth rate of JC27 under aerobic conditions. As expected, the growth rat e of the wild type was lower (0.604 h 1 59% of aerobic culture growth rate) in the absence of oxygen. However, a surprisingly lower growth rate (0.357 h 1 48% of aerobic culture growth rate) was seen in JC27 anaerobic cultures. The difference cannot be a ttributed to a lower ATP yield in JC27. Therefore, it is likely to be the result of limitations imposed by the lower cellular redox

PAGE 49

49 potential and the need to maintain the PMF via ATP hydrolysis driven proton pumping by the F 1 F O ATP synthase (ATPase). Anaer obic growth, under tested conditions, was not glucose limited for either of the strains. While ATCC 8739 consumed all glucose ( 50 mM initial conc.) by 30 h it reached its maximum cell density of approximately 0.90 gDCM/liter by 13 h JC27 cultures had onl y consumed 19 mM glucose an d reached a maximum cell density of approxima tely 0.17 gDCM/liter by 30 h The pH of the medium dropped from an initial value of 7.2 to 5.0 by 30 h in the ATCC 8739 culture, whereas the the pH of the medium in JC27 cultures o nly decreased to 6.7 by 30 h By the time JC27 gr owth reached a maximum (16 h ) the total concentration of reduced products measured (lactate, succinate and ethanol) was 8.5 mM. Total acids produced (pyruvate, succinate, lactate, formate and acetate) were 16.1 mM, from 7.1 mM of consumed glucose. These results indicate that the anaerobic growth of JC27 was limited by its inability to maintain an optimal cytoplasmic pH and the necessary proton motive force due to the lack of an ATP driven proton pump NADH Oxida se and NADH Dehydrogenase Activities M embrane vesicles were used to measure membrane associated NADH oxidation (Table 2 3). The rat e of oxidation of NADH by JC27 MVs was 2.5 fold higher than ATCC 8739. Rates of d NADH oxidation were at least 30 fold lower than NADH oxidation, suggesting that very little NDH 1 activity was present and that NDH 2 (non proton translocating) was primarily being used in the aerobic respiratory chains, under tested conditions. As the rate of NADH oxidation by JC27 MVs was higher than the parent, it was important to measure NDH 2 NADH d ehydrogenase activities in the MVs to determine whether higher NDH 2 levels were responsible for the elevated rate of NADH oxidtion by the entire electron transport chain. Potassium ferricyanide was used in the NADH dehydrogenase assays as the

PAGE 50

50 electron acceptor and cyanide was included to inhibit the terminal oxidases. NADH dehydrogenase activity of JC27 MVs was 1.7 fold that of ATCC 8739 (Table 2 3). The higher NADH dehydrogenase activity in JC27 is consistent with previous findings with an atp A mutant (86) Cytochrome Oxida se Activity There are two types of enzymes of the E. coli aerobic respiratory chain that are involved in the transfer of electrons from NADH to molecular oxygen (54, 118) The first is an NADH dehydrogenase (NDH 1 or NDH 2) and the second is a cytochrome oxidase (cytochrome bo cytochrome bd I, or cytochrome bd II). These two complexes represent the two energy coupling sites in the aerobic electron transport chain (Figure 2 6). Different levels of either type of enzyme could have been responsible for the higher r ate of NADH oxidation observed for JC27. Having determined that NDH levels were likely to be a m ajor factor in the difference, MV cytochrome oxidase activity was measured using TMPD as the substrate. TMPD oxidase activity in JC27 MVs was approximately 1.3 fold higher than that of the parent strain MVs (Table 2 3). The higher TMPD oxidase activity in JC27 MVs is probably due to higher levels of cytochrome bd (86) Discussion Deletions (or other disruptions) in the atp operon eliminate oxidative phosphorylation, lowering the ATP/ADP ratio and increasing the rate of glucose catabolism as a result. Targeted deletion of atpFH in ATCC 8739 increased the rate of glucose catabolism, but resulted in accumulation of acetate in aerobic batch cultures. Acetate overflow metabolism is thought to result from a redox imbalance, which makes additiona l flux through the TCA cycle (and its production of extra reducing equivalents) undesirable (124) Flux is shunted from acetyl coA through the PTA ACK pathway to acetate, providing one ATP per acetyl CoA, but avoiding

PAGE 51

51 production of additional reduci ng equivalents. This suggests that despite its higher rate of NADH oxidation, the respiratory chain of JC27 may still be rate limiting. However, the experimental evidence presented here does not clearly support that hy pothesis. The JC27 MV NADH oxidation r ate (2.5 fold that of the parent) should be sufficient to reoxidize the elevated rate of glucose catabolism (1.4 fold that of the parent). Perhaps, in vivo even the higher rate of reoxidation of NADH by the JC27 respiratory chain is rate limiting. The abs ence of the F 1 F O ATP synthase in the proton circuit of JC27 (Figure 2 6) may allow a higher respiratory rate, as reported in other atp mutants (52, 86) Although there are other components of the transmembrane proton circuit (such as transporters and flagella) that can dissipate the gradient formed by electron transport, JC27 1 F O ATP synthase) is missing. Higher NADH dehydrogenase activity provides a higher rate of electron transport, in conjunction with higher levels of terminal o xidase activity. Increases in NDH and cytochrome oxidase activity were reported to be primarily due to higher levels of non proton translocating NDH (NDH 2) and cytochrome oxidase (cyt bd ) in anot h er atp mutant (86) These changes could allow higher respiratory rates without being limited by an excessive proton gradient. However, in vivo electron trans port may not be able to compensate for the higher rate of glucose catabolism in JC27. Alternatively, pyridine nucleotide transhydrogenase (such as UdhA) activity (104) may limit the TCA cycle (NADP limitation of isocitrate dehydrogenase). The TCA cycle is subject to complex regulation and there may be other explanatio ns for the observed flux to acetate in JC27. However, reports of using cofactor engineering (manipulating NADH/NAD + ) to overcome the acetate overflow problem (124, 125) support the hypothesis that respiratory chain NADH

PAGE 52

52 reoxidation limits the TCA cycle. This work aims to test this hypothesis in ATCC 8739, as part of the overall goal to transform the stored energy in glucose to useful electrical energy, with high rate and yield.

PAGE 53

53 Table 2 1. Strains, plasmids, and PCR primers Strain/plasmid/primer Relevant features Reference/source E. coli ATCC 8739 Parent strain ATCC JC07 ATCC 87 atpD ::FRT kan FRT JC10 atpD ::FRT This study JC22 atpFH ::FRT cat FRT JC24 atpFH ::FRT cat FRT This study JC25 atpFHAGD ::FRT This study JC27 atpFH ::FRT This study Plasmids pCR2.1 TOPO Clon ing vector, bla kan Invitrogen pKD46 pSC101 rep. (temp. cond.), bla gam bet exo (30) pFTA pSC101 rep. (temp. cond.), bla tetR flp (79) pLOI2511 bla FRT kan FRT (120) pLOI4123 bla atpDC This study PLOI4125 bla atpDC kan FRT This study pLOI4131 bla FRT cat FRT (51) pLOI4133 bla atpFH This study pLOI4134 bla atpFH cat FRT This study pLOI4135 bla kan atpDC TOPO This study pLOI4139 bla kan atpFH TOPO This study Primers JM atpC C Kpn I AA GGTACC GCCGTGAGAGCTG CTAAT This study JM atpD N Nde I AA CATATG CCGCGCAGCACCTTCCTA This study JM atpD Stu I AAC AGGCCT TACATGGTCGGTTCCATC This study JM atpD Eco RI AAC GAATTC CTGACGGCTCAGTACCAC This study JM atpFH down Kpn I AA GGTACC GTCTGCAAGGCGCTCAAG This study JM atpFH up Nde I AA CATATG GAAG GCGCAGCGCGTCAA This study JM atpFH Stu I AAC AGGCCT TGCCGCACTGAGTGAACA This study JM atpFH Eco RI AAC GAATTC TGGCCTGGCCGAGGATTG This study Primer sequences are Apostrophes before or a fter genes indicate truncation.

PAGE 54

54 Table 2 2. Growth and glucose consumption of atp deletion strains Strain Growth rate 1 ] ) Max. cell density (gDCML 1 ) Biomass yield (gDCMmmol 1 ) Glucose flux (mmolgDCM 1 h 1 ) ATCC 8739 1.028 0.013 4.75 0.10 0.095 0.002 10.82 0.18 JC25 (F 1 ) 0.732 0.014 2.34 0.04 0.047 0.001 15.61 0.19 JC27 (SS ) 0.746 0.007 2.40 0.02 0.048 0.000 15.53 0.02 All values are reported as mean SD, n = 3

PAGE 55

55 Table 2 3. Respiratory chain activities Strain NADH oxidase (nmolmin 1 mg 1 ) d NADH oxidase (nmolmin 1 mg 1 ) NADH dehydrogenase (nmolmin 1 mg 1 ) d NADH dehydrogenase (nmolmin 1 mg 1 ) TMPD oxidase (nmolmin 1 mg 1 ) ATCC 8739 458.8 46.2 14.9 5.3 1938.0 90.6 467.2 23.9 71.0 11.1 JC27 1128.1 35.8 24.7 8.3 3315.8 215.4 774.2 37.2 94.7 8.3 All values are reported as mean SD, n = 3

PAGE 56

56 Figure 2 1. Genetic organization and gene products of the atpIBEFHAGDC operon wi th subunits shown above coding genes. Dotted lines represent region deleted from operon in the listed strains.

PAGE 57

57 Figure 2 2. Immunoblot of membrane fractions separated by SDS PAGE (12%, Tris glycine). Rabbit antisera ag ainst subunits a and c were used together as 1 antibodies (1:5000, 1:10,000, respectively) and goat anti rabbit AP 2 antibody was used, with NBT/BCIP substrate for detection Lane 4 contains BioRad Kaleidoscope prestained protein standards.

PAGE 58

58 Figure 2 3. (A) Growth of aerobic cultures in glucose mineral salts medium. ATCC8739 (wild type), squares; JC25 (F 1 ), triangles; JC27 (SS ), inverted triangles; JC16 (F 1 F o ), diamonds; JC29 (F o ), circles. Product formation and glucose consu mp tion by (B) ATCC 8739, (C) JC25 and (D) JC27. In (B) through (D ): glucose, squares; acetate, triangles; lactate, inverted triangles; pyruvate, diamonds; succinate, circles. Error bars represent standard error of the mean of triplicate values A B C D

PAGE 59

59 Figure 2 4. Net electron transfer from glucose as determined by methylene blue (MB ) reduction by aerobically grown whole cells of E. coli strains indicated Initial rates of reduction measured by decrease in A 609 backgroun d subtracted, calculated with MB standar d curve and normalized by dry cell mass. Error bars represent standard error of the mean of quadruplicate values.

PAGE 60

60 Figure 2 5. Growth rates of aerobically (open bars) and anaerobically (closed bars) grown cultures of E. coli str ains indicated Error bars represent standard error of the mean of triplicate values

PAGE 61

61 Figure 2 6. Circuit models for aerobically grown strains with glucose as carbon source. Thickness of arrows represents relative flux. NDH, NADH dehydrogenase; Cyt bo Cyt bd cytochrome bo or bd type oxidase; X, proton transporter The large complex shown to the right side of the top panel represents the F 1 F O ATP synthase of which only the c subunit ring is present in the inner membrane of the strains represented in the bottom panel.

PAGE 62

62 CHAPTER 3 DEVELOPMENT OF GENET IC TOOLS FOR STREAML INING CHROMOSOMAL MODIFICATION OF E COLI Introduction Development of a strain capable of efficient production of reducing equivalents at high r ates required several chromosomal modifications within a single genome. Chromosomal engineering approaches previously used in our laboratory were not optimal for construction of strains with multiple modifications. We con s tructed tools for use in a streaml ined chromosomal engineering method based upon Red recombinase catalyzed homologous recombination and levansucrase counterselection. Chromosomal Engineering Background Genetic engineering through modification of chromosomal DNA has been central to many ba sic and applied studies. Manipulation of genomes through targeted integration or deletion has made predictable and stable genotypic alteration possible. Other tools such as random chromosomal mutagenesis and gene expression from plasmids have been used wit h success. However, random mutagenesis, using chemicals, transposons, or other means (such as UV radiation), is not precise and involves much screening for the desired change. Furthermore, these methods can introduce additional unintended changes which can complicate research efforts and negatively affect project goals. Plasmids can facilitate introduction of recombinant material and permit high levels of expression, but are often difficult to stably maintain over many generations. A commonly used approach to making single gene deletions by homologous recombination in E. coli is a Red recombinase based method (30) Approximately 50 bp seq uences, homologous to regions flanking the site for deletion, are nase

PAGE 63

63 genes gam bet and exo are expressed from a helper plasmid with a conditional replicon in the strain that is to be modified. The Gam protein ( prevents degradation of the linear, dsDNA PCR product by E. coli nucleases (55) in the transformed cells. Exo is an exonuclease that the homologous recombination with the target sequence on the chromosome. Double crossovers result in the replacement of the targeted chromosomal sequ ence with the PCR amplified sequence containing the antibiotic resistance marker. The Red helper plasmid is cured from the deletion strain to prevent undesired effects from low level expression of the Red genes in the absence of inducer. The antibiotic res istance marker can be removed later, by the activity of FLP recombinase (127) which can be expressed from another helper plasmid. FLP recombinase recognizes the FRT (FLP recognition target) sites that flank the antibiotic resistance gene and catalyzes a site specific excision of the interveni ng sequence. An 82 to 85 bp scar sequence is left behind, which contains a single FRT site, along with stop codons in all reading frames. An alternative to FLP catalyzed antibiotic marker removal is the use of a cassette containing selective and counterse lective markers. The cassette is integrated into the target by homologous recombination, as described above. A second Red recombinase catalyzed recombination event is used to remove or replace the cassette, depending on whether seamless deletion or introdu ction new sequence is desired. Clones that lack the cassette are then recovered by counterselection. Counterselective markers used in this type of approach include tetAR (14) tolC (33) and sacB (66, 116) The sacB gene encodes a levansucrase, which transfers a fructosyl group from sucrose to make levan. In its native organism, Bacillus subtilis (and other levan producers), the levansucrase is secreted and the polymerization occurs outside the cell. In B. subtilis strains that are engineered to lack the ability to secrete the enzyme, the levan polymer

PAGE 64

64 accumulates in the cytoplasm and leads to cell lysis. Levansucrase counterselection is effecti ve in E. coli as it canno t secrete the levansucrase. The two step homologous recombination strategy using selection and counterselection offers two significant advantages over the FLP catalyzed marker removal. The first is the ability to make seamless de letions. The FRT containing scar sequences that are left behind by FLP catalyzed marker removal can be problematic, particularly if multiple deletions are to be made. Multiple FRT sites in the chromosome can result in rearra n gement or excision events (30) Also, the FRT scar sequence can serve as a target for undesired integration of FRT flanked marker sequences when deletions or replacements are attempted at othe r loci (30) Finally, for some applications, it is desirable to limit the amount of foreign sequence introduced into the chromosome in order to comply with gover nmental or industrial regulations. The selection counterselection method also permits replacement of chromosomal sequences. The popular one step PCR based deletion method provides a quick and simple means of making single deletions, but it is not suitable for more involved metabolic engineering projects. These projects may require multiple deletions and benefit from the ability to introduce new genes or regulatory elements into specific target sequences. Our study required multiple deletions to alleviate ne gative control of complete glucose oxidation and to eliminate fermentation pathways. We wanted to integrate naoX from S. mutans into the E. coli ATCC 8739 chromosome to increase the rate of electron transfer to an electrode via a diffusible mediator dye. F inally, we wanted to integrate S. oneidensis extracellular electron transfer genes to permit mediatorless electron transfer to an electrode. A streamlined method for making chromosomal gene replacements or seamless and nonpolar deletions was needed. Much t ime and effort is spent on the planning and construction of plasmids for specific deletions and replacements. pLOI4162

PAGE 65

65 was designed to provide a versatile tool to make this process faster and easier. Use of the cassette from this plasmid, in conjunction wi th a two step Red recombinase catalyzed homologous recombination procedure allows targeted gene deletions that leave behind a minimal sequence artifact (18 bp) containing stop codons in all reading frames (Figure 3 1). Alternately, a seamless deletion can be made, by using a different second recombination step. A new sequence can be used to replace the cassette in the second step, via a simple cloning step. While the above tools were designed to simplify new deletions and replacements, our lab had many str ains with multiple FRT scar sequences in their chromosomes from prior chromosomal engineering. We were interested in removing these sequences because they were a source of chromosomal instability and further modification was unsustainable because of the ef fects discussed earlier. Also, some of these strains were being used in industrial processes where the amount of foreign sequence in the chromosome needed to be minimal, due to regulatory requirements. One approach that we could have used to remove these s equences would have been to delete them using the same two step strategy described above, using the selective, counterselective cassette from pLOI4162. Another option was to construct a plasmid from which circular DNA, containing a single FRT site and lack ing an origin of replication, could be made easily. This circular DNA would contain both selectable and counterselectable markers and would be used to integrate into FRT sites in the chromosome, catalyzed by the FLP recombinase (30) Then the FRT flanked cassette could be removed in a Red catalyzed homologous recombination step. For this purpose, pLOI4151 was constructed and used to remove scar sequences from lact ate producing strains (139)

PAGE 66

66 Construction of Plasmids Construction of pLOI 4162 Containing a cat sacB Cassette for Markerless Gene Deletions To facilitate sequence replacement and sequential deletions in chromosomal DNA, plasmid pLOI4162 (Figure 3 2) was constructed with a removable cat sacB cassette and the option to include an 18 bp segment of synthetic DNA with stop codons in all reading frames. This plasmid is composed of synthetic sequences and parts of plasmids pLOI2228 (79) pLOI2511 (120) and pEL04 (66, 116) Using pEL04 as a template, inside out PCR was performed with the JMpEL04F1/R1 primers to eliminate unwanted Sma I and Bam HI sites between the cat and sacB genes. The amplified product was dig ested with BglII (within both primers) and self ligated to produce pLOI4152. Plasmid pLOI4131 was constructed by ligation of the FRT cat FRT fragment (Klenow treated Ban I, Cla I) from pLOI2228 into compatible sites of pLOI2511 (Klenow treated NheI, ClaI). P lasmid pLOI4131 was subsequently digested with Eco RI and self ligated to remove the FRT cat FRT fragment to produce pLOI4145, retaining single Kas I and Xma I sites. A polylinker segment (SfPBXPS) was prepared by annealing complementary oligonucleotides (Sf PBXPSsense and SfPBXPScomp). After digestion with Kas I and Xma I, this segment was ligated into corresponding sites of pLOI4145 to produce pLOI4153. The modified cat sacB cassette in pLOI4152 was amplified by PCR using the JM catsacB up3/down3 primer set. Aft er digestion with Bam HI and Xho I, this cassette was ligated into corresponding sites of pLOI4153 to produce pLOI4146. To create an 18 with stop codons in all six reading frames, pLOI4146 was digested with Pac I and self l igated to produce pLOI4154 (not shown), removing the cat sacB cassette. Two additional bases (T and A) were inserted between the Sfo I and Pac I sites of pLOI4154 using mutagenic primers (JM4161sense/comp) and linear plasmid amplification to produce pLOI4161 (not shown).

PAGE 67

67 Finally, the Pac I digested fragment from pLOI4146 containing the cat sacB cassette was ligated into the Pac I digested site of pLOI4161 to produce pLOI4162 (GenBank accession EU531506). Construction of pLOI4151 for Removal of FRT Sites The cat sacB cassette was PCR amplified from pEL04 (66, 116) with the JMcatsacBupNheI and JMcatsacBdownNheI primers. pLOI415 1 (Figure 3 3) was constru cted by ligating the Nhe I digested cat sacB PCR product into pLOI3421 (134) pLOI4162 based Chromosomal Modification Procedure The two step recombination method for making targeted chromosomal modification using the cat sacB cassette from pLOI4162 is gener ally described in this study (Figure 3 1) and in a previous report (51) However, as the method may be useful to those less familiar with the techniques involved, a few aspects are emphasized here. Careful design saves time later when designing PCR primers and planning construction of tools for gene deletions and replacements. The pLOI4162 method allows much flexibility in the selection of the chromosomal sequence to be deleted or replaced. Unlike the one step PCR method (30) where practical oligonucleotide synthesis limits the length of the homologous flanking sequence used for recombination to less than 100 bp, the fla nking region can be much longer. Longer homologous flanking sequence provides gr eater recombination efficiency and decreases the likelihood of integration into an undesi red location. F lanking sequences of approximately 200 to 500 bp are generally used with this method. The product yield from the inside out PCR step during plasmid construction may be negatively affected by designing very long flanking sequences. The pCR2.1 vector, which is used for PCR cloning the region of interest, is 3.9 kbp. Therefore fl anking sequences of 1 kbp each woul d require PCR amplification of a 6 kbp sequence, for example. While such reactions are done routinely, using much longer flanking sequences might make plasmid construction and

PAGE 68

68 recombination fragment preparation more diffi cult, without significantly affecting the efficiency or fidelity of recombination. The multiple cloning sites containing the Sma I and Sfo I restriction sites used to remove the cat sacB cassette from pLOI4162 contain extensive secondary structure, which ma y partially block restriction enzyme access. Yields from the double digest are generally around 50%. The inefficiency of the restriction digest makes it important to scale up to avoid the need for repetition. Gel purification at this step may help ligation efficiency, but is unnecessesary for downstream steps due to selection for desired transformed ligation products. Transformants containing the first step plasmid ( cat sacB cassette ligated into the inside out PCR product) can be selected for with a combin ation of kanamycin and chloramphenicol and screened for sucrose sensitivity. After PacI digest and self ligation (for deletion) or ligation with replacement gene, LB (no NaCl) kanamycin plates containing 6% sucrose should be used to select for transformant s lacking the cat sacB cassette. Sucrose sensitivity with a single chromosomal copy of sacB is not strong enough to permit plate based selection without liquid based selection first. High copy number (pUC based plasmid) allows plate based sucrose selection and counterselection during plasmid construction. It is important to have a sufficient number of doublings prior to sucrose selection to allow complete chromosomal segregation. Outgrowth time depends on the growth rate of each strain, likewise time before plating from LB sucrose liquid based selection. Low second step efficiency is generally due to problems in construction or due to high frequency of recombination (or lack of complexity) in target sequence region. If few of the colonies from the cat sacB r eplacement step have the desired genotype, reanalyze target sequence region and design new primers for reconstruction of tools. Target sequence problems aside, given enough time for chromosomal

PAGE 69

69 segregation and sucrose (liquid) selection, the second step ca t sacB replacement is highly efficient. Applications of pLOI4162 Method The above described pLOI4162 method was used in all chromosomal gene deletions and replacements in our study, with the exception of the atp deletion strains. Additionally the method h as been used for improvement of a succinate producing strain of E. coli (51) Several other projects in our lab oratory including those focusing on production of malate and ethanol are also using this method in their strain constructions. Recently, two papers were published describing the application of the pLOI4162 method in the modification of Enterobacter asburiae for the production of ethanol (13) and of optically pure D( ) lactic acid (12) demonstrating the utility of the method in the engineering of other, related organisms. Other res earch groups have requested pLOI4162 since the publication of the method in the succinate production improvement paper (51) Plasmid pL OI4162 promises to be a broadly applicable tool for facilitating chromosomal modification for basic and applied research. Applications of pLOI4151 The pLOI4162 method has been widely adopted by our lab for current and future chromosomal engineering effort s. However, prior to the introduction of pLOI4162, we used the FLP recombinase catalyzed removal of FRT flanked antibiotic markers (30) That method left FRT con complicated further chromosomal engineering efforts and also created problems due to some governmental regulations and industry preferences limiting the amount of foreign DNA se quence in biocatalyst strains. The pLOI4162 method can be used to remove the sequence artifacts if the genomic context is known. However, in a few cases, due to rearrangement or other issues, this context was not known and another approach was desired. pLO I4151 was designed to provide

PAGE 70

70 either a cat sacB or a sacB aac cassette containing, non replicating circle, upon removal of the origin of replication and unwanted antibiotic resistance gene. These circles are then used in a FLP recombinase catalyzed, single crossover homologous recombination, to integrate the cassette into the site. If the exact location in the genome is unknown, cassette specific primers can be used to sequence the flanking regions. A library of the genomic DNA from the integration strain c an be made for isolation of a plasmid containing the cassette, scar and flanking sequence. Inside out PCR and self ligation can be used to remove the cassette and scar sequence. The resulting plasmid can be used in a manner identical to the second step rec ombination of the pLOI4162 method, for removal of the cassette and scar sequence from the chromosome. The pLOI4151 based circle method was used to remove all foreign DNA sequence from a lactate producing strain of E. coli (139) pLOI4151 also served as selectable and counterselectable marker sequence donor for modified use in the m etabolic engineering of E. coli for succinate and malate production (50) as well as for production of L alanine (137) and other compounds.

PAGE 71

71 Table 3 1. Plasmids and PCR primers Plasmid/ prime r Relevant features Reference/source Plasmids pEL04 cat sacB (116) pLOI2228 FRT cat FRT (79) pLOI2511 b la ,FRT kan FRT (119) pLOI4131 bla ,FRT cat FRT (51) pLOI4145 bla ,FRT cat FRT (51) pLOI4146 bla cat sacB (51) pLOI4151 bla FRT cat sacB aac FRT (Figure 3 3) (137) pLOI4152 bla cat sacB (51) pLOI4153 bla SfPBXS polylinker (51) pLOI4154 bla Pac I digested pLOI4161, self ligation (51) pLOI4161 bla cat sacB Pac I cassette (51) pLOI4162 bla cat sacB Pac I cassette (Figur e 3 2) (51) Primers JM 4161sense ACCGCATCAGGCGCCTAATTAATTAATCCC GG (51) JM 4161comp CCGGGATTAATTAATTAGGCGCCTGATGCG GT (51) JMpEL04F1 CAGCAGATCTAAGTAAATCGCGCGGGTTTG (51) JMpEL04R1 CAGCAGATCTAGCGGCTATTTAACGACCCT (51) JM catsacB up3 TGTGCTGCAAGGCGATTAAG (51) JM catsacB down3 TTCGATCACGGCACGATCAT (51) SfPBXSsense ATGTAGGCGCCATTAATTAATGGATCCACT ATCTCGAGATTAATTAATCCCGGGACTAT (51) SfPBXScomp ATAGTCCCGGGATTAATTAATCTCGAGATA GTGGATCCATTAATTAATGGCGCCTACAT (51)

PAGE 72

72 Figure 3 1. A simple method for making chromosomal deletions or replacements. Steps 1 through 4 describe the integration of the cat sacB cassette into the region for deletion or replacement. The cat sacB cassette is produced b y digesting pLOI4162 with Sma I and Sfo I. The inside out PCR uses a proofreading polymerase that leaves blunt ended products. Steps 5A and 6A describe an option for the second recombination step in which a deletion can be made, leaving a 18 bp sequence cont aining stop codons in all reading frames. If a gene replacement is desired, the gene of interest can be amplified by PCR with Pvu I recognition sequences in the primers. Pvu I digestion provides compatible cohesive ends for ligation into the Pac I digested p lasmid. Steps 5B and 6B describe another option, which produces a seamless deletion. From (51) p. 886 (Figure 1), with permission

PAGE 73

73 Figure 3 2. Plasmid map of pLOI4162. Ap, beta lactamase gene; cat chloramphenicol acetyltransferase gene; sacB levansucrase gene. The complete sequence for this plasmid is available from GenBank (accession number EU531506)

PAGE 74

74 Figur e 3 3. Plasmid map of pLOI415 1. bla beta lactamase gene; cat chloramphenicol acetyltransferase gene; sacB levansucrase gene; aac apramycin acetyltransferase gene.

PAGE 75

75 CHAPTER 4 METABOLIC ENGINEERIN G OF E COLI ATCC 8739 FOR INCREA SED CURRENT AND C OULOMBIC EFFICIENCY IN AN ELECTROCHEMICA L CELL Introduction Microbial fuel cell technology offers many advantages as a means of converting stored chemical energy from organic materials into electrical energy. Due to global concern about limited fossil fuel resources and about the environmental impact of current practices, there is growing interest in energy efficiency and sustainability. MFCs can be important tools for creating a more efficient and sustainable energy infrastructure. Direct, microbially cata lyzed conversion of stored energy from almost any organic material into useful electrical energy provides versatility and a shorter carbon cycle. MFCs can increase efficiency through the elimination of the energy wasting intermediate steps required by most current methods of electricity production from stored chemical energy. However, widespread adoption of MFC technology is presently limited by low power output (92) Even for stationary generation of electricity from low or negative value feedstocks, such as wastewater treatment, low power output make s cost justification difficult. Much work is being done to improve MFC design to increase power output (1, 38, 69, 100, 101) While MFC design optimization is important, a major rate limiting factor in MFC power production is microbial catabolism of the substrate (88) coupled with the transfer of liberated electrons to the electrode. Since the discovery of mediatorless electron transfer to electrodes (59) very little work has been published on thorough study and e ngineering of MFC microbial biocatalyst metabolism. The bulk of the biological studies have focused on classification of the electrogenic microbial consortia enriched for in an MFC (49, 60, 61, 93, 99) MFC power increases should be possible through the engineering of microbial metabolic pathways and electron transport systems for MFCs, particularly in conjunction with MFC design optimiz ation.

PAGE 76

76 Our study endeavors to develop E. coli as a model for such improvements. E. coli was selected for its broad substrate range, ease of genetic manipulation, minimal nutrient requirements, and extensive background of physiological and genetic informati on. E. coli is capable of rapid catabolism of glucose, particularly under fermentative conditions. When no external electron acceptor is present, E. coli derives ATP from substrate level phosphorylation and achieve s redox balance by reducing endogenous me tabolic intermediates such as pyruvate in a fermentative metabolism. High glycolytic flux compensates for the lower ATP yield in the absence of oxidative phosphorylation. If electrons are harvested from reduced cofactors (such as NADH) in a microbial fuel cell (MFC), this high glycolytic flux can produce high electrical current. However, under anaerobic conditions, E. coli incompletely oxidizes glucose, resulting in a low electron recovery (low Coulombic efficiency) and side product formation, which can lim it the long term stability of MFC operation. For an MFC, both high rate and high Coulombic efficiency are desired. In order to develop an improved E. coli strain for bioelectricity production, we have used the targeted gene deletion and replacement strate gies described in the previous chapter to increase the rate and efficiency of reducing equivalent production. We focused on lowering cellular ATP/ADP and NADH/NAD + ratios in order to increase flux through the central metabolic pathways, while eliminating a cetate overflow metabolism and pathways competing for NADH. In addition to the targeted chromosomal modification approach, aerobic batch transfers in a pyruvate mineral salts medium were used to select for increase d flux through the PDH complex and through the TCA cycle.

PAGE 77

77 Materials and Methods Growth Conditions and Media Unless otherwise specified, all working cultures of E. coli ATCC 8739 ( 51) and its derivatives were grown in a minimal MOPS (4 morpholinopropanesulphonic acid, 0.1 M, pH 7.4) buffered NBSM medium previously described (20) containing 0.05 M glucose. Baffled 250 ml flasks were used for aerobically grown batch experiments with 50 ml culture volume at 37 C and 250 rpm agitation. In aerobic pyruvate transfer experiments, 0.1 M sodium pyruvate was used in place of gl ucose. LB broth and LB agar plates were used during plasmid and strain construction. Glucose was added (20 g/liter) to the LB broth and plates in most cases in strains containing the atpFH deletion. L arabinose was included (20 g/liter) in the medium to in duce expression of the Red recombinase genes on pKD46. LB broth with NaCl omitted and containing 100 g/liter sucrose was used in selection for loss of sacB (encoding levansucrase). Similarly plate based selection for removal of sacB used LB, no NaCl, 60 g/ liter sucrose. Antibiotics were included, as necessary, during plasmid and strain construction (kanamycin, 50 mg/liter; ampicillin, 100 mg/liter; chloramphenicol, 40 mg/liter). Genetic Methods Construction of strains used in this study was carried out as described previously (51) and in the preceding chapter, using the cat sacB cassette from plasmid pLOI4162 and two step, Red recombinase catalyzed double crossover homologous recombination for markerless deletions and integrations. Additional bacterial strains, plasmids and PCR primers, discussed in this section and not listed earlier, are listed (Tab le 4 1). To eliminate pathways that cou ld compete for NADH and limit current production, pflB adhE and ldhA were deleted from JC27 ( atpFH ) to produce JC62. Briefly, each gene was amplified by PCR from ATCC 8739 genomic DNA and cloned into pCR2.1 TOPO. Each of

PAGE 78

78 these plasmids was used as templ ate for an inside out PCR ( Pfu polymerase) with gene specific cat sacB cassette from pLOI4162 was ligated into the inside out PCR product and the ligation was transformed sucrose. This plasmid construct was linearized with an appropriate restric tion enzyme, diluted 1000 fold, and amplified by PCR, using the original gene cloning primers. The product was electroporated into the parent strain, which was induced for expression of the Red recombinase genes from pKD46. Successful integrants were chlor amphenicol resistant and ampicillin sensitive, after curing of pKD46 at 39 C or 40 C. The cat sacB cassette containing plasmid construct was Pac I digested to remove the cassette and self ligated. The resulting plasmid was digested, diluted and PCR amplifie d for use in a second homologous recombination step, to remove the cat sacB from the chromosome. T he replacement of the cassette resulted in an 18 bp sequence containing stop codons in all reading frames. The same procedure as used to delete ackA from JC62 to make JC66, to prevent flux to acetate via phosphotransacetylase and acetate kinase. The native lpd (encoding dihydrolipoamide dehydrogenase) in JC101 was replaced by a copy ( lpd101 ) (62) containing a single nucleotide substitution, making an acidic to basic amino acid change (E354K). ATCC 8739 genomic DNA was used as the template for PCR amplification of lpd The product was cloned into p CR2.1 TOPO to produce pLOI4179. Complementary mutagenic oligonucleotides JME354Ksense and JME354Kcomp were used to introduce the lpd101 mutation using linear plasmid amplification (51) to produce pLOI4180. Several of the resulting clones were verified by sequencing. The mutated gene was amplified by

PAGE 79

79 PCR and used to replace the cat sacB cassette as described above. The lpd101 mutation in the new strain, JC80, was verified by sequencing. Reoxidation of NADH during aerobic respiration can be rate limiting and it is likely to be more so in an MFC, with a less positive (than oxygen) electron acceptor. The gene encoding a water forming NADH ox idase from Streptococcus mutans ( naoX Sm ) (80) was PCR amplified with primers containing Pvu naoX F1 and JM naoX R1) an d cloned into pCR2.1 TOPO The gene was integrated into two chromosomal loci, providing different levels of transcription from chromosomal promoters (or from its own promoter). A variation of the two step homologous recombination procedure (described above ) was used in the integrations. During construction of the plasmid for the second recombination step ( cat sacB replacement), Pvu I digested naoX Sm PCR product was ligated into the Pac I digested ( cat sacB cassette removed) vector. In this way, naoX Sm was int egrated into ackA in JC62 to produce JC68 and in JC80 to produce JC85. The integrated naoX Sm in JC68 was oriented in the opposite orientation to ackA such that the gene was not transcribed from the ackA promoter. JC85 had naoX Sm integrated into ackA so th at it could be transcribed from the ackA promoter. To increase expression levels of naoX Sm it was also integrated into pflB in JC85 (under focA pflAB operon promoter transcriptional control) to produce JC93. Electrochemical Analytical Methods Coulombic yi elds from glucose were measured using a three electrode, poised potentia l electrochemical cell (Figure 4 1). The cell was a bulk electrolysis cell (BASi, West Lafayette, IN) with a reticulated vitreous carbon working electrode, a platinum wire counter elec trode in a fritted glass chamber and an Ag/AgCl reference electrode. The anode potential was poised at +100 mV vs. Ag/AgCl using eDAQ (Colorado Springs, CO) EA161 potentiostats connected to an eDAQ e corder 821 and Chart software. The anolyte was stirred w ith a small magnetic stir bar

PAGE 80

80 at 250 rpm and the headspace of the working electrode chamber was continuously flushed with 100 ml/min argon, to maintain anaerobic conditions. The anolyte was composed of NBS medium (20) containing 0.1 M MOPS and 0.1 M NaCl, with 0.001 M thionin Measurement of current began with 60 ml anolyte. Aerobically grown cells (50 ml) were harvested (5000g, 5 min, 22 C) at stationary phase (16.5 h culture from 3 mg/liter initial cell density). The cells were washed once (30 ml anolyte, lacking thionin), centrifug ed again and resuspended to a cell density of 50 mg in 15 ml anolyte The cell suspension was added to the wor king electrode chamber (upon stable current baseline) and the current was monitored until a ll charge transfer from endogenous reserves completed and a stable baseline (no change for at least 30 min) or an arbitrarily determined 1 mA threshold level was rea ched. At this point, anolyte containing 0.01 M glucose was added by syringe and the resulting current was measured over time until it returned to the starting level of the first glucose pulse and the next pulse was added. In this way, multiple pulses of gl ucose were added and the Coulombic efficiencies determined by integrating the area under each peak above the threshold value (the current immediately before addition of glucose) and dividing by the theoretical Coulombic yield from the amount of glucose add ed in each pulse. All additions to the working electrode chamber were pre sparged with at least 100 volumes of argon and were transferred anaerobically to the vessel using a stoppered flask and positive argon pressure to initiate the transfer (argon sparge d glucose transferred by 1 ml syringe). NADH Oxidation Assay Total cytoplasmic NADH oxidase activity was measured to determine the effect of integrations of naoX Sm into ackA and pflB Aerobically grown cells (20 ml of a 25 ml, ~0.33 g/liter culture in a 2 50 ml flask ) were harvested (5000g, 5 min, 4 C ) after chilling briefly on ice. The cells were washed once with cold TM buffer (0.05 M Tris HCl, 0.01 M MgCl, pH 7.5)

PAGE 81

81 and resuspended in TM buffer (1 ml, containing 0.01 mg/liter DNase I). The cells were lyse d by mechanical disruption using a Fast Prep 24 system from Molecular BioProducts (San Diego, CA). NADH oxidase assay reactions contained NADH (0.25 mM) in TM buffer. The reaction was initiated by the addition of cell lysate (0.1 ml in a 1 ml reaction vo l ). NADH oxidation was followed at A 340 at room temperature. Protein concentration of the lysates was determined using the bicinchoninic acid method (BCA Protein Assay Kit, Thermo Scientific, Rockford, IL) with bovine serum albumin as a protein standard. S pecific activity (nmol min 1 mg protein 1 ) was calculated using the molar extinction coefficient for NADH (6220 M 1 cm 1 ). Results Elimination of Competition for NADH In the presence of an electron acceptor with a sufficiently positive reduction potential, NADH should be reoxidized quickly enough to prevent the involvement of fermentative pathways during sugar metabolism in E. coli The fermentative enzymes are affected by cellular redox state through regulation of their gene expression, through direct or i ndirect inactivation of catalytic activity, and through K m values for NADH that are lower than the NADH dehydrogenases Ideally, NADH reoxidation should not be rate limiting in an MFC. However, the anode potential is not as positive as oxygen, due to overp otentials, and it is subject to fluctuation if it is not kept constant with a potentiostatic system. Production of fermentation products in an MFC lowers Coulombic efficiency and their accumulation can limit long term current stability. Therefore, it is de sirable to eliminate the primary fermentation pathways through targeted gene deletion. Genes encoding alcohol dehydrogenase ( adhE ), lactate dehydrogenase ( ldhA ), and pyruvate formate lyase ( pflB ) were deleted to prod uce JC62 (Figure 4 2), using the markerl ess gene deleti on method described in a previous chapter This method

PAGE 82

82 left only an 18 bp sequence (containing stop codons in all reading frames) in each deleted region and was developed to permit multiple chromosomal modifications within a single strain, w ithout increasing the likelihood of undesired rearrangements and without hindering the introduction of further modifications. Gene deletions were verified by PCR and by phenotype. Deletion of adhE ldhA and pflB did not have a significant effect upon growt h or product formation (data not shown) under aerobic conditions, nor did it significantly affect current production or Coulombic yield in poised potential electrochemical cell testing (T able 4 2). Deletion of ackA The higher rate of glycolysis in atpFH de letion strains leads to increased acetate production under aerobic conditions. This overflow metabolism is due to the inability of the respiratory chain to maintain redox balance. The redox imbalance limits flux through the PDH complex and through the TCA cycle. Production of acetate, via phosphotransacetylase and acetate kinase activity, yields one ATP per pyruvate and prevents the buildup of reducing equivalents that would otherwise occur through further oxidation of pyruvate via the TCA cycle. Glucose c atabolism to acetate by this pathway yields four reducing equivalents, while complete oxidation of the glucose by the TCA cycle adds eight more. This overflow mechanism results in almost ten percent of glucose carbon going to acetate in aerobically grown J C27 cultures. That much incompletely oxidized side product represents a six percent reduction in Coulombic yield. The decrease in Coulombic yield due to acetate production could be even greater under MFC operating conditions. An ackA deletion was made (in JC62) to decrease acetate production, resulting in JC66. Th is strain produced less acetate. H owever, the downstream rate limitation remained and pyruvate accumulated as a result of rate limitation at the PDH complex node (data not shown). Elimination of t he phosphotransacetylase/acetate kinase pathway created the backup of carbon to

PAGE 83

83 the pyruvate node because available catabolic pathways from pyruvate caused a redox imbalance. Oxidation of pyruvate to acetate via pyruvate oxidase yields no ATP and involves the transfer of electrons to ubiquinone. This process would decrease the amount of ubiquinone available to reoxidize NADH (catalyzed by NADH dehydrogenase) from glycolysis. Flux through the PDH complex and the TCA cycle would provide one ATP per pyruvate, but would also create a redox imbalance through production of excess reducing equivalents. Excess NADH limits flux through this pat KGDH complexes (62) and of other TCA cycle enzymes, particularly citrate synthase (4) Heterologous Expres sion of naoX Sm The water forming NADH oxidase from S. mutans catalyzes the reoxidation of NADH in the cytoplasm via a two electron transfer to molecular oxygen (48) Heterologous expression of a gene (from S. pneumoniae ) encoding a water forming NADH oxidase in E. coli increases flux through the PDH complex and TCA cycle by helping to overcome the rate limitation of NADH reoxidation by the respiratory chain (124, 125) Initially, naoX from S. mutans was cloned into pCR2.1 TOPO and expressed in JC62 to evaluate its effect on growth and metaboli sm (Figure 4 3). The control strain (JC62/pCR2.1) produced a maximum of 17.7 mM pyruvate and 23.9 mM acetate from 50 mM glucose in aerobic NBSM batch culture, while JC62 transformed with pLOI4174 ( naoX Sm ) accumulated a maximum of 3.9 mM pyruvate and 23.9 m M acetate. This result suggested that expression of naoX Sm in JC62 allowed increased flux through the PDH complex and to a lesser extent, through the TCA cycle, consistent with the previous studies (124, 125) However, much of the acetyl CoA was still converted to acetate, indicating that citrate synthase remained as a rate limiting step for complete glucos e oxidation via the TCA cycle. JC62 retained an intact Pta/AckA pathway. In order to be able to harvest all electrons from glucose through its complete oxidation by the TCA cycle, the carbon flow to acetate needed to

PAGE 84

84 be redirected. ackA was replaced with naoX Sm by chromosomal integration to eliminate the primary route to acetate, while simultaneously enabling higher flux through the PDH complex and perhaps the TCA cycle. The resulting strain, JC68, grew slightly faster than JC66 ( ackA ) presumably due to higher flux through the TCA cycle (Figure 4 4). The ATP available via succinyl CoA synthetase of the TCA cycle could help to offset the ATP that was no longer available from acetate production in the ackA strains. Despite the growth rate difference, both JC66 and JC68 accumulated similar concentrations of pyruvate. Without the overflow pathway to acetate, a redox imbalance was likely caused by higher flux through the TCA cycle. The NADH produced by the TCA cycle was not reoxidized fast enough, despite the presence of the water forming NADH oxidase in JC68. A high NADH/NAD + ratio negatively controls PDH activity (62) and the negative redox feed back control circuit caused the pyruvate accumulation. Different levels of naoX Sm expression were tested by integrating the gene in different orientations in ackA and by integrating an additional copy into pflB (gene replacement). JC68 had naoX Sm replacing ackA but in the opposite orientation, so that it was not transcribed from the ackA promoter. Expression of naoX Sm in JC68 was presumed to be from its own promoter. The gene was replaced in the other orientation during construction of JC85, to allow its t ranscription from the ackA promoter. Next, an additional copy was used to replace pflB under the control of the focA promoter, resulting in JC93. NADH oxidase (NOX) activity of the soluble fractions from these strains was measured to assess the relative e xpression of naoX Sm (Figure 4 5). JC68 and JC92 (JC85, pflB::cat sacB ) soluble fractions had almost twice the NOX activity of JC62, and JC93 had approximately three fold higher NOX activity than JC62. The NOX activity measured for the JC62 soluble fraction may have been due to other soluble dioxygenases or fr om other native activities. As will be described, JC85 and JC93 had additional chromosomal

PAGE 85

85 modifications and their progenitors had been selectively transferred, making direct metabolic comparisons of JC92 and JC93 with JC68 difficult. Aerobic Batch Transf ers in Pyruvate for Increased TCA Cycle Flux The pyruvate accumulation in aerobically grown JC68 cultures was indicative of negative control of PDH by an elevated NADH/NAD + ratio. The preferred solution to this problem would be to more rapidly reoxidize N ADH, by using the water forming NADH oxidase. The NADH oxidase activity in JC68 was not sufficient. JC68 was therefore serially transferred in NBSM pyruvate (aerobic batch cultures) to select for changes that would permit higher flux through the TCA cycle despite the high NADH/NAD + ratio. This selection was based upon the extra ATP available per pyruvate oxidized via the TCA cycle. Initial doubling times of the JC68 cultures grown in pyruvate w ere approximately 16 h (Figure 4 6A). After 20 transfers the dou bling time had de creased to just over 3 h and the resulting strain was named JC100. After construction of arcA ), doubling times increased to 7 h. Another 20 transfers (no significant change after transfer 10) brought the generation time down to that of JC100 (Figure 4 6B). The resulting strain (JC101) was tested in aerobic batch culture and found to completely oxidize glucose without the accumulation of pyruvate observed for the previous strains. Introduction of lpd101 for Lower Sensitivity of PDH an KGDH Complexes to NADH Kim et al (62) describe a mutation in lpd (encoding dihydrolipoamide dehydrogenase) which causes the PDH complex to be less sensitive to NADH. Since dihydrolipoamide KGDH) complex, the mutation ( lpd101 ) may also allow greater flux through the TCA cycle when NADH levels are elevated. Evolved strain JC101 lpd was sequenced to determine whether such a mutation could explain the higher flux observed through the PDH complex. The lpd sequence of JC101 was identical to that of ATCC 8739.

PAGE 86

86 A single nucleotide substitution (G to A) was made in ATCC 8739 lpd by an oligo directed mutagenesis method, resulting in an E354K amino acid change. The mutated copy was used to replace the native copy in JC101, resulting in JC80. Aerobic JC80 batch culture growth and glucose catabolism was similar to the parent (data not shown). This was expected, because JC101 was already capable of completely oxidizing glucose aerobically, with no significant product formation. The lpd mutation could not provide any additional benefit under these conditions, but might under more reducin g conditions (such as in an MFC). Deletion of arcA An additional approach used to circumvent negative control by NADH was to delete arcA Since the phosphorylated ArcA negatively regulates expression of the genes e ncoding the PDH complex (as well as many T CA cycle genes) under reducing conditions, deletion of arcA can allow higher aerobic flux through PDH and the TCA cycle in conjunction with heterologous expression of naoX (125) JC72 was constructed by deleting arcA from JC100. JC72 accumulated a small amount of pyruvate during aerobic growth on glucose. The pyruvate accumulat ion was eliminated by aerobic transfers in pyruvate mineral salts medium (JC101). Comparison of Contribution of Individual Genetic Modifications to Complete Aerobic Glucose Oxidation Starting with JC85, individual chromosomal modifications were repaired on e at a time to provide an isogenic set for determination of the relative contribution of each change to the observed phenotype. JC86 grew similarly to ATCC 8739, due to repair of the atpFH deletion (Figure 4 7). Repair of the arcA deletion (JC91) resulted in increased pyruvate accumulation during aer obic growth on glucose (Figure 4 8B). Repair of the ackA deletion (JC89) resulted in accumulation o f acetate, as expected (Figure 4 8C). Apparently, integration of naoX Sm into ackA (JC85 vs JC84) had little effe ct (Figures 4 7 and 4 8), consistent with the NADH oxidase assay

PAGE 87

87 results (Figure 4 5). The lpd101 mutation (JC85 vs JC90) had no effect under the tested conditions. Current Production and Coulombic Efficiencies Using Engineered Strains in Poised Potential Electrochemical Cell The engineered strains were tested in a three electrode, poised potential electrochemical cell, with glucose as the electron donor (and sole carbon source) and using thionin as a diffusible mediator. The poised potential, bulk electrol ysis cell system was chosen to limit non biological variables. A high surface area RVC electrode and excess mediator were used to further ensure that bacterial activity would be limiting. As glucose was catabolized by the bacterial cells, electrons were tr ansferred to the anode (via the thionin) from reduced cofactors in the cells. The resulting current was measured over time (Figure 4 9A F) and used to compare maximum current and Coulombic effi ciencies of the strains (Table 4 2). Maximum current values fro m ATCC 8739, JC27 and JC62 were similar. C E s were also similar (approximately 50%). Maximum current and C E values f or all tested strains with the naoX Sm integration (JC85, JC91 and JC93) were similar. Maximum current values for these strains were approxima tely 20% higher than wild type and C E s were 75%. Discussion We have cons tructed a strain (JC93, Figure 4 10) that is capable of significantly higher current output and Coulombic efficiency in the conversion of the stored energy in glucose to electricity. W hile an atpFH deletion (JC27) increased the rate of glycolysis, downstream rate limitation prevented the complete oxidation of glucose. The incomplete oxidation of glucose by JC27 was evident in aerobic culture (Figure 2 3) with the production of acetate. In the electrochemical cell, the incomplete catabolism of glucose was observed as a low Coulombic yield. We attempted to overcome the downstream metabolic rate limitations through several

PAGE 88

88 targeted chromosomal modifications and via aerobic transfers in a py ruvate minimal medium. We initially tested the modified strains in aerobic batch culture. We hypothesized that these conditions might be similar to an anaerobic electrochemical cell (EC) environment with a sufficiently positive reduction potential and the aerobic batch cultures were simpler for preliminary investigations. The poised potential electrochemical cell experiments were in good general agreement with the results obtained from the aerobic batch experiments. Integration of naoX Sm into ackA appeared to have the greatest effect upon redirecting flux through the TCA cycle and upon increasing the Coulombic yield. No significant change in current of C E was observed upon integration of an additional copy of naoX Sm However, the additional NOX activity coul d be beneficial under different conditions. The increase in MFC current with NOX activity suggests that reduction of thionin via the native electron transport chain was limiting and that the cytoplasmic NADH oxidase helped to overcome that limitation. Simi larly, the elimination of aerobic overflow metabolism upon introduction of the cytoplasmic NOX implicates a respiratory limitation as the primary cause of the overflow. Interestingly, deletion of arcA appeared to result in higher flux through PDH and the TCA cycle un der aerobic conditions (Figure 4 8B), but did not significantly affect maximum current and C E in t he electrochemical cell (Table 4 2). ArcA P is known to repress transcription of some genes involved in anaerobic respiration while it activates t ranscription of genes involved in aerobic metabolism. It is possible that some of the genes involved in anaerobic respiratory metabolism are important under the tested EC conditions. It would be interesting to determine whether restoring arcA in JC93 provi des any benefit. The metabolic engineering of JC93 was done in a way that complicated comparison of the effects of individual modifications. In particular, the aerobic pyruvate transfers that were done in between targeted modifications made

PAGE 89

89 the genetic bac kgrounds of subsequent strains uncertain. The decreases in generation times resulting from the transfers were likely due to subtle flux redistributions and readapta t ion to mineral salts medium culture conditions after chromosomal modifications were made (i n rich media). In order to make strain comparisons, it was necessary to restore individual modifications in JC85. It may be useful to return to JC62 and integrate naoX Sm into ackA and into pflB (or another site that would provide a higher level of expressi on). Transfers could be made (to select for higher C E if necessary) after the changes were made in the cleaner genetic background. The results of our study suggest that oxygen is not necessary for the complete oxidation of glucose, only sufficiently oxidi zing conditions are required. It also provides further evidence that the respiratory chain limits PDH complex and TCA cycle flux and leads to overflow metabolism atpFH ). The higher rate of NADH oxidation (by oxygen or by thionin) catalyzed by the cytoplasmic NOX from S. mutans alleviates negative control of downstream catabolic steps in JC93 and allows a high er Coulombic yield at a high metabolic rate. These attributes make JC93 an attractive candidate for application in a mediated MFC or in a mediatorless MFC after introducing heterologous genes for extracellular electron transfer.

PAGE 90

90 Table 4 1. Strains, plasmids, and PCR primers Strain/plasmid/primer Relevant features Reference/source E. coli JC62 pflB, adhE, ldhA This study JC66 ackA This study JC68 ackA :: naoX Sm (opposite orientation) This study JC72 arcA This study JC80 JC101, lpd101 This study JC84 ackA :: cat sacB This study JC85 ackA :: naoX Sm This study JC86 JC85, atpFH + This study JC89 JC85, ackA + This study JC90 JC85, lpd wt + This study JC91 JC85, arcA + This study JC92 pflB :: cat sacB This study JC93 pflB :: naoX Sm This study JC100 JC68, 20 transfers +O 2 NBSM pyruvate This study JC101 JC72, 20 transfers +O 2 NBSM pyruvate This study Plasmids pLOI4147 adhE in pCR2.1 This study pLOI4148 adhE :: cat sacB (from pLOI4162) This study pLOI4149 adhE ::6RFstop This study pLOI4158 ackA in pCR2.1 This study pLOI4159 ackA :: cat sacB (fr om pLOI4162) This study pLOI4160 ackA ::6RFstop This study pLOI4163 pflB in pCR2.1 This study pLOI4164 ldhA in pCR2.1 This study pLOI4165 ldhA :: cat sacB (from pLOI4162) This study pLOI4166 ldhA ::6RFstop This study pLOI4171 pflB :: cat sacB (from pLOI416 2) This study pLOI4172 pflB ::6RFstop This study pLOI4175 ackA :: naoX Sm This study pLOI4176 arcA in pCR2.1 This study pLOI4177 arcA :: cat sacB (from pLOI4162) This study pLOI4178 arcA ::6RFstop This study pLOI4179 lpdA in pCR2.1 This study pLOI4180 lpd1 01 This study

PAGE 91

91 Table 4 1. Continued Plasmid/primer Relevant features Reference /source pLOI4181 lpdA :: cat sacB (from pLOI4162) This study pLOI4182 naoX Sm in pCR2.1 This study pLOI4184 ackA :: naoX Sm This stu dy pLOI4185 ackA :: naoX Sm (opp. orient.) This study pLOI4190 pflB :: naoX Sm This study Primers JM adhE up TCGCTGAACTTAACGCACTC This study JM adhE down ACGACCGTAGTAGGTATCCA This study JM adhE iodown2 GACCGTACTGCTGCTAAGAT This study JM adhE ioup2 GTGTCGTCTTCAG ACAGAAC This study JM ackA F1 GCCTGAAGGCCTAAGTAGTA This study JM ackA R1 GCACGATAGTCGTAGTCTGA This study JM ackA down1 GCCGCAATGGTTCGTGAACT This study JM ackA up1 GTTGAGCGCTTCGCTGTGAG This study JM pflB F1 CCGGTTACGATCGGCAACAT This study JM pflB R1 TCGAAGGCTACGT CGAGTCT This study JM pflB down1 ATGCACGGTCGTGACCAGAA This study JM pflB up1 GGAAGCAACAGCGGTGTCAA This study JM ldhA F1 AAGGTTGCGCCTACACTAAG This study JM ldhA R1 GCGATGATGCTGTAGCTGTT This study JM ldhA down1 CGCGTCAAGGTCGACGTTAT This study JM ldhA up1 TCTCAGGCA GCAATTGAAGC This study JM arcA F1 GCTCAACTCTGCCGATAGC This study JM arcA R1 CAACTTATTACGCGGTGCGA This study JM arcA down1 TATCGCTTCTGCGGTGATCT This study JM arcA up1 CGTGTTACCAACTCGTCTTC This study lpdA A first 18 bases of lpdA ORF Sigma lpdA C last 18 bases of lpdA ORF Sigma JME354Ksense CGTCCATCGCCTATACCAAACCAGAAGTTGCATGG This study JME354Kcomp CCATGCAACTTCTGGTTTGGTATAGGCGATGGACG This study JM lpdA down1 TTCTGGCCGTGCTATCGCTT This study JM lpdA up1 AGCAGGCGTTCTGGTACTTC This study JM naoX F1 CACGATCGGCTAGCAATC AGGAGCTTAT This study JM naoX R1 CACGATCGGCTGTGGTTCTCTTAGAAGT This study

PAGE 92

92 Table 4 2. Coulombic efficiencies and maximum current from engineered strains in glucose fed bulk electrolysis cell ( BEC ) testing Strain Coulombic efficiency (%) Maximum curren t (mA) ATCC 8739 48.6 2.8 3.90 0.23 atpFH ) 55.9 10.1 3.96 0.67 ldhA adhE pflB ) 53.0 4.3 3.65 0.19 ackA :: naoX Sm ) 81.5 16.2 4.48 0.26 JC85 (JC91 arcA ) 74.7 9.7 4.71 0.34 pflB :: naoX Sm ) 76.0 6.6 4.64 0.26 All values are reported as mean SD, n = 3

PAGE 93

93 Figure 4 1. Photograph (A) and expanded schematic (B) of the bulk electrolysis cell used in this study for electrochemical measuremen ts. Adapted from BASi product literature (Bioanalytical Systems, Inc., West Lafayette, IN)

PAGE 94

94 Figure 4 2. Summary of strain construction. ackA :: naoX *, naoX Sm not transcribed from ackA promoter.

PAGE 95

95 Figure 4 3. Decreased pyruvate accumulation with overexpression of naoX from Streptococcus mutans atpFH ldhA adhE pflB ) (A) vs. JC62 with empty vector (B )

PAGE 96

96 Figure 4 4. Part ial restoration, through naoX Sm integration (JC68) and pyruvate transfers atpFH ldhA adhE pflB ) growth rate after deletion of ackA (JC66)

PAGE 97

97 Figure 4 5. Cytoplasmic NADH oxidase activity from aerobically grown strains with different naoX Sm integrations. atpFH ldhA adhE pflB ), JC68 (JC62, ackA :: naoX Sm transcription from native naoX Sm promoter), JC92 ( ackA :: naoX Sm naoX Sm transcription from ackA promoter), JC93 (JC92, pflB :: naoX Sm naoX Sm transcri ption from focA promoter

PAGE 98

98 Figure 4 6. Aerobic batch culture arcA)(B) in pyruvate mineral salts med ium for shorter generation times. Final transferred strain designation indicated with an arrow

PAGE 99

99 Figure 4 7. Growth curves of aerobic glucose mineral salts medium cultures of isogenic set of strains with restoration of each chromosomal modif ication present in JC85. ATCC 8739 growth is included as a reference

PAGE 100

100 Figure 4 8. Glucose utilization and product formation of aerobic cultures (glucose mineral salts medium) of isogenic set of strains with restoration of each chromosomal modifica tion present in JC85

PAGE 101

101 Figure 4 9. Current production over time with glucose bulk electrolysis cell (poised potential at +100mV vs Ag/AgCl)

PAGE 102

102 Figure 4 10. Simplified metabolic overview of JC9 3. Uncoupled electron transfer to exogenously supplied redox dye (blue and white circles) via the native electron transport system (ETS) and NADH oxidase from S. mutans and proton leakage across cytoplasmic membrane shown in inset panel. Electron (e ) and ATP (~P) maximum theoretical net yields from substrates are given for JC93

PAGE 103

103 CHAPTER 5 HETEROLOGOUS EXPRESS ION OF GENES ENCODIN G EXTRACYTOPLASMIC CYTOCHROMES FROM DIS SIMILATORY METAL RED UCING BACTERIA IN E COLI FOR EXTRACELLULAR EL ECTRON TRANSFER Introduction MFC operation using E. coli as the biocatalyst requires the addition of exogenous, soluble electron carriers. The E. coli outer membrane lacks the terminal reductase cytochromes of the DMRB and prevents direct contact between insoluble exrace llular electron acceptors and the electron transport chain. E. coli does not produce its own soluble electron carriers as do Shewanella (78, 128) an d Pseudomonas (47, 98) species, for example. Due to these limitations, mediatorless E. coli MFCs do not produce significant electrical current. There have been a few reports of evolved populations of E. coli in MFCs which were capable of mediatorless electrode reduction (129, 136) but no independent confirmation has been reported. One study found quinones in the anolyte and concluded that selection for extracellular electron transfer had led to production and secretion of small molecule electron shuttles (97) However, it is also possible that the low levels of these compounds were present because of cell lysis. More evidence is needed to confirm extracellular electron transfer by evolved strains of E. coli Another approach to enabling mediatorless electron transfer by E. coli is the heterologous expression of electron carrier encoding genes from DMRB. S. oneidensis extends its electron transport chain to the extracellular environment via several multiheme containing cytochromes. The periplasmic decaheme cytochrome MtrA from S. oneidensis is capable of accepting electrons from the native NapC (a cytoplasmi c membrane located c type cytochrome involved in periplasmic nitrate reduction) in E. coli and reducing soluble ferric compounds (94) A separate study showed that S. oneidensis outer membrane terminal reductase OmcA can localize to the outer membrane of E. coli B and can be oxidized by insoluble ferric oxide (34) Proper

PAGE 104

104 localization of OmcA was absent in a K12 strain of E. coli E. coli K12 strains do not express the necessary gsp genes encoding the components of the type II secretion system (T2SS) under any teste d conditions (39) The gsp genes in E. coli K12 strains (and in ATCC 8739) are organized into two divergent ly transcribed operons (Figure 5 1). The histone like nucleoid structuring protein (H NS) reportedly binds to the region between the operons and represses their transcription (39) The T2SS of E. coli is responsible for secre tion of some exotoxins (114) and chitinase (39) It is similar to the Pul (pullulunase secretion) system of Klebsiella strains (31) the Out (pectate lyase secretion) system of Erwinia chrysanthemi (45) and the Gsp system of S. oneidensis (responsible for secretion of outer membrane cytochromes) (107) While T2SS systems from various bacteria share organizational and functional characteristics, most are specific for their substrates. Surprisingly, E. coli Type II secr etion is sufficient and necessary for functional localization of the heterologously produced OmcA (34) Reconstitution of a functional S. oneidensis MR 1 extra cellular electron transport system in E. coli would establish the minimal set of genes needed for reduction of insoluble acceptors. We attempted to achieve this goal and to enable mediatorless electrode reduction through genet ic engineering of JC85 (Figure 5 2). Materials and Methods Growth Conditions and Media E. coli cultures were grown as described elsewhere in this study, unless otherwise specified. Antibiotics were included in growth medium, where appropriate, to maintain plasmids. S. oneidensis MR 1 was acquired from ATCC (ATCC 700550) and grown aerobically at 30 C in LB broth, for genomic DNA extraction using the Qiagen DNeasy procedure.

PAGE 105

105 Integration of Heterologous Genes and Replacement of gsp Promoters A combinatorial approach was used in the chro mosomal integration of omcA and mtrCAB from S. oneidensis MR 1 into adhE and pflB of JC85. S. oneidensis MR1 genomic DNA was used as a template for PCR amplification of the region containing omcA and mtrCAB with the primer set JM omcA F2/JM mtrB R2 (all strain s, plasmids and primers introduced in th is chapter are listed in Table 5 1). The product was used for nested PCR to amplify the entire region (with primer set JM omcA F1/JM mtrB R1), omcA (JM omcA F1/R1), and mtrCAB (JM mtrC F1/ mtrB R1). These primers contained Pvu I restriction sites and Pvu I digests left ends that were compatible with Pac I digested ( cat sacB cassette removed) pLOI4171 and pLOI4148. Ligation of each Pvu I digested product into each Pac I digested plasmid resulted in pLOI4192 and pLOI5009 through pLOI5 013. These plasmids were linearized with an appropriate restriction enzyme, diluted and PCR amplified for use as integration fragments to replace the cat sacB cassette in the second step of the two step chromosomal sequence replacement strategy described p reviously (51) The entire omcAmtrCAB So region was integrated into pflB and adhE (JC94 and JC123, respectively). In another strain, omcA So was integrated into pflB and mtrCAB So into adhE (JC121). For construction of JC122, omcA So was integrated into adhE and mtrCAB So into pflB The gspA and gspC promoters in strains JC94, JC121, JC122 and JC123 were replaced with constitutive promoters. Th e entire intergenic region between gspA and gspC (containing the divergent promoters P gspA and P gspC and the H NS binding region) was replaced (in the same manner as the replacements of adhE and pflB ) by a construct containing divergent constitutive promot ers (P bla and P cat ). Inside out PCR was used (JM41XX cat modF2/R1 primer set) to amplify pLOI5000, omitting P cat The primers contained Nhe I sites and the product was self ligated after digest to produce pLOI5001. pLOI5000 was digested with Nde I and Nhe I, Kl enow fragment treated and self lig ated to produce pLOI5003. P lasmid pLOI5003 was amplified by

PAGE 106

106 PCR (JM41X5promF1/R1 primers), digested with Nhe I/ Cla I and ligated into compatible ends of PCR amplified (JM41X3 bla modprodF1/R1 primers) pLOI5001. The resulting plasmid (pLOI5002) contained bla and cat genes, divergently transcribed, with a minimal sequence between them. gspA and gspC and the intergenic region between them were amplified by PCR (JMgspCAF2/R2 primers) from ATCC 8739 genomic DNA and c loned into pCR2.1 TOPO (resulting in pL OI5006). A ppropriate constructs were made (in the manner described above and using primers JM gspA up1/JM gspC up2 for the inside out PCR) for the two step replacement of the gspA / gspC intergenic region with the P bla /P cat sequence. SDS PAGE and Heme Staining Cultures for analysis by SDS PAGE with heme staining were grown (250 ml flasks, 30 ml LB broth, 30 C 120 rpm) to stationary phase (48 h, 0.5 gDCM /liter), harvested (25 ml, 8,000g 4 C) and stored at 20 C. Cells were disrupted by two passages through a French pressure cell (14,000 psi). Un broken cells were removed (5,000g 4 C ) and a crude membrane fraction was prep ared from 1.5 ml supernatant (16,9 00 g microcentrifuge, 4 C ). Crude membrane fraction samples were in cubated for 1h at room temperature in SDS sample buffer (46) and resolved on a 4 to 15% gradient polyacrylamide gel with Tris glycine running buffer (65) for 35 min at 200 V. The heme staining procedure was modified from the method of Francis and Becker (40) Gels were washed with 12.5 % trichloroacetic acid for 30 min with rocking, followed by a 30 min rinse w ith deionized water. The gel was developed using a peroxidase staining method. The gel was developed in (final concentrations): 10% acetic acid, 1 g/liter o dianisidine, and 0.06% hydrogen peroxide in 0.05 M citrate buffer, pH 4.4.

PAGE 107

107 Electrochemical Analyti cal Methods Recording of current output over time with the engineered strains was done with the poised potential BEC system described in the previous chapter, with the following modifications. Riboflavin (10 M) and homogentisic acid (HGA, 0.1mg/ml) were i ncluded in the anolyte (first trial) to facilitate electron transfer. A second trial used each compound at half concentration. Glucose (10 mol) was added to the anolyte (without cells) upon reaching a stable baseline and bacterial cells (grown to late sta tionary phase at 100rpm, 0.67g/liter, final in BEC) were added approximately 1 h after glucose addition. Results and Discussion Chromosomal Integration of omcA So and mtrCAB So and Replacement of gsp Promoters in JC85 The omcA and mtrCAB genes from S. oneide nsis MR 1 were integrated into the chromosome of JC85 by targeted gene replacement to construct a functional extracellular electron transfer system. JC85 was chosen as the parent strain because while the presence of the cytoplasmic water forming NADH oxida se (from S. mutans ) might be beneficial (for oxygen scavenging), higher level expression (JC93) would be unnecessary in the mediatorless system. A combinatorial approach was used for integration into the native adhE or pflB to increase the probability of f inding the optimal expression pattern of the heterologous genes. JC94 and JC123 had the entire gene cluster ( omcAmtrCAB ) integrated into pflB or adhE (respectively). Since omcA and mtrCAB are transcribed separately in S. oneidensis MR 1 (10) o mcA transcription was from the pflB or a dhE promoters, while the mtrCAB was transcribed from its own promoter. For JC121 and JC122, omcA and mtrCAB were integrated into the sites separately. Integrations were confirmed by PCR and by DNA sequence analysis. The chromosomal sequence between the gsp operons (of JC94, JC121, JC122 and JC123) was replaced with constitutive promoters (P bla and

PAGE 108

1 08 P cat ). Replacement of the gsp promoters resulted in strains JC128 through JC131. Cell pellets from strains with integration of the heterologous genes were observe d to be a darker orange color than the parental strain, characteristic of cytochrome c production (34) Preliminary SDS PAGE (with heme staining) results sugg ested that the integrated cytochrome genes from S. oneidensis MR 1 were expressed in the tested strains (Figure 5 3). E. coli requires expression of cytochrome c maturation genes ( ccmABCDEFGH ) to make functional c type cytochromes (117) such as the native NapC (or MtrA, MtrC and OmcA from S. oneidensis MR 1). The ccm genes are normally only expressed under nitrate respiring conditions, so strains were transformed with pEC86 (constitutive expression of ccm genes) for testing. T here appeared to be high molecular weight staining for the integration strains (lanes 1 4) that corresponded with the heme staining seen for S. oneidensis MR 1 (lane 5) The poorly resolved high molecular weight staining was from outer membran e cytochromes OmcA and MtrC ( 75 kDa) (106) No 35 kDa band for MtrA (periplasmic decaheme cytochrome c ) (94) was visible in any of the E coli samples. No heme staining at all was seen for the parent strain sample (lane 6 ) The heme staining high molecular weight bands from the integration strain samples were faint and the results were not consistently reproducible. Expression levels may have been suboptimal due to copy number or induction conditions. Other studies reporti ng heterologous expression of the S. oneidensis MR 1 cytochrome c genes used similar induction conditions (aerobic cultures grown at 30 C to late stationary growth phase), but the genes were overexpressed from plasmids (34) Further study using higher expression levels and anaerobic respiratory growth conditions may be beneficial. Production of Current by JC131 The EC testing (with addition of ribofl avin and h omogentisic acid ) results were also inconclusive. Based upon the preliminary heme staining observations, JC131 (transformed with

PAGE 109

109 pEC86) was selected for comparison of current production with JC85/pEC86. In the first trial, JC131/pEC86 produced 50% higher a verage current from glucose versus the parent, but the average current produced by each was low (40 A vs. 10 A, respectively, baseline corrected). The C E values were extremely low, less than 7% and 2% (respectively). The average current produced by JC131 /pEC86 in a second trial (5M thionin, 0.05 mg/ml HGA) was 76A (baseline corrected) vs. 66 A for JC85/pEC86. C E s were 3.7% and 3.5% (respectively). At the end of the experiments the JC85/pE C86 containing anolyte (Figure 5 4B) was observed to be darker ( brown) in color than the anolyte containing J C131/pEC86 (Figure 5 4A). Oxidized homogentisic acid polymerize s to form pyomelanin, a brown pigment (25) The lighter pigmentation observed in the JC131/pEC86 MFC anolyte indicated that the homogentisic acid was being kept in a more reduced state (less pyomelanin formed) possibly by extracellular electron transfer from JC131. Overall, the results show that electron transfer from the cells was inefficient and that it was limiting MFC current production. Further investigation and optimization of expressio n are necessary. It is also possible that the gsp genes were not properly expressed, leading to problems with secretion of the outer membrane cytochromes.

PAGE 110

110 Table 5 1. Strains, plasmids, and PCR primers Strain/plasmid/primer Relevant features Refere nce/source Escherichia coli JC94 pflB :: omcAmtrCAB So This study JC95 pflB :: omcA So This study JC96 pflB :: mtrCAB So This study JC99 adhE :: cat sacB This study JC121 adhE :: mtrCAB So This study JC122 adhE :: omcA So This study JC123 adhE :: omcAmt rCAB So This study JC128 gsp C / P gsp A )::P bla / P cat This study JC129 gsp C / P gsp A )::P bla / P cat This study JC130 gsp C / P gsp A )::P bla / P cat This study JC131 gsp C / P gsp A )::P bla / P cat This study Shewanella oneidensis MR1 ATCC (700550) Plasmids pEC86 cat, ccm (6) pLOI4192 bla kan, pflB :: omcAmtrCAB So This study pLOI5000 bla, cat self ligation of (Klenow treated) pLOI4151, Sap I/ Not I frag. removed This study pLOI5001 bla cat cat ) This study pLOI5002 bla cat (divergent P bla / P cat ) This study pLOI5003 bla bla ), cat This study pLOI5006 bla kan P gsp C / P gsp A in pCR2.1 This study pLOI5007 gsp C / P gsp A ):: cat sacB This study pLOI5008 gsp C / P gsp A )::P bla / P cat (di vergent) This study pLOI5009 bla kan pflB :: omcA So This study pLOI5010 bla kan adhE :: omcA So This study pLOI5011 bla kan pflB :: mtrCAB So This study pLOI5012 bla kan adhE :: mtrCAB So This study pLOI5013 bla kan adhE :: omcAmtrCAB So This study Prime rs JM41XX cat modF2 CGCTGCTAGCAAGGAAGACCTGCGATG GAGAAAAAAATCACTGG This study JM41XX cat modR1 CGCTGCTAGCTTAGACGTCAGGTGGCAC TT This study JM41X5promF1 CGCTGCTAGCGTCTCATGAGCGGATACA This study JM41X5promR1 CGCTATCGATGCTTCCTTAGCTCCTGAA This study

PAGE 111

111 Table 5 1. Continued Primer Relevant features Reference/source JM41X3 bla modF1 CGCCATCGATAAGGAAGGCTCTTCATGA GTATTCAACATTTCCG This study JM41X3 bla modR1 CGCTATCGATCTAAGCTAGCAAGGAAGA C This study JM gsp CAF2 CTGCCACGGGATTTGCATCT This study JM gsp CAR2 TATGCGGCGGTGATTCAGGT This study JM gsp Aup1 CGCCATCGATAAGGAAGTTCTATGTCTA CGAGAAGAG This study JM gsp Cup2 CGCTGCTAGCTCCTTCCTTACATCGTGCC CACACTACGTTTCC This study JM41X5promF Pvu I CGCTCGATCGGTCTCATGAGCGGATACA This study JM41X5promR Pvu I CGCTCGATCGGCTTCCTTAGCTCCTGAA This study JM om cA F2 ACCTCTCGCGCTTAACAATG This study JM mtrB R2 TATCAAGGCGCTCAGTGGTA This study JM omcA F1 CACGATCGAGGCCTGCAACTGCCAAT This study JM omcA R1 CGTTCGATCGCGACTTAGTTACCGTGTG CTTCCA This study JM mtrC F1 CGTTCGATCGCCCTTGTGGTTTAACTACC This study Primer sequences are

PAGE 112

112 Figure 5 1. Arrangement of gsp (general secretory pathway, T2SS) genes in representative strains. H NS (histone like nucleoid structuring protein) is shown binding to the operator region between gspA and gspC of E. coli K12, blocking tran scription from divergent promoters

PAGE 113

113 Figure 5 2. A model for extracellular electron transfer by an engineered E. coli strain producing periplasmic (MtrA) and outer membrane cytochromes (MtrC, OmcA) from Shewanella oneidensis MR 1

PAGE 114

114 Figure 5 3. Heme stained crude membrane fractions (after SDS PAGE) of JC128/pEC86, lane 1; JC129/pEC86, lane 2; JC130/pEC86, lane 3; JC131/pEC86, lane 4; S. oneidensis MR 1, lane 5; JC85/pEC86, lane 6; MW markers (BioRad Kaleidoscope protein s tandards) lane7

PAGE 115

115 Figure 5 4. Electrochemical cell (EC) anolytes (post run) from the first trial with JC131/pEC86 (A) and JC85/pEC86 (B). Anolytes contained riboflavin (5M) and homogentisic acid (1 mg/ml). ECs were disassembled and electrodes (top) were removed from the anode chamber in the photo

PAGE 116

116 CHAPTER 6 OPTIMIZATION OF ELEC TRON TRANSFER TO THE ANODE IN A MICROBIAL FUEL CELL USING SINGLE WA LL CARBON NANOTUBES AND MIXED LENGTH CAR BON FIBERS Introduction Microbial fuel cells are devices that tran sform stored chemical energy in organic materials into electrical energy for useful work. The MFC uses microbial whole cell biotransformation, versus the immobilized enzyme biocatalysis used in other types of biological fuel cells. Using whole cells has ad vantages, particularly low cost of materials and the potential for indefinite stability and regeneration of biocatalyst. This approach also brings challenges, primarily the lower power densities achieved versus enzymatic and non biological fuel cells. The important roles of microbial metabolic rates and transfer of electrons across biological membranes in MFC power output was addressed in previous chapters. We anticipated that once the biological current production and efficiency of electron transfer to the exterior of the bacterium were optimized, it would be important to minimize overpotential losses. Single walled carbon nanotubes (SWNT) (35) and mixed length bulk carbon fibers (111) were therefore tested in the anode compartment of a dual chambered MFC, using E. coli W3110 as the whole cell biocatalyst and methylene blue as the diffusible electron carrying mediator. These micro and nanoscale carbon structures were included in a suspended, conductive netwo rk in order to decrease overpotentials due to mediator diffusion and anode reduction limitations. Minimization of losses downstream of bacterial electron transfer through anode optimization would complement the genetic engineering efforts. SWNTs are cylind rical structures whose walls consist of a graphitic sheet of carbon atoms (one atom thick). SWNT diameters can be as small as 0.5 nm, while their length depends upon the method of synthesis, but provide high aspect ratios and surface areas. This geometry, along with their high electrical conductivity, make SWNTs ideal for increasing MFC anode surface

PAGE 117

117 area and decreasing diffusional distance of the mediator. The SWNT suspension density could be optimized to maximize conductivity without limiting bacterial lo ading. It is also possible that the hydrophobic SWNTs could infiltrate the E. coli membranes and allow direct electron transfer to the anode. Micro scale carbon fibers can also be used to limit concentration overpotentials. While nanotubes have higher aspe ct ratios, surface area densities, and conductivities (in some cases), their synthesis and purification are still more difficult and expensive than the larger carbon fibers. It is important for MFCs to be constructed with inexpensive materials and to be sc alable, which would be difficult (at this time) with SWNTs. Bulk carbon fibers may help to optimize electron transfer to the electrode, are less expensive than SWNTs and are available in large quantities. Single walled carbon nanotubes or mixed length carb on fibers were suspended in the anode compartment of a two chambered MFC for the purpose of increasing stable current production. The MFC employed E coli W3110 as the microbial biocatalyst in the anode compartment and methylene blue as the redox mediator Current produced from glucose catabolism was measured over time and the performance of the SWNT MFCs compared with the control MFCs, containing only a graphite felt electrode. The effect of SWNT or CF inclusion on the magnitude and stability of electrica l current production under different MFC operating conditions was evaluated. Materials and Methods Materials, Instrumentation and MFC Assembly The basic MFC assembly (NCBE, The University of Reading, UK) consisted of Perspex (acrylic) chambers, neoprene gaskets, carbon fiber electrodes (cut to 2.4 cm 3.3 cm) and a cation exchange membrane. Assembly of these components was performed as outlined by the

PAGE 118

118 supplier and as described by Bennetto (11) In addition, platinum wire (from Mini Subcell GT electrophoresis apparatus electrodes, BioRad, Hercules, CA) was used to connect the carbon fiber electrod es to the external circuit, consisting of a 1000 ohm resistor and an ammeter (46 Range Digital Multimeter with PC interface and data logging MeterView 1.0 software, RadioShack, Fort Worth, TX), in series. Argon was bubbled into the anode compartment using 1 mm outer diameter, 0.2 mm inner diameter PTFE tubing passing through the same aperture as the platinum wire of the anode down to the bottom of the chamber. This aperture was sealed to divert exiting gases to the feeding/venting tube, located in the oth er aperture of the chamber The argon flow rate was maintai ned at approximately 3.5 ml/min as measured at the venting tube using a J&W Scientific (Folsom, CA) ADM 2000 flow meter. The cathode chamber was modified in the s ame manner as the anode chamber; however, no gas was delivered to the cathode chamber and the electrode aperture was not sealed. The catholyte (9 ml total volume added per MFC after assembly) consisted of 50 mM potassium ferricyanide in 100 mM sodium phosphate buffer, pH 7.0. A cloth la yer (75% rayon, 25% polyester, EasyWipe, Magla Products L.L.C., Morristown, New Jersey) was used to separate the electrodes from the cation exchange membrane in each chamber. Both assembled MFCs were attached to a weight, placed in a one gallon plastic fr eezer bag and immersed (to the top of the MFCs) in a 37 o C water bath. Media and Growth of Cultures E coli W3110 (ATCC 27325) cultures were inoculated from a single colony on a NBSM (with 20 g/liter glucose) agar plate into 50 ml NBSM glucose medium (me dium contained 30 g/liter glucose unless otherwise noted) in a 250 ml baffled Erlenmeyer flask. Cultures were grown aerobically (37C, 250 rpm) and 10.9 ml was harvested (5000 g 5 min, 22C) at a density of approximately 36.3 mg DCM/liter) and resuspended in 0.5 ml NBSM glucose medium.

PAGE 119

119 SWNT MFC Anolyte Preparation and Inoculation The SWNT suspension (0.06 mg/ml) to be used in the MFC was washed in an equal volume of NBSM glucose medium to minimize carryover of detergent used in its preparation as well as t o limit the dilution of the medium. The tube was inverted to mix and was spun down in a clinical centrifuge for 2 min This caused aggregation of the SWNTs and they were trans ferred in a minimal volume (2 ml ) to a fresh tube. To this tube was added an addi tional 4 ml of medium. 2 ml of the SWNT wash was transferred to a fresh tube and 4 ml medium was added to it for use in the no SWNT control MFC. The resuspended cells were added to each of these tubes and briefly sparged with argon, using a 25 5 / 8 gauge ne edle on a 1 ml syringe with plunger removed and an argon filled balloon on the other end. The suspensions were then transferred to the anode chamber of their res pective MFCs by means of a 10 ml syringe equipped with an 18 gauge cannula. The final cell den sity in the a nolytes was approximately 60 mg DCM/liter. Methylene blue (480 l of a 10 mM stock) was added to each anolyte by syringe. CF MFC Preparation The procedure for preparation of CF MFCs was as described for the SWNT MFCs, but a mixed length carbon fiber suspension was used in place of the nanotubes. During the first trial it was found that the mixed carbon fibers could not be transferred by syringe. This necessitated the feeding of the aggregated fibers through the feeding/vent tube in the anode chamber, after the addition of the cell suspension and prior to addition of me diato r. The CF transfer was done by pushing the fibers through the tube with toothpicks. For the second trial, the carbon fiber suspension was transferred as several aggregates (from which most of the liquid had been removed) to the anode compartment duri ng assembly. E coli W3110 cells were resuspended ( 60 mgDCM/liter in 6.5 ml total volume NBSM glucose

PAGE 120

120 medium) for each MFC (CF MFC and control MFC) and cell suspensions were transferred to the MFC anode chambers, followed by addition of methylene blue. Re sults SWNT MFC Current Production After addition of cells and mediator (and after a brief current spike in which the no SWNT MFC surpassed that of the SWNT MFC), the SWNT MFC sust ained a higher current (Figure 6 1). After approximately 7 h the SWNT MFC c urrent dropped to the same level as the control MFC (~ 0.1 mA). Argon flow was stopped 19 h into the run to determine the current produced by each MFC in the abs ence of anolyte mixing (Figure 6 2). The SWNT MFC produced a current that was over 2.5 fold gr eater than that of the no SWNT control MFC. The current levels were stable for 1.5 h of the control. Brief mixi ng with argon, performed 2 h after this observed drop, greatly increas ed the SWNT MFC current and resulted in a stable 3 fold higher cur rent for an additional 1.5 h A second experiment was performed in which argon mixing was halted soon after inoculation of the MFCs and addition of mediator. The same MFC setup procedure was followed as in the initial experiment. However, the inoculated anolyte cell densi ty was slightly higher (~ 66 mg DCM/liter). The SWNT MFC current was about 50% higher than t hat of the control MFC (Figure 6 3). CF MFC Current Production Upon addition of al l components, the CF MFC was observed to produce a current that was nearly twice t hat of the control MFC (Figure 6 4). Initial current was only slightly higher from th e CF MFC. However over a 12 h period the CF MFC produced a stable current that averaged at least 40% higher than t hat of the control MFC (Figure 6 5).

PAGE 121

121 Discussion Minimizing concentration overpotentials (and ohmic losses) and increasing active electrode surface area are critical factors for increasing MFC power output. These improvements are o f particular importance as the limitations of bacterial metabolic rate and efficiency of extracellular electron transport are alleviated. The results of the SWNT and CF MFC testing showed that the CFs and SWNTs also provided higher stable current productio n without continuous mixing. These results suggest that the dynamic anode matrix composed of bacterial cells and SWNTs (or CFs) was able to form a stable, active and conductive matrix after mixing was stopped. In contrast, the MFCs without the SWNT or CF a ddition experienced significant decreases in current production after stirring was stopped. In those MFCs, planktonic cells sank to the bottom of the vessel and were not as readily accessible to the mediator. Thus, concentration overpotentials were likely greater in the control MFCs. Further MFC design improvements that incorporate the type of high surface area, robust electrode design discussed here will help to increase MFC power output and reliability in mediatorless and mediated systems.

PAGE 122

122 Figure 6 1. Initial hour (mixing) average current with and without addition of single walled carbon nanotubes (SWNT)

PAGE 123

123 Figure 6 2. Avera ge current with and without addition of single walled carbon nanotubes (SWNT).

PAGE 124

124 Figure 6 3. Initial hour (post mixing) ave rag e current with and without addition of single walled carbon nanotubes (SWNT).

PAGE 125

125 Figure 6 4. Average cur rent with and without addition of micro scale carbon fibers (CF).

PAGE 126

126 Figure 6 5. Second trial ave rage current with and without addition of micro scale carbon fibers (CF).

PAGE 127

127 CHAPTER 7 CONCLUSIONS AND FUTU RE DIRECTIONS A sustainable energy infrastructure depends upon increased efficiency and use of renewable sources. Microbial fuel cells can contribute by adding value to byprod ucts (currently considered waste streams) from various agricultural, industrial and municipal processes. At present, wastewater treatment strategies are energetically and economically costly. MFCs can be used to remove most of the organic material from the wastewater, while providing electricity to run the facility. Wastewater treatment is one prominent example of how MFCs can increase energy efficiency and decrease costs. The processes at breweries, food processing plants, green waste disposal and biofuel processing plants could also be complemented by MFC based energy recovery from wastes. Solar, wind, geothermal and hydroelectric energy are important components of a comprehensive renewable energy strategy, but all have specific siting requirements and som e can only provide intermittent power. MFCs can be run continuously and located virtually anywhere, without requiring sunlight, wind, drilling, or large bodies of water. Depending upon the organisms used, they can be operated at a wide range of temperature s, with many types of fuels. Despite all of these comp elling positive attributes, MFC technology is still in its early stages and is not ready for widespread adoption as part of our energy infrastructure. MFC designs need to be tested at larger scale, cost of materials must be minimized, and power densities need to be increased. Further study is needed of the biological factors that influence MFC power output. Metabolic engineering of microorganisms for use in MFCs can help make generation of electricity wi th MFCs more practical. Our study focused on improvement of E. coli ATCC 8739 as a biocatalyst for production of electricity in an MFC. We used a cofactor engineering approach to increase the rate and yield

PAGE 128

128 of production of electron s from glucose. Eliminat ion of oxidative phosphorylation lowered the ATP/ADP ratio and led to a 44% increase in the rate of glycolysis. However, transfer of electrons from NADH to oxygen or to the anode (via thionin) was enzyme limited (by dehydrogenases of the ETS). This caused overflow metabolism (acetate production) and limited the electric current and C E in the poised potential MFC system. Heterologous expression of naoX Sm provided an alternate electron route from NADH to the terminal electron acceptor. This removed the electr on transfer rate limitation and lowered the NADH/NAD + ratio, resulting in a 19% increase in current (JC93 versus wild type) and an increase in C E from 49% to 76%. Future directions for this research include testing the engineered strains in design optimize d MFCs, further study of expression of extracellular ETS genes, and the metabolic engineering of other MFC biocatalysts (such as DMRB). A poised potential electrochemical cell allowed us to study processes upstream of the anode without the influence of int ernal resistance or cathodic overpotentials. It is important to test the engineered strains under non idealized MFC conditions to see if high rates and yields can be achieved. Some limitation of current may have occurred in the poised potential system due to mass transfer limitations. At high current low pH may have affected the electrochemical activity of the bacteria. Reoxidation of thionin may also have become limiting. Coulombic efficiency may have been limited at high current due to diffusion of oxyge n (and H 2 O 2 ) into the anolyte from the cathode. Testing the improved strains in an optimized MFC design may help to eliminate concentration overpotentials. In our study, we have shown that supplementation of the anode with mixed length carbon fibers and SW NTs can result in increased current. Optimization of expression of the omcA mtrCAB So genes in the engineered strains may minimize activation overpotentials and enable mediatorless MFC operation with high power densities. Following the

PAGE 129

129 S. oneidensis model, overproduction of flavin electron shuttles may also be beneficial. We investigated the incorporation of an extracellular ETS into an improved E. coli MFC biocatalyst. However, our genetic engineering strategy could also be applied to optimization of the me tabolism of an exoelectrogen for increased MFC power density and broader fuel range.

PAGE 130

130 REFERENCES 1. Aelterman, P., K. Rabaey, H. T. Pham, N. Boon, and W. Verstraete. 2006. Continuous electricity generation at high voltages and currents using stacked microbial fuel cells. Environmental Science & Technology 40: 3388 3394. 2. Ahmad, M., and V. K. Jindal. 2006. An automatic procedure for rapid online estimation of raw milk quality. LWT Food Science and Technology 39: 432 436. 3. Allen, R. M., and H. P. Bennetto. 1993. Microbial fuel cells Electricity production from carbohydrates. Applied Biochemistry and Biotechnology 39: 27 40. 4. Anderson, D. H., L. J. Donald, M. V. Jacob, and H. W. Duckworth. 1991. A mutant of Escherichia coli citrate syn thase that affects the allosteric equilibrium. Biochemistry and Cell Biology Biochimie Et Biologie Cellulaire 69: 232 238. 5. Aoki, R., M. Wada, N. Takesue, K. Tanaka, and A. Yokota. 2005. Enhanced glutamic acid production by a H + ATPase defective mutant of Corynebacterium glutamicum Bioscience Biotechnology and Biochemistry 69: 1466 1472. 6. Arslan, E., H. Schulz, R. Zufferey, P. Kunzler, and L. Thony Meyer. 1998. Overproduction of the Bradyrhizobium japonicum c type cytochrome subunits of the cbb(3) oxidas e in Escherichia coli Biochemical and Biophysical Research Communications 251: 744 747. 7. Babul, J., and V. Guixe. 1983. Fructose bisphosphatase from Escherichia coli. Purification and characterization. Archives of Biochemistry and Biophysics 225: 944 949. 8. Bapat, P., S. K. Nandy, P. Wangikar, and K. V. Venkatesh. 2006. Quantification of metabolically active biomass using Methylene Blue dye Reduction Test (MBRT): Measurement of CFU in about 200 s. Journal of Microbiological Methods 65: 107 116. 9. Behera, M., P. S. Jana, and M. M. Ghangrekar. 2010. Performance evaluation of low cost microbial fuel cell fabricated using earthen pot with biotic and abiotic cathode. Bioresource Technology 101: 1183 1189. 10. Beliaev, A. S., D. A. Saffarini, J. L. McLaughlin, an d D. Hunnicutt. 2001. MtrC, an outer membrane decahaem c cytochrome required for metal reduction in Shewanella putrefaciens MR 1. Molecular Microbiology 39: 722 730. 11. Bennetto, H. P. 1990. Electricity generation by microorganisms. Biotechnology Education 1: 163 168. 12. Bi, C., X. Zhang, J. D. Rice, L. O. Ingram, and J. F. Preston. 2009. Genetic engineering of Enterobacter asburiae strain JDR 1 for efficient D( ) lactic acid production from hemicellulose hydrolysate. Biotechnology Letters 31: 1551 1557.

PAGE 131

131 13. Bi, C. H., X. L. Zhang, L. O. Ingram, and J. F. Preston. 2009. Genetic engineering of Enterobacter asburiae strain JDR 1 for efficient production of ethanol from hemicellulose hydrolysates. Applied and Environmental Microbiology 75: 5743 5749. 14. Bochner, B. R., H. C. Huang, G. L. Schieven, and B. N. Ames. 1980. Positive selection for loss of tetracycline resistance. Journal of Bacteriology 143: 926 933. 15. Bowman, J. P., S. A. McCammon, D. S. Nichols, J. H. Skerratt, S. M. Rea, P. D. Nichols, and T. A. Mc Meekin. 1997. She wanella gelidimarina sp. nov. and Shewanella frigidimarina sp. nov., novel Antarctic species with the ability to produce eicosapentaenoic acid (20:5 omega 3) and grow anaerobically by dissimilatory Fe(III) reduction. International Journal of Systematic Bacteriology 47: 1040 1047. 16. Boyer, P. D. 1993. The binding change mechanism for ATP synthase Some probabilities and possibilities. Biochimica et Biophysica Acta 1140: 215 250. 17. Calhoun, M. W., K. L. Oden, R. B. Gennis, M. J. T. Demattos, and O. M. Neijssel. 1993. Energetic efficiency of Escherichia coli Effects of mutations in components of the aerobic respiratory chain. Journal of Bacteriology 175: 3020 3025. 18. Campbell, J. E., D. B. Lobell, and C. B. Field. 2009. Greater transportati on energy and GHG offsets from bioelectricity than ethanol. Science 324: 1055 1057. 19. Causey, T. B., K. T. Shanmugam, L. P. Yomano, and L. O. Ingram. 2004. Engineering Escherichia coli for efficient conversion of glucose to pyruvate. Proceedings of the Na tional Academy of Sciences of the United States of America 101: 2235 2240. 20. Causey, T. B., S. Zhou, K. T. Shanmugam, and L. O. Ingram. 2003. Engineering the metabolism of Escherichia coli W3110 for the conversion of sugar to redox neutral and oxidized pr oducts: Homoacetate production. Proceedings of the National Academy of Sciences of the United States of America 100: 825 832. 21. Chang, I. S., H. Moon, O. Bretschger, J. K. Jang, H. I. Park, K. H. Nealson, and B. H. Kim. 2006. Electrochemically active bact eria (EAB) and mediator less microbial fuel cells. Journal of Microbiology and Biotechnology 16: 163 177. 22. Chaudhuri, S. K., and D. R. Lovley. 2003. Electricity generation by direct oxidation of glucose in mediatorless microbial fuel cells. Nature Biotec hnology 21: 1229 1232. 23. Cheng, S. A., and B. E. Logan. 2007. Ammonia treatment of carbon cloth anodes to enhance power generation of microbial fuel cells. Electrochemistry Communications 9: 492 496. 24. Clauwaert, P., P. Aelterman, T. H. Pham, L. De Scham phelaire, M. Carballa, K. Rabaey, and W. Verstraete. 2008. Minimizing losses in bio electrochemical systems: the road to applications. Applied Microbiology and Biotechnology 79: 901 913.

PAGE 132

132 25. Coon, S. L., S. Kotob, B. B. Jarvis, S. J. Wang, W. C. Fuqua, and R. M. Weiner. 1994. Homogentisic acid is the product of MelA, which mediates melanogenesis in the marine bacterium Shewanella colwelliana D. Applied and Environmental Microbiology 60: 3006 3010. 26. Coursolle, D., D. B. Baron, D. R. Bond, and J. A. Gralnick 2009. The Mtr respiratory pathway is essential for reducing flavins and electrodes in Shewanella oneidensis Journal of Bacteriology (Online ahead of print). 27. Cozzone, A. J., and M. El Mansi. 2005. Control of isocitrate dehydrogenase catalytic activit y by protein phosphorylation in Escherichia coli Journal of Molecular Microbiology and Biotechnology 9: 132 146. 28. Cunningham, L., D. Georgellis, J. Green, and J. R. Guest. 1998. Co regulation of lipoamide dehydrogenase and 2 oxoglutarate dehydrogenase s ynthesis in Escherichia coli : characterisation of an ArcA binding site in the lpd promoter. FEMS Microbiology Letters 169: 403 408. 29. D'Alessandro, M., P. Turina, and B. A. Melandri. 2008. Intrinsic uncoupling in the ATP synthase of Escherichia coli Bioc himica Et Biophysica Acta Bioenergetics 1777: 1518 1527. 30. Datsenko, K. A., and B. L. Wanner. 2000. One step inactivation of chromosomal genes in Escherichia coli K12 using PCR products. Proceedings of the National Academy of Sciences of the United States of America 97: 6640 6645. 31. Denfert, C., A. Ryter, and A. P. Pugsley. 1987. Cloning and expression in Escherichia coli of the Klebsiella pneumoniae genes for production, surface localization and secretion of the lipoprotein pullulanase. EMBO Journal 6: 35 31 3538. 32. Devillebonne, D., and J. R. Garel. 1992. A conformational transition involved in antagonistic substrate binding to the allosteric phosphofructokinase from Escherichia coli Biochemistry 31: 1695 1700. 33. DeVito, J. A. 2008. Recombineering with tolC as a selectable/counter selectable marker: remodeling the rRNA Operons of Escherichia coli Nucleic Acids Research 36: e4. 34. Donald, J. W., M. G. Hicks, D. J. Richardson, and T. Palmer. 2008. The c type cytochrome OmcA localizes to the outer membran e upon heterologous expression in Escherichia coli Journal of Bacteriology 190: 5127 5131. 35. Dresselhaus, M. S., and H. Dai. 2004. Carbon nanotubes: Continued innovations and challenges. MRS Bulletin 29: 237 239. 36. Du, Z. W., H. R. Li, and T. Y. Gu. 200 7. A state of the art review on microbial fuel cells: A promising technology for wastewater treatment and bioenergy. Biotechnology Advances 25: 464 482.

PAGE 133

133 37. Dubiel, M., C. H. Hsu, C. C. Chien, F. Mansfeld, and D. K. Newman. 2002. Microbial iron respiration can protect steel from corrosion. Applied and Environmental Microbiology 68: 1440 1445. 38. Fan, Y. Z., H. Q. Hu, and H. Liu. 2007. Enhanced Coulombic efficiency and power density of air cathode microbial fuel cells with an improved cell configuration. Jour nal of Power Sources 171: 348 354. 39. Francetic, O., D. Belin, C. Badaut, and A. P. Pugsley. 2000. Expression of the endogenous type II secretion pathway in Escherichia coli leads to chitinase secretion. EMBO Journal 19: 6697 6703. 40. Francis, R. T., and R R. Becker. 1984. Specific indication of hemoproteins in polyacrylamide gels using a double staining process. Analytical Biochemistry 136: 509 514. 41. Futai, M., and H. Kanazawa. 1983. Structure and function of proton translocating adenosine triphosphatas e (F O F 1 ) Biochemical and molecular biological approaches. Microbiological Reviews 47: 285 312. 42. Gorby, Y. A., S. Yanina, J. S. McLean, K. M. Rosso, D. Moyles, A. Dohnalkova, T. J. Beveridge, I. S. Chang, B. H. Kim, K. S. Kim, D. E. Culley, S. B. Reed, M. F. Romine, D. A. Saffarini, E. A. Hill, L. Shi, D. A. Elias, D. W. Kennedy, G. Pinchuk, K. Watanabe, S. Ishii, B. Logan, K. H. Nealson, and J. K. Fredrickson. 2006. Electrically conductive bacterial nanowires produced by Shewanella oneidensis strain MR 1 and other microorganisms. Proceedings of the National Academy of Sciences of the United States of America 103: 11358 63. 43. Gralnick, J. A., and D. K. Newman. 2007. Extracellular respiration. Molecular Microbiology 65: 1 11. 44. Harlow, E., and D. Lane. 1 988. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 45. He, S. Y., C. Schoedel, A. K. Chatterjee, and A. Collmer. 1991. Extracellular secretion of pectate lyase by the Erwinia chrysanthemi Out pathway is depen dent upon Sec mediated export across the inner membrane. Journal of Bacteriology 173: 4310 4317. 46. Hermolin, J., and R. H. Fillingame. 1995. Assembly of F O sector of Escherichia coli H + ATP synthase Interdependence of subunit insertion into the membrane Journal of Biological Chemistry 270: 2815 2817. 47. Hernandez, M. E., A. Kappler, and D. K. Newman. 2004. Phenazines and other redox active antibiotics promote microbial mineral reduction. Applied and Environmental Microbiology 70: 921 928.

PAGE 134

134 48. Higuchi, M ., M. Shimada, Y. Yamamoto, T. Hayashi, T. Koga, and Y. Kamio. 1993. Identification of two distinct NADH oxidases corresponding to H 2 O 2 forming oxidase and H 2 O forming oxidase induced in Streptococcus mutans Journal of General Microbiology 139: 2343 2351. 49. Holmes, D. E., D. R. Bond, R. A. O'Neill, C. E. Reimers, L. R. Tender, and D. R. Lovley. 2004. Microbial communities associated with electrodes harvesting electricity from a variety of aquatic sediments. Microbial Ecology 48: 178 190. 50. Jantama, K., M J. Haupt, S. A. Svoronos, X. L. Zhang, J. C. Moore, K. T. Shanmugam, and L. O. Ingram. 2008. Combining metabolic engineering and metabolic evolution to develop nonrecombinant strains of Escherichia coli C that produce succinate and malate. Biotechnology and Bioengineering 99: 1140 1153. 51. Jantama, K., X. Zhang, J. C. Moore, K. T. Shanmugam, S. A. Svoronos, and L. O. Ingram. 2008. Eliminating side products and increasing succinate yields in engineered strains of Escherichia coli C. Biotechnology and Bioen gineering 101: 881 893. 52. Jensen, P. R., and O. Michelsen. 1992. Carbon and energy metabolism of ATP mutants of Escherichia coli Journal of Bacteriology 174: 7635 7641. 53. Jurtshuk, P., and J. K. Liu. 1983. Cytochrome oxidase analyses of Bacillus strains Existence of oxidase positive species. International Journal of Systematic Bacteriology 33: 887 891. 54. Kajie, S., R. Ideta, I. Yamato, and Y. Anraku. 1991. Molecular cloning and DNA sequence of dniR a gene affecting anaerobic expression of the Escheri chia coli hexaheme nitrite reductase. FEMS Microbiology Letters 83: 205 211. 55. Karu, A. E., Y. Sakaki, H. Echols, and S. Linn. 1975. Gamma protein specified by bacteriophage Lambda Structure and inhibitory activity for recBC enzyme of Escherichia coli Journal of Biological Chemistry 250: 7377 7387. 56. Kashefi, K., and D. R. Lovley. 2000. Reduction of Fe(III), Mn(IV), and toxic metals at 100 degrees C by Pyrobaculum islandicum Applied and Environmental Microbiology 66: 1050 1056. 57. Katz, E., B. Filanov sky, and I. Willner. 1999. A biofuel cell based on two immiscible solvents and glucose oxidase and microperoxidase 11 monolayer functionalized electrodes. New Journal of Chemistry 23: 481 487. 58. Kelly, I. 2003. The design of a robotic predator: The SlugBo t. Robotica 21: 399 406. 59. Kim, B. H., H. J. Kim, M. S. Hyun, and D. H. Park. 1999. Direct electrode reaction of Fe(III) reducing bacterium, Shewanella putrefaciens Journal of Microbiology and Biotechnology 9: 127 131.

PAGE 135

135 60. Kim, B. H., H. S. Park, H. J. Ki m, G. T. Kim, I. S. Chang, J. Lee, and N. T. Phung. 2004. Enrichment of microbial community generating electricity using a fuel cell type electrochemical cell. Applied Microbiology and Biotechnology 63: 672 681. 61. Kim, G. T., G. Webster, J. W. T. Wimpenny B. H. Kim, H. J. Kim, and A. J. Weightman. 2006. Bacterial community structure, compartmentalization and activity in a microbial fuel cell. Journal of Applied Microbiology 101: 698 710. 62. Kim, Y., L. O. Ingram, and K. T. Shanmugam. 2008. Dihydrolipoamid e dehydrogenase mutation alters the NADH sensitivity of pyruvate dehydrogenase complex of Escherichia coli K 12. Journal of Bacteriology 190: 3851 3858. 63. Koebmann, B. J., H. V. Westerhoff, J. L. Snoep, D. Nilsson, and P. R. Jensen. 2002. The glycolytic f lux in Escherichia coli is controlled by the demand for ATP. Journal of Bacteriology 184: 3909 3916. 64. Kolker, E., A. F. Picone, M. Y. Galperin, M. F. Romine, R. Higdon, K. S. Makarova, N. Kolker, G. A. Anderson, X. Y. Qiu, K. J. Auberry, G. Babnigg, A. S Beliaev, P. Edlefsen, D. A. Elias, Y. A. Gorby, T. Holzman, J. A. Klappenbach, K. T. Konstantinidis, M. L. Land, M. S. Lipton, L. A. McCue, M. Monroe, L. Pasa Tolic, G. Pinchuk, S. Purvine, M. H. Serres, S. Tsapin, B. A. Zakrajsek, J. H. Zhou, F. W. Lari mer, C. E. Lawrence, M. Riley, F. R. Collart, J. R. Yates, R. D. Smith, C. S. Giometti, K. H. Nealson, J. K. Fredrickson, and J. M. Tiedje. 2005. Global profiling of Shewanella oneidensis MR 1: Expression of hypothetical genes and improved functional annot ations. Proceedings of the National Academy of Sciences of the United States of America 102: 2099 2104. 65. Laemmli, U. K. 1970. Cleavage of structural proteins during assembly of head of bacteriophage T4. Nature 227: 680 685. 66. Lee, E. C., D. G. Yu, J. M. de Velasco, L. Tessarollo, D. A. Swing, D. L. Court, N. A. Jenkins, and N. G. Copeland. 2001. A highly efficient Escherichia coli based chromosome engineering system adapted for recombinogenic targeting and subcloning of BAC DNA. Genomics 73: 56 65. 67. Li ao, J. C., Y. P. Chao, and R. Patnaik. 1994. Alteration of the biochemical valves in the central metabolism of Escherichia coli Biochemical Engineering VIII 745: 21 34. 68. Litster, S., and G. McLean. 2004. PEM fuel cell electrodes. Journal of Power Source s 130: 61 76. 69. Logan, B., S. Cheng, V. Watson, and G. Estadt. 2007. Graphite fiber brush anodes for increased power production in air cathode microbial fuel cells. Environmental Science & Technology 41: 3341 3346. 70. Logan, B. E. 2009. Exoelectrogenic ba cteria that power microbial fuel cells. Nature Reviews Microbiology 7: 375 381.

PAGE 136

136 71. Logan, B. E. 2007. Microbial Fuel Cells. John Wiley & Sons, Hoboken. 72. Logan, B. E., B. Hamelers, R Rozendal, U. Schrorder, J. Keller, S. Freguia, P. Aelterman, W. Verstraete, and K. Rabaey. 2006. Microbial fuel cells: Methodology and technology. Environmental Science & Technology 40: 5181 5192. 73. Logan, B. E., and J. M. Regan. 2006. Electricity produ cing bacterial communities in microbial fuel cells. Trends in Microbiology 14: 512 518. 74. Lovley, D. R. 2006. Bug juice: harvesting electricity with microorganisms. Nat Rev Microbiol 4: 497 508. 75. Lovley, D. R. 1993. Dissimilatory Metal Reduction. Annual Review of Microbiology 47: 263 290. 76. Luli, G. W., and W. R. Strohl. 1990. Comparison of growth, acetate production, and acetate inhibition of Escherichia coli strains in batch and fed batch fermentations. Applied and Environmental Microbiology 56: 1004 1 011. 77. Malpica, R., B. Franco, C. Rodriguez, O. Kwon, and D. Georgellis. 2004. Identification of a quinone sensitive redox switch in the ArcB sensor kinase. Proceedings of the National Academy of Sciences of the United States of America 101: 13318 13323. 78. Marsili, E., D. B. Baron, I. D. Shikhare, D. Coursolle, J. A. Gralnick, and D. R. Bond. 2008. Shewanella secretes flavins that mediate extracellular electron transfer. Proceedings of the National Academy of Sciences of the United States of America 105: 3968 3973. 79. Martinez Morales, F., A. C. Borges, K. Martinez, K. T. Shanmugam, and L. O. Ingram. 1999. Chromosomal integration of heterologous DNA in Escherichia coli with precise removal of markers and replicons used during construction. Journal of Bact eriology 181: 7143 7148. 80. Matsumoto, J., M. Higuchi, M. Shimada, Y. Yamamoto, and Y. Kamio. 1996. Molecular cloning and sequence analysis of the gene encoding the H 2 O forming NADH oxidase from Streptococcus mutans Bioscience Biotechnology and Biochemist ry 60: 39 43. 81. Matsushita, K., T. Ohnishi, and H. R. Kaback. 1987. NADH ubiquinone oxidoreductases of the Escherichia coli aerobic respiratory chain. Biochemistry 26: 7732 7737. 82. McLachlin, D. T., A. M. Coveny, S. M. Clark, and S. D. Dunn. 2000. Site d irected cross linking of b to the alpha, beta, and a subunits of the Escherichia coli ATP synthase. Journal of Biological Chemistry 275: 17571 17577. 83. Miller, K. 2004. Waste Not. In T. Phillips (ed.), NASA News: Technology. Science@NASA.

PAGE 137

137 84. Mohan, Y., S M. M. Kumar, and D. Das. 2008. Electricity generation using microbial fuel cells. International Journal of Hydrogen Energy 33: 423 426. 85. Nealson, K. H., A. Belz, and B. McKee. 2002. Breathing metals as a way of life: geobiology in action. Antonie Van L eeuwenhoek International Journal of General and Molecular Microbiology 81: 215 222. 86. Noda, S., Y. Takezawa, T. Mizutani, T. Asakura, E. Nishiumi, K. Onoe, M. Wada, F. Tomita, K. Matsushita, and A. Yokota. 2006. Alterations of cellular physiology in Esche richia coli in response to oxidative phosphorylation impaired by defective F 1 ATPase. Journal of Bacteriology 188: 6869 6876. 87. Noji, H., R. Yasuda, M. Yoshida, and K. Kinosita. 1997. Direct observation of the rotation of F 1 ATPase. Nature 386: 299 302. 88 Noll, K. 2006. Microbial Fuel Cells, p. 277 296. In N. Sammes (ed.), Fuel Cell Technology. Springer London, London. 89. Ogasawara, H., Y. Ishida, K. Yamada, K. Yamamoto, and A. Ishihama. 2007. PdhR (pyruvate dehydrogenase complex regulator) controls the respiratory electron transport system in Escherichia coli Journal of Bacteriology 189: 5534 5541. 90. Ozaki, Y., T. Suzuki, Y. Kuruma, T. Ueda, and M. Yoshida. 2008. UncI protein can mediate ring assembly of c subunits of F O F 1 ATP synthase in vitro Bioche mical and Biophysical Research Communications 367: 663 666. 91. Patnaik, R., W. D. Roof, R. F. Young, and J. C. Liao. 1992. Stimulation of glucose catabolism in Escherichia coli by a potential futile cycle. Journal of Bacteriology 174: 7527 7532. 92. Pham, T H., K. Rabaey, P. Aelterman, P. Clauwaert, L. De Schamphelaire, N. Boon, and W. Verstraete. 2006. Microbial fuel cells in relation to conventional anaerobic digestion technology. Engineering in Life Sciences 6: 285 292. 93. Phung, N. T., J. Lee, K. H. Kan g, I. S. Chang, G. M. Gadd, and B. H. Kim. 2004. Analysis of microbial diversity in oligotrophic microbial fuel cells using 16S rDNA sequences. FEMS Microbiology Letters 233: 77 82. 94. Pitts, K. E., P. S. Dobbin, F. Reyes Ramirez, A. J. Thomson, D. J. Rich ardson, and H. E. Seward. 2003. Characterization of the Shewanella oneidensis MR 1 decaheme cytochrome MtrA. Journal of Biological Chemistry 278: 27758 27765. 95. Posfai, G., M. D. Koob, H. A. Kirkpatrick, and F. R. Blattner. 1997. Versatile insertion plasm ids for targeted genome manipulations in bacteria: Isolation, deletion, and rescue of the pathogenicity island LEE of the Escherichia coli O157:H7 genome. Journal of Bacteriology 179: 4426 4428.

PAGE 138

138 96. Potter, M. C. 1911. Electrical effects accompanying the de composition of organic compounds. Proceedings of the Royal Society of London Series B Containing Papers of a Biological Character 84: 260 276. 97. Qiao, Y., C. M. Li, S. J. Bao, Z. S. Lu, and Y. H. Hong. 2008. Direct electrochemistry and electrocatalytic me chanism of evolved Escherichia coli cells in microbial fuel cells. Chemical Communications : 1290 1292. 98. Rabaey, K., N. Boon, M. Hofte, and W. Verstraete. 2005. Microbial phenazine production enhances electron transfer in biofuel cells. Environmental Scie nce & Technology 39: 3401 3408. 99. Rabaey, K., N. Boon, S. D. Siciliano, M. Verhaege, and W. Verstraete. 2004. Biofuel cells select for microbial consortia that self mediate electron transfer. Applied and Environmental Microbiology 70: 5373 5382. 100. Rabae y, K., P. Clauwaert, P. Aelterman, and W. Verstraete. 2005. Tubular microbial fuel cells for efficient electricity generation. Environmental Science & Technology 39: 8077 8082. 101. Ringeisen, B. R., E. Henderson, P. K. Wu, J. Pietron, R. Ray, B. Little, J. C. Biffinger, and J. M. Jones Meehan. 2006. High power density from a miniature microbial fuel cell using Shewanella oneidensis DSP10. Environmental Science & Technology 40: 2629 2634. 102. Ross, D. E., S. S. Ruebush, S. L. Brantley, R. S. Hartshorne, T. A Clarke, D. J. Richardson, and M. Tien. 2007. Characterization of protein protein interactions involved in iron reduction by Shewanella oneidensis MR 1. Applied and Environmental Microbiology 73: 5797 5808. 103. Sambrook, J., E. F. Fritsch, and T. Maniatis 1989. Molecular cloning: A laboratory manual, 2nd ed, vol. I. Cold Spring Harbor Press, Cold Spring Harbor, N.Y. 104. Sauer, U., F. Canonaco, S. Heri, A. Perrenoud, and E. Fischer. 2004. The soluble and membrane bound transhydrogenases UdhA and PntAB hav e divergent functions in NADPH metabolism of Escherichia coli Journal of Biological Chemistry 279: 6613 6619. 105. Seitarides, T., C. Athanasiou, and A. Zabanlotou. 2008. Modular biomass gasification based solid oxide fuel cells (SOFC) for sustainable deve lopment. Renewable & Sustainable Energy Reviews 12: 1251 1276. 106. Shi, L., B. W. Chen, Z. M. Wang, D. A. Elias, M. U. Mayer, Y. A. Gorby, S. Ni, B. H. Lower, D. W. Kennedy, D. S. Wunschel, H. M. Mottaz, M. J. Marshall, E. A. Hill, A. S. Beliaev, J. M. Zac hara, J. K. Fredrickson, and T. C. Squier. 2006. Isolation of a high affinity functional protein complex between OmcA and MtrC: Two outer membrane decaheme c type cytochromes of Shewanella oneidensis MR 1. Journal of Bacteriology 188: 4705 4714.

PAGE 139

139 107. Shi, L ., S. Deng, M. J. Marshall, Z. M. Wang, D. W. Kennedy, A. C. Dohnalkova, H. M. Mottaz, E. A. Hill, Y. A. Gorby, A. S. Beliaev, D. J. Richardson, J. M. Zachara, and J. K. Fredrickson. 2008. Direct involvement of type II secretion system in extracellular tra nslocation of Shewanella oneidensis outer membrane cytochromes MtrC and OmcA. Journal of Bacteriology 190: 5512 5516. 108. Shi, L., T. C. Squier, J. M. Zachara, and J. K. Fredrickson. 2007. Respiration of metal (hydr)oxides by Shewanella and Geobacter : a ke y role for multihaem c type cytochromes. Molecular Microbiology 65: 12 20. 109. Singh, A., M. D. Lynch, and R. T. Gill. 2009. Genes restoring redox balance in fermentation deficient E. coli NZN111. Metabolic Engineering 11: 347 354. 110. Sinha, P. K., J. Tor res Bacete, E. Nakamaru Ogiso, N. Castro Guerrero, A. Matsuno Yagi, and T. Yagi. 2009. Critical roles of subunit NuoH (ND1) in the assembly of peripheral subunits with the membrane domain of Escherichia coli NDH 1. Journal of Biological Chemistry 284: 9814 9823. 111. Smith, J. J., and G. A. McFeters. 1997. Mechanisms of INT (2 (4 iodophenyl) 3 (4 nitrophenyl) 5 phenyl tetrazolium chloride), and CTC (5 cyano 2,3 ditolyl tetrazolium chloride) reduction in Escherichia coli K 12. Journal of Microbiological Metho ds 29: 161 175. 112. Sorgen, P. L., T. L. Caviston, R. C. Perry, and B. D. Cain. 1998. Deletions in the second stalk of F 1 F O ATP synthase in Escherichia coli Journal of Biological Chemistry 273: 27873 27878. 113. Steed, P. R., and R. H. Fillingame. 2009. Aq ueous Accessibility to the Transmembrane Regions of Subunit c of the Escherichia coli F 1 F O ATP Synthase. Journal of Biological Chemistry 284: 23243 23250. 114. Tauschek, M., R. J. Gorrell, R. A. Strugnell, and R. M. Robins Browne. 2002. Identification of a protein secretory pathway for the secretion of heat labile enterotoxin by an enterotoxigenic strain of Escherichia coli Proceedings of the National Academy of Sciences of the United States of America 99: 7066 7071. 115. Tender, L. M., C. E. Reimers, H. A. Stecher, D. E. Holmes, D. R. Bond, D. A. Lowy, K. Pilobello, S. J. Fertig, and D. R. Lovley. 2002. Harnessing microbially generated power on the seafloor. Nature Biotechnology 20: 821 825. 116. Thomason, L., D. L. Court, M. Bubunenko, N. Constantino, H. Wil son, S. Datta, and A. Oppenheim. 2005. Recombineering: Genetic Engineering in Bacteria Using Homologous Recombination, p. 1.16.1 1.16.21. In F. M. Ausubel, R. Brent, R. E. Klingston, D. D. Moore, J. G. Deidman, J. A. Smith, and K. Struhl (ed.), Current Pr otocols in Molecular Biology. John Wiley & Sons Inc., New York.

PAGE 140

140 117. Thony Meyer, L., F. Fischer, P. Kunzler, D. Ritz, and H. Hennecke. 1995. Escherichia coli genes required for cytochrome c maturation. Journal of Bacteriology 177: 4321 4326. 118. Unden, G. and J. Bongaerts. 1997. Alternative respiratory pathways of Escherichia coli : Energetics and transcriptional regulation in response to electron acceptors. Biochimica Et Biophysica Acta Bioenergetics 1320: 217 234. 119. Underwood, S. A., M. L. Buszko, K. T Shanmugam, and L. O. Ingram. 2002. Flux through citrate synthase limits the growth of ethanologenic Escherichia coli KO11 during xylose fermentation. Applied and Environmental Microbiology 68: 1071 1081. 120. Underwood, S. A., S. Zhou, T. B. Causey, L. P. Yomano, K. T. Shanmugam, and L. O. Ingram. 2002. Genetic changes to optimize carbon partitioning between ethanol and biosynthesis in ethanologenic Escherichia coli Applied and Environmental Microbiology 68: 6263 6272. 121. Valentini, G., M. Stoppini, M. L Speranza, M. Malcovati, and G. Ferri. 1991. Bacterial pyruvate kinases have a shorter N terminal domain. Biological Chemistry Hoppe Seyler 372: 91 93. 122. van Hees, W. 1965. A bacterial methane fuel cell. Journal of the Electrochemical Society 112: 258 26 2. 123. Veit, A., T. Polen, and V. F. Wendisch. 2007. Global gene expression analysis of glucose overflow metabolism in Escherichia coli and reduction of aerobic acetate formation. Applied Microbiology and Biotechnology 74: 406 421. 124. Vemuri, G. N., E. A ltman, D. P. Sangurdekar, A. B. Khodursky, and M. A. Eiteman. 2006. Overflow metabolism in Escherichia coli during steady state growth: Transcriptional regulation and effect of the redox ratio. Applied and Environmental Microbiology 72: 3653 3661. 125. Vemu ri, G. N., M. A. Eiteman, and E. Altman. 2006. Increased recombinant protein production in Escherichia coli strains with overexpressed water forming NADH oxidase and a deleted ArcA regulatory protein. Biotechnology and Bioengineering 94: 538 542. 126. Venka teswaran, K., D. P. Moser, M. E. Dollhopf, D. P. Lies, D. A. Saffarini, B. J. MacGregor, D. B. Ringelberg, D. C. White, M. Nishijima, H. Sano, J. Burghardt, E. Stackebrandt, and K. H. Nealson. 1999. Polyphasic taxonomy of the genus Shewanella and descripti on of Shewanella oneidensis sp. nov. International Journal of Systematic Bacteriology 49: 705 724. 127. Vetter, D., B. J. Andrews, L. Robertsbeatty, and P. D. Sadowski. 1983. Site specific recombination of yeast 2 micron DNA in vitro Proceedings of the Nat ional Academy of Sciences of the United States of America Biological Sciences 80: 7284 7288.

PAGE 141

141 128. von Canstein, H., J. Ogawa, S. Shimizu, and J. R. Lloyd. 2008. Secretion of flavins by Shewanella species and their role in extracellular electron transfer. Ap plied and Environmental Microbiology 74: 615 623. 129. Wang, Y. F., S. Tsujimura, S. S. Cheng, and K. Kano. 2007. Self excreted mediator from Escherichia coli K 12 for electron transfer to carbon electrodes. Applied Microbiology and Biotechnology 76: 1439 14 46. 130. Weber, J. 2006. ATP synthase: Subunit subunit interactions in the stator stalk. Biochimica Et Biophysica Acta Bioenergetics 1757: 1162 1170. 131. Wilkinson, S. 2000. "Gastrobots" Benefits and challenges of microbial fuel cells in food powered rob ot applications. Autonomous Robots 9: 99 111. 132. Wilkinson, S. 2001. Hungry for success future directions in gastrobotics research. Industrial Robot an International Journal 28: 213 219. 133. Willner, I., Y. M. Yan, B. Willner, and R. Tel Vered. 2009. In tegrated enzyme based biofuel cells A review. Fuel Cells 9: 7 24. 134. Wood, B. E., L. P. Yomano, S. W. York, and L. O. Ingram. 2005. Development of industrial medium required elimination of the 2,3 butanediol fermentation pathway to maintain ethanol yield in an ethanologenic strain of Klebsiella oxytoca Biotechnology Progress 21: 1366 1372. 135. Yokota, A., M. Henmi, N. Takaoka, C. Hayashi, Y. Takezawa, Y. Fukumori, and F. Tomita. 1997. Enhancement of glucose metabolism in a pyruvic acid hyperproducing Esch erichia coli mutant defective in F 1 ATPase activity. Journal of Fermentation and Bioengineering 83: 132 138. 136. Zhang, T., C. Z. Cui, S. L. Chen, X. P. Ai, H. X. Yang, S. Ping, and Z. R. Peng. 2006. A novel mediatorless microbial fuel cell based on direct biocatalysis of Escherichia coli Chemical Communications : 2257 2259. 137. Zhang, X., K. Jantama, J. C. Moore, K. T. Shanmugam, and L. O. Ingram. 2007. Production of L alanine by metabolically engineered Escherichia coli Applied Microbiology and Biotechno logy 77: 355 366. 138. Zhao, F., R. C. T. Slade, and J. R. Varcoe. 2009. Techniques for the study and development of microbial fuel cells: an electrochemical perspective. Chemical Society Reviews 38: 1926 1939. 139. Zhou, S., L. O. Ingram, K. T. Shanmugam, L Yomano, T. B. Grabar, and J. C. Moore. 2006. Materials and methods for efficient lactic acid production. US, Application number 11/501,137.

PAGE 142

142 140. Zilberstein, D., V. Agmon, S. Schuldiner, and E. Padan. 1984. Escherichia coli intracellular pH, membrane pot ential, and cell growth. Journal of Bacteriology 158: 246 252. 141. Zuo, Y., S. Cheng, D. Call, and B. E. Logan. 2007. Tubular membrane cathodes for scalable power generation in microbial fuel cells. Environmental Science & Technology 41: 3347 3353.

PAGE 143

143 BIOGR APHICAL SKETCH Jonathan Moore was born in Concord, Massachusetts in 1974. His family moved to Santa Barbara, California in 1983 and in 1992, Jonathan moved to La Jolla to attend the University of California, San Diego. Jonathan graduated from UC San Diego, Revelle College, with a Bachelor of Arts degree in biochemistry and cell biology in 1996. After receiving his undergraduate degree, Jonathan started a career in the biotechnology industry in the San Diego area. He worked as a researcher at Diversa Corpora tion until 2003, when his interest in renewable energy led him to join the the laboratory of Dr. Lonnie O. Ingram as a graduate student in the Department of Microbiology and Cell Scien ce at the University of Florida.