Production of Butyric Acid and Hydrogen by Metabolically Engineered Mutants of Clostridium Tyrobutyricum

Home , Acidogenesis, Butyrate kinase, Butyric acid, Metabolic pathway

PRODUCTION OF BUTYRIC ACID AND HYDROGEN BY METABOLICALLY ENGINEERED MUTANTS OF CLOSTRIDIUM TYROBUTYRICUM

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By Xiaoguang Liu, M.S. *****

The Ohio State University

2005

Dissertation committee: Approved by

Professor Shang-Tian Yang, Adviser

Professor Barbara E Wyslouzil Adviser

Professor Hua Wang Department of Chemical Engineering

ABSTRACT

Butyric acid has many applications in chemical, food and pharmaceutical

industries. The production of butyric acid by fermentation has become an increasingly

attractive alternative to current petroleum-based chemical synthesis. Clostridium tyrobutyricum is an anaerobic bacterium producing butyric acid, acetic acid, hydrogen and carbon dioxide as its main products. Hydrogen, as an energy byproduct, can add value to the fermentation process. The goal of this project was to develop novel bioprocess to produce butyric acid and hydrogen economically by Clostridial mutants.

Conventional fermentation technologies for butyric acid and hydrogen production are limited by low reactor productivity, product concentration and yield. In this project, novel engineered mutants of C. tyrobutyricum were created by gene manipulation and cell adaptation. Fermentation process was also optimized using immobilizing cells in the fibrous-bed bioreactor (FBB) to enhance butyric acid and hydrogen production.

First, metabolic engineered mutants with knocked-out acetate formation pathway were created and characterized. Gene inactivation technology was used to delete the genes of phosphotransacetylase (PTA) and acetate kinase (AK), two key enzymes in the acetate-producing pathway of C. tyrobutyricum, through homologous recombination. The metabolic engineered mutants were characterized by Southern hybridization, enzyme assay, protein expression and metabolites production. The enzyme assays showed that

PTA and AK activities in the pta-deleted mutant (PPTA-Em) were reduced by 44% and

91%, respectively, whereas AK activity in the ack-deleted mutant (PAK-Em) decreased by 50%. Meanwhile, the activity of BK in PPTA-Em increased by 44%, and hydrogenase activity in PAK-Em increased by 40%. The SDS-PAGE showed that the expression of the proteins around 32 kDa and 70 kDa had significant changes in the mutants. Two dimensional protein electrophoresis gels showed that both PTA and AK were deleted from mutants. Butyric acid tolerances were improved significantly in the mutants, indicating high butyric acid productivity.

The free cell fermentation by PPTA-Em produced 40 g/L of butyric acid using glucose with lower specific cell growth rate, lower acetate yield (0.058 g/g), higher butyric acid yield (0.38 g/g), butyrate productivity (0.63 g/L·h), and consequent higher selectivity of butyrate over acetate as compared with that of wild type. In order to improve butyric acid and hydrogen production, a novel fibrous-bed bioreactor (FBB) was applied to immobilize the metabolically engineered mutants. The immobilization fermentation using PPTA-Em showed that butyric acid concentration was improved to 50 g/L with higher butyric acid yield compared with that of free cell fermentation. The effect of sugar sources and cell adaptation on the fermentation was also studied.

As compared with wild type, the specific growth rate of PAK-Em from glucose decreased by 50% (from 0.24 h-1 to 0.14 h-1) because of the impaired PTA-AK pathway.

Meanwhile, butyric acid production by the mutant was improved greatly, with higher

butyric acid yield (0.42 g/g vs. 0.34 g/g) and final concentration (42 g/L vs. 20 g/L), Also,

hydrogen production by PAK-Em mutant increased significantly, with higher hydrogen

yield (2.61 vs. 1.35 mol/mol glucose) and H2/CO2 ratio (1.44 vs. 1.04). Free cell

iii

fermentation using various sugar sources, including glucose, xylose and fructose, were carried out to evaluate their abilities to produce butyric acid and hydrogen. Repeated fed- batch immobilization fermentations using FBB were optimized to improve butyric acid and hydrogen production further. Through adaptation in the FBB fibrous matrix, a high butyric acid concentration of 81 g/L was obtained at pH 6.3 with PAK-Em. This concentration was the highest ever attained in butyric acid fermentation to date. The butyrate yield was also increased to ~0.45 g/g due to the reduced cell growth in the immobilized-cell fermentation. A new adaptation mutant that produced even more hydrogen with a H2/CO2 ratio of 2.69 and with very fast growth rate was discovered from

the FBB matrix.

In order to better understand the metabolic mechanism of butyrate production by

the metabolically engineered mutant of C. tyrobutyricum, metabolic flux analysis was

applied to quantitatively describe the global cellular metabolism. Different pH values,

from pH 5.0 to pH 7.0, were applied with the FBB fermentation by PAK-Em using

glucose and xylose to evaluate the metabolic changes. The stoichiometric analysis

indicated the PTA-AK metabolic pathway to produce acetic acid was blocked completely

in the mutant.

The novel metabolic engineered mutants and the FBB application are important to

the development of an economical bioprocess for butyric acid and hydrogen production

from biomass by C. tyrobutyricm.

Dedicated to my parents and my husband Lufang Zhou

ACKNOWLEDGMENTS

My great appreciation goes to my advisor, Dr. Shang-Tian Yang, for his

enormous help both academically and financially throughout my Ph.D. study. I am

eternally grateful for his inspiring advice, warm support and encouragement, and the flexibility he gave me in conducting my research. I gained a lot from his insights and I have greatly enjoyed my staying here as his student.

I would also like to acknowledge with sincere gratitude to the members of my dissertation committee, Dr. Hua Wang and Dr. Barbara Wyslouzil. I am grateful for their helpful suggestions and advice offered on a variety of topics.

I would like to thank Dr. Ying Zhu and Ms. Yali Zhang for their help in the work

reported in Chapters 3 and 8, respectively. The technical assistance provided by Dr. Ying

Zhu at the initial stage of my research is gratefully acknowledged.

My colleagues within my research group in the Department of Chemical

Engineering offered many kinds of support over the last four years. I have been fortunate

in working in such a good team.

Financial support from the Department of Energy-STTR Program, the Consortium

for Plant Biotechnology Research, and the U.S. Department of Agriculture-NRI Program

to the various phases of this work is acknowledged.

Special thanks go to my husband Lufang Zhou and my parents for their sustained

support over the last four years. vi

VITA

November 3, 1975 ...……………………………….Born – Shouguang, China July, 1997…...………………………………………B.S. Chemical Engineering Shandong (Tech) University Shandong, China

April, 2000 …...………………………………..……M.S. Biochemical Engineering Tianjin University Tianjin, China August, 2000 – July, 2001...………………………...University Research Assistant University of Connecticut September, 2001 – August, 2002...…………………University Fellow The Ohio State University September, 2002 – September, 2003..……………... Graduate Research Assistant Biochemical Engineering The Ohio State University October, 2003 – September, 2004 ………………….CPBR Research Fellow The Ohio State University October, 2004 – August, 2005..……………...... Graduate Research Assistant Biochemical Engineering The Ohio State University June, 2005..……………………………...…...... Alumni Grants for Graduate Research and Scholarship (AGGRS) The Ohio State University

vii

PUBLICATION

1. Liu, X., Zhu, Y., Yang, S.T. 2005. Butyric acid and hydrogen production by Clostridium tyrobutyricum ATCC 25755 and mtants. Enzyme Microbial Technolo. In press.

2. Zhu, Y., Liu, X., Yang, S.T. 2005. Construction and characterization of pta gene deleted mutant of Clostridium tyrobutyricm for enhanced butyric acid fermentation. Biotechnol. Bioeng. 90, 154-166.

3. Dong, X., Wang, Y., Liu, X., Y. Sun. 2001. Kinetic model of lysozyme renaturation with the molecular chaperone GroEL. Biotechnol. Lett. 23, 1165-1169.

4. Dong, X., Bai, S., Liu, X., Sun, Y. 2001. Kinetics of lysozyme refolding facilitated by molecular chaperone GroEL. Huagong Xuebao. 52, 1049-1053.

5. Liu, X., Dong, X. 2000. Molecular chaperone and protein renaturation. Chem. Ind. Eng. 17, 120-124.

6. Liu, X., Dong, X., Zhou, L., Wang, Y., Zeng, K., Sun, Y. 2000. Kinetics of lysozyme refolding assisted by chaperonin GroEL. Proceeding paper at 10th National Conference on Chemical Engineering.

FIELD OF STUDY

Major Field: Chemical Engineering Specialty: Biochemical Engineering

viii

TABLE OF CONTENTS

Page

ABSTRACT ....…...……………….……………………………………….…….….……ii DEDICATION .…..…...………………………..……………………………...…….…...v

ACKNOWLEDGEMENT ....….……………………………………………….…….....vi VITA ...………………...…………………………………….……………………...... vii LIST OF TABLES ….…………….……………….……….…………...………….…viii LIST OF FIGURES …....…………………………………..…………………………..xiv Chapters:

1. Introduction ...………….…………………………………………………..…….1

2. Literature Review………………………………………………………..……..13

2.1 Metabolic Engineering and Mutant Development ….……...……………13 2.1.1 Gene Manipulation..…………...….………………………….....…..…14 2.1.2 Other Methods...…………………….….…………………….....…..…19 2.1.3 Mutant Characterization..….………..….…………………….....…..…22 2.2 Butyric Acid and Hydrogen Production from Fermentation ..……...…23 2.2.1 Butyric acid and Hydrogen…….……….…………………….....…..…23 2.2.2 Microorganisms for Fermentation…..………………………….....…..…25 2.2.3 Immobilization Cell Fermentation ..………………………….....…..…29 2.3 Metabolic Pathway and Flux Analysis ..…...……………..…………..…31 2.3.1 Metabolic Pathway ..……….……….……………………….....…..…31 2.3.2 Metabolic Flux Analysis ..………………….……………….....…..…33 2.4 Functional Genomics and Proteomics …………..………………….……36 2.4.1 Functional Genomics ………….…………………………….…..……37 2.4.2 Proteomics .………………………………………………….…..……38 2.4.3 Application of Function Genomics and Proteomics …..………………40 2.5 References ..….……………………………………………………………42

3. Construction and Characterization of pta Deleted Mutant…...... ….52

Summary…...……………………..…….…………………………...…….....…52 3.1 Introduction...…………….…..……………………………..………...……54 3.2 Materials and Methods...………..….………………………………...……56 3.2.1 Bacterial Strains and Plasmids ….…….……………...…………...... …56 3.2.2 DNA Manipulations ……….. .….…….……………………….……....57 3.2.3 Transformation ….………………………………….……..…....…...... 59 3.2.4 Southern Hybridization ………..………….…………………….……59 3.2.5 Characterization of Mutants … ..………….…………………….……60 3.3 Results.…………...……………………..…….…………………..…...……63 3.3.1 PCR Amplification and Sequence Analysis ………...…………...... …63 3.3.2 Transformation …… ……….….…….……………………….……....64 3.3.3 Southern Hybridization …………………………….……..…....…...... 65 3.3.4 Protein Expression …....…………………………….……..…....…...... 66 3.3.5 Enzyme Activity …..……………………………….……..…....…...... 67 3.3.6 Fermentation Kinetics….. ………………………….……..…....…...... 68 3.3.7 Butyrate Tolerance….. ……………….…………….……..…....…...... 69 3.4 Discussion……………………………….……..……………….……...……70 3.4.1 Cloning of C. tyrobutyricum ………… …………...…………...... …70 3.4.2 Protein Expression ………….….…….……………………….……....72 3.4.3 Effects of Integrational Mutagenesis……………….……..…....…...... 72 3.5 Conclusion …………………..………….……..……………….……...……75 3.6 References …...... ………...…………………...……………………...……..77

4. Butyric Acid Production by PPTA-Em Mutant …………..…………….……91

Summary…...... …………..………...………………………...……....…...91 4.1 Introduction ...…………………...…..…………..…………..………...……93 4.2 Materials and Methods………….....………..………………………...……95 4.2.1 Culture and Medium.…………….…...…..…………...…………...... …95 4.2.2 Fermentation Kinetics ………….…….……………..………….……....95 4.2.3 Fermentation in Fibrous-bed Bioreactor.……....………....…...... …...... 96 4.2.4 Analytical Methods ……………………………...…..…..…..……..….……97 4.2.5 Fermentation …..………………………..….….………………….……97 4.3 Results and Discussion..………………………….………………..…...……99 4.3.1 Fermentation Kinetics.………..…………..…………...…………...... …99 4.3.2 Effects on Cell Growth…....…..….…………..……………….……....100 4.3.3 Effects of Butyric and Acetic Acids Production……..…....…....…...... 101 4.3.4 Effects on Gas Production…………………..…..……..……..….……102 4.3.5 SDS-PAGE and 2-DE Analysis of Protein Expression..……..….……102 4.3.6 Effects of Carbon Source…..………………..…..……..……..….……104 4.3.7 Immobilized Cell Fermentation in FBB..……..………………………104

4.4 Conclusion ..….……………………………….……….……….……...……105 4.5 References ...……………………………..…………………………...……106

5. Construction and Characterization of ack Deleted Mutant...... ….116

Summary.....……………………..…….…………………………...…….....…116 5.1 Introduction.…………….…..……………………………..………...……118 5.2 Materials and Methods.………..….………………………………...……121 5.2.1 Culture and Medium…………...…….……………...…………...... …121 5.2.2 Mutant Development……….….…….……………………….……....121 5.2.3 Enzyme Assay and SDS-PAGE …..……………….……..…....…...... 123 5.2.4 Butyric Acid Tolerance Study....………….…………………….……124 5.2.5 Fermentation ..………………………….. .…………………….……125 5.3 Results and Discussion………………..…….…………………..…...……126 5.3.1 Cloning and Mutant Characterization…….………...…………...... …126 5.3.2 Enzyme Assay. …………….….…….……………………….……....127 5.3.3 Protein Expression ….….………………………….……..…....…...... 128 5.3.4 Butyric Acid Tolerance….. .……………………….……..…....…...... 129 5.3.5 Butyric Acid and Hydrogen Fermentation…….…..……..…....…...... 129 5.3.6 Metabolic Shift Analysis………….……………….……..…....…...... 132 5.3.7 Effect of Gene Manipulation and Cell Adaptation...….…..…....…...... 133 5.4 Conclusion………………..………….……....……………….……...……134 5.5 References…...... ………...…………………...……………………...……136

6. Butyric Acid and Hydrogen Production by PAK-Em Mutant………..…....148

Summary ……..……………………………..……………………...…….....…148 6.1 Introduction……………………...……..…………………..………...……150 6.2 Materials and Methods..…..…..…………………...………………...……152 6.2.1 Culture and Media …..………………………….….……..…....…...... 152 6.2.2 Fermentation Kinetic Studies……...…………….….……..…....…...... 153 6.2.3 Adapted Mutant Screening…..…….……………….…………….…....154 6.2.4 Preparation of Cell Extract and SDS-PAGE …….…………….…...... 154 6.2.5 Analytical Methods …………………………………………….…...... 155 6.3 Results and Discussions ……………………….………………..…...……155 6.3.1 Free Cell Fermentation Using Different Sugar Sources………..…...... 155 6.3.2 Fermentation Using Fibrous-bed Bioreactor..…………….…....…...... 159 6.3.3 Fermentation by Mutant at Different pHs.………….……..…....…...... 161 6.3.4 Fermentation by the Adapted Mutant.…………..………....…....…...... 163 6.3.5 Protein Expression… ……………………..…….….……..…....…...... 163 6.4 Conclusion ..….……………………………….……….……….……...……165 6.5 References ...……………………………..…………………………...……166

7. Functional Genomics and Future Work ……………………………………178

7.1 Conclusions ....…………..……………...……………………...…….....…178 7.1.1 Fermentation by PPTA-Em ……...... ………..…..…….…..…...... 178 7.1.2 Fermentation by PAK-Em ….………....…….……….…….……...... 179 7.1.3 Protein Expression with Gene Manipulation ….....…..…….…..…...... 181 7.1.4 Metabolic Flux and Pathway ……...…….....………..…….…..…...... 181

7.2 Recommendations ..…………..……………..……………..………...……182 7.2.1 Functional Genomics by DNA Microarray ..…………...... ……….....183 7.2.2 Protein Electrophoresis and Proteomics ……………………..……....184 7.2.3 Re-Engineer Metabolically Mutant..…………………………..……....185

BIBLIOGRAPHY………………………………………..………………………..…..189

Appendices:

Appendix A Medium Compositions………………………….…...…….....…202 Appendix B Analytial Methods………………….…………….………...……203 B.1 High Performance Liquid Chromatography ...……….….………....203 B.2 Gas Production by Micro-oxymax …………….…..………………203 B.3 Enzyme Assays and SDS-PAGE…..…………….……………...…204 B.4 2D Protein Electrophoresis …………….………………..……....…205 B.5 Southern Hybridization ………………..………………..……....…206 B.6 Metabolic Flux Analysis….. …………..………………..……....…207 Appendix C Bioreactor Construction and Operation …….…………..……209 C.1 Construction of Immobilized Cell Bioreactor .…...... ……209 C.2 Bioreactor Start-up and Operation ...... …………….…...…...……209 Appendix D Protocols of Genetic Experiments …………….………….……211 D.1 Genome Library Construction.……………………………..………211 D.2 DNA Microarray Construction ...... ….212 D.3 RNA Extraction, Labeling and Chip Hybridization ..…..…………214 D.4 Scanning and Data Analysis…….………………………………….215 D.5 DNA Sequence Analysis ...………………….………..……………216 D.6 2D-PAGE Data Analysis ………………...…………….….………216 D.7 Protein Separation and Purification …...... ….217 D.8 Protein Sequencing by Mass Spectrometer Analysis ….….………217 D.9 Mutant Library Construction…… ….….…………………….……218 Appendix E Reagents and Buffers …...……………..………...………...……210

xii

LIST OF TABLES

Table Page

3.1 Bacterial strains and plasmids ……………………………………..….…….…..81

3.2 Free cell fermentation by wild type and PPTA-Em ……….………………………...82

4.1 Comparison of fermentation with glucose by PPTA-Em and wild type .…..…..107 4.2 Comparison of fermentation with xylose by PPTA-Em and wild type .…..……108 4.3 Comparison of protein expression in PPTA-Em and wild type ……………..…109 5.1 Comparison of fermentation results by wild type and PAK-Em..………………138

5.2 Metabolic shift by wild type and PAK-Em at pH 5.0 …… .….…………………139 6.1 Comparison of free cell fermentation by wild type and PAK-Em using different sugar sources ……………………………………………....…..168 6.2 Comparison of immobilized cell fermentation by wild type and PAK-Em using different sugar sources ……………………………………………....…..169 6.3 Effect of pH on FBB fermentation by PAK-Em using glucose……...…..……..170

xiii

LIST OF FIGURES

Figure Page

1.1 Research objectives and scope of this study ……………...………………..…….12 3.1 Construction of integrational plasmid pPTA-Em……..…..……………………..83

3.2 Amino acid alignment… ………………………...…………………..…………..84 3.3 Southern hybridization of PPTA-Em and wild type……….....………….…….... 85 3.4 SDS-PAGE of PPTA-Em and wild type .……………………...... ……..…....…86 3.5 2D protein electrophoresis of PPTA-Em and wild type .………….….....… ….…87 3.6 Enzyme assay .………….…...... …..……88 3.7 Fermentation kinetics of free wild type and PPTA-Em from glucose....…….. …89

3.8 Butyric acid inhibition model....………………………..……………….…..… …90 4.1 Fermentation kinetics of free wild type and PPTA-Em using glucose....………………………………………….…………… ….……110 4.2 Fermentation kinetics of immobilized wild type and PPTA-Em using xylose....………………………………………...……………… ….……111 4.3 SDS-PAGE by free and immobilized PPTA-EM and wild type cell ……..…....112 4.4 2D protein electrophoresis of PPTA-Em and wild type using glucose …….…..113 4.5 2D protein electrophoresis of PPTA-Em and wild type using xylose …..….…..114 4.6 Kinetics of immobilization fermentation with glucose using PPTA-Em and wild type using xylose …..…………….…………………………..115 5.1 Construction of pAK-Em plasmid …………………. …………………………140 5.2 Amino acid sequence pf ack alignment………………………...……………… 141 5.3 Enzyme assay of PAK-Em …………………………………..…………………142 5.4 SDS-PAGE of wild type and PAK-Em …...……………………………………143

xiv

5.5 Butyric acid inhibition of wild type and PAK-Em ………………..……………144 5.6 Fermentation kinetics by free cell of wild type and PAK-Em …...………….…146 5.7 Immobilization Fermentation by PAK-Em at pH 5.0………………….….…….147 6.1 Fermentation kinetics by free PAK-Em using different sugar sources………….172 6.2 Fermentation kinetics by PAK-Em in FBB using different sugar sources ………173 6.3 Effect of pH on FBB Fermentation Kinetics by PAK-Em………………….…. 175 6.4 Fermentation kinetics by free HydEm using glucose…………………...………176

6.5 SDS-PAGE of immobilized wild type and PAK-Em …...………………………177 A.1 Shotgun DNA microarray ….……………………………………………..…....212 A.2 Mutant library construction……………………………………………….…….219

CHAPTER 1

INTRODUCTION

Clostridium tyrobutyricum is a gram-positive, rod-shaped, spore-forming,

obligate anaerobic bacterium that produces butyric acid, acetic acid, hydrogen, and

carbon dioxide as its main fermentation products from various carbohydrates, including

glucose and xylose (Wu and Yang, 2003). Butyric acid (CH3CH2CH2COOH) is used to

synthesize butyryl polymers in chemical industry and to enhance butter-like note in food

flavors in the food industry. Esters of butyrate are used as additives for increasing fruit

fragrance and as aromatic compounds for the production of perfumes. Butyric acid, as

one of the short-chain fatty acids generated by the anaerobic fermentation of dietary

substrates, is known to have therapeutic effects on colorectal cancer and

hemoglobinopathies (Willims, et al., 2003). The production of butyric acid from

renewable resources has become an increasingly attractive alternative to the current

petroleum-based chemical synthesis because of public concerns about the environmental pollution caused by the petrochemical industry and consumer’ preference for bio-based natural ingredients for foods, cosmetics, and pharmaceuticals.

Hydrogen, having a high energy content per unit weight (141.86 kJ/g or 61,000

Btu/lb), can be used as a clean fuel and easily converted to electricity by fuel cells.

Hydrogen is thus considered the most promising future fuel if its production cost can be greatly reduced (Dunn, 2002). Hydrogen can be generated in several ways, including catalytic fuel reforming and electrolysis of water. However, the present chemical routes of hydrogen production are energy intensive and expensive than fossil fuels. On the other hand, biological hydrogen production, either by photosynthesis with algae and photosynthetic bacteria or by fermentation with anaerobic bacteria, can be operated at ambient temperature and pressure (Momirlan and Veziroglu, 1999; Das and Veziroglu,

2001). Hydrogen production by anaerobic fermentation offers an attractive method to produce energy when low-cost renewable biomass is available as the feedstock (Mizuno, et al., 2000). Hydrogen, as an energy byproduct from butyric acid fermentation, can add value to the fermentation process.

The production of butyric acid and hydrogen by fermentation is both environmentally and economically significant, but conventional fermentation processes suffer from low productivity, low final product concentration, and low yield. The development of an effective fermentation technology for the economical production of butyric acid appears to be extremely necessary. In order to economically produce butyric acid from biomass, it is desirable to further improve the final product yield and concentration of the fermentation process. It is also desirable to reduce acetate production in order to facilitate the separation and purification of the final product, butyric acid.

In general, glucose (hexose) is catabolized via EMP (Embden-Meyerhof-Parnas) pathway and xylose (pentose) is catabolized by HMP (Hexose Monophosphae) pathway to pyruvate, which is then oxidized to acetyl-CoA and carbon dioxide with concomitant reduction of ferredoxin (Fd) to FdH2. FdH2 is then oxidized to Fd to produce hydrogen, which is catalyzed by hydrogenase, with excess electron released to convert NAD+ to

NADH. Acetyl-CoA is the key metabolic intermediate at the node dividing the acetate-

formation branch from butyrate-formation branch. Phosphotransacetylase (PTA) and

acetate kinase (AK) are two enzymes that convert acetyl-CoA to acetic acid, whereas

phosphotransbutyrylase (PTB) and butyrate kinase (BK) catalyze the production of

butyric acid from butyryl-CoA. Integrational mutagenesis, a genetic engineering

technique that can selectively inactivate undesired genes from host chromosome, has

been developed and successfully used to create metabolically engineered mutants of

Clostridial strains (Green et al., 1996). In this technique, a fragment of the target gene is

cloned into a non-replicative vector with a selection marker, resulting in the non-

replicative integrational plasmid. The partial gene in the non-replicative plasmid can

recombine with the internal homologous region of the original target gene in the parental

chromosome, which results in the insertional inactivation of the target gene. Previous studies using integrational mutagenesis to improve clostridial fermentation product yield have focused on solventogenic C. acetobutylicum. A non-replicative plasmid (pJC4) with partial the pta gene (encoding PTA) was constructed and integrated into the homologous region of pta gene on the chromosome of C. acetobutylicum ATCC 824, resulting in reduced PTA and AK activities and acetate production (Green and Bennett, 1998; Green et al., 1996). The same metabolic engineering approach can be used to reduce or

eliminate acetate production in butyric acid fermentation by C. tyrobutyricum. In this study, integrational mutagenesis was applied to inactivate the pta or ack gene by non- replicative plasmid to develop metabolically engineered mutants of C. tyrobutyricum, which were used to enhance butyric acid and hydrogen production by fermentation.

Current fermentation processes for butyric acid have poor efficiency partly due to the physiological characteristics of the fermentation, the dependence of acid selectivity on cell growth rate and product inhibition (Michel-Savin et al., 1990a and 1990b).

Extensive work has been done to improve fermentation productivity and product concentration by using immobilized cells, continuous bioreactors, cell recycling, and extractive fermentation processes. However, no competitive fermentation process is available for efficient butyrate production. The application of fibrous-bed bioreactor

(FBB) in several organic acid fermentations has shown to give high rate production with more than tenfold increase in productivity and stable operation over a long period. Cells in the FBB were able to quickly adapt themselves to tolerate high concentrations of inhibitory fermentation products and toxic substrates (Huang and Yang, 1998; Huang et al., 1998). Therefore, the FBB operated in a recycle batch mode should have the advantages of producing high product concentration and fermentation productivity, thus making the butyrate fermentation process attractive for industrial application. The superior performance of the FBB is dependent on the high density of active cells maintained in the fibrous bed.

Meanwhile, elucidating the patterns of metabolic pathways in butyric acid fermentation at the biochemical level is useful for investigating and controlling the complex metabolism of the bacterium. A quantitative understanding of flux distributions

among various metabolic pathways would provide a guideline for process optimization to direct flux distribution to desirable product (butyrate) and to eliminate or reduce byproducts (acetate and lactate). The effective way to alter the flux distribution in a metabolic network is by changing the metabolic pathway through metabolic engineering, or more specifically, by knocking out certain undesirable genes and over-expressing certain desirable genes in the pathway. Presently, most studies on the metabolic engineering of clostridia have been limited to solventogenic C. acetobutylicum ATCC

824. The genetic or metabolic information from previous C. acetobutylicum research would be helpful to develop proper protocol for gene manipulation in acidogenic C. tyrobutyricum.

Recent years have seen an explosion in the number of complete or almost complete genomic sequences of organisms including several bacteria that are important to agriculture and food processing (de Vos et al., 2004; de Vos and Hugenholtz, 2004;

Mill, 2003). The vast genome sequence information can be used to manipulate the metabolism or physiology of the producer organism, and eventually this information can be integrated into the development of more efficient production strains. Functional genomic microbes, based on rapidly emerging genome sequence information, generate valuable knowledge that can be used for metabolic engineering, improving cell factories, and developing functional foods and novel preservation methods (Kuipers, 1999; de Vos et al., 2004). DNA microarray is increasingly used for the functional analysis of genes.

However, a previous knowledge of the DNA sequence usually is a prerequisite for gene chip construction from already known genes, which requires one to undertake expensive and time-consuming genome sequencing projects in order to study the function of any

particular set of genes. But for most of industrially important bacterial strains, including

C. tyrobutyricum, there is either very little or no genomic sequencing information available. In this case, a shotgun DNA microarray strategy has been successfully used for gene expression analysis (Hayward et al., 2000), bypassing the need for previous knowledge of the genome sequence. In a shotgun approach, the genome is represented by random DNA fragments of unknown sequence that are arrayed on a glass slide. DNA fragments that contain regulated genes are identified by differential hybridization with two labeled cDNAs. The identity of these genes need to be elucidated by subsequent sequencing. Shotgun DNA microarrays have been constructed and used to study the metabolism of archaeon Haloferax volvanii (Zaigler et al., 2003), gene function in environmental isolates of Leptospirillum ferrooxidans (Parro and Moreno-Paz, 2003), and phylogenetic lineages of Listeria monocytogenes (Zhang et al., 2003). Shotgun DNA microarrays can be developed and used for functional genomic studies of Clostridium tyrobutyricum.

Conventional methods for the improvement of industrial microorganisms range from the random approach of classical strain improvement (CSI) to the highly rational methods of metabolic engineering. Although this rational metabolic engineering method produced greatly improved mutants for the production of butyric acid, it is information and tool intensive and relies on methods that simplify biological systems for the industrial bacterial mutant development. For example, we can only work with one or two genes in the time consuming gene manipulation. Besides these conventional strain improvement methods, we also applied culture adaptation to develop new mutants, such as the high acid tolerance mutant and high productivity mutant. But adaptation has a

relatively low efficiency and uncertain results as compared with the genetic or metabolic method. Recently, the combinatorial approach, genome shuffling, for the determination of optimal genetic configuration in microbes for industrial application has been developed (Patnaik, et al., 2002; Zhang, et al., 2002). “In this method, the cells that are most useful with respect to the phenotype to interest are screened or selected from a genotypically diverse collection (or library) of candidates. Such libraries have been generated in the past by random mutation of cells, successive rounds of mutation and screening/selection of individual genes (directed evolution), or shuffling of gene homologs from different strains” (Stephanopoulos, 2002). In this study, the mutant libraries were constructed based on the random genome library of C. tyrobutyricum.

Objectives

The overall goal of this study was to develop and characterize novel metabolic engineered mutants to improve butyric acid and hydrogen production. The bioprocess was also developed and optimized to economically produce butyric acid by using C. tyrobutyricum mutants immobilized in a fibrous-bed bioreactor. To reach this goal, the following specific objectives were proposed:

1. To develop and characterize the metabolic engineered mutants of C. tyrobutyricum from integrational mutagenesis

To improve the production of butyric acid and hydrogen by C. tyrobutyricum, two metabolically engineered mutants with inactivated acetate kinase gene (ack) and phosphotransacetylase gene (pta) were created by insertional inactivation of the gene(s) on the chromosome by homologous recombination with integrational plasmids. It was

expected that by knocking out ack and pta genes, the acetate formation pathway would be impaired, resulting in global changes in the metabolic pathway leading to more production of butyric acid and hydrogen. Integrating the plasmid into the homologous region on the chromosome inactivated the target gene and produced the pta-deleted mutant (PPTA-Em) and ack-deleted mutant (PAK-Em), which was confirmed by

Southern hybridization. SDS-PAGE and two-dimensional protein electrophoresis results indicated that protein expressions were changed in the mutants. Enzyme activity assays using the cell lysate showed that the activities of PTA and AK in the mutant were reduced greatly.

2. To characterize PPTA-Em for butyric acid and hydrogen production

Both free cell fermentation and FBB fermentation were carried out to evaluate butyric acid and hydrogen production ability by PPTA-Em. Different sugar sources, including glucose and xylose, were applied to the fermentation to understand the metabolite production changes. The SDS-PAGE experiment was used to study the protein expression changes of the free cell, the immobilization cell, and the effects of glucose and xylose.

3. To improve butyric acid and hydrogen production by PAK-Em

The kinetics of glucose, xylose, and grape juice fermentations using free cells of the mutant were studied to evaluate its fermentation ability from different substrates. The

FBB was applied to immobilize mutant cells to further improve the acid and hydrogen production from glucose fermentation. One adaptation mutant (HydEm) was screened

from the FBB cotton matrix and the glucose fermentation kinetics was studied. The results of improving the fermentation production by developing metabolically engineered mutants, applying novel bioreactors and optimizing the operation conditions were compared and discussed in this paper, which will be helpful for directing future large- scale industry production. The protein expressions of PAK-Em under different conditions were studied by SDS-PAGE.

4. To understand the metabolic pathway of C. tyrobutyricum by metabolic shift and metabolic flux analysis

Finally, in order to improve the product yield, butyric acid selectivity and concentration, glucose fermentations under different pH values, in the range of 5.0~7.0, were performed optimize the fermentation process using FBB and better understand the metabolic changes in the PAK-Em mutant. The metabolic shift results by PAK-Em were compared with the wild type to understand the metabolic pathway changes and clarify the metabolic pathway to produce butyric acid in C. tyrobutyricum. Metabolic flux analysis was done to understand the quantitative metabolic flux distribution changes in the PAK-

Em mutant. The result from this study will provide guidelines for metabolic engineering to eliminate acetate formation in butyrate fermentation.

5. To study the functional genomics and proteomics of C. tyrbutyricum by DNA microarray

The main objectives of this section are to globally study the bacterial genomic function, analyze and identify the key genes and proteins present in wild type and mutant

strains of C. tyrobutyricum under different environmental conditions by developing and utilizing shotgun DNA microarrays and proteomics; design metabolically engineered industrial mutants by rational genetic methods based on the genomic and proteomic understanding obtained from the shotgun microarray experiment; and improve the bacterial strains by combining all the merits of the currently available mutants in our lab into one mutant using novel genome shuffling. The focus of the study will be on the improvement of cells with high butyric acid production, the development of mutants with high tolerance to extreme environmental conditions, and cell surviving mechanisms in response to environmental changes in, e.g., pH, temperature, and salt and product concentrations. The results from the functional genomic studies can be used in the genetic and metabolic engineering of mutants for industrial applications.

Figure 1.1 illustrates the main research objectives and the scope of this study. The main body of this dissertation includes eight chapters. Chapter 1 gives an overall statement of this research. Chapter 2 provides the background information from the literature on gene manipulation for the development of engineered mutants, butyric acid and hydrogen fermentation, metabolic pathway and metabolic flux analysis, functional genomic and proteomics studies, and the construction of mutant library. The detailed construction process of metabolically engineered mutant PPTA-Em by pta gene inactivation is given in Chapter 3, along with the characterization of the metabolic engineered mutants. The ability to produce butyric acid and hydrogen using PPTA-Em and the metabolic flux analysis are discussed in Chapter 4. Chapter 5 presents the construction of ack gene knocked out mutant PAK-Em and the mutant characterization by enzyme assay, protein expression and metabolites analysis. The results of butyric acid

and hydrogen production by PPTA-Em using both free-cell and immobilized-cell fermentations from different sugars sources and their metabolic flux analysis results are reported in Chapter 6. The effect of pH on immobilized PAK-Em cells in FBB bioreactor, metabolic shift, metabolic flux analysis and metabolic pathway analysis are described in

Chapter 7. Chapter 8 discusses the conclusions of this research and future research recommendations, including random genome library construction, functional genomic studies by shotgun microarray, and recommendations for future work.

Objective Improve the butyric acid and hydrogen production by metabolic engineered industrial mutants of C. tyrobutyricum

Gene manipulation Functional genomics and

Develop pta and ack gene deleted proteomics analysis mutants by integrational mutagenesis • Random genome library construction • Shotgun DNA microarray • 2D protein electrophoresis Butyric acid and hydrogen • Genomic and proteomic analysis production • Free cell fermentation by PPTA-Em • FBB fermentation by PPTA-Em • Free cell fermentation by PAK-Em Novel industrial mutants • FBB fermentation by PAK-Em • Adaptation mutant screening development • Mutant library construction • High throughput mutant screening • Mutant improvement by genome shuffling Metabolic pathway and flux analysis • Metabolic flux analysis • Metabolic shift analysis • Metabolic pathway analysis

Figure 1.1 Research objectives and scope of this study

CHAPTER 2

LITERATURE REVIEW

2.1 Metabolic Engineering and Mutant Development

Metabolic engineering has been widely used to design engineered strains to achieve higher efficiencies in metabolite overproduction through alterations in the metabolic flux distribution. Most metabolic engineering work to date is related to the production of secondary metabolites (e.g., antibiotics), amino acids (e.g., lysine), and heterologous proteins using organisms with well studied genetics and physiology (e.g., E. coli, yeast, and hybridoma cells). Stoichiometric analysis of metabolic flux distributions provides a guide to metabolic engineering, optimal medium formulation and feeding strategies, and bioprocess optimization. However, very little research has been done on butyric acid fermentations of C. tyrobutyricum. Genetic modification of the byproduct, acetic acid, formation pathway can be one effective method for reducing acetate formation and directing more carbon towards butyrate production. Since both the acetic acid and butyric acid formation pathways are responsible for generating energy (ATP) in cells, reducing acetate production as a result of gene disruption of the acetate-forming enzymes may impose a metabolic burden on cells. A feasible cellular response to this

metabolic burden is elevation of the flux through the alternate ATP-generation pathway, namely butyrate formation, which might be beneficial for butyric acid fermentation.

2.1.1 Gene Manipulations to Develop Mutant

2.1.1.1 Butyric Acid and Hydrogen Producing Bacteria

Butyric acid can be produced by several bacterial species such as Clostridium,

Butyrvibrio, Butyribacterium, Sarcina, Eubacterium, Fusobacterium, and Megasphera

(Playne, 1985; Sneath, 1986) under anaerobic process. Clostridial species are preferred for commercial butyric acid production because they can form resistant endospores under harsh environments. Most of these butyric acid-producing bacteria produce acetic acid in addition to butyric acid as their major fermentation products. There are two kinds of microorganisms to produce biological hydrogen: one is anaerobic bacteria (obligate anaerobes or facultative anaerobes), such as C. acetobutylicum, C. butyricum, C. pasteurianum, Enterobacter aerogenes, and Enterobacter cloacae, Desulfovibrio vulgaris,

Magashaera elsdenii, Citrobacter intermedius, and the other is photosynthetic bacteria, such as Rhodobacter sphaeroides, and Rhodobacter capsulatus. The production of hydrogen through anaerobic fermentation technique has some advantages because it is environmentally friendly, low energy intensive, and waste reusinge, thus decreasing environmental pollution. There is a wide range of clostridia species, including pure and mixed cultures, such as C. acetobutyricum, C. butyricum (Chin, et al., 2003; Heyndrickx,

1987; Zigova, et al. 1999), Clostridium sp. no. 2, which produce hydrogen.

C. tyrobutyricum produces large amounts of acetate, butyrate, CO2, and H2 from

glucose fermentation (Huang et al., 2002). Some work has been done for butyric acid

production from C. tyrobutyricum, but hydrogen production from the fermentation of C.

tyrobutyricum has not yet been studied. In this project, C. tyrobutyricum ATCC 25755

was used to improve the production of butyric acid and hydrogen.

2.1.1.2 Genes and Enzymes Involved in the Acids Formation Pathways

Currently, there is no proteomics and genomics information about C. tyrobutyricum. Some general protein information is available from protein bank,

including pyruvate-ferredoxin oxidoreductate (EC 1.2.7.1), hydrogenase (EC 1.2.7.2),

NADH-ferredoxin oxidoreductase (EC 1.12.7.2), phosphotransacetylase (EC 2.3.1.8),

acetate kinase (EC 2.7.2.1), acetyl-CoA-acetyltransferase (EC 2.3.1.9), L(+)-β- hydroxybutyryl-CoA dehydrogenase (EC 1.3.99.3), L-3-hydroxyacyl-CoA hydrolase (EC

4.2.1.17), butyryl-CoA dehydrogenase (EC 1.3.99.2), CoA transferase (EC 2.8.3.9), phosphotransbutyrylase, butyrate kinase (EC 2.7.2.7), butyryl-CoA synthetase (EC

6.2.1.2), lactate dehydrogenase, and NADH-independent lactate dehydrogenase (EC

1.1.1.27). Only partial pta gene encoding phosphotransacetylase (#AAS77870, our lab), partial ack gene encoding acetate kinase (#AY706093, our lab) and complete enr gene encoding enoate reductase (#Y09960, with poor homology analysis) are listed in the

Genebank.

Phosphotransacetylase (PTA) and acetate kinase (AK) are two key enzymes involved in the pathway leading to acetic acid formation, whereas

phosphotransbutyrylase (PTB) and butyrate kinase (BK) catalyze the formation of butyric acid from butyryl-CoA. Phosphotransacetylase gene (pta) and acetate kinase gene (ack) encode PTA and AK enzymes involved in the metabolic pathway that forms acetic acid from acetyl-CoA, which plays an important role for the metabolic flux distribution of carbon and energy. To date, PTA and AK from several microorganisms has been purified and characterized (Boyton, et al., 1996). PTA is a monomer with a molecular weight of approximate 36.2 kDa in C. acetobutylicum, 88 kDa in C. thermoaceticum, and 27 kDa in

S. pyogenes. AK is a dimer of two identical subunits with various molecular masses in different species, 44.3 kDa in C. acetobutylicum and 60 kDa in C. thermoaceticum. The

PTB is about 31 kDa and the BK is approximately 39 kDa in C. acetobutylicum. Recently, two acetate kinase isozymes from spirochete MA-2 cell extracts (Harwood and Canale-

Parola, 1982) and a butyrate kinase isoenzyme (BKII) in C. acetobutylicum ATCC 824

(Huang, et. al, 2000) have been reported. The ack gene encoding AK and the pta gene encoding PTA have been cloned, sequenced and characterized from Escherichia coli

(Matsuyama et al., 1989; Kakuda, 1994), Methanosarcina thermophila (Latimer and

Ferry, 1993), and C. acetobutylicum (Boynton et al, 1996). The sequences for pta from

Paracoccus denitrificans, Bacillus subtilis, and Mycoplasma genitalium, and the sequences for ack from Haemophilus influenzae and Mycoplasma genitalium are also available in the genome database, but only defined by sequence homology (Boynton et al,

1996). Previous studies suggest that these two sets of acid-forming genes, ack and pta, buk and ptb exist in operons on the chromosome, with pta preceding ack and ptb preceding buk in other clostridial bactera.

2.1.1.3 Integrational Mutagenesis and Overexpression

In this project, the purpose is to improve the butyric acid and hydrogen production using metabolically engineered mutants developed by gene manipulation and other novel methods. To date, most of the gene manipulation in C. tyrobutyricum focused on its identification as the causative agent of late blowing in cheese by using PCR amplification of rRNA sequences (Klijn et al., 1995; van der Meer et al., 1993; Herman et al., 1995) or its flagellin gene for immunoenzymatic counting (Arnold et al., 1999).

One common problem in butyric acid fermentation is the co-production of acetate, which not only lowers the yield of butyric acid but also causes difficulty in product purification. Based on a stoichiometric analysis of the metabolites from fermentation, it is found that more butyrate can be produced if the PTA-AK metabolic pathway to acetate is blocked and the carbon source is redistributed via metabolic engineering. The theoretical maximum propionate yield is 48.8% (w/w) if there is no intermediate accumulation, no biomass is produced, and no acetate is produced. It is thus desirable to reduce or eliminate acetate formation by knocking out the genes associated with the acetate formation pathway.

Integrational mutagenesis, a genetic engineering technique that can selectively inactivate undesired genes from the host chromosome was developed and successfully used to create metabolically engineered mutants of Clostridial strains (Green et al., 1996).

In this technique, a fragment of the target gene was cloned into a non-replicative vector with a selection marker, resulting in the non-replicative integrational plasmid. The partial gene in the non-replicative plasmid recombined with the internal homologous region of the original target gene in the parental chromosome, which resulted in insertional 17

inactivation of the target gene. Many genetic engineering studies to improve product pattern and overall yield of clostridial fermentation were carried out in solventogenic C. acetobutylicum. Its genome sequence was determined (Nölling et al., 2001). Previous studies using integrational mutagenesis to improve clostridial fermentation product yield focused on solventogenic C. acetobutylicum. A non-replicative plasmid (pJC4) with partial pta gene (encoding PTA) was constructed and integrated into the homologous region of pta gene on the chromosome of C. acetobutylicum ATCC 824, resulting in reduced PTA and AK activities and acetate production (Green and Bennett, 1998; Green et al., 1996). The same metabolic engineering approach could be used to reduce or eliminate acetate production in butyric acid fermentation by C. tyrobutyricum.

Based on the metabolic stoichiometric analysis, it is possible to block the PTA-

AK acetate metabolic pathway and still maintain good cell growth and butyrate production. Therefore, eliminating acetate formation can be applied as a method to enhance butyrate yield from the fermentation of glucose. Recently, we have developed an integrational mutagenesis technique for disrupting genes in C. tyrobutyricum. Non- replicative integrational plasmid constructs containing either acetate kinase (ack) or phosphotransacetylase (pta) gene fragments and an erythromycin resistance cassette were introduced into C. tyrobutyricum. Inactivation of ack or pta occurred as a result of the integration of the plasmid into the homologous regions on the chromosome. Cell growth rate and acetate yield decreased in these mutants as compared to the wild-type, but final butyrate concentration and productivity increased significantly.

In addition to gene inactivation, gene overexpression was also employed in genetic modifications, which were carried out by replicative plasmid. Results from the

overexpression of ack and pta in C. acetobutylicum showed that the final ratios of acetate to other major products were higher and that there was a greater proportion of two- versus four-carbon-derived products (Boynton et al., 1996). Similar effects of overproduction of butyrate formation enzymes indicated that the butyrate-forming pathway enzymes were in excess in wild-type cells (Walter et al., 1994; Mermelstein et al., 1992) and in mutant strains lacking specific enzyme activities (Green and Bennett,

1998). Also, overexpression of genes (ptb and buk) in the butyrate production pathway was carried out via replicative plasmid pTHBUT containing a butyrate operon from C. acetobutylicum.

2.1.2 Other Methods for Developing Engineered Mutants

Fermentation-based bioprocesses rely extensively on strain improvement for commercialization. Whole-cell biocatalyst is commonly limited by low tolerance for extreme process conditions, such as temperature, pH, and solute concentration.

Conventional methods for the improvement of industrial microorganisms range from the random approach of classical strain improvement (CSI) to the highly rational methods of metabolic engineering. Although CSI is robust, it is time and resource intensive. To improve the butyric acid production yield, final concentration and productivity, we have applied the rational metabolic engineering approache, insertional mutagenesis, for developing novel mutants. The gene manipulation through insertional mutagenesis to decrease the production of the by-product by deleting of one or two specific genes encoding enzymes involved in the metabolic pathway of the byproduct has been

developed (Zhu, et al., 2005). The fermentation results using these novel mutants developed by gene inactivation manipulations show that the final product concentration, acid tolerance, yield, and productivity of these mutants were improved greatly as compared with the wild type bacteria.

Although the rational metabolic engineering method produced greatly improved mutants for the production of butyric acid, it was information and tool intensive and relied on methods that simplify biological systems for industrial bacterial mutant development. For example, we can only work with one or two genes in one gene manipulation. Although the resulting mutant reached our original goal to some extent, the genetic method resulted in some unexpected effects on the mutant at the same time, such as slow growth rate and little decrease of byproduct, due to the complex and rather poorly understood metabolic pathway in the bacteria. We also applied culture adaptation to develop new mutants, such as the highly acid tolerant mutant and highly productive mutant. But the adaptation was in efficient and unstable as compared with the genetic or metabolic method. Because the metabolic method is limited by the fact that the production of acid using bacterial cells depends on multiple of genes, which are poorly understood, mostly unknown, and broadly distributed throughout the whole genome, it is not easy to apply the direct genetic engineering to strain improvement.

Recently, the combinatorial approach, genome shuffling, for determining optimal genetic configuration in microbes for industrial application was developed (Patnaik, et al.,

2002; Zhang, et al., 2002). “In this method, the cells that are most useful with respect to the phenotype to interest are screened or selected from a genotypically diverse collection

(or library) of candidates. Such libraries have been generated in the past by random

mutation of cells, successive rounds of mutation and screening/selection of individual genes (directed evolution), or shuffling of gene homologs from different strains”

(Stephanopoulos, 2002). Using comparative genomic analysis to identify beneficial mutations, genome breeding could be applied to reconstruct a robust production strain, a minimally mutated strain with useful mutations and no unnecessary mutations (Ikeda and

Nakagawa, 2003). Genome shuffling is a process that combines the advantages of multi- parental crossing allowed by DNA shuffling with the recombination of entire genomes normally associated with conventional breeding. This method has been applied with several bacteria. Genome shuffling was applied with Streptomyces fradiae to improve the production of tylosin, which shows that two rounds of genome shuffling were sufficient to achieve results that had previously required 20 rounds of classical strain improvement

(CSI) (Zhang, et al., 2002). The acid-tolerant strains of Lactobacillus were isolated using this method (Patnaik, et al., 2002), and one newly shuffled lactobacilli that grew at a substantially lower pH than did the wild type strain was produced. Also one additional shuffled strain was identified that produced threefold more lactic acid than the wild type at pH 4.0. Recently, a new L-lysine producing mutant of Corynebacterium glutamicum consisting of characterization and reconstitution of a mutation set essential for high level production by genome shuffling has been developed (Ohnishi, et al., 2003). The fermentation at 40oC allowed an increase in yield of about 20% with a concomitant

decrease in final growth level as compared with the traditional 30oC fermentation, suggesting a significant transition of carbon flux distribution in glucose metabolism.

DNA array analysis of metabolic changes between the 30oC and 40oC fermentations

identified several differentially expressed genes in the central carbon metabolism.

2.1.3 Mutant Characterization Methods

After the improved mutants of C. tyrobutyricum were obtained, quantitative proteomic and transcriptomic profiling could be applied tothe fermentation studies of the mutants under different conditions. The comparison includes the changes between the parental strain and different mutants, different carbon sources, substrate exhaustion and the presence of alternate carbon sources, the entry into different growth phases, pH variation, temperature shift, cell adaptation in the fibrous-bed bioreactor, tolerance to fermentation products, and different toxic factors inactivating cell growth. The characterization includes several methods: i) The key acid-forming enzymes in the central metabolic pathway can be assayed

following previous methods (Zhu and Yang, 2003), including acetic acid forming

enzymes (PTA and AK) and butyric acid producing enzymes (PTB and BK); ii) SDS-PAGE and two-dimensional protein electrophoresis are powerful methods

for studying the protein expression difference between the wild type and different

mutants to confirm and further understand the metabolic engineered mutants

following the procedure described above; iii) Batch and fed-batch fermentations with free or immobilized-cells of C.

tyrobutyricum performed in a 5 L stirred-tank fermentor containing 2 L of

clostridial growth medium (CGM) supplemented with sugars could be used to

directly characterize the butyric acid and hydrogen production of the mutants. A

high performance liquid chromatograph (HPLC) can used to analyze the organic

compounds, including glucose, butyrate, and acetate in the liquid samples (Wu

and Yang, 2003). 22

iv) The identification of the functional genomic and proteomic differences between

the wild type and mutant strains is another important characterization method.

The cell samples can be taken at different growth phases during the fermentation

and are used for microarray analysis and proteomic analysis to find the genomic

and proteomic differences of the mutants; v) The metabolic changes resulted from the gene manipulation can be studied by the

metabolic flux analysis of the mutants based on the biomass, glucose consumption,

and acids production data from the fermentation study; vi) Product tolerance can be studied by cultivating the C. tyrobutyricum mutants in

serum tubes containing 10 mL of media with various concentrations of butyrate to

evaluate the inhibition effect of butyrate on cell growth, which was followed by

measuring the optical density at 600 nm (OD600) with a spectrophotometer (Zhu,

et al., 2002).

The characterization of the mutants obtained from metabolic engineering is helpful for

understanding the effect of the gene manipulations and the genome shuffling in C. tyrobutyricum, and for directing future industrial scale up.

2.2 Butyric Acid and Hydrogen Fermentation

2.2.1 Butyric Acid and Hydrogen

Butyric acid has many important applications in the chemical, food, and

pharmaceutical industries. The consumption of butyric acid to produce cellulose acetate

butyrate thermoplastics is a major use in the chemical industry. Glycerol tributyrate and other esters also play an important role in the plastic materials (Playne, 1985). It is also used as a raw material for the production of biodegradable polymer β–hydroxybutyrate

(Lefranc and Cie, 1923). Butyric acid is used to supply butter-like-notes in food flavors and its esters are widely used as additives to increase fruit fragrance in the food industry

(Sharpell, 1985; Zigová et al., 1999). It is one of the short-chain fatty acids generated by the bacterial fermentation of dietary fibers in the colon. It is used as a main source of energy for human body and also marked as a suppressor of colon cancer (Hara, 2002). Its biological effects have been widely studied and it is considered to have a therapeutic nature for the treatment of hemoglobinopathies, cancer, and gastrointestinal diseases

(Pouillart, 1998), it has been shown to have an anticancer effect and can be used as a neutraceutical or even as a drug to cure colo-rectal cancers (Williams et al., 2003). A family of acyloxyalkyl butyrate prodrugs is presently in clinical development (Rephaeli et al., 2000). Butyric acid derivatives have been developed to produce antithyroid and vasoconstrictor drugs and used in anaesthetics (Playne, 1985).

Hydrogen is a non-greenhouse gas with a high energy content per unit weight

(141.86 kJ/g or 61,000 Btu/lb) that can be used as a clean fuel and easily converted to electricity by fuel cells. Hydrogen is a clean and renewable energy and generates non- toxic by-product, water, and has various uses, such as efficient energy carrier and fuel cells using hydrogen to drive many electronic devices or electric automobile (Dunn,

2002). As an energy-efficient and low-polluting fuel, hydrogen appears to be one of the most promising energy carriers for the future if its production cost can be greatly reduced and the storage technologies can be developed.

2.2.2 Butyric Acid and Hydrogen Production

Butyric acid is currently produced mainly by oxidation of butyraldehyde, which has been obtained from propylene by oxosynthesis (Kroschwitz and Howe-Grant, 1978;

Pryde, 1978). Some anaerobic bacteria (clostridia), such as C. tyrobutyricum, C. beijerinckii, C. populeti, C. butyricum, C. barkeri, and C. thermobutyricum, produce butyric acid as their main product on a variety of substrates (Zigova and Sturdik, 2000).

Butyric acid production by fermentation from natural resources has become an increasingly attractive alternative to the petroleum-based chemical synthesis. In order to improve the production of butyric acid, a fibrous-bed bioreactor (FBB) was successfully used for butyrate fermentation with increased reactor productivity and final product concentration (Zhu and Yang, 2003). With high cell densities immobilized in the fibrous matrix, the FBB can improve the reactor productivity, final product concentration, and product yields. However, the butyrate yield was only ~0.423 g/g or ~0.9 mol/mol and acetate yield was ~0.27 mol/mol of glucose fermented in the fermentation (Zhu, et al.,

2002). Another novel extractive fermentation process for butyric acid fermentation by

FBB was also developed by our group, using 10% (v/v) Alamine 336 in oleyl alcohol as the extractant contained in a hollow-fiber membrane extractor for selective removal of butyric acid from fermentation broth (Yang, 1996; Wu and Yang, 2003).

Hydrogen can be generated by various ways, including thermochemical methods from fossil fuels (steam reforming of natural gas, thermal cracking of natural gas, coal gasification), electrochemical processes from water (electrolysis, photolysis), and biological production (photosynthesis and anaerobic fermentation) (Momirlan and

Veziroglu, 1999; Das and Veziroglu, 2001). Both thermochemical and electrochemical 25

hydrogen generation processes are energy intensive and not always environment friendly, but biological hydrogen production processes are usually performed at ambient temperatures and pressures, and thus are less energy intensive. Also the biological method can use various waste materials. As reviewed before, there are two kinds of microorganisms that produce biological hydrogen. One is anaerobic bacteria (obligate anaerobes or facultative anaerobes), such as C. acetobutylicum, C. butyricum, C. pasteurianum, Enterobacter aerogenes, and Enterobacter cloacae, Desulfovibrio vulgaris, Magashaera elsdenii, Citrobacter intermedius, and the other is photosynthetic bacteria, such as Rhodobacter sphaeroides, and Rhodobacter capsulatus. Some low value substrate includes sterile or non-sterile organic waste such as cane juice, corn pulp, saccharified cellulose, tofu manufacturing waste, damaged wheat grains, sugar factory wastewater, and so on (Mizuno, et al., 2000). The hydrogen production from the fermentation of C. tyrobutyricum has not been studied by now. Because anaerobic fermentation can produce organic chemicals along hydrogen and the organic acids produced by the anaerobic bacteria can be utilized by photosynthetic bacteria to produce more hydrogen, two-stage system using both anaerobic fermentation and photosynthesis has been developed (Kataoka et al., 1997). Some hybrid systems using photosynthetic and fermentative bacteria have also been developed (Kataoka, et al. 1997). Hybrid systems are comprised of both non-photosynthetic and photosynthetic bacteria; theyt can enhance hydrogen production. Varity of carbohydrates can be digested by hydrogen producing clostridium strains. These bacteria produce hydrogen by degrading carbohydrates without using light, resulting in organic acids, which are used by photosynthetic bacteria to produce more hydrogen. Anaerobic bacteria decompose

carbohydrates to obtain both energy and electron. Complete degradation of glucose to hydrogen and carbon dioxide is impossible by anaerobic digestion. Various Clostridium strains, such as C. butyricum, C. pasteurianum, C. beijerinckii AM21B, and so on, were studied for hydrogen production with yield of 1.3~2.36 moles hydrogen per mole of glucose. The maximum hydrogen yield reported was 2.52 mole/mole carbohydrate consumed using continuous culture (Kumar et al., 1995). Some processes were developed for hydrogen production from fermentation. For example, hydrogen had been continuously produced by a combined procedure in two reactor systems performing continuous enzymatic hydrolysis of Avicel in an aqueous two-phase system and hydrogen fermentation from Avicel hydrolysate by stain no. 2 producing 2.14 mol of hydrogen from 1 mol of glucose and 4.46 hydrogen from 1 mol of Avicel hydrolysate

(Taguchi, et al., 1996). The combination of anaerobic bacteria and photosynthetic bacteria also increased hydrogen production (Das and Veziroglu, 2001). Some important parameters including pH, temperature, concentrations of phosphate and glucose, intermittent purging of culture broth by argon gas, and types of sugars, have been investigated.

Biotechnological production of butyric acid and hydrogen is not commercially competitive with chemical production because of low productivity and low yield.

However, food and pharmaceutical manufacturers prefer food additives or pharmaceutical products that are produced biologically. Hydrogen, as an energy byproduct from butyric acid fermentation, can add value to the fermentation process. As a result, improving in the economics and efficiency of butyrate and hydrogen fermentation process is necessary. If the fermentation process can use a low-grade

environmentally benign biomass as the feedstock, commercially competitive production of butyric acid from fermentation would be obtained; this has a great market potential.

Hydrogen production by anaerobic fermentation offers an attractive method to produce energy when low-cost renewable biomass is available as the feedstock (Mizuno, et al.,

2000). Several anaerobic bacteria can produce butyric acid and hydrogen as the major fermentation products from a wide range of substrates. Among them, Clostridium tyrobutyricum has many advantages over other species, including simple medium for cell growth and relatively high product purity and yield (Michel-Savin et al., 1990ab).

Clostridium tyrobutyricum is a gram-positive, rod-shaped, spore-forming, obligate anaerobic bacterium capable of fermenting a wide variety of carbohydrates to butyric and acetic acids. Various carbon sources can be used as substrates, including glucose, xylose, fructose, lactose, sucrose, starch, potato wastes, and cellulose (Grobben et al., 1993; Reid et al., 1996; Vandak, 1995; Champan et al., 1992, Heyndrickx et al., 1991). The optimal cultivation conditions of these strains are 30-37°C, pH 6.5-7.0, with CO2 or N2, or the

mixture of N2 and CO2, in a ratio of 1:9 (Sneath, 1986; van Andel et al., 1985).

Historically, butyric acid fermentation in cheese (late blowing) caused by the outgrowth

of Clostridial spores present in raw milk, most commonly originating from silage, can

result in considerable product loss (Klijin et al., 1995). DNA probes based on specific

16S and 23S rRNA sequences have also been developed for the detection and

identification of C. tyrobutyricum (Klijin et al., 1995; Klijin et al., 1994; Van Der Meer et

al., 1993). On the other hand, butyric acid has many applications in chemical, food, and

pharmaceutical industries (Zigova, et al. 1999; Vandak et al. 1997; Williams et al., 2003).

As described above, there has been increasing interest in the production of butyric acid

from biomass using C. tyrobutyricum (Michel-Savin, et al., 1990; Zhu et al., 2002; Zhu and Yang, 2003). However, like other acidogenic bacteria, butyric acid bacteria are strongly inhibited by their acid products (Michel-Savin et al., 1990b). Some work has focused on developing separation process to decrease final product inhibition in the fermentation. Micro-filtration, ultra-filtration, and permeate electrodialysis have also been integrated with fermentation of lactic acid (Boyaval, et al., 1987). The extractive and pertractive fermentation processes using Hostarex (20% w/w) in oleylalcohol as organic phase to produce butyric acid by C. butyricum have been developed. The experimental results show that the integration of extraction and pertraction with fermentation can improve the butyric acid concentration from 7.3 g/L to 10.0 g/L

(extractive fermentation) and 20.0 g/L (pertractive fermentation) and that the butyric acid production yield was increased from 0.24 g/g to 0.30 g/g (pertractive fermentation).

Many fermentation studies have been directed towards increasing cell density, reactor productivity and final butyric acid concentration, but with only limited success. Also, butyric acid is not the sole fermentation product. Acetic acid is also produced as a byproduct, which not only reduces butyric acid yield, but also makes the final product recovery more difficult and challenging.

2.2.3 Cell Immobilization and Culture Adaptation

The problems associated with conventional butyric acid fermentations (and many other carboxylic acid fermentations) can be partially addressed by cell immobilization.

Recently, fibrous materials have been developed for cell immobilization because of their

high specific surface area, high void volume, low cost, high mechanical strength, and high permeability. The spiral wound terry cloth was packed loosely inside a glass column to construct the immobilized fibrous-bed bioreactor (FBB) (Huang and Yang, 1998; Yang,

1996; Yang et al., 1994, 1995). Its unique packing structure allows for free flow of gas, liquid, and solid in the reactor bed. A simple in situ immobilization of cells can be carried out by circulating cell suspension through the FBB. The fibrous-bed bioreactor has been successfully used for several organic acid fermentations, including lactic acid, propionic acid and acetic acid (Huang and Yang, 1998; Huang et al., 1998; Silva and Yang, 1995;

Yang, 1996; Yang et al., 1994, 1995) with greatly increased reactor productivity, final product concentration, and product yield. It gave stable, high-rate production of the fermentation product for a long period because of the high density of active cells maintained in the fibrous bed. It can also tolerate a low contaminant level. The most important finding was that the FBB had the ability to quickly adapt and enrich cultures with high tolerance to the inhibitory fermentation products, resulting in an increase in the final product concentration by 2 to 3-fold (Huang and Yang, 1998; Huang et al., 1998).

Cells adapted in the FBB had higher specific grown rate than the original culture used to seed the bioreactor.

Using cell immobilization in the fibrous-bed bioreactor (FBB), we have obtained acid-tolerant strains of C. tyrobutyricum. The C. tyrobutyricum mutant grew well under high butyrate concentrations (>30 g/L) and had better fermentative ability as compared to the wild-type strain used to seed the bioreactor. Kinetic analysis of butyrate inhibition on cell growth, acid-forming enzymes, and ATPase activity showed that the adapted cells from the FBB are physiologically different from the original wild type. It is shown that

cell immobilization in FBB provides an effective means for the in-process adaptation and selection of mutant with higher tolerance to inhibitory fermentation product.

Further improvement of the fermentation in terms of product yield, concentration, purity and production rate may be achieved through metabolic engineering, which requires a thorough knowledge of the organism, both wild types and acid-tolerant mutants, at the molecular biology (genomic) and biochemical (proteomic) levels.

2.3 Metabolic Pathway and Flux Analysis

2.3.1 Metabolic Pathway

C. tyrobutyricum has several end products, including butyrate, acetate, CO2, H2 and lactate. The metabolic pathway from glucose to pyruvate is different from the pathway from xylose to pyruvate. + Glucose + 2 NAD + 2ADP 2 Pyruvate + 2 NADH2 + 2 ATP + 3 Xylose + 5 NAD 5 ADP 5 Pyruvate + 5 NADH2 + 5 ATP The metabolic pathway and the reaction in the pathway can be found in previous study

(Zhu and Yang, 2004). Following transport into the cytoplasma, glucose is metabolized

to pyruvate via the Embden-Meyerhof-Parnas (EMP) pathway. Pentoses, such as xylose,

are metabolized by pentose-phosphate pathway. This is a combination of phosphorylation,

isomerization, and epimerization, and the resulting phosphorylated intermediates are

converted to fructose 6-phosphate and glyceraldehydes-3-phosphate (Papoutsakis and

Meyer, 1985). Some lactate may be formed by reduction of pyruvate. There are two

enzymes involved in lactate production. Under most conditions, pyruvate is

predominantly cleaved by pyruvate–ferredoxin (Fd) oxidoreductase to form acetyl-CoA,

CO2, and reduced ferredoxin (FdH2) (Mitchell, 2001). Electrons from reduced Fd can be

used in the reduction of protons to form H2, or can be transferred to NAD(P). Acetyl-

CoA is a branch point intermediate located along the central metabolic pathway of acidogenic clostridia at the node dividing the acetate-formation branch from the butyrate

formation branch. Two analogous pathways lead to the formation of acetic and butyric

acids, respectively. Each pathway consists of two reactions: first, the conversion of the respective acyl-CoA into acyl-phosphate, and second, the formation of the acid from acyl-phosphate with concomitant phosphorylation of ADP. The two sequential reactions are catalyzed respectively by an acyltransferase and a kinase, but distinct enzymes are responsible for the analogous reactions of the parallel pathways. ATP generation is involved in the formation pathways of both acids. During acetate production, 4 moles of

ATP are formed from 1 mole of glucose. During butyrate production, only 3 moles of

ATP are formed. The theoretical glucose fermentation with complete conversion of butyrate is shown as follows (Zhu et al., 2002): 1 glucose → 3 ATP + 1 butyrate +2

CO2+ 2 H2

It is apparent that the major characteristic of butyric acid fermentation is the

concomitant production of acetate. It should be noted that not only PTA-AK pathway for

acetate formation and the PTB-BK pathway for butyrate formation is connected by

acetyl-CoA as a key interdedium, but butyrate and acetate can also be converted via the

CoA transferase from acetyl-CoA and butyryl-CoA in C. tyrobutyricum. The situation

raises the interesting physiological question of the repartition of carbon flow between the

acetate and butyrate pathways involved.

2.3.2 Metabolic Flux Analysis and Application in Clostridia

Cellular metabolism is the “chemical engine” that drives the life process. The regulation and control of metabolic pathways have been prime concerns of scientists and chemical engineers who are attempting to optimize bioprocesses by manipulation of cellular metabolism using modern molecular biology techniques. This is metabolic engineering (Bailey, 1991; Lee and Papoutsakis, 1999; Koffas et al., 1999;

Stephanopoulos and Sinskey, 1993). Therefore, a quantitative description of metabolism is not only important for our fundamental understanding of cell biology, but also helpful for designing and implementing genetic modifications. In recent years, an approach was reported that used metabolic pathway balances to develop a model of cellular metabolism

(Papoutsakis, 1983, 1984; Papoutsakis and Meyer, 1985a, b; Venkatesh, 1997). It involves the calculation of in vivo flux from substrate and product data by using a system of linear equations developed from reaction stoichiometry. Based on the fundamental law of mass conservation, the flux balance method is able to assess the roles of individual steps in a network of metabolic pathways and provide insightful information about the systematic constraints placed on metabolic function.

In general, the metabolism of any organism can be represented as a network of chemical reactions that result in the conversion of substrate to biomass and products. The system of mass balance equations developed from the stoichiometric relationships of these reactions can be represented in matrix form as:

Ar = x where A is a matrix containing stoichiometric information, r is a vector of flux on the metabolic matrix, and x is a species accumulation vector (Desai et al., 1999a, b; Varma 33

and Palsson, 1994a; Venkatesh, 1997). Species can be metabolic intermediates and exchangeable species. The assumption of pseudo-steady state approximation on metabolic intermediates is typically incorporated into metabolic intermediates based on the fact that the accumulation of metabolic intermediates is very small compared to the cellular growth rates or the rates of their production and consumption. The accumulation term of exchangeable species, such as nutrients and products should be measured.

Metabolic flux analysis then becomes a problem of determining a vector of in vivo fluxes, r, which best fits the equation above. Singularities in the systems of linear equations are usually present and prevent the calculation of some critical pathway flux in the metabolism. In order to resolve singularities, some researches have incorporated optimality principles to seek an optimal value of a stated objective, such as maximal growth (Varma and Palsson, 1994b), maximal ATP generation (Majewski and Domach,

1990), and minimal norm of flux vector (Bonarius et al., 1996), which indicates that cells can channel the metabolites as efficiently as possible through the metabolic pathways.

Others have utilized in vitro enzyme activity information to remove insignificant pathways or to develop correlation relating some of the fluxes involved (Desai et al.,

1999b). Madron’s statistical method of maximum likelihood estimation can be applied to solve the overdetermined system and analyze the bioreactor data (Tsai and Lee, 1988).

Papoutsakis was the first to recognize the use of linear equations for the study of cell metabolism (Papoutsakis, 1983 and 1984). Up to now, most studies on butyric acid bacteria (Clostridia) have been attributed to his group. A stoichiometric equation that interrelates the biomass, the substrate uptake, and the by-product secretion rate was derived in fermentation of butyric acid bacteria. This model was based on metabolic

stoichiometry, the carbon weight fraction, the degree of reductance of the biomass, and an assumed ATP yield. The validity of the equation was demonstrated using experimental data from the literature. Papoutsakis and Meyer also derived stoichiometric equations for butanediol and mixed-acid fermentations (Papoutsakis and Meyer, 1985a) and for propionic acid bacteria and the production of assorted oxychemicals (butanol, acetone, isopropanol, butanediol, butyrate, acetate, propionate, succinate, lactate and acrylate) from various sugars, including pentoses, hexoses, and cellobiose (Papoutsakis and Meyer,

1985b). These flux balance equations were derived based on the same assumptions and simplified network as those in Papoutsakis’s earlier work. It was shown that the equations are useful not only for checking the consistency of experimental data, for calculating maximal yields and selectivities for the fermentation products, but also for calculating the extent of various intracellular reactions, such as the utilization of the Embden-Meyerhof-

Parnas pathway versus the Hexose Monophophate pathway of glucose utilization. In situ intracellular NAD(P)H fluorescence measurement during the fermentation was also used to provide more information about the electron flow and metabolic control for solvent production in the metabolic pathways of C. acetobutylicum (Reardon et al., 1987).

Recently, this approach has been extended to the analysis of C acetobutylicum fermentation in a more accurate and simplified way (Desai et al., 1999b). An equation relating the acetate and butyrate uptake fluxes was developed to reformulate the stoichiometric model as a non-linear constrained minimization problem.

Metabolic flux analysis has also been applied to the fermentations of other clostridia, such as 1,3–propanediol production by C. butyricum (Abbad-Andaloussi et al.,

1996; Zeng, 1996) and succinate production by C. thermosuccinogenes (Sridhar and

Eiteman, 2001). Metabolic flux analysis not only gives a quantitative interpretation of metabolic physiology, but also provides a guide to metabolic engineering and enables the design and optimization of bioprocesses.

2.4 Functional Genomics and Proteomics

Recent years have seen an explosion in the number of complete or almost

complete genomic sequences of organisms, including several bacteria that are important

to agriculture and food processing (de Vos et al., 2004; de Vos and Hugenholtz, 2004).

The vast genome sequence information can be used to manipulate the metabolism or

physiology of the producer organism and eventually to integrate this information into the

development of more efficient production strains. Functional genomics of food microbes

based on rapidly emerging genome sequence information generated valuable knowledge

that could be used for metabolic engineering, improving cell factories, and developing

functional foods and novel preservation methods (Kuipers, 1999; de Vos et al., 2004). In

addition to genomic sequence information, transcriptomic, proteomic, and metabolomic

data was generated at an ever-increasing rate from high-throughput technologies, such as

DNA microarrays, two-dimensional gel electrophoresis combined with tandem mass spectrometry, and isotopic label distributions probing the metabolic phenotype, that can impact both biomedical research and industrial bioprocesses. Functional genomics is

currently the most effective approach for increasing the knowledge at the molecular level

of metabolic and adaptive processes in whole cells, providing detailed knowledge of the

properties of food microbes (and pathogens) in their industrial, food and consumer environments.

2.4.1 Functional Genomics

Despite the industrial importance of these two bacteria, there is very little genomic information available for them. Some work has been done to study some genes involved in the central metabolic pathway of C. tyrobutyricum (Zhu, 2002). It is still not enough to design better mutant because the genomic sequence and the global genomic function of these two bacteria are not now available. DNA microarray is increasingly used for the functional analysis of the newly and already characterized genes. Today, it is possible to spot all ORFs of complete genomes from bacteria to human on a gene chip.

Consequently, most functional genomic studies are performed with eukaryotic species, and DNA microarrays are used primarily in human medical research. Few studies have been performed with bacteria, e.g., Bacillus subtilis, Clostridium acetobutyricum,

Corynebacterium glutamicum, Escherichia coli, and a very few additional species

(Alsaker and Papoutsakis, 2004; Tomas, 2003; Ikeda and Nakagawa, 2003; Gaertner et al., 2004). For most of the industrially important bacterial strains, very limited or no genomic sequencing information is available. In this case, a shotgun DNA microarray strategy has been successfully used for gene expression analysis (Hayward et al., 2000), bypassing the need for previous knowledge of the genome sequence. In a shotgun approach, the genome is represented by random DNA fragments of unknown sequence that are arrayed on a glass slide. DNA fragments that contain regulated genes were

identified by differential hybridization with two labeled cDNAs. The identity of these genes was elucidated by subsequent sequencing. Shotgun DNA microarrays were constructed and used to study the metabolism of archaeon Haloferax volvanii (Zaigler et al., 2003), gene function in environmental isolates of Leptospirillum ferrooxidans (Parro and Moreno-Paz, 2003), and phylogenetic lineages of Listeria monocytogenes (Zhang et al., 2003). In this project, shotgun DNA microarrays can be developed and used for functional genomic studies of Clostridium tyrobutyricum.

2.4.2 Proteomics

Proteomics is the use of quantitative protein-level measurements of gene expression to characterize biological processes, including disease processes, cellular response to drugs, and deciphering the mechanisms of gene expression control.

Simultaneously monitoring the proteome and transcriptome profiles of microorganism during fermentation under various conditions using quantitative proteomics, transcriptomics, and bioinformatics in time-resolved experiments is possible. Proteome signatures can be used to predict the physiology changes that occur in the gene expression profile during the fermentation growth in complex medium and to detect the significance of quantitative proteome changes that were identified relative to a threshold of scatter in replicate samples. The same samples can be submitted to transcriptional profiling using DNA microarrays and comparing the RNA profiles to the proteome data.

Functional groups of proteins can be clustered according to significant expression changes. Previous work has been done to Bacillus subtilis (Gaertner, 2004).

The post-genomics era has lead to proteomics, where millions of proteins are known to exist and await systemic study. Western blotting and other immunological methods have been successfully used to study the protein expression of various microorganisims, cells, and tissues. However, these techniques are constrained by the limited number of proteins that can be studied in each experiment and the availability of specific antibodies. Proteome analysis allows the profiling of a large number of proteins in a given organism on a two-dimensional polyacrylamide gel. Proteomic techniques date to 1975, when two-dimensional PAGE was simultaneously described by O’Farrell and

Klose (O’Farrell, 1975; Klose and Kobalz, 1995) and applied to the study of a large number of proteins simultaneously. In two-dimensional PAGE, proteins are separated by differential isoelectric point (pI) for the first dimension and by differential weight average molecular weight (Mw) for the second dimension. Using this technique, 1100 protein

components have been resolved from Escherichia coli (O’Farrell, 1975). Recently, up to

10, 000 protein forms have been visualized by high resolution two-dimensional PAGE

(Klose and Kobalz, 1995). The analysis involves the systematic sand simultaneous

eparation, identification, and quantification of cellular proteins. By profiling and

comparing of the proteomes obtained, the changes in the levels of protein expression

under different environmental conditions can be used to identify proteins (enzymes) to be

manipulated to achieve the desired goals (Han and Lee, 2003). There have been

significant advances in proteome analysis with the aid of two-dimensional gel

electrophoresis, capillary electrophoresis, and mass spectrometry. High throughput

analysis by mass spectrometry of proteins separated with two-dimensional PAGE has

permitted the analysis of proteins on a “genomic” scale (Jungblut and Wittmann-Liebold,

1995).

During quantitative proteomics studies, the amounts of proteins in a particular biological system were measured by comparison with a reference system. Relative levels of proteins were measured, rather than absolute amounts. Many quantitative proteomics studies relied on the comparison of the amounts of proteins in the system under study with the amounts in a reference system. This is true for the mass spectrometric approaches known as multidimensional protein identification technology (MudPIT) and comparative shotgun proteomics using isotope-coded affinity tags (ICAT) (Flory et al.,

2002). Matrix-assisted laser-desorption ionization mass spectrometry (MALDI-MS) and electrospray ionization mass spectrometry (ESI-MS) of tryptic peptide fragments were used to identify the modified C-termini of the longer protein species. Recently, MALDI-

TOF was used to rapidly identify different strains of the flu virus, leading to the production of more effective vaccines. MALDI-TOF research provides a better understanding of the building blocks of life and will lead to the development of new treatments for cancer, diabetes, emphysema, and immune system disorders. MALDI-TOF mass spectrometry is one of the fastest growing segments of the analytical instrumentation industry.

2.4.3 Application of Functional Genomics and Proteomics

Both functional genomics and proteomics offer abundant information on gene expression patterns and phenotypes of biological system at different states of cells and

cellular metabolism. By combining gene sequence, gene expression, protein interactions and metabolic control mechanisms, the genomic function analysis of microorganism can have many applications.

How can traditional fermentation benefit from functional genomics and proteomics? In the last decade, there has been rapid development in the field of metabolic engineering. Many examples of metabolic engineering leading to successful yield improvement have been demonstrated in the production of useful metabolites (Shimizu,

2002). Quantitative proteomic and transcriptomic profiling during fermentation of pleiotropic Bacillus subtilis mutants was studied (Gaertner et al., 2004). Proteome signatures were used to predict the physiology changes that occur in the gene expression profile during fermentation growth in complex medium and detected the significance of quantitative proteome changes. The same samples were also submitted to transcriptional profiling using DNA microarrays, and the RNA profiles were compared to the proteome data. After studying the physiology changes, gene expression, and proteomic changes during the fermentation process using DNA microarrays and proteome signatures, it is possible to decide the strategies for improving the production of final products and decreasing the byproducts by comparing the fermentation results under different conditions and gene manipulations.

2.5 References

Abbad-Andaloussi, S. Durr, C., Raval, G., and Petitdemange, H. Carbon and electron flow in Clostridium butyricum grown in chemostat culture on glycerol and on glucose. Microbiology 1996, 142, 1149-1158.

Alsaker, K., and Papoutsakis, E. T. Microarray-based analysis of the stress response in Clostridium acetobutylicum cultures. Abstracts of papers, 22. 7th ACS National Meeting, Anaheim, C. A., United States, March 28-April 1, 2004, BIOT-335.

Arnold, F., Bedouet, L., Batina, P., and Robreau, G. Cloning and sequencing of the central region of the flagellin gene from the gram-positive bacterium Clostridium tyrobutyricum ATCC 25755. Microbiol. Immunol. 1999, 43, 1-8.

Bailey, J. E. Toward a science of metabolic engineering. Science 1991, 252, 1668-1675.

Bonarius, H. P., Hatzimanikatis, V., Meesters, K. P. H., de Gooijer, C. D., Schmid, G., and Tramper, J. Metabolic flux analysis of hybridoma cells in different culture media using mass balances. Biotechnol. Bioeng. 1996, 50, 299-318.

Boyaval, P., Corre, C., and Terre, S. Continuous lactic acid fermentation with concentrated product recovery by ultrafiltration and electrodialysis. Biotechnol. Lett. 1987, 9, 207-212.

Boynton, Z. L., Bennett, G. N., and Rudolph, F. B. Cloning, sequencing, and expression of genes encoding phosphotransacetylase and acetate kinase from Clostridium acetobutylicum ATCC 824. Appl. Environ. Microbiol. 1996, 62, 2758-2766.

Campbell, A. M. Episomes. Adv. Genet. 1962, 11, 101-146.

Champan, S. J., Veal, D. A., and Lynch, J. M. Effect of oxygen concentration on dinitrogen fixation and volatile fatty acids production by C. butyricum growing in association with fungi on cellulose and wheat stray. J. Appl. Bacteriol. 1992, 72, 9-15.

Chin, H. L., Chen, Z. S., and Chou, P. Fedbatch operation using Clostridium acetobutylicum suspension culture as biocatalyst for enhancing hydrogen production. Biotechnol. Prog. 2003, 19, 383-388.

Das, D, and Veziroglu T. N. Hydrogen production by biological processes: a survey of literature. Int. J. Hydrogen Energy. 2001, 26, 13-28. 42

de Vos, W. M., and Bron, P. A., and Kleerebezem, M. Post-genomics of lactic acid bacteria and other food-grade bacteria to discover gut functionality. Current Opinion in Biotechnol. 2004, 15, 86-93. de Vos, W. M., and Hugenholtz, J. Engineering metabolic highways in Lactococci and other lactic acid bacteria. Trends Biotechnol. 2004, 22, 72-79.

Desai, N. P., Harris, L. M., Welker, N. E., and Papoutsakis, E. T. Metabolic flux analysis elucidates the important of the acid-formation pathways in regulating solvent production by Clostridium acetobutylicum. Metab. Eng. 1999a, 1, 206-213;

Desai, R. P., Nielsen, L. K. and Papoutsakis, E. T. Stoichiometric modeling of Clostridium acetobutylicum fermentation with non-linear constraints. J. Biotechnol. 1999b, 71, 191-205.

Dunn, S. Hydrogen futures: toward a sustainable energy system. Int. J. Hydrogen Energy. 2002, 27, 235-264.

Flory, M. R., Griffin, T. J., Martin, D., and Aebersold, R. Advances in quantitative proteomics using stable isotope tags. Trends Botechnol. 2002, 20, 23-29.

Gaertner, A. L., Antelmann, H., Sapolsky, R., Ferrari, E., Chotani, G., Miller, B., and Hecker, M. Qantitative proteomic and transcriptomic profiling during fermentation of pleiotropic Bacillus subtilis mutants. Abstracts of Papers, 227th ACS National Meeting, Anaheim, CA, March 28-April 1, 2004.

Green, E. M. and Bennett, G. N. Inactivation of an aldehyde/alcohol dehydrogenase genes from Clostridium acetobutylicum ATCC 824. Appl. Biochem. Biotechnol. 1996, 57-58, 213-221.

Green, E. M. and Bennett, G. N. Genetic manipulation of acid and solvent formation in Clostridium acetobutylicum ATCC 824. Biotechnol. Bioeng. 1998, 58, 215-221.

Green, E. M., Boynton, Z. L., Harris, L. M., Rudolph, F. B., Papoutsakis, E. T., and Bennett, G. N. Genetic manipulation of acid formation pathways by gene inactivation in Clostridium acetobutylicum ATCC 824. Microbiology 1996, 142, 2079-2086.

Grobben, N. C., Egging, G., Cuperus, F. P., and Huizing, H. J. Production of ABE from potato wastes: fermentation with integrated membrane extraction. Appl. Microbiol. Biotechnol. 1993, 39, 494-498.

Han, M.J., and Lee, S. Y. Proteome profiling and its use in metabolic and cellular engineering. Proteomics. 2003, 3, 2317-2324.

Hara, H. Physiological effects of short-chain fatty acid produced from prebiotics in the colon. Bioscience Microflora 2002, 21, 35-42.

Harwood, C. S., and Caale-Parola E. Properties of acetate kinase isozymes and a branched-chain fatty kinase from a spirochete. J. Bacteriol. 1982, 152, 246-254.

Haynes, P. A., Gygi, S. P., Figeys, D., and Aebersold, R. Electrophoresis. 1998, 19, 1862.

Hayward, R. E., DeRisi, J. L., Alfadhli, S., Kaslow, D. C., Brown, P. O., and Rathod, P. K. Shotgun DNA microarrays and stage-specific gene expression in Plasmodium falciparum malaria. Molecular Microbiol. 2000, 35, 6-14.

Herman, L. M., De Block, J. H. G. E. and Waes, G. M. A. V. J. A direct PCR detection method for Clostridium tyrobutyricum spores in up to 100 milliliters of raw milk. Appl. Environ. Microbiol. 1995, 61, 4141-4146.

Heyndrickx, M., DeVos, P., Thibau, B., Stevens, P., and Delay, J. Effect of various external factors on the fermentative production of hydrogen gas from glucose by Clostridium butyricum strains in batch culture. Syst. Appl. Microbiol. 1987, 9, 163- 168.

Heyndrickx, M., and DeVos, P., and DeLey, J. Fermentation of D-xylose by Clostridium butyricum LMG 1213t1 in chemostats. Enzyme Microb. Technol. 1991, 13, 893-897.

Huang, K. X., Huang, S., Rudolph, F. B., and Bennett, G. N. Identification and characterization of a second butyrate kinase from Clostridium acetobutylicum ATCC 824. J. Mol. Microbiol. Biotechnol. 2000, 2(1), 33-38.

Huang, Y. L., Mann, K., Novak, J. M., and Yang, S. T., Acetic acid production from fructose by Clostridium formicoaceticum immobilized in a fibrous-bed bioreactor. Biotechnol. Prog. 1998, 14(5), 800-806.

Huang, Y. L., Wu, Z, Zhang, L., Cheuang, C. M., and Yang, S.-T. Production of carboxylic acids from hydrolyzed corn meal by immobilized cell fermentation in a fibrous-bed bioreactor. Bioresource Technol. 2002, 82(1), 51-59.

Huang, Y., and Yang, S.-T., Acetate production from whey lactose using co-immobilized cells of homolactic and homoacetic bacteria in a fibrous-bed bioreactor. Biotechnol. Bioeng. 1998, 60, 499-507. 44

Ikeda, M., and Nakagawa, S. The Corynebacterium glutamicum genome: features and impacts on biotechnological processes. Appl. Microbiol. Biotechnol. 2003, 62, 99-109.

Jungblut, P., and Wittmann-Liebold, B. J. Biotechnol. 1995, 41, 111-120.

Kakuda, H., Hosono, K., Shiroshi, K., and Ichihara, S. Identification and characterization of the ackA (acetate kinase A) - pta (phosphotransacetylase) operon and complementation analysis of acetate utilization by an ackA-pta deletion mutant of Escherichia coli. J. Biochem. 1994, 116, 916-922.

Kataoka, N., Miya, A., and Kiriyama, K. Studies on hydrogen production by continuous culture system of hydrogen-producing anaerobic bacteria. Water Sci. Technol. 1997. 36, 41-47.

Klijn, N., Bovie, C., Dommes, J., Hoolwerf, J. D., van der Waals, C. B., Weerkamp, A. H., and Nieuwenhof, F. F. J. Identification of Clostridium tyrobutyricum and related species using sugar fermentation, organic acid formation and DNA probes based on specific 16S rRNA sequences. System. Appl. Microbiol. 1994, 17, 249-256.

Klijn, N., Nieuwenhof, F. F. J., Hoolwerf, J. D., van der Waals, C. B., and Weerkamp, A. H. Identification of Clostridium tyrobutyricum as the causative agent of late blowing in cheese by species-specific PCR amplification. Appl. Environ. Microbiol. 1995, 61, 2919-2924.

Klose, J., and Kobalz, U. Electrophoresis. 1995. 16, 1034-1059.

Koffas, M., Roberge, C., Lee, K., and Stephanopoulos, G. Metabolic engineering. Annu. Rev. Biomed. Eng. 1999, 1, 535-557.

Kuipers, O. P. Genomics for food biotechnology: prospects of the use of high-throughput technologies for the improvement of food microorganisms. Current Opinion in Biotechnol. 1999, 10, 511-516

Kumar, A., Jain, S. R., Sharma, C. B., Joshi, A. P., and Kalia, V. C. Increased H2 production by immobilized microorganisms. World J. Microbiol. Biotechnol. 1995, 11, 156-159.

Latimer, M. T. and Fery, J. G. Cloning, sequence analysis, and hyperexpression of the genes encoding phosphotransacetylase and acetate kinase from Methanosarcina thermophila. J. Bacterial. 1993, 175, 6822-6829.

Lee, S. Y. and Papoutsakis, E. T. The challenges and promise of metabolic engineering. In: Metabolic engineering, eds., S. Y. Lee and E. T. Papoutsakis, Marcel Dekker Inc., NY, 1999, pp 1-12.

Lefranc, L. and Cie, E. A process for the manufacture of butyric acid and other fatty acids with recovery of the gases of fermentation. Br. Patent 1923, No. 186572.

Maharbiz, M. M., Holtz, W. J., Howe, R. T., and Keasling, J. D. Microbioreactor arrays with parametric control for high-throughput experimentation. Biotechnol. Bioeng. 2004, 85, 376-381.

Majewski, R. A. and Domach, M. M. Simple constrained optimization view of acetate overflow in E. coli. Biotechnol. Bioeng. 1990, 35, 732-738.

Matsuyama, A. H., Yamamoto, H., and Nakano, E. Cloning, expression, and nucleotide sequences of the Escherichia coli K-12 ackA gene. J. Bacteriol. 1989, 171, 577-580.

Mermelstein, L. D., Welker, N. E., Bennett, G. N., and Papoutsakis, E. T. Expression of cloned homologous fermentative genes in Clostridium acetobutylicum ATCC 824. Bio/Technology 1992, 10, 190-195.

Mclaren, J. S., and Jeffries, T. W. Genetics and genomics in bioenergy and bioproducts. Appl. Biotechnol. Biochem. 2003. 107, 631-632.

Michel-Savin D., Marchal R., and Vandecasteele J. P. Control of the selectivity of butyric acid production and improvement of fermentation performance with Clostridium tyrobutyricum. Appl. Microb. Biotechnol. 1990a, 32, 387-392.

Michel-Savin D., Marchal R., and Vandecasteele J. P. Butyrate production in continuous culture of Clostridium tyrobutyricum: effect of end-product inhibition. Appl. Microb. Biotechnol. 1990b, 33, 127-131.

Mitchell, W. J. Biology and physiology. In: Clostridia, Biotechnology and medical applications, eds., H. Bahl and P. Dürre, Wiley-VCH, Weinheim, 2001, pp 49-104.

Mizuno, O., Dinsdale, R., Hawkes, F. R., Hawkes, D. L., and Noike, T. Enhancement of hydrogen production from glucose by nitrogen gas sparging. Bioresour. Technol. 2000, 73, 59-65.

Momirlan, M., and Veziroglu, T. Recent directions of world hydrogen production. Energy Rev. 1999, 3, 219-31.

Nölling, J., Breton, G., Omelchenko, M. V., Makarova, K. S., Zeng, Q., Gibson, R., et al. Genome sequence and comparative analysis of the solvent-producing bacterium Clostridium acetobutylicum. J. Bacteriol. 2001, 183, 4823-4838.

O’Farrell, P. H. J. Biol. Chem. 1975, 250, 4007-4021.

Ohnishi, J., Hayashi, M., Mitsuhashi, S., and Ikeda, M. Efficient 40oC fermentation of L- lysine by a new Corynebacterium glutamicum mutant developed by genome breeding. Appl. Microbiol. Biotechnol. 2003, 62, 69-75.

Papoutsakis, E. T. A useful equation for fermentation of butyric acid bacteria. Biotechnol. Lett. 1983, 5(4), 253-258.

Papoutsakis, E. T. Equations and calculations for fermentations of butyric acid bacteria. Biotechnol. Bioeng. 1984, 26, 174-187.

Papoutsakis, E. T. and Meyer, C. L. Equations and calculations of product yields and preferred pathways for butanediol and mixed-acid fermentations. Biotechnol. Bioeng. 1985a, 27, 50-66.

Papoutsakis, E. T. and Meyer, C. L. Fermentation equations for propionic acid bacteria and production of assorted oxychemicals from various sugars. Biotechnol. Bioeng. 1985b, 27, 67-80.

Parro, V., and Moreno-Paz, M.. Gene function analysis in environmental isolates: The nif regulon of the strict iron oxidizing bacterium Leptospirillum ferrooxidans. Proc. Natl. Acad. Sci. 2003, 100, 7883-7888.

Patnaik, R., Louie, S., Gavrilovic, V., Perry, K., Stemmer, W. P. C., Ryan, C. M. and del Cardayre, S. Genome shuffling of Lactobacillus for improved acid tolerance. Nat. Biotechnol. 2002, 20, 707-712.

Playne, M. J. Propionic acid and butyric acids. In: Comprehensive biotechnology, vol. 3, eds., M. Moo-Young, Pergamon Press, Oxford, 1985, pp 731-759.

Pouillart, P. R. Role of butyric acid and its derivatives in the treatment of colorectal cancer and hemoglobinopathies. Life Sci. 1998, 63(20), 1739-1760.

Reardon, K., Scheper, T.-H., and Bailey, J. Metabolic pathway rates and culture fluorescence in batch fermentations of Clostridium acetobutylicum. Biotechnol. Prog. 1987, 3, 153-167.

Reid, C. A., Hillman, K., Henderson, C., and Glass, H. Fermentation of native and processed starches by the porcine caecal anaerobe Clostridium butyricum (NCIMB 7423), J. Appl. Bacteriol. 1996, 80, 191-198,

Rephaeli, A., Zhuk, R., and Nudelman, A. Prodrugs of butyric acid from bench to bedside: synthetic design, mechanisms of action, and clinical applications. Drug Development Research 2000, 50, 379-391.

Riehle, M. M., Bennett, A. F., and Long, A. D. Genetic architecture of thermal adaptation in Escherichia coli. Proc. Natl. Acad. Sci. 2001, 98, 899-910.

Schulz, C. M., and Ruzicka, J. Real-time determination of glucose consumption by live cells using a lab-on-valve system with an integrated microbioreactor. Analyst. 2002, 127, 1293-1298.

Sharpell, F. H. J. Microbial flavors and fragrances. In: Comprehensive biotechnology, eds., M. Moo-Young. Pergamon Press, New York, 1985, pp 965-981.

Shimizu, H. Metabolic engineering integrating methodologies of molecular breeding and bioprocess systems engineering. J. Biosci. Bioeng. 2002, 94, 563-573.

Silva, E. M. and Yang, S. T. Kinetics and stability of a fibrous-bed bioreactor for continuous production of lactic from unsupplemented acid whey. J. Biotechnol. 1995, 41(1), 59-70.

Sneath, P. H. A. Bergey’s manual of systematic bacteriology 2, Waverly Press, Baltimore, 1986, pp 1141-1200.

Sridhar, J. and Eiteman, M. A. Metabolic flux analysis of Clostridium thermosuccinogenes, effects of pH and culture redox potential. Appl. Biochem. Biotechnol. 2001, 94, 51-69.

Stephanopoulos, G. Metabolic engineering by genome shuffling. Nat. Biotechnol. 2002, 20, 66-667.

Stephanopoulos, G. and Sinskey, A. J. Metabolic engineering – methodologies and future prospects.Trends Biotech. 1993, 11, 392-396.

Taguchi, F., Yamada, K., Hasegawa, K., Taki-saito, T., and Hara, K. Continuous hydrogen production by Clostridium sp. stain no. 2 from cellulose hydrolysate in an aqueous two-phase system. J. Ferment. Bioeng. 1996, 82, 80-83.

Tomas, C. A., K. V. Alsaker, H. P. J. Bonarius, W. T. Hendriksen, H. Yang, J. A. Beamish, C. J. Paredes, and E. T. Papoutsakis. DNA array-based transcriptional analysis of asporogenous, nonsolventogenic Clostridium acetobutylicum strain SKO1 and M5. J. Bacteriol. 2003, 185, 4539-4547. van Andel, J. G., Zouttberg, G. R. Crabbendam P. M., and Breure, A. M. Glucose fermentation by Clostridium butyricum grown under a self generated gas atmosphere in chemostat culture. Appl. Microbiol. Biotechnol. 1985, 23, 21-26. van der Meer, J. R., Ludwig, W., and De Vos W. M. Characterization of a ribosomal RNA gene cluster from Clostridium tyrobutyricum: phylogenetic positioning based on the 16S and 23S nucleotide sequences. System. Appl. Microbiol. 1993, 16, 201-207.

Vandak, D., Tomaska, M., Zigová, J., and Šturdík, E. Effect of growth supplements and whey pretreatment on butyric acid production by Clostridium butyricum. World J. Microbiol. Biotechnol. 1995a, 11(3),363.

Vandak, D., Telgarsky, M., and Sturdik, E. Influence of growth factor components on butyrate production from sucrose by Clostridium butyricum. Folia Microbiol. 1995b, 40(6), 669-672.

Vandak, D., Zigová, J., Šturdík, E., and Schlosser, S. Evaluation of solvent and pH for extractive fermentation of butyric acid. Proc. Biochem. 1997, 32, 245-251.

Varma, A. and Palsson, B. O. Metabolic flux balancing: basic concepts, scientific and practical use. Bio/Technology 1994a, 12, 994-998.

Varma, A. and Palsson, B. O. Stoichiometric flux balance models quantitatively predict growth and metabolic byproduct secretion in wile-type Escherichia coli W3110. Appl. Environ. Microbiol. 1994b, 60, 3724-3731.

Venkatesh, K. V. Metabolic flux analysis of lactic acid fermentation: effects of pH and lactate ion concentration. Proc. Biochem. 1997, 32(8), 651-655.

Walter, K. A., Mermelstein, L. D., and Papoutsakis, E. T. Studies of recombinant Clostridium acetobutylicum with increased dosage of butyrate formation genes. Ann. N. Y. Acad. Sci. 1994, 721, 69-72.

Williams, E. A., Coxhead, J. M., and Mathers, J. C. Anti-cancer effects of butyrate: use of micro-array technology to investigate mechanisms. Proceeding of the Nutrition Society. 2003, 62, 107-115.

Wiseman, A. Novel food-proteomics by control of bioprocess-enzymology outcomes? Trends Food Science & Technol. 2004, 15, 44-45.

Wong, T. K. K., Philp, R., Nissom, P. M., Zheng, L., Lau, L. Y. A., Wong, C. L., and Yap, M. G. S. (2004), The study of cell adaptation to different growth conditions using quantitative proteomics and DNA microarray. Abstracts of Papers, 227th ACS National Meeting, Anaheim, CA, March 28-April 1, 2004.

Wu, Z. and Yang, S. T. Extractive fermentation for butyric acid production from glucose by Clostridium tyrobutyricum. Biotechnol. Bioen. 2003, 82, 93-02.

Yang, S.-T. Extractive fermentation using convoluted fibrous bed bioreactor. U.S. Patent 1996, No. 5563069.

Yang, S.-T., Huang, Y., and Hong, G. A novel recycle batch immobilized cell bioreactor for propionate production from whey lactose. Biotechnol. Bioeng. 1995, 45, 379-386.

Yang, S. T., Zhu, H., Li, Y., and Hong, G. Continuous propionate production from whey permeate using a novel fibrous bed bioreactor. Biotechnol. Bioeng. 1994, 43(11), 1124-1130.

Yu, B. J., Sung, B. H., Koob, M. D., Lee, C. H., Lee, H., Lee, W. S., Kim, M. S, Kim, S. C. Minimization of the Escherichia coli genome using a Tn-5 targeted Cre/loxP excision system. Nat. Biotechnol. 2002, 20, 1018-1023.

Zaigler, A., Schuster, S. C., and Soppa, J. Construction and usage of a onefold-coverage shotgun DNA microarray to characterize the metabolism of the archaeon Haloferax volcanii. Molecul. Microbiol. 2003, 48, 1089-1105.

Zeng, A.-P. Pathway and kinetic analysis of 1,3-propanediol production from glycerol fermentation by Clostridium butyricum. Bioprocess Engineering 1996, 14, 169-175.

Zhang, C., Zhang, M., Ju, J., Nietfeldt, J., Wise, J., Terry, P. M., Olson, M., Kachman, S. D., Wiedmann, M., Samadpour, M., and Benson, A. K. Genome diversification in phylogenetic lineages I and II of Listeria monocytogenes: Identification of segments unique to lineage II populations. J. Bacteriol. 2003, 185, 5573-5584.

Zhang, Y-X, K. Perry, V.A. Vinci, K. Powell, W. P. C. Stemmer, and S. B. Cardayre. Genome shuffling leads to rapid phenotypic improvement in bacteria. Nature. 2002, 415, 644-646.

Zhu, Y., Wu, Z., and Yang, S.-T. Butyric acid production from acid hydrolysate of corn fiber by Clostridium tyrobutyricum in a fibrous bed bioreactor. Proc. Biochem. 2002, 38, 657-666.

Zhu, Y. and Yang, S. T. Adaptation of Clostridium tyrobutyricum for enhanced tolerance to butyric acid in a fibrous-bed bioreactor. Biotechnol. Prog. 2003, 19, 365-372.

Zhu, Y. and Yang, S. T. Effect of pH on metabolic pathway shift in fermentation of xylose by Clotridium tyrobutyricum. J. Biotechnol. 2004, 110, 143-157.

Zigová, J. and Šturdík, E. Advances in biotechnological production of butyric acid. J. Ind. Microbiol. Biotechnol. 2000, 24, 153-160

Zigová, J., Šturdík, E., Vandak D., and Schlosser S. Butyric acid production by Clostridium butyricum with integrated extraction and pertraction. Proc. Biochem. 1999, 34, 835-843.

CHAPTER 3

CONSTRUCTION OF PTA GENE DELETED OF CLOSTRIDIUM TYROBUTYRICUM FOR ENHANCED BUTYRIC ACID FERMENTATION

Summary

Clostridium tyrobutyricum ATCC 25755 is an acidogenic bacterium, producing butyrate and acetate as its main fermentation products. In order to decrease acetate and increase butyrate production, integrational mutagenesis was used to disrupt gene associated with the acetate formation pathway in C. tyrobutyricum. A non-replicative integrational plasmid containing phosphotransacetylase gene (pta) fragment cloned from

C. tyrobutyricum by using degenerate primers and an erythromycin resistance cassette was constructed and introduced into C. tyrobutyricum by electroporation. Integration of the plasmid into the homologous region on the chromosome inactivated the target pta gene and produced the pta-deleted mutant (PPTA-Em), which was confirmed by

Southern hybridization. SDS-PAGE and two-dimensional protein electrophoresis results indicated that protein expressions were changed in the mutant. Enzyme activity assays using the cell lysate showed that the activities of PTA and AK in the mutant were reduced by more than 60% for PTA and 80% for AK. The mutant grew slower in batch 52

fermentation with glucose as the substrate, but produced 15% more butyrate and 14% less acetate as compared to the wild type strain. Its butyrate productivity was approximately two fold higher than the wild type strain. Moreover, the mutant showed much higher tolerance to butyrate inhibition and the final butyrate concentration was improved by

68%. However, inactivation of pta gene did not completely eliminate acetate production in the fermentation, suggesting the existence of other enzymes (or pathways) also leading to acetate formation. This is the first reported genetic engineering study demonstrating the feasibility of using gene inactivation technique to manipulate the acetic acid formation pathway in C. tyrobutyricum in order to improve butyric acid production from glucose.

3.1 Introduction

Clostridium tyrobutyricum ATCC 25755 is a gram-positive, rod-shaped, spore-

forming, obligate anaerobic bacterium capable of fermenting a wide variety of

carbohydrates to butyric and acetic acids. There has been increasing interest in the

production of butyric acid from agricultural commodities and processing wastes using C. tyrobutyricum (Zhu et al., 2002). Butyric acid has many applications in chemical, food, and pharmaceutical industries (Zigova, et al. 1999; Vandak et al. 1997). Conventional butyric acid fermentation process is not yet economically competitive because it produces butyric acid at a relatively low concentration, yield and rate. Recently, C. tyrobutyricum cells immobilized in a fibrous-bed bioreactor were successfully used for butyrate fermentation with increased reactor productivity and final product concentration (Zhu et al., 2002; Wu and Yang, 2003). However, the butyrate yield was only ~0.5 g/g or 0.9 mol/mol and acetate yield was ~0.27 mol/mol of glucose fermented in the fermentation

(Zhu et al., 2002). To improve the economics of the fermentation process, it is desirable to increase butyrate production while reducing acetate production, which also reduces the product separation cost. Several factors influencing the selectivity for butyrate over acetate in the fermentation have been identified, including cell growth rate, glucose concentration or supply mode, and the partial pressure of H2 (van der Lelie, et al., 1988;

Michel-Savin, et al., 1990; Michel-Savin, et al., 1990). Complete selectivity for butyrate production was shown to be possible in glucose-limited fed-batch cultures (Michel-Savin, et al., 1990), but the reactor productivity and final product concentration were not high enough for economical production purpose. 54

The metabolic pathways for acids production in some clostridial strains, such as

Clostridium acetobutylicum, have been extensively studied (Rogers and Gottschalk,

1993). The breakdown of hexose to pyruvate proceeds via the Embden-Meyerhof-Parnas pathway in C. acetobutylicum. Phosphotransacetylase (PTA) and acetate kinase (AK) are two key enzymes involved in the pathway leading to acetic acid formation, whereas phosphotransbutyrylase (PTB) and butyrate kinase (BK) catalyze the formation of butyric acid from butyryl-CoA. C. tyrobutyricum produces butyrate with acetate as its main co- product, indicating similar acids formation metabolic pathways possibly exist in this bacterium. The detection of enzyme activities of PTA, AK, PTB, and BK in C. tyrobutyricum demonstrates that the production of butyrate and acetate can also be catalyzed by these enzymes (Zhu and Yang, 2003).

Integrational mutagenesis, a genetic engineering technique that can selectively inactivate undesired genes from the host chromosome has been developed and successfully used to create metabolically engineered mutants of Clostridial strains (Green et al., 1996). In this technique, a fragment of the target gene is cloned into a non- replicative vector with a selection marker, resulting in the non-replicative integrational plasmid. The partial gene in the non-replicative plasmid can recombine with the internal homologous region of the original target gene in the parental chromosome, which results in insertional inactivation of the target gene. Previous studies using integrational mutagenesis to improve clostridial fermentation product yield have focused on solventogenic C. acetobutylicum. A non-replicative plasmid (pJC4) with partial pta gene

(encoding PTA) was constructed and integrated into the homologous region of pta gene on the chromosome of C. acetobutylicum ATCC 824, resulting in reduced PTA and AK

activities and acetate production (Green and Bennett, 1998; Green et al., 1996). The same metabolic engineering approach can be used to reduce or eliminate acetate production in butyric acid fermentation by C. tyrobutyricum.

The main objectives of this study were to genetically modify the acetate formation pathway by inactivating pta gene in C. tyrobutyricum and to study its effect on butyric acid fermentation. The pta gene has been cloned and characterized for several microorganisms, including C. acetobutylicum (Boynton, et al., 1996), Escherichia coli

(Kakuda et al., 1994; Matsuyama, et al., 1989), and Methanosarcina thermophila

(Latimer and Fery, 1993). However, to date no genetic engineering study has been reported for C. tyrobutyricum or similar butyric acid bacteria, and very little is known about the pta gene in C. tyrobutyricum. In this work, the partial pta gene from C. tyrobutyricum was cloned and sequenced. Gene inactivation by integrational plasmid was then carried out to develop a pta-deleted mutant. The protein expressions and the enzyme activities of the mutant were examined to understand how the gene manipulation worked.

Finally, the effects of the mutation on cell growth and fermentation kinetics were studied and are discussed in this paper.

3.2 Materials and Methods

3.2.1 Bacterial Strains and Plasmids

Table 3.1 lists all bacterial strains and plasmids used or created in this work along with their characteristics and sources. C. tyrobutyricum ATCC 25755, designated as the

wild type, was grown anaerobically at 37°C in clostridial growth medium (CGM) (Huang et al., 1998). Colonies were maintained on Reinforced Clostridial Medium (RCM; Difco) plates in the anaerobic chamber. These media were supplemented with 40 µg/ml erythromycin (Em) for C. tyrobutyricum mutant selection. E. coli was grown aerobically at 37°C in Luria-Bertani (LB) medium supplemented with ampicillin (100 µg/ml) and erythromycin (200 µg/ml).

3.2.2 DNA Manipulations

Plasmid DNA from E. coli was isolated using QIAprep Miniprep plasmid purification kit (Qiagen, Valencia, CA) for sequencing and transformation purposes.

Chromosomal DNA from C. tyrobutyricum was prepared using QIAGEN genomic DNA kit. DNA fragment was purified from gel using QIAGEN gel extraction kit.

3.2.2.1 PCR Amplification.

Synthetic oligonucleotides were designed as primers for PCR amplification based on the homology alignment analysis of PTA from E. coli, C. acetobutylicum, B. subtilis,

M. thermophila, P. denitrificans, and Mycoplasma genitalium (Boynton et al., 1996) and the codon usage preference for C. tyrobutyricum (http://www.kazusa.or.jp/codon). The

highest homologous region of the amino acid sequence was selected as the degenerate

primers. The DNA sequences of the non-specific primers for pta gene were 5’–GA(A/G)

(C/T)T(A/T/G) AG(A/G) AA(A/G) CA(T/C) AA(A/G) GG(A/T) ATG AC–3’ (upstream)

and 5’–(A/T)GC CTG (A/T)(G/A)C (A/T)GC(A/T/C) GT(A/T) AT(A/T) GC–3’

(downstream). Amplification of partial pta sequence from wild type C. tyrobutyricum chromosomal DNA (template) was performed with optimized PCR buffer containing 0.5 mM (each) dNTPs, 300 nM (each) primers, 2.5 mM MgCl2 and 2.5 U Taq DNA

polymerase (Invitrogen, Carlsbad, CA) in a DNA engine (MJ Research, Reno, NV).

Thermal cycling was conducted under the following conditions: initial denaturation (94oC

for 3 min); 40 cycles program with template denaturation (94oC for 50 s), primers

annealing (48oC for 50 s), and extension (72oC for 1 min); deoxyadenosine (A) adding to

the 3’ ends of PCR products (72oC for10 min). The PCR product with expected size of

730 bp was cloned into PCR vector pCR 2.1 (3.9 kb) using TA cloning kit (Invitrogen), and the produced plasmid pCR-PTA (4.65 kb) was then sequenced to determine the DNA sequence of the cloned pta gene fragment.

3.2.2.2 Construction of Integrational Plasmid.

Figure 3.1 shows the general design in constructing the integrational plasmid pPTA-Em. First, a 1.5-kb fragment was removed from pCR-PTA (4.65 kb) with SphI

digestion, and the remaining pCR-PTA was religated to form plasmid pCR-PTA1 (3.15

kb). Then, a 1.6-kb HindIII fragment containing the Emr cassette from pDG 647

(Guérout-Fleury et al., 1995) was ligated with HindIII digested pCR-PTA1, forming the integrational plasmid pPTA-Em (4.75 kb) for use in pta gene inactivation.

3.2.3 Transformation

Plasmid transformation to E. coli was performed according to the manufacturer’s instruction (Invitrogen). Transformation of integrational plasmid into C. tyrobutyricum was carried out using a Bio-Rad Gene pulser (Model II). All manipulations were operated in an anaerobic chamber equipped with an incubator and a centrifuge. The competent cells of C. tyrobutyricum were prepared as follows: after overnight growth, a 50 ml culture at late log-growth phase was used to inoculate 40 ml CGM medium supplied with

40 mM DL-threonine. Cells were grown for 4 h (OD600 = ~0.8), harvested, washed twice

and suspended in ice-cold electroporation buffer, referred as SMP buffer (270 mM

sucrose, 1 mM MgCl2, 7 mM sodium phosphate, pH 7.4). About 0.5 ml of cell suspension

was chilled on ice for 5 min in a 0.4-cm electroporation cuvette (Bio-Rad, Hercules, CA),

and plasmid DNA (10~15 µg of non-replicative plasmid pPTA-Em) was added into the

cold competent cell suspension. After the pulse (2.5 kV, 600 Ω, 25 µF) was applied to the

cuvette, the transformed cells were transferred to 5 ml pre-warmed CGM and incubated

at 37°C for 3 h prior to plating on RCM plates containing 40 µg/ml Em. Plates were

incubated to develop the mutant colonies in 37°C anaerobic incubator for 3-5 days.

3.2.4 Southern Hybridization

Restriction enzyme SmaI was used to digest the chromosomal DNA of both wild

type and mutant completely at 30oC. After being separated on a 1% agarose gel with low voltage, all digested DNA fragments were transferred from the agarose gel to a Hybond-

N+ nylon membrane (Amersham, Piscataway, NJ) by upward Southern capillary transfer. 59

Pre-hybridization of blotted nylon membrane was performed at 50oC for 1 h. Two probes

were used separately for the hybridization of Emr gene and pta gene. The Emr probe was

prepared from HindIII-digested pPTA-Em followed with SacI digestion, resulting in a

partial Emr gene fragment of ~345 bp. The pta probe was the same as the cloned pta

fragment from PCR. Both probes and the HindIII-digested λ DNA, which was used as

DNA-size marker, were labeled with alkaline phosphatase (Amersham). The

hybridization with the probes was carried out with gently shaking at 62oC overnight.

HyperfilmTM ECL (Amersham) was then used for the detection.

3.2.5 Characterization of Mutants

The protein expression pattern and the enzyme activities of PTA, AK, PTB and

BK in the pta-deleted mutant (PPTA-Em) were studied. Fermentations were performed to

further characterize PPTA-Em in its butyrate production and sensitivity to butyrate

inhibition.

3.2.5.1 Preparation of Cell Extract

Bacteria were grown in 50 ml CGM at 37°C to the exponential phase (OD600 =

~1.5). Cells were harvested, washed and suspended in 5 ml of 25 mM Tris/HCl (pH 7.4).

The cell suspension was sonicated, and cell debris was removed by centrifugation. The

protein content of extracts was determined following standard Bradford protocol (Bio-

Rad).

3.2.5.2 SDS-PAGE and Two-Dimensional Protein Electrophoresis

Protein samples for SDS-PAGE electrophoresis were prepared following standard protocol (Bio-Rad). Total protein samples (24 µg each) were loaded into wells and SDS-

PAGE gel was run at 100 V for 3 h with PROTEAN II xi Cell (Bio-Rad). For two- dimensional protein electrophoresis (2DE) analysis, cell extract was concentrated by acetone and then dissolved in rehydration buffer (8 M urea, 4% CHAPS, 10 mM DTT,

0.2% (w/v) Bio-Lytes 3/10) for sample preparation. The first dimension was performed on a 7 cm IPG strip with a nonlinear immobilized pH 3-10 gradient (Amersham). The

IPG strip was rehydrated in rehydration buffer with 6 µg protein sample at 50 V for 12 h using PROTEAN IEF Cell (Bio-Rad). After rehydration, the protein was focused on IPG strip by preset method, at 250 V for 15 min to remove excess salts, then ramped linearly from 250 V to 4000 V for 2 h, and finally maintained at 4000 V for 5 h for focusing purpose. After isoelectric focusing (IEF), the strip was equilibrated in equilibrated buffer

I (6 M urea, 2% SDS, 0.375 M Tris-HCl, pH 8.8, 20% glycerol and 130 mM DTT) for

10-15 min and in equilibrate buffer II (6 M urea, 2% SDS, 0.375 M Tris-HCl, pH 8.8,

20% glycerol and 135 mM iodoacetamide) for 10-15 min. The equilibrated strip was applied to a polyacrylamide/PDA SDS gel to run the second dimension electrophoresis at

100 V for 90-120 min with Mini-PROTEAN 3 Cell (Bio-Rad). The protein spots were developed using silver staining kit (Amersham). The two-dimensional protein electrophoresis gels were analyzed using Phoretix 2D AdvancedTM software (Nonlinear

Dynamics Ltd, Newcastle upon Tyne, UK).

3.2.5.3 Enzyme Assays

The activities of acetate kinase (AK) and butyrate kinase (BK) were measured by monitoring the formation of acyl phosphate from acetate and butyrate, respectively, at

540 nm (Rose, 1955). Enzyme activity was calculated on the basis of a molar extinction coefficient of 0.169 mM-1cm-1 (Cary et al., 1988). One unit of AK and BK activity was

defined as the amount of enzyme that produces 1 µmol of hydroxamic acid per minute.

Phosphotransacetylase (PTA) and phosphotransbutyrylase (PTB) were assayed by detecting the liberation of CoA from acetyl-CoA and butyryl-CoA at 405 nm, respectively (Andersch et al., 1983). An extinction coefficient of 13.6 mM-1cm-1 was used for activity calculation. One activity unit of PTA and PTB was defined as the amount of enzyme converting 1 µmol of acyl-CoA or butyryl-CoA per minute under the reaction conditions. Specific activity of all enzymes was defined as the unit of activity per mg of protein.

3.2.5.4 Fermentation Kinetic Study

Batch and fed-batch fermentations of C. tyrobutyricum were performed in a 5 L stirred-tank fermentor (Marubishi MD-300) containing 2 L of clostridial growth medium

(CGM) supplemented with glucose (30 g/L) and 40 µg/ml erythromycin (Em) as required.

Anaerobiosis was reached by initially sparging the medium with nitrogen. The medium pH was adjusted to ~6.0 with 6 N HCl before inoculation with ~100 ml of cell suspension prepared in a serum bottle. Experiments were carried out at 37°C, 150 rpm, and pH 6.0 controlled by NH4OH. The fed-batch mode was operated by pulse feeding concentrated 62

substrate solution when the sugar level in the fermentation broth was close to zero. The feeding was continued until the fermentation ceased to produce butyrate due to product inhibition. Gas (H2 and CO2) production in the fermentation was monitored using an on-

line respirometer system equipped with both H2 and CO2 sensors (Micro-oxymax,

Columbus Instrument). Samples were taken at regular intervals from the fermentation

broth for the analyses of cell, substrate and products. A high performance liquid

chromatograph (HPLC) was used to analyze the organic compounds, including glucose,

butyrate, and acetate in the liquid samples (Wu and Yang, 2003).

3.2.5.5 Inhibition Effect of Butyrate on Cell Growth

Cultures of C. tyrobutyricum were grown in serum tubes containing 10 ml of

media with various concentrations of butyrate (0 – 15 g/L) to evaluate the inhibition

effect of butyrate on cell growth, which was followed by measuring the optical density at

600 nm (OD600) with a spectrophotometer (Sequoia-Turner, Model 340). Specific growth

rates were estimated from the OD600 data.

3.3 Results

3.3.1 PCR Amplification and Sequence Analysis

One DNA fragment with expected size of ~730 base nucleotides was generated

with degenerate primers by PCR amplification. This DNA fragment was cloned into pCR

2.1 vector. The resulting plasmid, designated as pCR-PTA, was then transformed into E. coli. Positive clones were identified, and nucleotide sequencing of the pta fragment showed 732 nucleotides, encoding for 244 amino acids, which can be found in GenBank

(GBAN AY572855). The partial amino acid sequence of PTA of C. tyrobutyricum was then compared with the known sequences of complete PTA from several other microorganisms. As shown in the homology alignment (Figfure 3.2), there are high degrees of identities and similarities between C. tyrobutyricum and C. acetobutylicum

(70%), Methanosarcina thermophila (56%), E. coli (51%), B. subtilis (47%),

Mycoplasma genitalium (47%), and P. denitrificans (47%), confirming that the PCR product was from the pta gene in C. tyrobutyricum.

3.3.2 Transformation

The non-replicative plasmid pPTA-Em (4.75 kb) was constructed and used to transform

C. tyrobutyricum by electroporation. Before electroporation, the protoplast of C. tyrobutyricum was prepared to examine the presence of restriction system on this plasmid.

The inability to detect any digestion of the plasmid suggested that no restriction enzymes were present in C. tyrobutyricum, similar to the results previously obtained with

Clostridium pasteurianum ATCC 6013 (Richards et al., 1988). After electroporation, a selective pressure of Em was used to detect mutant cells containing the non-replicative plasmids. A total of ~10 Em-resistant colonies were obtained after electroporation with a transformation efficiency of 1 colony per µg DNA, which was similar to those obtained for the integrational plasmids in C. acetobutylicum ATCC 824 (Green and Bennett, 1998; 64

Green et al., 1996). As a negative control, non-replicative plasmid with Emr cassette but

without the pta fragment was also used to transform the C. tyrobutyricum cells. As

expected, no transformed mutant was obtained since the plasmid cannot be replicated in

the cells without first integrating into the chromosome via homologous recombination.

Therefore, the pPTA-Em must have been integrated into the chromosome by homologous

recombination in the transformed cells. Since the homologous region in pPTA-Em is the

internal DNA sequence of pta, the transformed cells should be mutagenic and the original

pta gene on the chromosome should have been disrupted.

3.3.3 Southern Hybridization

DNA hybridization was performed to localize the integration site of the non-

replicative plasmid on the parental chromosome. Both SmaI digested PPTA-Em and wild

type chromosomal DNA was identified with two probes (partial Emr and pta gene)

following the method described by Green et al. (Green et al., 1996). It is noted that the

non-replicative plasmid pPTA-Em had unique SmaI restriction site in the backbone and

there was no SmaI site in the pta insert. As shown in Figure 3.3, the Emr probe only

hybridized to one SmaI fragment (6.3 kb) from mutant PPTA-Em but none from the wild

type strain, indicating that the integrational plasmid was inserted into the chromosomal

DNA of the mutant since the 6.3-kb fragment contained the antibiotics gene from the

plasmid pPTA-Em. Meanwhile, two SmaI fragments (approximate 4.5 kb and 6.3 kb) from the mutant strain and a 6-kb SmaI fragment from the wild type strain were detected by the pta probe (Figure 3.3). The total size (10.8 kb) of the pta probe hybridized

fragments from the mutant (4.5 kb + 6.3 kb) was equal to the size of pta hybridized fragment from the wild type (6.0 kb) plus the size of the plasmid pPTA-Em (4.8 kb), indicating that pPTA-Em had been inserted into the parental pta gene on the chromosome in the mutant through homologous recombination (Campbell, 1962), as illustrated in

Figure 3.3. Since pPTA-Em was internal to the pta gene, the original pta gene on the chromosome was disrupted and should have lost its function, producing the pta-deleted mutant PPTA-Em.

3.3.4 Protein Expression

The effects of pta gene disruption on protein expression in the mutant cells were evaluated with SDS-PAGE and 2DE analyses. The SDS-PAGE gel clearly showed that the highly expressed protein in the wild type with molecular weight of ~32 kDa diminished in PPTA-Em mutant (Figure 3.4). Figure 3.5 shows the two-dimensional protein analysis of wild type and mutant PPTA-Em grown at 37oC, pH 6.0, and with

glucose as the carbon source. The 2DE gels were analyzed with the Phoretix 2D

AdvancedTM software, which also normalized the different intensities of the protein spots

on these gels for easy comparison. As can be seen in Figure 3.5, the number of proteins

and their expression levels were altered in the mutant PPTA-EM. At least two proteins

with molecular weight of ~32 kDa (spot #57, PI ≈ 6.2 and spot #60, PI ≈ 6.8) in the wild

type were missing and one protein (#58) highly expressed in wild type was dramatically

down regulated in the mutant. This result is consistent with the finding from the SDS-

PAGE analysis. The missing proteins in the mutant were probably AK and PTA.

Disrupting the pta gene might have resulted in the deletion of both AK and PTA from the mutant PPTA-Em since both pta and ack genes are most likely to be in the same operon with pta gene being upstream of the ack gene, as found in several microorganisms including C. acetobutylicum (Boynton et al., 1996).

3.3.5 Enzyme Activities

Exponential-phase cultures of C. tyrobutyricum wild type and PPTA-Em mutant were harvested, and the cell extracts were assayed for acetate and butyrate-producing enzymes PTA, AK, PTB, and BK. As shown in Figure 3.6, although the activities of

PTA and AK were reduced dramatically, the mutant PPTA-Em still had some activities of PTA (40%) and AK (20%). The smaller decrease in the PTA activity is probably because the cloned pta fragment is near the stop codon of the gene and after recombination, more than 40% of the pta gene is still intact on the chromosome.

However, the lowered PTA and AK activities in the mutant were more likely from other enzymes that also can produce acetate from the same substrates (Rogers and Gottschalk,

1993). The mutant also had a higher BK activity (~135%) and similar PTB activity, as compared to the wild type strain. The greatly reduced AK enzyme activity in PPTA-Em also indicates that the ack gene lies downstream from pta gene in the same operon.

After growing the mutant in the medium without Em for ~10 generations (48 h), no differences in the key enzyme activities were detected as compared to the mutant in the antibiotics-containing medium, indicating no revertants (data not shown). Also, there was no obvious change in cell growth and acid production kinetics in repeated batch

fermentations even in the absence of the antibiotics (data not shown). It thus can be concluded that the mutant with gene mutation resulted from the homologous recombination on the chromosome is stable and can be maintained without using the antibiotics in long-term fermentation.

3.3.6 Fermentation Kinetics

Figure 3.7 shows the kinetics of fed-batch fermentations with the mutant and the wild type strains. The mutant grew exponentially in the first fed-batch and then entered the stationary phase. It continued to produce butyrate until the butyrate concentration reached 32.5 g/L, which was much higher than that obtained in the wild-type fermentation (20.2 g/L). However, a significant amount of acetate was still produced by the mutant and reached ~4.28 g/L. It was found that the mutant had a lower specific growth rate than that of the wild type, although its cell yield appeared to be higher. Table

3.2 compares the fermentation results from the mutant and the wild type. It is clear that pta deletion resulted in lower acetate yield, higher butyrate yield, final concentration, and productivity, and consequently, higher selectivity of butyrate over acetate. Besides the greatly increased final butyrate concentration produced by the mutant, the butyrate yield was also increased, from 0.33 g/g by the wild type to 0.38 g/g by the mutant, and the acetate yield decreased from 0.067 g/g by the wild type to 0.058 g/g by the mutant. There was significant improvement in butyrate productivity by the mutant (0.63 g/L·h), which was about 1.9 fold higher than that of the wild type (0.33 g/L·h). However, inactivation of the pta gene did not completely eliminate acetate formation, indicating that additional

acetate-forming enzymes were probably present in C. tyrobutyricum. Also, the higher butyrate concentration produced by the mutant in the fed-batch fermentation suggested a phenotypic change in its butyrate tolerance, which was unexpected in the original experimental design. It is noted that the carbon balance in the fermentation was close to

100% for both the wild type (95.3%) and the mutant (100%) (see Table 3.2), indicating a complete recovery of carbon in the fermentation products and cells from the substrate, glucose.

3.3.7 Butyrate Tolerance

To determine the phenotypic change about butyrate tolerance in mutant strain, cells were grown as free-cell suspension cultures at different initial butyrate concentrations (0 ~ 15 g/L). The specific growth rates determined from the growth data are shown as the relative growth rate with the rate at zero initial butyrate concentration being 100%. As shown in Figure 3.8, the pta-deleted mutant had a much higher tolerance to butyric acid than the wild type. At 15 g/L of butyric acid, the mutant retained ~30% of its maximum growth rate as compared to less than 10% in the wild type. The growth inhibition by butyric acid followed the non-competitive inhibition kinetics with the inhibition rate constants KP equal to 1.59 g/L and 5.56 g/L for the wild type and mutant,

respectively. It is clear that butyric acid strongly inhibited cell growth of the wild type

but not as strongly to the mutant.

3.4 Discussion

3.4.1 Cloning of C. tyrobutyricum

The partial pta gene in C. tyrobutyricum ATCC 25755 was successfully cloned

and sequenced using degenerate primers. One integrational plasmid (pPTA-Em) was

constructed with the cloned pta fragment, which was then introduced into the

chromosomal DNA of the parental bacterium via electroporation and homologous

recombination, and the pta gene on the chromosome was successfully inactivated. This is the first time a genetically engineered mutant of C. tyrobutyricum has been

successfully created for butyric acid fermentation. Although similar cloning work has

been done with other Clostridia species, the cloning procedures for C. tyrobutyricum used

in this work were developed after careful optimization. For PCR amplification, the

optimal Mg2+ concentration in the buffer was found to be 2.5 mM and the optimal primer

annealing temperature was 48°C. Decreasing the annealing temperature reduced the

reaction specificity and enhanced the incorrect anneal of the degenerate primers.

The electroporation method to transform plasmids into C. tyrobutyricum was

developed after optimizing the conditions used in preparing the competent cells, the

selection of an appropriate electroporation buffer, and the electroporation gene pulser

parameters. The choice of electroporation buffer was important to transform C. tyrobutyricum. Several electroporation buffers referred in the literature (Phillips-Jones,

1995), including 15% glycerol, 10% (w/v) PEG 8000, SP buffer (270 mM sucrose, 5 mM

NaH2PO4, pH 7.4) and the SMP buffer (270 mM sucrose, 1 mM MgCl2, 7 mM sodium

phosphate, pH 7.4) were evaluated, and the last one was found to work the best for C.

tyrobutyricum. The highest transformation efficiency was obtained by using mid-log phase cells (OD600 = 0.8-1.0) instead of late-log phase (OD600 = ~2.0). This result is

similar to that observed with C. botulinum (Zhou and Johnson, 1993) and C. acetobutylicum DSM 792 (Nakotte, 1998), but different from the result obtained in C.

perfringens, in which the optimal transformation efficiency was found with cells

harvested from the late-log phase (Allen and Blaschek, 1990) or the stationary phase

(Phillips-Jones, 1990). The gene pulser parameters, including the voltage 2.5 kV with the

field strength of 6.25 kV/cm and the electrical resistance 600 Ω with a capacitance 25 µF,

were applied in electroporation. It is found that decreasing the discharge to 2.0 kV and

increasing the resistance to ∞ did not improve the transformation efficiency with C.

tyrobutyricum. The pulse duration under the optimized conditions was 6.3-7.5 ms. Also,

the transformation efficiency of C. tyrobutyricum by electroporation was enhanced with

the addition of DL-threonine in the medium, which helped to weaken the cell wall of

gram-positive microorganism by incorporation of the D-isomer (van der Lelie et al.,

1988). It is noted that the optimized transformation conditions are somewhat different

from the ones previously reported and commonly used for other clostridia. No

transformed colonies of C. tyrobutyricum could be obtained under the previously reported

electroporation conditions. The improved electroporation method provides an efficient

way to introduce foreign genes into C. tyrobutyricum with a transformation efficiency of

1 colony per µg DNA for non-replicative plasmids. It opens the way for future molecular

genetic studies in C. tyrobutyricum.

3.4.2 Protein Expression

In this work, both SDS-PAGE gel and two-dimensional protein electrophoresis

(2DE) maps showed that at least two proteins (spot #57 and #60) found in the wild type were not expressed in the pta-deleted mutant. These missing proteins are most likely to include PTA and AK, but cannot be identified in this work due to lack of proteomic information for C. tyrobutyricum. Phosphotransacetylase (PTA) and acetate kinase (AK) from several microorganisms have been purified and characterized (Boyton, et al., 1996).

PTA is a monomer with a molecular weight of approximate 36.2 kDa in C. acetobutylicum, 88 kDa in C. thermoaceticum, and 27 kDa in S. pyogenes. AK is a dimer of two identical subunits with various molecular masses in different species, 44.3 kDa in

C. acetobutylicum and 60 kDa in C. thermoaceticum. Only the PI values of PTA and AK for some E. coli strains are available in the literature (Kirkpatrick et al., 2001; Peng and

Shimizu, 2003). The large variations in PI and molecular weight for AK and PTA from different microorganisms make it difficult to identify AK and PTA on the 2DE gels.

Further studies and analysis of the proteins isolated from the 2DE gels will be necessary to confirm their identities.

3.4.3 Effects of Integrational Mutagenesis

Gene inactivation has been proved to be a feasible genetic engineering technique for studying molecular genetics in C. acetobutylicum (Harris, et al., 2000), but has not been applied to C. tyrobutyricum. In this work, the pta gene in the acetate-forming pathway in C. tyrobutyricum was inactivated by integration of a non-replicative plasmid

on the chromosome. The enzyme activity assays showed that both PTA and AK activities were reduced greatly in the pta-deleted mutant, suggesting that deletion of the pta gene also resulted in the inactivation of ack gene. This finding is consistent with the study of the pta-deleted mutant of C. acetobutylicum (Green et al., 1996), and suggests that the acetic acid-forming genes, pta and ack, exist in the same operon with pta preceding ack (Boynton, et al., 1996). However, PTA and AK activities were still detected in the mutant, which probably came from other enzymes working with the same substrates as PTA and AK. The specific activities of PTB and BK enzymes involving in the formation of butyrate were either unaffected or increased. It is possible that the deletion of the acetate formation pathway had pushed more carbon and energy sources to flow through the butyrate formation pathway, which required a higher BK enzyme activity for the increased flux.

Fermentation study showed that the pta-deleted mutant produced more butyrate from glucose but was still capable of producing acetate. The butyrate yield increased

~15% while acetate yield decreased only ~14%, even though the specific activities of

PTA and AK in the mutant had been reduced by more than 50% as compared to the wild type. The chance for a revertant to appear in the fermentation was low even when Em was not included in the fermentation medium. Therefore, there must be other enzymes besides PTA and AK in C. tyrobutyricum that also can produce acetate from acetyl-CoA and perhaps other substrates as well. For example, CoA transferase can catalyze the formation of acetate from acetyl-CoA. This enzyme has been found in some clostridia bacteria (Rogers and Gottschalk, 1993), and could also be present in C. tyrobutyricum. It also has been reported that PTB and BK exhibited broad substrate specificities with C2-

to C6-chained acyl-CoA and carboxylic acid compounds in C. acetobutylicum

(Wiesenborn et al., 1989; Hartmanis, 1987). The butyrate-producing enzymes PTB and

BK in C. tyrobutyricum mutant might have been responsible for the observed acetate production in the fermentation. It is also possible that there is additional acetate formation pathway catalyzed by other enzymes in C. tyrobutyricum, as was reported for some species (Lindmark, 1976). Possible presence of PTA and AK isozymes also could not be ruled out. Two acetate kinase isozymes from spirochete MA-2 cell extracts (Harwood and Canale-Parola, 1982) and a butyrate kinase isoenzyme (BKII) in C. acetobutylicum

ATCC 824 (Huang, et. al, 2000) have been reported.

Inactivation of pta also had a significant effect on cell growth. Compared to the wild-type, the mutant PPTA-Em had a reduced specific growth rate, but had a higher cell yield from glucose (Table 3.2). This observation is similar to that found in pta or buk inactivated C. acetobutylicum (Green et al., 1996). Since both acid-formation pathways are responsible for generating energy (ATP) for cells, a reduced acetate production may impose a metabolic burden on cells. A feasible cellular response to this metabolic burden is the elevation of the flux through the alternate ATP-generation pathway, namely butyrate formation, to avoid any significant loss in overall cell growth. This metabolic change might have resulted in the observed increase in cellular BK enzyme activity.

Since the acetate formation pathway can produce more ATP per glucose metabolized than the butyrate formation pathway, the elimination of acetate metabolic pathway would result in reduced energy production efficiency, which in turn could direct more substrates towards the butyrate formation pathway to offset loss in the ATP generation and overall

cell growth. This also could explain why biomass production was increased in the mutant even though the specific growth rate was lowered.

Compared to the wild type, the butyrate tolerance of pta-deleted mutant had significantly enhanced butyrate tolerance (Fig 3.8), which is unexpected but can explain the observed higher butyrate productivity and final butyrate concentration in the fermentation with the mutant. It had been noted that PTA in C. tyrobutyricum was more strongly inhibited by butyric acid than PTB (Zhu and Yang, 2003). It is thus possible that by disrupting the butyrate-sensitive PTA and acetate-forming pathway, the mutant became less sensitive to butyrate inhibition since they used mainly the butyrate-forming pathway to generate ATP needed for biosynthesis and maintaining a functional pH gradient across the cell membrane. It should be noted, however, that cultures adapted and grown in a fibrous-bed bioreactor (FBB) were much more tolerant to butyrate inhibition and produced more butyric acid from glucose than those obtained in free-cell fermentations (Zhu and Yang, 2003). With the FBB, butyric acid can be produced at a concentration of higher than 50 g/L with a high butyrate yield of more than 0.5 g/g glucose.

3.5 Conclusions

This is the first genetic engineering study of C. tyrobutyricum for enhanced butyric acid fermentation. In this work, the cloning procedures for C. tyrobutyricum were optimized and gene inactivation experiments were carried out to develop mutant strains of C. tyrobutyricum for butyrate production from glucose with improved productivity, 75

yield, final product concentration, and butyrate tolerance. The manipulation of acid- forming pathways by gene inactivation proved to be feasible for obtaining metabolically advantageous mutants for butyrate production from glucose. However, gene manipulations in the metabolic pathway can lead to unexpected changes in protein expression pattern and other phenotypes that require further studies.

3.6 References

Allen, S. P., and Blaschek, H. P. Factors involved in the electroporation-induced transformation of Clostridium perfringens. FEMS Microbiol. Lett. 1990, 70, 217-220.

Andersch, W., Bahl, H., and Gottschalk, G. Levels of enzymes involved in acetate, butyrate, acetone and butanol formation by Clostridium acetobutylicum. Eur. J. Appl. Microbiol. Biotechnol. 1983, 17, 327-332.

Campbell, A. M. Episomes. Adv. Genet. 1962, 11, 101-146.

Cary, J. W., Peterson, D. J., Papoutsakis, E. T., and Bennett, G. N. Cloning and expression of Clostridium acetobutylicum phosphotransbutyrylase and butyrate kinase genes in Escherichia coli. J. Bacteriol. 1988, 170, 4613-4618.

Drake, H. L. Demonstration of hydrogenase in extracts of the homoacetate-fermenting bacterium Clostridium thermoaceticum. J. Bacteriol. 1982, 150, 702-709.

Green, E. M., and Bennett, G. N. Genetic manipulation of acid and solvent formation in Clostridium acetobutylicum ATCC 824. Biotechnol. Bioeng. 1998, 58, 215-221.

Guérout-Fleury, A. M., Shazand, K., Frandsen, N., and Stragier, P. Antibiotic-resistance cassettes for Bacillus subtilis. Gene 1995, 167, 335-336.

Harris, L. M., Desai, R. P., Welker, N. E., and Papoutsakis, E. T. Characterization of recombinant strains of the Clostridium acetobutylicum butyrate kinase inactivation mutant: need for new phenomenological models for solventogenesis and butanol inhibition? Biotechnol. Bioeng. 2000, 67, 1-11.

Hartmanis, M. G. N. Butyrate kinase from Clostridium acetobutylicum. J. Biol. Chem. 1987, 262, 617-621.

Harwood, C. S. and Caale-Parola, E. Properties of acetate kinase isozymes and a branched-chain fatty kinase from a spirochete. J. Bacteriol. 1982, 152, 246-254.

Huang, Y. L., Mann, K., Novak, J. M., and Yang, S. T. Acetic acid production from fructose by Clostridium formicoaceticum immobilized in a fibrous-bed bioreactor. Biotechnol. Prog. 1998, 14, 800-806.

Kirkpatrick, C., Maurer, L. M., Oyelakin, N. E., Yoncheva, Y. N., Mauer, R., Slonczewski, J.L. Acetate and formate stress: opposite responses in the proteome of Escherichia coli. J. Bacterial. 2001, 183, 6466-6477.

Latimer, M. T., and Fery, J. G. Cloning, sequence analysis, and hyperexpression of the genes encoding phosphotransacetylase and acetate kinase from Methanosarcina thermophila. J. Bacterial. 1993, 175, 6822-6829.

Lindmark, D. G. Acetate production by Tritrichomonas foetus. Biochem. Parasites Host- Parasite Relat. Proc. Int. Sym. 1976, 10, 15-21.

Matsuyama, A. H., Yamamoto, H., and Nakano, E. Cloning, expression, and nucleotide sequences of the Escherichia coli K-12 ackA gene. J. bacterial. 1989, 171, 77-580.

Michel-Savin, D., Marchal, R., and Vandecasteele, J.P. Control of the selectivity of butyric acid production and improvement of fermentation performance with Clostridium tyrobutyricum. Appl. Microbiol. Biotechnol. 1990, 32, 387-392.

Michel-Savin, D., Marchal, R., and Vandecasteele, J. P. Butyrate production in continuous culture of Clostridium tyrobutyricum: effect of end-product inhibition. Appl. Microbiol. Biotechnol. 1990, 33, 127-131.

Nakotte, S., Schaffer, S., Böhringer, M., and Dürrre, P. Electroporation of, plasmid isolation from and plasmid conservation in Clostridium acetobutylicum DSM 792. Appl. Microbiol. Biotechnol. 1998, 50, 564-567.

Papoutsakis, E. T., and Meyer, C. L. Equations and calculations of product yields and preferred pathways for butanediol and mixed-acid fermentations. Biotechnol. Bioeng. 1985, 27, 50-66.

Peng, L., and Shimizu, K. Global metabolic regulation analysis for Escherichia coli K12 based on protein expression by 2-dimensional electrophoresis and enzyme activity measurement. Appl. Microbiol. Biotechnol. 2003, 61, 163-178.

Phillips-Jones, M. K. Plasmid transformation of Clostridium perfringens by electroporation methods. FEMS Microbiol. Lett. 1990, 66, 221-226.

Phillips-Jones, M. K. Introduction of recombinant DNA into Clostridium spp. In: Nickoloff JA, editor. Methods in Molecular Biology, vol. 47. Electroporation protocols for microorganisms. New York: Hunana Press Inc. 1995, p 227-235.

Richards, D. F., Linnett, P. E., Oultram, J. D., and Young, M. Restriction endonucleases in Clostridium pasteurianum ATCC 6013 and C. thermohydrosulfuricum DSM 568. J. Gen. Microbiol. 1988, 134, 3151-3157.

Rogers, P., and Gottschalk, G. Biochemistry and regulation of acid and solvent production in Clostridia. In: Woods DR, editor. The clostridia and biotechnology. Massachusetts: Butterworth-Heinemann. 1993, p 25-50.

Rose, I. A. Acetate kinase of bacteria (acetokinase). Methods Enzymol. 1955, 1, 591-595. van Andel, J. G., Zouttberg, G. R., Crabbendam, P. M., and Breure, A. M. Glucose fermentation by Clostridium butyricum grown under a self generated gas atmosphere in chemostat culture. Appl. Microbiol. Biotechnol. 1985, 23, 21-26. van der Lelie, D., van der Vossen, J. M. B. M., and Venema, G. Effect of plasmid incompatibility on DNA transfer to Streptococcus cremoris. Appl. Environ. Microbiol. 1988, 54, 865-871.

Vandak, D., Zigova, J., Sturdik, E., and Schlosser, S. Evaluation of solvent and pH for extractive fermentation of butyric acid. Process Biochem. 1997, 32, 245-251.

Wiesenborn, D. P., Rudolph, F. B., and Papoutsakis, E. T. Phosphotransbutyrylase from Clostridium acetobutylicum ATCC 824 and its role in acidogenesis. Appl. Environ. Microbiol. 1989, 55, 317-322.

Wu, Z., and Yang, S. T. Extractive fermentation for butyric acid production from glucose by Clostridium tyrobutyricum. Biotechnol. Bioeng. 2003, 82, 93-102. 79

Zhou, Y., and Johnson, E. A. Genetic transformation of Clostridium Botulinum Hall A by electroporation. Biotechnol. Lett. 1993, 15, 121-126.

Zhu, Y., Wu, Z., Yang, S. T. Butyric acid production from acid hydrolysate of corn fiber by Clostridium tyrobutyricum in a fibrous-bed bioreactor. Process Biochemistry 2002, 38, 657-666.

Zhu, Y. and Yang S. T. Adaptation of Clostridium tyrobutyricum for enhanced tolerance to butyric acid in a fibrous-bed bioreactor. Biotechnol. Prog. 2003, 19, 365-72.

Zigova, J., Sturdik, E., Vandak, D., and Schlosser, S. Butyric acid production by Clostridium butyricum with integrated extraction and pertraction. Process. Biochemistry. 1999, 34, 835-843.

Strain/plasmid Characteristic Source/reference

Strain

C. tyrobutyricum Ems ATCC ATCC 25755 PPTA-Em pta- Emr Insertion of pPTA-Em into ATCC 25755; this study

E. coli INVαF’ recA1 Apr Invitrogen pCR 2.1 Apr Kmr Invitrogen pCR-PTA a Apr Kmr This study pCR-PTA1 Apr This study pDG 647 Apr Emr Guérout-Fleury et al. (1995) pPTA-Em b Apr Emr This study apCR-PTA was constructed by insertion of 0.73-kb pta fragment into pCR 2.1. bpPTA-Em is the non-replicative integrational plasmid. Note: Abbreviations: pta-, phosphotransacetylase gene deleted; Apr, ampicillin resistant; Kmr, kanamycin resistant; Emr, erythromycin resistant; recA1, homologous recombination abolished.

Table 3.1 Bacterial strains and plasmids.

Products Wild-type PPTA-Em

µ (h-1) 0.19 ± 0.02 0.12 ± 0.01 Cell (g/g) 0.11 ± 0.02 0.16 ± 0.02 Yield (mole C/mole)* 0.77 ± 0.14 1.12 ± 0.14

Max Concentration (g/L) 20.2 32.5

(g/g) 0.33 ± 0.03 0.38 ± 0.03 Butyrate Yield (mole/mole) 0.68 ± 0.06 0.78 ± 0.06

Productivity (g/L·h) 0.33 0.63

Max. Concentration (g/L) 3.57 4.28

Acetate (g/g) 0.067 ± 0.003 0.058 ± 0.004 Yield (mole/mole) 0.20 ± 0.01 0.17 ± 0.01

Butyrate/Acetate Ratio (g/g) 4.95 6.55

Carbon dioxide (mole/mole) 1.83 ± 0.12 1.51 ± 0.09

Carbon balance (%) 95.3 ± 4.22 101 ± 3.81

*Based on that the carbon content of cell biomass is 46.7% for Clostridia (Papoutsakis and Meyer, 1984)

Table 3.2 Comparison of fermentation results with the pta-deleted mutant PPTA- EM and the wild type strain of C. tyrobutyricum in fed-batch cultures controlled at pH 6.0, 37 ºC.

EcoRI EcoRI

pta’ (0.7 kb) pCR-PTA (4.65 kb) HindIII SphI EcoRI EcoRI HindIII SphI pta’

DELETE f1 pCR-PTA1 pCR 2.1 (3.15 kb) ColE1 (3.9 kb)

ColE1 Apr

Kmr Apr

SphI HindIII Digestion SacI

r EcoRI SmaI Em (1.6 kb) HindIII EcoRI HindIII HindIII pta’ SphI SacI

pDG647 (4.3 kb)

Emr pPTA-Em (4.75 kb) Apr

ColE1 SmaI HindIII

Figure 3.1 Construction of integrational plasmid pPTA-Em with one 0.7-kb pta fragment cloned from C. tyrobutyricum ATCC 25755. The directions of each gene are shown by arrows. Some important restriction sites used in this work are indicated. Abbreviations: pta, partial pta gene; f1, f1 filamentous phage origin of replication with helper phage; Kmr, kanamycin resistance gene; Apr, ampicillin resistance gene; Emr, erythromycin resistance gene; ColE1, compatibility group origin of replication in E. coli.

Cac 1 ELRKHKGMTIEKSEKMVRDPLYFATMALKDGYVDGMVSGAVHTTGDLLRPGLQIIKTAPG Cty 1 ELRKHKGMTPDKANKIVRDPLYFATMMVKLGDADGLVSGSIHTTGDLLRPGLQIVKTAPG Mes 1 ELRKHKGITLENAAEIMSDYVYFAVMMAKLGEVDGVVSGAAHSSSDTLRPAVQIVKTAKG Pde 1 RMRAARGMTAERALTEMRDPIRQAAMRVRLGQADGTVGGAVATTADTVRAALQIIGKAPG Eco 1 ELRKNKGMTETVAREQLEDNVVLGTLMLEQDEVDGLVSGAVHTTANTIRPPLQLIKTAPG Bsu 1 -ERRKGKATEEQARKALLDENYFGTMLVYKGLADGLVSGAAHSTADTVRPALQIIKTKEG Myc 1 EKRKHKGMDLKEAQKFVRDPSSLAATLVALKVVDGEVCGKEYATKDTLRPALQLLATG--

Cac 61 VKIVSGFFVMIIPDCDYGEEGLLLFADCAVNPNPTSDELADIAITTAETARKLCNVEPKV Cty 61 TSVVSSIFMMEVPNCDLGDNGFLLFSDCAVNPVPNTEQLAAIAISTAETAKSLCGMDPKV Mes 61 AALASAFFIISVPDCEYGSDGTFLFADSGMVEMPSVEDVANIAVISAKTFELLVQDVPKV Pde 61 AGIVSSFFLMLSCGPGAPVRGGMIFADCGLVIQPDARELAAIALSAADSCRRILAEEPRV Eco 61 SSLVSSVFFMLLP------EQVYVYGDCAINPDPTAEQLAEIAIQSADSAAAFG-IEPRV Bsu 60 VKKTSGVFIMARG------EEQYVFADCAINIAPDSQDLAEIAIESANTAKMFD-IEPRV Myc 59 -NFVSSVFIMEKG------EERLYFTDCAFAVYPNSQELATIAENTFNFAKSLNEDEIKM

Cac 121 AMLSFSTMGGAKGEMVDKVKNAVEITKKFRPH--LAIDGELQLDAAIDSEVAALKAPSSN Cty 121 AMLSFSTKGSAQHENVDKVREATKLAKQMQPD--LKIDGELQLDASLIQEVANLKAPGSP Mes 121 AMLSYSTKGSAKSKLTEATIASTKLAQELAPD--IAIDGELQVDAAIVPKVAASKAPGSP Pde 121 ALLSFSTAGSAEHPSLGRIREALALIRAAAPG--LEVDGEMQFDAALDEAIRARKAPESP Eco 114 AMLSYSTGTSGAGSDVEKVREATRLAQEKRPD--LMIDGPLQYDAAVMADVAKSKAPNSP Bsu 113 AMLSFSTKGSAKSDETEKVADAVKIAKEKAPE--LTLDGEFQFDAAFVPSVAEKKAPDSE Myc 112 AFLSYSTLGSGKGEMVDKVVLATKLFLEKHPELHQSVCGELQFDAAFVEKVRLQKAPQLT

Cac 179 VAGNANVLVFPDLQTGNIGYKLVQRFAKAKAIGPICQGFAKPINDLSRSCSSEDIVNVVA Cty 179 VAGKANVLIFPELQAGNIGYKLVQRFAKAEAIGPICQGFAKPINDLSRGCSSDDIVNVVA Mes 179 VAGKANVFIFPDLNCGNIAYKIAQRLAKAEAYGPITQGLAKPINDLSRGCSDEDIVGAVA Pde 179 LTGRPNVFVFPDLADGNIGYKIAERLAGLTAIGPILQGLAKPANDLSRACSVKDIVNATA Eco 172 VAGRATVFIFPDLNTGNTTYKAVQRSADLISIGPMLQGMRKPVNDLSRGALVDDIVYTIA Bsu 171 IKGDANVFVFPSLEAGNIGYKIAQRLGNFEAVGPILQGLNMPVNDLSRGCNAEDVYNLAL Myc 172 WKNSANIYVFPNLDAGNIAYKIAQRLGGYDAIGPIVLGLSSPVNDLSRGASVSDIFNVGI

Cac 239 ITVVQA Cty 239 ITAAQA Mes 239 ITCVQA Pde 239 ITAMQT Eco 232 LTAIQS Bsu 231 ITAAQA Myc 232 ITAAQA

Figure 3.2 Alignment of partial amino acid sequences of PTA from C. acetobutylicum (Cac; GBAN U38234), C. tyrobutyricum (Cty; GBAN AY572855), Methanosarcina thermophila (Mes; GBAN L23147), P. denitrificans (Pde; GBAN U08864), E. coli (Eco; GBAN D21123), B. subtilis (Bsu; GBAN; X73124), and Mycoplasma genitalium (Myc; GBAN L43967).

Erm gene

Non-replicative plasmid pPTA-Em SmaI 4.8 kb

pta fragment Homologous recombination SmaI SmaI 6.0 kb pta gene Wild type chromosome

SmaI 4.5 kb SmaI 6.3 kb SmaI pta 1 pta frag. 2 Erm pta frag. 1 pta 2

pta gene deleted mutant 10.8 kb = 4.5 kb + 6.3 kb = 6 kb + 4.8 kb

Emr pta 1 1

Figure 3.3 DNA hybridization. (A) Homologous recombination of plasmid pPTA- Em containing partial pta gene with the chromosome, resulting in the disruption of pta gene. (B) Southern blots with Emr probe and pta probe. Wild type (Lane 1) showed no Emr and a 6.0 kb pta-containing DNA. The mutant PPTA-Em (Lane 2) showed both Emr and 2 pta-containing DNAs (6.3 kb and 4.5 kb).

1 2 3 kDa . 116 80

51.8

34.7

32 30

Figure 3.4 SDS polyacrylamide gel electrophoresis of cellular proteins from C. tyrobutyricum. (Lane 1: PPTA-Em mutant; Lane 2: Molecular weight markers; Lane 3: Wild type.)

kDa Wild 66

PI 4.5 5.1 5.5 5.9 6.6 7.0 8.5

kDa Mutant 66

PI 4.5 5.1 5.5 5.9 6.6 7.0 8.5 Figure 3.5 Two-dimensional protein electrophoresis maps of cell extracts of C. tyrobutyricum wild type (top) and mutant PPTA-Em (bottom).

160

140

120

100

Relative activity (%) 40

0 PTA AK PTB BK

Figure 3.6 Relative activities of key enzymes in acetate and butyrate-forming pathways in the PPTA-Em mutant as compared with the wild type.

40 8 Wild Type

30 6

20 4 600 OD Glucose Acetate Concentration (g/L) 10 Butyrate 2 OD600

0 0 0 1020304050607080 Time (h)

50 8 PPTA-Em

40 6

4 600 OD 20 Glucose Acetic Concentration (g/L) 2 10 Butyric OD

0 0 0 102030405060708090100 Time (h)

Figure 3.7 Fermentation kinetics of C. tyrobutyricum wild-type (top) and pta-deleted mutant (bottom).

100

Mutant 40 Relative Growth Rate (%) 20 Wild Type

0 0 5 10 15 20 Butyrate Concentration (g/L)

Figure 3.8 Noncompetitive inhibition of butyric acid on cell growth of C. tyrobutyricum wild type (∆) and PPTA-Em (○).

CHAPTER 4

BUTYRIC ACID PRODUCITON BY PPTA-EM MUTANT

Summary

The kinetics of butyric acid fermentation by Clostridium tyrobutyricum at pH 6.0

and 37°C were studied with the wild type ATCC 25755 and its mutant PPTA-Em, which

was obtained from integrational mutagenesis to inactivate the chromosomal pta gene,

encoding phosphotransacetylase (PTA). The potential of using this mutant to improve

butyric acid production from glucose and xylose was evaluated in both free and

immobilized cell fermentations. Compared to the wild type, in free-cell fermentations

PPTA-Em produced 15% more butyrate (0.38 g/g vs. 0.33 g/g) at a much higher

concentration (62% to 72% higher) from both glucose and xylsoe. The increased butyrate

production in the mutant can be attributed to the reduced acetate production as well as

reduced specific growth rate. The acetate yield in the mutant was reduced by 13.5%

(0.058 g/g vs. 0.067 g/g) and 32% (0.045 g/g vs. 0.066 g/g) from glucose and xylose,

respectively. The mutant’s specific growth rate was reduced by 36% (0.137 h-1 vs. 0.214 h-1) on glucose and 26% (0.086 h-1 vs. 0.116 h-1) on xylose, respectively. A fibrous bed bioreactor (FBB) was used to immobilize PPTA-Em mutant cells and further improve

butyric acid production. The final butyric acid concentrations in fed-batch fermentations reached 49.9 g/L from glucose and 51.6 g/L from xylose, with the butyrate yield increased to ~0.45 g/g. As evidenced by the increased butyrate/acetate ratio (from ~5.5 to

~8.9) in the final product profile, it is concluded that the mutant’s metabolic pathway has been shifted to favor butyrate production due to the knock-out of pta gene even though acetate production remains at a significant level. The observed metabolic shift is corroborated by the changed protein expression patterns as seen in two-dimensional protein electrophoresis and SDS-PAGE.

4.1 Introduction

Clostridium tyrobutyricum, a gram-positive, rod-shaped, spore-forming, and

obligate anaerobic bacterium, can produce butyric acid, acetic acid, hydrogen and carbon

dioxide from various carbohydrates including glucose and xylose (Wu and Yang, 2003).

Butyric acid is a short-chain fatty acid naturally generated by anaerobic fermentation of

dietary substrates in intestines. It is currently produced by petrochemical routes and has

many applications; in the chemical industry, it is mainly used to synthesize butyryl

polymers; in the food industry, it is used to enhance butter-like note in food flavors; and

in the pharmaceutical industry, it can be used to treat colorectal cancer and

hemoglobinopathies (Willims, et al., 2003). Also, the esters of butyrate are used as

additives for increasing fruit fragrance and as aromatic compounds in perfumes.

There has been increasing interest in the production of butyric acid from biomass using C. tyrobutyricum and C. butyricum (Michel-Savin, et al., 1990 a b; Zhu, et al.,

2002). However, the conventional butyric acid fermentation process is not yet economically competitive because of its low reactor productivity, low final product concentration, and low product yield. Recently, C. tyrobutyricum cells immobilized in a fibrous-bed bioreactor were successfully used for butyrate fermentation with increased reactor productivity and final product concentration (Zhu, et al., 2002). A novel extractive fermentation process has also been developed, using Alamine 336 in oleyl alcohol as the extractant contained in a hollow-fiber membrane extractor for selective removal of butyric acid from the fermentation broth, that can further enhance product

concentration and purity while lowering the recovery and purification costs (Wu and

Yang, 2003; Yang, 1996).

Creating metabolically engineered mutants of Clostridial strains is another approach for improving fermentation. Recently, integrational mutagenesis has been successfully used to create gene knock-out mutants (Green, et al., 1996; Zhu, et al., 2005).

In this technique, a non-replicative integrational plasmid containing a fragment of the target gene and a selection marker is constructed and then inserted into the parental chromosome by homologous recombination to knock out the target gene, resulting in mutants with altered metabolic pathways and fermentation characteristics. Using a non- replicative plasmid containing a partial pta gene, encoding PTA (phosphotransacetylase),

Green et al. obtained mutants of solventogenic C. acetobutylicum ATCC 824 with much reduced PTA and AK (acetate kinase) activities and acetate production (Green, et al.,

1996). Similarly, by using the integrational plasmid pPTA-Em, a pta-deleted C. tyrobutyricum mutant (PPTA-Em) with decreased PTA activity and increased butyrate production was obtained (Zhu, et al., 2005).

The main objective of this work was to evaluate the potential of using the metabolically engineered C. tyrobutyricum mutant PPTA-Em for butyric acid production from glucose and xylose. In this study, free-cell and immobilized-cell FBB fermentations were carried out at 37oC and pH 6.0 with PPTA-Em and the wild type for comparison

purpose. The effects of integrational mutagenesis, sugar source, and cell adaptation in the

FBB on butyric acid fermentation kinetics were studied and are discussed in this paper.

Finally, SDS-PAGE and two-dimensional protein electrophoresis were used to study and

characterize protein expression changes as affected by the mutation in PPTA-Em and different sugar sources used in the fermentations.

4. 2 Materials and Methods

4.2.1 Culture and Medium

C. tyrobutyricum ATCC 25755 was maintained on Reinforced Clostridial

Medium (RCM; Difco) plates in an anaerobic chamber (95% N2, 5% H2). Working

cultures were grown at 37°C in a previously described synthetic medium (CGM) with

glucose as the substrate (Huang, et al., 1998). The metabolically engineered mutant,

PPTA-Em, was obtained by transforming C. tyrobutyricum competent cells with non-

replicative plasmids containing erythromycin (Em) resistant gene and partial pta

fragment obtained from PCR amplification (Zhu, et al., 2005). The pta knock-out mutant,

created through homologous recombination of the plasmids with the chromosome, were

selected and stored on the RCM plates containing 40 µg/mL Em.

4.2.2 Fermentation Kinetics

Fed-batch fermentations of C. tyrobutyricum were performed in a 5-L stirred-

tank fermentor (Marubishi MD-300) containing 2 L of the medium with either glucose or

xylose as the substrate. Erythromycin (Em) was not used in the fermentation study as the

mutant PPTA-Em was genetically stable without the selection marker. Anaerobiosis was

reached by initially sparging the medium with nitrogen. The medium pH was adjusted to

~6.0 with 6 N HCl before inoculation with ~100 ml of cell suspension prepared in a serum bottle. Experiments were carried out at 37°C, 150 rpm for agitation, and pH 6.0 ±

0.1 controlled by adding NH4OH. The fed-batch mode was operated by pulse feeding

concentrated substrate solution when the sugar level in the fermentation broth was close

to zero. The feeding was continued until the fermentation ceased to produce butyrate due

to product inhibition. Samples were taken at regular intervals from the fermentation broth

for the analyses of cell, substrate and acid products.

4.2.3 Fermentation in Fibrous-bed Bioreactor

The fibrous-bed bioreactor (FBB) was made of a glass column packed with spiral

wound cotton towel and had a working volume of ~480 ml. Detailed description of the

reactor construction has been given before (Huang, 1998). The reactor was connected to

the 5-L stirred-tank fermentor containing the medium through a recirculation loop (~1 m

long, tubing ID: 3.1 mm; Microflex Norprene 06402-16, Cole Parmer, Chicago, IL) and

operated under well-mixed conditions with pH and temperature controls. The FBB was

first operated at a repeated batch mode to increase the cell density in the fibrous bed to a

stable, high level (>50 g/L). To adapt the culture to tolerate a higher butyrate

concentration, the reactor was then operated at fed-batch mode by pulse-feeding a

concentrated substrate solution whenever the sugar level in the fermentation broth was

close to zero. At the end of each fed-batch experiment, the medium in the fermentor was

completely replaced with 2 L of fresh medium to start new fed-batch fermentation.

4.2.4 Analytical Methods

Cell density was analyzed by measuring the optical density of the cell suspension at a wavelength of 600 nm (OD600) with a spectrophotometer (Sequoia-turner, Model

340). One unit of OD600 corresponded to 0.68 g/L of cell dry weight for cells grown in

the glucose medium. A high-performance liquid chromatography (HPLC) was used to

analyze the organic compounds, including glucose, xylose, lactate, acetate, and butyrate,

in the fermentation broth. The HPLC system consisted of an automatic injector (Shimazu

SIL-10Ai), a pump (Shimadzu LC-10A), organic acid analysis column (Bio-Rad HPX-

87H), a column oven at 45oC (Shimadzu RID-10A), and a refractive index detector

(Shimadzu RID-10A). The eluent was 0.01 N H2SO4, at a flow rate of 0.6 mL/min.

Hydrogen and carbon dioxide production during the fermentation was monitored using an

on-line respirometer (Micro-oxymax system, Columbus Instrument, Columbus, OH). The

fermentor was connected to a tightly sealed bottle (volume: 5400 mL) for the collection

of the gas products, which was flushed with nitrogen periodically. The change of gas

composition inside the bottle was monitored and used to estimate the gas production,

which was also measured by collecting the produced gas in an inverted four-liter plastic

graduate cylinder in a water trough.

4.2.5 Preparation of Cell Extract and Protein Electrophoresis

Cells cultivated in 100 mL of CGM at 37°C were allowed to grow to the exponential phase (OD600 = ~1.5), and then harvested and washed. The cell pellets

suspended in 10 mL of 25 mM Tris-HCl buffer (pH 7.4) were sonicated, and the protein extract was collected by centrifugation.

Protein samples for sodium dodecyl sulfate-polyacrylamide gel electrophoresis

(SDS-PAGE) were prepared from the cell extract after sonication and centrifugation. The cell extract (10 mL) was concentrated using four volumes of acetone (40 mL) to precipitate protein at -20oC overnight, and re-dissolved in 2 mL of 25 mM Tris-HCl

buffer (pH 7.4), following the standard protocol (Bio-Rad, Hercules, CA). Protein

samples, 24 µg per well, were loaded into 12.5% SDS-PAGE gel, and run at 100 V for

2.5 h with PROTEAN II xi Cell (Bio-Rad) and stained following the instruction of the

manufacturer.

For two-dimensional protein electrophoresis (2DE), cell extract was concentrated

by acetone and then dissolved in the rehydration buffer (8 M urea, 4% CHAPS, 10 mM

DTT, 0.2% (w/v) Bio-Lytes 3/10) for sample preparation. The first dimension was

performed on an 11 cm IPG strip with a nonlinear immobilized pH 3-10 gradient

(Amersham, Piscataway, NJ). The IPG strip was rehydrated in the rehydration buffer

with 20 µg protein sample at 50 V for 12 h using PROTEAN IEF Cell (Bio-Rad). After

rehydration, the protein was focused on IPG strip by preset method, at 250 V for 15 min

to remove excess salts, then ramped linearly from 250 V to 5000 V for 2 h, and finally maintained at 8000 V for 4.5 h for focusing purpose. After isoelectric focusing (IEF), the strip was equilibrated in Equilibrate Buffer I (6 M urea, 2% SDS, 0.375 M Tris-HCl, pH

8.8, 20% glycerol and 130 mM DTT) for 10-15 min and in Equilibrate Buffer II (6 M urea, 2% SDS, 0.375 M Tris-HCl, pH 8.8, 20% glycerol and 135 mM iodoacetamide) for

10-15 min. The equilibrated strip was applied to a polyacrylamid/PDA SDS gel to run the

second dimension electrophoresis at 100 V for 120-180 min with Mini-PROTEAN 3 Cell

(Bio-Rad). The protein spots were developed using silver staining kit (Amersham). The two-dimensional protein electrophoresis maps were analyzed by using Phoretix 2D software (Nonlinear Dynamics Ltd, Newcastle upon Tyne, UK).

4.3 Results and Discussion

4.3.1 Fermentation Kinetics

Typical fed-batch fermentation kinetics with C. tyrobutyricum wild type and

PPTA-Em mutant grown on glucose and xylose are shown in Figures 4.1 and 4.2,

respectively. In general, there was a short lag phase with little gas (H2 and CO2) and acid

(acetic acid and butyric acid) production, which then increased during the exponential phase. For the wild type, acid production either stopped or slowed down dramatically when cells entered the stationary phase, although gas production and glucose (or xylose) consumption continued for a substantial period. For the PPTA-Em mutant, acetic acid production also stopped in the stationary phase; however, butyric acid production continued to reach a much higher level. Consequently, the mutant produced much more butyric acid and reached a higher final butyric acid concentration when the fed-batch fermentations were finally stopped. The fermentations were stopped when the sugar substrate was no longer consumed by the cells due to inhibition by butyric acid (Zhu and

Yang, 2003).

It is clear that PPTA-Em is more tolerant to butyric acid inhibition, a result from the genetic manipulation of the pta gene in the mutant (Zhu, et al., 2005). The specific growth rates and product yields in these fermentations were estimated and are listed in

Tables 4.1 and 4.2. The effects of the mutation on C. tyrobutyricum and the butyric acid fermentation are further discussed in the following sections.

4.3.2 Effects on Cell Growth

The mutant’s specific growth rate was reduced by 36% (0.137 h-1 vs. 0.214 h-1) on glucose (Table 4.1) and 26% (0.086 h-1 vs. 0.116 h-1) on xylose (Table 4.2), respectively.

However, the cell biomass yields for the mutant appeared to be higher than those for the

wild type, although the difference was within the error range. The lower specific growth

rate for the mutant can be attributed to the metabolic burden on cells caused by possibly

less energy (ATP) generation in the sugar metabolism due to pta knock-out. This is

because that pyruvate oxidation to acetate generates more ATP than its oxidation to

butyrate. The lower specific growth rate and biomass yield with xylose as the substrate is

because that xylose fermentation gives lower energy efficiency as compared with glucose fermentation. The different specific growth rate and biomass yield from the mutant suggest that the carbon and energy fluxes have been redistributed in the metabolic pathways of the mutant, which also led to the significant changes in acids production.

100

4.3.3 Effects of Butyric and Acetic Acid Production

Compared to the wild type, the PPTA-Em mutant not only can tolerate and produce a much higher concentration of butyric acid, it also gives higher butyric acid yields from glucose and xylose. As can be seen in Tables 4.1 and 4.2, PPTA-Em produced 15% more butyrate (0.38 g/g vs. 0.33 g/g) from both glucose and xylsoe. The increased butyrate production in the mutant can be attributed to the reduced acetate production as well as reduced specific growth rate. The acetate yield in the mutant was reduced by 13.5% (0.058 g/g vs. 0.067 g/g) from glucose and 32% (0.045 g/g vs. 0.066 g/g) from xylose, respectively. Although the mutant still produced a significant amount of acetate, the inactivation of pta gene reduced acetic acid and increased butyric acid production because more substrates were directed toward the butyric acid formation pathway, as evidenced by the increased butyrate/acetate (B/A) ratio (from ~5.5 to ~8.9 g/g) in the fermentations. This metabolic shift was the direct result of pta knock-out mutation.

The final butyric acid concentrations produced from glucose and xylose by

PPTA-Em increased by 62% (from 22.9 g/L to 37.2 g/L) and 72% (from 19.4 g/L to 33.5 g/L), respectively. Our previous study has shown that the acetic acid-forming enzymes

(PTA and AK) are more sensitive to butyric acid inhibition than butyric acid-forming enzymes (PTB and BK) (Zhu and Yang, 2003). The mutant’s reduced sensitivity to butyric acid inhibition thus can be attributed to the reduced carbon flux through the PTA-

AK pathway. However, the final acetic acid concentration produced in the fermentation by PPTA-Em was not much affected even though the acetic acid yield from sugar was reduced significantly. This is because that there may be other enzymes or pathways 101

present in C. tyrobutyricum that can also produce acetate from pyruvate or acetyl CoA

(Zhu, et al., 2005).

4.3.4 Effects on Gas Production

As discussed before, both hydrogen and carbon dioxide were produced throughout the fermentation. The pta knock-out mutation did not appear to have any significant effect on gas production by the PPTA-Em mutant. The average hydrogen and carbon dioxide yields were found to be ~0.017 g/g and ~0.37 g/g, respectively, for both the mutant and the wild type using either glucose or xylose as the substrate.

4.3.5 SDS-PAGE and 2-DE Analysis of Protein Expression

The effects of pta knock-out and fermentation conditions on protein expression were first studied with SDS-PAGE. As can be seen in Figure 4.3, there are notable differences in the SDS-PAGE protein profiles obtained from the wild type and the mutant. For example, the level for the proteins with molecular weight (MW) of ~32 kDa was significantly lower for PPTA-Em mutant grown on glucose. Interestingly, this group of ~32 kDa proteins in the mutant was much higher when xylose was used as the growth substrate. This substrate effect, however, was not observed with the wild type. It is clear that a single gene knock out can cause rather complicated responses in protein expression by the cells due to gene and metabolic regulatory networks.

102

The effects on protein expression were further studied with two-dimensional protein electrophoresis (2DE). There were ~200 protein spots on the 2DE maps obtained for both the wild type and mutant PPTA-Em grown on glucose (Figure 4.4). A smaller number of protein spots were obtained for growth on xylose (Figure 4.5). These 2DE protein maps were analyzed with the Phoretix 2D AdvancedTM software, which compared

and identified proteins with changed expression levels after normalizing the different

intensities of the protein spots on these 2DE gels. As expected, the number of protein

spots and their intensities (expression levels) on the 2DE gels were different between the

wild type and the mutant. For example, in the region circulated on the gels (PI: 5.5–7.5,

MW: ~32 kDa) there are two more protein spots for the wild type than for the mutant.

These missing protein spots are very likely to include PTA and AK, but cannot be identified in this work due to lack of proteomic information for C. tyrobutyricum. Table

4.3 shows the results of 2DE analysis of proteins in this region. These proteins are identified in terms of their PI and MW, and their expression levels are compared based on the normalized protein expression volume. With the wild type grown on glucose as the reference, above 100% indicates up regulation while below 100% indicates down regulation of the protein expression in the mutant or growth on xylose. For growth on glucose, two proteins (spot #89, PI: 6.3, MW: 31.1; spot #93, PI: 6.77, MW: 33.6) found in the wild type were missing and at least one protein (spot #95, PI: 6.90 and MW: 33.55) highly expressed in the wild type (N Volume = 0.802) was dramatically down-regulated in the mutant. Similarly, for growth on xylose, the protein spot #93 (PI: 6.77, MW: 33.6) found in the wild type was also missing in the mutant. It is noted that protein expression is also significantly affected by the carbon source used in the fermentation.

103

4.3.6 Effects of Carbon Source

C. tyrobutyricum is one of a few bacteria that can use xylose to produce butyric acid. As observed in this work, both glucose and xylose can be efficiently used in butyric acid fermentation and there is no obvious preference for either glucose or xylose as the carbon source, although cell growth and biomass yield from xylose were significantly lower than those from glucose. The lower growth rate from xylose should not be a concern for immobilized cell fermentation using the FBB, which is discussed below.

4.3.7 Immobilized Cell Fermentation in FBB

Immobilized cell bioreactors have the potential in improving a fermentation process by increasing cell density, reactor productivity and final product concentration.

We have previously developed a fibrous-bed bioreactor (FBB) for immobilized-cell fermentations to produce several organic acids from biomass with significantly improved productivity, yield, and final product concentration (Huang, et al., 1998; Zhu and Yang,

2003; Silva and Yang, 1995). The FBB was thus applied to the PPTA-Em mutant in this work to further improve butyric acid production from glucose and xylose. As shown in

Figure 4.6, the FBB fermentations produced much more butyric acid from glucose and xylose than those obtained in free-cell fermentations. As cells adapted in the FBB became more tolerant to butyrate inhibition, they were able to produce butyric acid at a final concentration of ~50 g/L with a butyrate yield of ~0.45 g/g sugar (see Tables 4.1 and

4.2). The improvements were over 16.5% in butyrate yield and 34%−54% in the final butyrate concentration. The improved butyrate yield can be attributed to the reduced cell 104

growth in the immobilized cell fermentation, whereas the improved butyrate tolerance is the result of adaptation and possible physiology changes in the FBB environment (Zhu and Yang, 2003). SDS-PAGE protein expression profiles (see Figure 4.3) for cells from the FBB showed similar patterns to those in the free-cell fermentations, however.

4.4 Conclusion

The potential of using the C. tyrobutyricum mutant obtained from integrational

mutagenesis that selectively inactivated the chromosomal pta gene to produce butyrate

from glucose and xylose was studied. Compared with the wild type, butyric acid

production by this mutant is improved with higher butyrate yields, final product

concentrations, and product purity (B/A ratio). The butyric acid fermentation was further

improved by immobilizing the mutant in the fibrous-bed bioreactor to facilitate cell

adaptation to attain even higher product yields and concentrations. Both glucose and

xylose can be used to produce butyric acid with similar fermentation performance. It is

concluded that butyric acid production from glucose and xylose can be efficiently carried

out in the FBB using the metabolically engineered mutant of C. tyrobutyricum.

105

4.5 References

Green, E. M., Boynton, Z. L., Harris, L. M., and Rudolph, F. B., Papoutsakis, E. T., and Bennett, G. N. Genetic manipulation of acid formation pathways by gene inactivation in Clostridium acetobutylicum ATCC 824. Microbiology 1996, 142, 2079-2086.

Huang, Y. L., Mann, K., Novak, J. M., and Yang, S. T. Acetic acid production from fructose by Clostridium formicoaceticum immobilized in a fibrous-bed bioreactor. Biotechnol. Prog. 1998, 14, 800-806.

Michel-Savin, D., Marchal, R., and Vandecasteele, J. P. Control of the selectivity of butyric acid production and improvement of fermentation performance with Clostridium tyrobutyricum. Appl. Microbiol. Biotechnol. 1990a, 32, 387-392.

Willims, E. A., Coxhead, J. M., and Mathers, J. C. Anti-cancer effects of butyrate: use of micro-array technology to investigate mechanisms. Proc. Nutr. Soc. 2003, 62, 107-115.

Wu, Z., and Yang, S. T. Extractive fermentation for butyric acid production from glucose by Clostridium tyrobutyricum. Biotechnol. Bioeng. 2003, 82, 93-102.

Yang, S. T. Extractive fermentation using convoluted fibrous bed bioreactor, U. S. Patent No. 5563069. 1996.

Zhu, Y., Liu, X., and Yang, S. T. Construction and characterization of pta gene deleted mutant of Clostridium tyrobutyricm for enhanced butyric acid fermentation. Biotechnol. Bioeng. 2005, 90, 154-166.

Zhu, Y., Wu, Z., and Yang, S. T. Butyric acid production from acid hydrolysate of corn fiber by Clostridium tyrobutyricum in a fibrous-bed bioreactor. Proc. Biochem. 2002, 38, 657-666.

Zhu, Y., and Yang, S. T. Adaptation of Clostridium tyrobutyricum for enhanced tolerance to utyric acid in a fibrous-bed bioreactor. Biotechnol. Prog. 2003, 19, 365-372.

106

Wild type PPTA-Em

Free cell Free cell Immobilized cell

Cell Growth

Specific growth rate (h-1) 0.214 ± 0.044 0.137 ± 0.032 0.095 ± 0.036

Biomass yield (g/g) 0.109 ± 0.019 0.141 ± 0.024 0.087 ± 0.008

Acid Production

Butyric acid concentration (g/L) 22.90 ± 4.06 37.22 ± 4.80 49.85 ± 0.51

Butyric acid yield (g/g) 0.33 ± 0.01 0.38 ± 0.01 0.44 ± 0.01

Acetic acid concentration (g/L) 4.15 ± 0.59 4.19 ± 0.013 8.74 ± 0.63

Acetic acid yield (g/g) 0.067 ± 0.012 0.058 ± 0.004 0.081 ± 0.005

Butyrate/Acetate ratio (g/g) 5.52 8.88 5.70

Gas Production

H2 yield (g/g) 0.017 ± 0.002 0.018 ± 0.001 0.016 ± 0.001

CO2 yield (g/g) 0.360 ± 0.035 0.389 ± 0.027 0.388 ± 0.005

H2/CO2 ratio (mole/mole) 1.05 ± 0.03 1.05 ± 0.03 0.93 ± 0.02

Table 4.1 Comparison of fermentations of glucose by C. tyrobutyricum wild type and PPTA-Em at 37 oC, pH 6.0.

107

Wild type PPTA-Em

Free cell Free cell Immobilized cell

Cell Growth

Specific growth rate (h-1) 0.116 ± 0.009 0.086 ± 0.020 0.048 ± 0.006

Biomass yield (g/g) 0.095 ± 0.003 0.109 ± 0.013 0.069 ± 0.003

Acid Production

Butyric acid concentration (g/L) 19.42 ± 1.195 33.49 ± 2.89 51.546 ± 3.49

Butyric acid yield (g/g) 0.33 ± 0.02 0.38 ± 0.02 0.45 ± 0.02

Acetic acid concentration (g/L) 3.31 ± 0.01 3.89 ± 0.26 6.42 ± 1.60

Acetic acid yield (g/g) 0.066 ± 0.006 0.045 ± 0.001 0.045 ± 0.016

Butyrate/Acetate ratio (g/g) 5.87 8.61 8.02

Gas Production

H2 yield (g/g) 0.017 ± 0.001 0.017 ± 0.001 0.015 ± 0.001

CO2 yield (g/g) 0.365 ± 0.001 0.373 ± 0.031 0.348 ± 0.004

H2/CO2 ratio (mole/mole) 1.07 ± 0.02 1.05 ± 0.02 0.96 ± 0.01

Table 4.2 Comparison of fermentations of xylose by C. tyrobutyricum wild type and PPTA-Em at 37 oC, pH 6.0.

108

Glucose Xylose Protein Wild type PPTA-Em Wild type PPTA-Em

MW N. Volume R Volume R Volume (%) R Volume (%) R Volume (%) PI (kD) (spot #) (%) (spot #) (spot #) (spot #) 5.89 35.51 0.166 (87) 100 221.7 (86) 0 0

5.99 34.26 0.090 (88) 100 482.2 (87) 157.8 (100) 211.1 (98)

6.17 34.02 0.261 (90) 100 167.4 (88) 386.6 (101) 243.7(99)

6.32 31.10 0.302 (89) 100 0 184.4 (105) 42.7 (102)

6.49 33.86 0.727 (92) 100 77.0 (91) 139.3 (106) 132.7 (100)

6.65 34.02 0.347 (91) 100 34.0 (90) 213.8 (107) 24.8 (105)

6.77 33.63 0.343 (93) 100 0 147.8 (103) 0

6.90 33.55 0.802 (95) 100 32.3 (92) 52.1 (104) 38.9 (103)

7.19 33.63 0.051 (94) 100 27.5 (89) 0 0

1) Only parts of proteins around 32 kD are listed here; 2) The protein number is the same number on the two-dimensional protein electrophoresis gels shown in Figures 4.3 and 4.4. 3) The expression data reported here are the average values of two gels.

Table 4.3 Comparison of protein expression in the wild type and mutant of C. tyrobutyricum using either glucose or xylose as the carbon source.

109

35 25

Wild Type, Free cells 30 Glucose 20 25

15 20 OD 15 Glucose Acetate 10 Butyrate

10 H2 (L) Production Gas

Concentration (g/L); OD (g/L); Concentration CO2 5 5

0 0 0 1020304050607080 A Time (h)

50 40

PPTA-Em, Free cells Glucose 40 30

20 OD Glucose 20 Acetate Butyrate H2 (L) Production Gas 10 Concentration (g/L); OD 10 CO2

0 0 0 102030405060708090100 B Time (h)

Figure 4.1 Kinetics of fed-batch fermentations of glucose by free cells of C. o tyrobutyricum wild type (A) and mutant PPTA-Em (B) at pH 6.0 and 37 C. OD600 (×), glucose (■), butyrate (○), acetate (∆), hydrogen (●), and carbon dioxide (▲).

110

35 25 Wild Type, Free cells 30 Xylose 20 25 Xylos e Acetate 15 20 Butyrate OD600 15 H2 10 CO2

10 (L) Production Gas

Concentration (g/L); OD 5 5

0 0 0 102030405060708090 A Time (h)

45 35

40 PPTA-Em, Free cells 30 Xylose 35 25 30 OD 25 Xylos e 20 Acetate Butyrate 20 15 H2 15 CO2

10 (L) Production Gas

Concentration (g/L); OD 10 5 5

0 0 0 102030405060708090100 B Time (h)

Figure 4.2 Kinetics of fed-batch fermentations of xylose by free cells of C. o tyrobutyricum wild type (A) and mutant PPTA-Em (B) at pH 6.0 and 37 C. OD600 (×), glucose (■), butyrate (○), acetate (∆), hydrogen (●), and carbon dioxide (▲).

111

1 2 3 4 5 6 7 kDa

Figure 4.3 SDS-PAGE of cellular proteins from C. tyrobutyricum. (Lane 1: wild type, free cell with glucose; Lane 2: wild type, free cell with xylose; Lane 3: PPTA-Em, free cell with glucose; Lane 4: PPTA-Em, free cell with xylose; Lane 5: PPTA-Em, Immobilized cell with glucose; Lane 6: PPTA-Em, Immobilized cell with xylose; Lane 7: Molecular size marker).

112

KDa

97 66

14 A KDa

97 66

14 B PI 4.5 5.1 5.5 5.9 6.6 7.0 8.5

Figure 4.4 Two dimensional protein electrophoresis for wild type (A) and mutant (B) grown on glucose.

113

KDa

97 66

14 A KDa

97 66

14 B PI 4.5 5.1 5.5 5.9 6.6 7.0 8.5

Figure 4.5 Two dimensional protein electrophoresis for wild type (A) and mutant (B) grown on xylose.

114

70 70 PPTA-Em, FBB 60 Glucose 60

50 50

40 OD 40 Glucose Acetate 30 30 Butyrate H2

20 CO2 20 (L) Production Gas Concentration (g/L); OD

10 10

0 0 0 20 40 60 80 100 120 140 160 180 200 220 240 260 A Time (h)

70 70 PPTA-Em, FBB 60 Xylose 60

50 50

40 40 OD xylos e 30 Acetate 30 Butyrate H2

20 20 (L) Production Gas CO2

Concentration (g/L); OD 10 10

0 0 0 20 40 60 80 100 120 140 160 180 200 220 240

B Time (h)

Figure 4.6 Kinetics of fed-batch fermentations by C. tyrobutyricum mutant PPTA- Em immobilized in the FBB with glucose (A) and xylose (B) as the substrate, at pH o 6.0 and 37 C. OD600 (×), glucose (■), butyrate (○), acetate (∆), hydrogen (●), and carbon dioxide (▲).

115

CHAPTER 5

CONSTRUCTION AND CHARACTERIZATION OF ACK DELETED MUTANT OF CLOSTRIDIUM TYROBUTYRICUM

Summary

Clostridium tyrobutyricum ATCC 25755 is an acidogenic bacterium, producing butyrate, acetate, H2 and CO2 as its main fermentation products. In this work, one

metabolically engineered mutant PAK-Em with inactivated ack gene, encoding acetate

kinase (AK) which is associated with acetate formation, was created by integrational

mutagenesis to improve butyric acid and hydrogen production from glucose and xylose.

The non-replicative plasmid pAK-Em containing the acetate kinase gene (ack) fragment

was introduced into C. tyrobutyricum by electroporation to inactivate the target ack gene

through homologous recombination, producing the ack-deleted mutant (PAK-Em). The

AK activity in PAK-Em decreased by ~50%; meanwhile, the PTA and hydrogenase

activities each increased by ~40%. The SDS-PAGE results indicated that protein

expression was changed in the mutant. The noncompetitive butyric acid inhibition model

demonstrated that the mutant had a much higher butyrate tolerance than the wild type.

Fed-batch fermentations with free cells of the mutant were carried out at pH 6.0 and 37oC

to evaluate its ability to produce butyric acid and hydrogen. Compared with the wild type, 116

the specific growth rate of the mutant decreased by 42%, from 0.24 h-1 to 0.14 h-1, due to the impaired PTA-AK pathway. Butyric acid production by the mutant was improved greatly, with a yield of 0.42 g/g and a concentration of 41.65 g/L. Also, the mutant produced a high yield of hydrogen (0.024 of g/g) with H2/CO2 ratio of 1.44. The final

butyric acid concentration increased to 50.11 g/L, with a yield of 0.45 g/g fromimmobilized-cell fermentation using a fibrous-bed bioreactor (FBB). Metabolic shift

analysis with glucose and xylose showed that both acetic acid and lactic acid were

eliminated from xylose fermentation by PTA-Em mutant at pH 5.0. This result suggested

that the acetic acid and lactic acid pathways were blocked in the PAK-Em mutant, so

butyric acid was produced as the only product from xylose at pH 5.0.

117

5.1 Introduction

Butyric acid and hydrogen can be produced by Clostridium tyrobutyricum as main fermentation products. Due to its wide applications in food and pharmaceutical industries, butyric acid production from natural resources has become an increasingly attractive alternative to petroleum-based chemical synthesis (Zigova, et al. 1999; Vandak et al.

1997; Williams et al., 2003). Hydrogen, with high energy content per unit weight (141.86 kJ/g or 61,000 Btu/lb), is considered one of the most promising future fuels (Dunn, 2002).

Recently, C. tyrobutyricum cells immobilized in a fibrous-bed bioreactor were successfully used for butyrate fermentation with increased reactor productivity and final product concentration (Zhu et al., 2002; Wu and Yang, 2003). However, this process still produces a high level of acetic acid as a byproduct and is not yet economically competitive. To improve the economics of the fermentation process, it is desirable to increase butyrate production while reducing acetate production, which also reduces the product separation cost. Complete selectivity for butyrate production is possible in glucose-limited fed-batch culture (Michel-Savin, et al., 1990ab), but the reactor productivity and final product concentration were not high enough for economical production. Hydrogen can be generated through catalytic fuel reforming and water electrolyzing, but these methods are energy intensive and expensive. Biological production of hydrogen, such as photosynthesis with algae and photosynthetic bacteria or by fermentation with anaerobic bacteria, has been developed for its advantage of easy operation at ambient temperature and pressure (Momirlan and Veziroglu, 1999; Das and

Veziroglu, 2001). It is possible to produce hydrogen by anaerobic fermentation from low- 118

cost renewable biomass using C. tyrobutyricum, which can add value to the butyric acid fermentation process.

The metabolic pathways for butyric acid production in C. tyrobutyricum have been studied (Liu, et al., 2005). In general, glucose (hexose) is catabolized via EMP

(Embden-Meyerhof-Parnas) pathway and xylose (pentose) is catabolized by HMP

(Hexose Monophosphae) pathway to pyruvate, which is then oxidized to acetyl-CoA and carbon dioxide with concomitant reduction of ferredoxin (Fd) to FdH2. FdH2 is then

oxidized by hydrogenase to Fd, producing hydrogen, releasing the excess electron and

converting NAD+ to NADH (Papousakis and Meyer, 1985). Acetyl-CoA is the key

metabolic intermediate at the node dividing the acetate-forming branch, catalyzed by

phosphotransacetylase (PTA) and acetate kinase (AK), from the butyrate-forming branch,

catalyzed by phosphotransbutyrylase (PTB) and butyrate kinase (BK). The acetate can

also be produced by CoA transferase from acetyl-CoA. Integrational mutagenesis has

been successfully applied to create metabolically engineered mutants of Clostridial

strains (Green et al., 1996; Zhu, et al., 2005). It can be used to selectively inactivate

undesired genes from the host chromosome with non-replicative integrational plasmid.

Previous studies show that the inactivation of the chromosomal pta gene which encodes

PTA, using integrational mutagenesis resulted in reduced PTA and AK activities and

acetate production. A fibrous-bed bioreactor (FBB) has been developed to immobilize C.

tyrobutyricum and improve butyric acid production from fermentation. Through

adaptation in FBB, the butyric acid concentration and yield can be improved greatly. It is

found that the butyrate-forming enzyme activities of the adapted cells were increased and

that the tolerance to butyric acid was significantly improved (Zhu and Yang, 2003).

119

It is well known that the medium pH can change final product yield and purity by inducing metabolic shift. It has been reported that C. acetobutylicum is able to direct its metabolic pathway from acid to solvent production associated with some physiological changes when the pH falls below a critical point. A similar metabolic pathway shift from xylose by C. tyrobutyricum has been reported at pH 5.0 (Zhu and Yang, 2004). Butyric acid fermentation at different pH values by wild type C. tyrobutyricum using FBB has been studied and the results show that changing the medium pH induced a metabolic shift from predominant butyrate production at pH 6.0 to predominant lactate and acetate production at pH 5.0 (Zhu and Yang, 2004). It is thus possible to understand the metabolic pathway through metabolic shift studies with different pH values.

The main objectives of this study were to improve butyric acid and hydrogen production by constructing a metabolically engineered C. tyrobutyricum mutant. The ack gene fragment was cloned, characterized, and used to construct the integrational plasmid.

The gene inactivation approach was applied to modify the acetate formation pathway by inactivating the ack gene and to evaluate its effect on butyric acid fermentation. The enzyme activities and protein expressions of the mutant were examined to understand how well the gene manipulation worked. The effects of the mutation on butyric acid and hydrogen fermentation were studied and are discussed. Finally, metabolic shift analysis was performed to better understand the pathway changes caused by gene manipulation in this paper.

120

5.2 Material and Methods

5.2.1 Culture and Media

Wild type C. tyrobutyricum ATCC 25755 was cultured anaerobically at 37°C in a

previously described synthetic clostridial growth medium (CGM) (Huang et al., 1998).

Colonies were maintained on Reinforced Clostridial Medium (RCM; Difco) plates in an

anaerobic chamber. The stock culture was maintained in serum bottles under anaerobic

conditions at 4°C. The C. tyrobutyricum mutant was selected by supplying with 40 µg/ml

of erythromycin (Em). E. coli was grown on Luria-Bertani (LB) medium supplemented

with ampicillin (100 µg/ml) aerobically at 37°C.

5.2.2 Mutant Development

The isolation of plasmid DNA from E. coli was undertaken using QIAprep

Miniprep plasmid purification kit (Qiagen, Valencia, CA), chromosomal DNA from C. tyrobutyricum was prepared using QIAGEN genomic DNA kit, and DNA was purified from gel using QIAquick gel extraction kit. Restriction enzymes, T4 ligase, and shrimp alkaline phosphatase were used in accordance with the supplier’s instruction (Amersham,

Piscataway, NJ).

The amplification of the partial ack sequence from C. tyrobutyricum genomic

DNA and the construction of non-replicative plasmid was performed according to previous work (Boynton et al., 1996). Synthetic oligonucleotides were designed as degenerate primers using the codon usage preference for C. tyrobutyricum

121

(http://www.kazusa.or.jp/codon) following previous study (Boynton et al., 1996). The sequences of the PCR primers for ack gene amplification were 5’– GAT AC(A/T)

GC(A/T) TT(C/T) CA(C/T) CA(A/G) AC –3’ (forward) and 5’– (G/C)(A/T)(A/G)

TT(C/T) TC(A/T) CC(A/T) AT(A/T) CC(A/T) CC –3’ (reverse). The partial ack fragment was amplified using wild type C. tyrobutyricum chromosomal DNA as template.

The PCR amplification was performed in a DNA engine (MJ Research, Reno, NV) with a previously developed PCR buffer (Zhu et al., 2005). Thermal cycling was initiated with a denaturation step (94oC for 3 min) followed by a 40 cycle program with template

denaturation (94oC for 50 s), primers annealing (42oC for 50 s), and extension (72oC for 1

min). At the end of the program, the deoxyadenosine (A) was added to the 3’ end of the

PCR products at 72oC for 10 min. The 560 bp PCR product was cloned into PCR vector

pCR 2.1 (3.9 kb) using TA cloning kit (Invitrogen), and the produced plasmid pCR-PAK

(4.5 kb) was used to determine the ack gene fragment DNA sequence.

As shown in Figure 5.1, a 1.6-kb HindIII fragment containing the Emr cassette was obtained from pDG 647 as antibiotic gene for mutant selection (Guérout-Fleury et al.,

1995). The size of pCR-PAK (4.5 kb) was reduced by removing a 1.5-kb fragment with

SphI digestion and the left vector backbone was relegated to form plasmid pCR-PAK1

(3.0 kb). The HindIII digested plasmid pCR-PAK1 was ligated with the HindIII ended

Emr cassette, producing the non-replicative integrational plasmid pAK-Em (4.6 kb) to be

used to inactivate the ack gene in C. tyrobutyricum.

The transformation of E. coli was performed according to the manufacturer’s

instruction of T-A cloning kit (Invitrogen, Carlsbad, CA). The competent cells of C.

tyrobutyricum were prepared following previous study (Zhu, et al., 2005). The protoplast

122

was prepared to examine the presence of the restriction system in C. tyrobutyricum on pAK-Em before transformation (Richards, 1988). To prepare the competent cell, the 40 mM DL-threnine was added into the CGM medium and helped to weaken the cell wall of gram-positive microorganism. The optimized electroporation SMP buffer (270 mM sucrose, 1 mM MgCl2, 7 mM sodium phosphate, pH 7.4) was used to wash the harvested

cells. The transformation of the integrational plasmid into C. tyrobutyricum was carried

out using a Bio-Rad Gene pulser (Model II) in an anaerobic chamber equipped with an

incubator and a centrifuge. The optimized gene pulser parameters, 2.5 kV, 600 Ω and 25

µF, were applied in electroporation. As a negative control, a non-replicative plasmid with

a Emr cassette but without the ack fragment was also used to transform the C.

tyrobutyricum cells. The positive mutants containing the non-replicative plasmids were

selected by 40 µg/mL of Em on RCM plates.

5.2.3 Enzyme Assay and SDS-PAGE

The bacterial cells were cultivated in serum bottles with 100 mL of CGM at 37°C,

and harvested by centrifuge after having grown to the exponential phase (OD600 = ~1.5).

The cell pellets were washed, suspended in 10 mL of 25 mM Tris-HCl buffer (pH 7.4), and broken by sonication. The protein extract was then collected for protein electrophoresis and PTA and AK activity assays under ambient conditions. However, the protein sample preparation and enzyme assay was carried out in the anaerobic chamber for hydrogenase activity assay. The cultured cells were suspended in 1 mL of TE buffer

(10 mM Tris-HCl, 1 mM EDTA, pH 8.0), lysed at 37oC for 30 min with mutanolysin

123

(100 µg/mL; Sigma), and centrifuged to remove the cell debris. Standard Bradford protocol (Bio-Rad, Hercules, CA) was used to measure protein content in the cell extract sample.

The PTA activity was measured by monitoring the liberation of CoA from acetyl-

CoA 405 nm using spectrophotometer (Andersch, et al., 1983). One unit of PTA activity was defined as the amount of enzyme converting 1 µmol of acetyl-CoA per minute. The activity of AK was assayed by the method of Rose using potassium acetate as the substrate (Rose, 1955). One unit of AK was defined as the amount of enzyme producing

1 µmol of hydroxamic acid per minute. Hydrogenase activity was detected using the procedure developed by Drake (Drake, 1982). One unit of hydrogenase activity is defined as 2 µmol of methyl viologen reduced (equivalent to 1 µmol of H2 oxidized) per minute.

Specific enzyme activity was calculated as the units of activity per mg of protein. The

relative enzyme activity (%) was reported by comparing the specific enzyme activities in

the mutant with the corresponding specific enzyme activities of the wild type.

Protein samples for SDS-PAGE electrophoresis were prepared following standard

protocol (Bio-Rad). Total protein samples (24 µg each) were loaded into wells and 12.5%

SDS-PAGE gel was run at 100 V for 2.5 h with PROTEAN II xi Cell (Bio-Rad). The gel

was stained and distained following the manufacturer instruction.

5.2.4 Butyric Acid Tolerance Study

To evaluate the inhibition effect of butyrate on cell growth, various concentrations of butyrate (0 – 15 g/L) were applied to C. tyrobutyricum culture in serum tubes

124

containing 10 mL of media. Cell growth was monitored by measuring the optical density at 600 nm (OD600) with a spectrophotometer (Sequoia-Turner, Model 340), and the

specific cell growth rates were estimated from the OD600 data.

5.2.5 Fermentation

Fed-batch fermentations of C. tyrobutyricum were performed in a 5-L stirred-tank

fermentor (Marubishi MD-300) containing 2 L of the CGM medium and substrate agitated at 150 rpm with pH and temperature controls. For the repeated-batch fermentations with immobilized cells, a 0.5-L fibrous-bed bioreactor (FBB) with a working volume of ~500 mL was made of a glass column packed with spiral wound cotton towel, as described elsewhere (Silva and Yang, 1995). The anaerobiosis was achieved by sparging the fermentor medium and the FBB with N2. The 100 mL of cell

suspension prepared in serum bottle was inoculated into the fermentor. After 3 days of growth, the cells with medium in the fermentor were circulated through the fibrous bed

and immobilized in the fibrous matrix. Most of the cells were immobilized, and there was

no change in cell density in the fermentor broth after 2 days; the spent medium was then

replaced with fresh medium. The fed-batch mode was operated by pulse feeding

concentrated substrate solution when the sugar level in the fermentation broth was close

to zero. The Micro-oxymax system (Columbus Instrument, Columbus, OH) connected to the fermentor was used to measure the production of H2 and CO2. Samples were taken at

regular intervals from the fermentor for the analyses of cell density, substrate, and

product concentration.

125

At the end of the FBB fermentation study, cells immobilized in the FBB were washed off from the fibrous matrix, collected and maintained as suspended cultures in the serum bottles with glucose as the carbon source. The fresh culture of the adapted cells was spread on the RCM plates with 40 µg/ml of Em and isolated colonies with good growth were chosen and purified on RCM plates. Theses purified colonies were cultivated with 15 mL CGM medium supplied with glucose as substrate. Samples were taken from the 15 mL culture for product analysis and the cell density was measured to monitor the bacterial growth. The purified culture of the adapted mutant was stored in -

80oC freezer.

5.3 Results and Discussions

5.3.1 Cloning and Mutant Characterization

The partial ack gene fragment in C. tyrobutyricum ATCC 25755 was successfully

amplified by degenerate primers using homologous alignment. The Mg2+ concentration in

the PCR buffer and the primer annealing temperature were optimized to be 2.5 mM and

42°C, respectively. The low primer annealing temperature resulted in multiple PCR

products. The PCR product with expected size of ~560 base nucleotides was cut and

purified from gel and cloned to T-A cloning vector pCR2.1, which was used to sequence the ack fragment cloned from wild type chromosomal DNA. The partial DNA sequence of ack gene in C. tyrobutyricum published in GenBank (GBAN AY706093) had 560

nucleotides encoding for 188 amino acids. The deduced partial amino acid sequence of

126

AK was then compared with the known sequences of complete AK from other microorganisms by homologous alignment. As shown in Figure 5.2, high degrees of identities of amino acid sequences were found between the partial AK of Clostridium tyrobutyricum and the AK of Escherichia coli (52%), Methanosarcina thermophila

(53%), Bacillus subtilis (48%), Mycoplasma genitalium (40%), Haemophilus influenzae

(47%) and Clostridium acetobutylicum (44%). The non-replicative integrational plasmid

(pAK-Em) containing the cloned ack fragment and antibiotics selection marker (Em) was inserted into the chromosomal DNA of the parental bacterium via electroporation and inactivated the original ack gene through homologous recombination. Transformation conditions followed previous study (Zhu, et al., 2005), producing the ack inactivated mutant PAK-Em. It is found that the transformation efficiency with C. tyrobutyricum was very low, about 1 colony per µg pAK-Em. There was no mutant obtained in the control experiment using negative non-replicative plasmid, which suggested that plasmid pPAK-

Em was integrated into the chromosome of the metabolically engineered mutant.

5.3.2 Enzyme Assay

The wild type and mutant of C. tyrobutyricum cultured untile the exponential- phase were harvested, and the cell extracts were assayed. The specific enzyme activities for acetic acid forming enzymes (PTA and AK) and the hydrogen catalyzing enzyme hydrogenase in the PAK-Em mutant were measured and their relative activities as compared with those of the wild type are shown in Figure 5.3. It is noted that the AK activity was significantly decreased by 50% as compared with the wild type due to the

127

gene manipulation. The remaining AK enzyme activity in PAK-Em was probably from some enzymes that can used the same substrate. Meanwhile, the PTA activity was increased by 42% in the mutant, indicating that pta gene lies upstream from ack gene in the same operon. The increase of PTA activity was probably due to the positive control on protein expression by the accumulation of acetyl-CoA that resulted from the (partial) inactivation of the PTA-AK pathway. However, the hydrogenase enzyme activity was increased by 40% resulting in the high production of hydrogen by PAK-Em, which was unexpected and is discussed later in this paper.

5.3.3 Protein Expression

Protein expression was studied by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) to better understand the effect of ack gene disruption in the mutant cell. The SDS-PAGE gel clearly showed that the highly expressed protein in the wild type with molecular weight of ~32 kDa (likely the deleted acetate kinase, AK) diminished in PAK-Em mutant, and the expression of the protein with ~70 kDa molecular weight (very likely to be the hydrogenase) was higher in PAK-Em (Figure

5.4). The result was consistent with the protein expression in previous session, suggesting a reduction in acetic acid production and an improvement in hydrogen production in the

PAK-Em mutant.

128

5.3.4 Butyric Acid Tolerance

Final product inhibition is one of the key factors limiting the improvement of organic acid production by fermentation. To determine the butyric acid tolerance in the mutant, PAK-Em cells were grown as suspension cultures supplied with different initial butyric acid concentrations (0 ~ 15 g/L). The specific growth rate was defined as the relative growth rate of the culture added with butyric acid and the culture at zero initial butyrate concentration. As can be seen in Figure 5.5, the mutant retained more than 30% as compared to less than 10% in the wild type at 15 g/L of butyric acid, indicating that the mutant had a much higher tolerance to butyric acid with gene manipulation. It is found that the growth inhibition by butyric acid in C. tyrobutyricum followed noncompetitive inhibition kinetics with different inhibition rate constants KP, 1.59 g/L for

wild type and 6.66 g/L for mutant.

5.3.5 Butyric Acid and Hydrogen Fermentation

Figure 5.7 shows the kinetics of fed-batch fermentations at pH 6.0 and 37oC with wild type free cells (A), PAK-Em (B), and immobilized PAK-Em cells (C) using glucose.

Most of the butyric acid and hydrogen were produced during the exponential and short stationary phases, and the production kinetics of hydrogen and carbon dioxide were similar in the wild type. The fermentation period of PAK-Em was much longer than that of wild type. It is noted that butyric acid and hydrogen production did not ceased even the mutant cells reached death phase, and the hydrogen production rate was much faster than the production rate of carbon dioxide. The fibrous-bed bioreactor (FBB) with high cell

129

densities immobilized in the fibrous matrix was also used in this work to characterize the butyric acid and hydrogen production of the immobilized metabolically engineered mutant due to cell adaptation in the FBB (Zhu, et al., 2005). The immobilized cell fermentation using PAK-Em mutant converted more glucose to butyric acid and hydrogen.

The fermentation results in terms of cell growth and the production of acid and gas are summarized in Table 5.1. The maximum cell density of the free cell mutant

(OD600 = 4.42) was lower than that of the wild type (OD600 = 7.05), resulting in the

lowered final biomass yield (0.06 g/g) from glucose. The PAK-Em mutant grew slower

than the wild type from glucose with specific growth rate of 0.14 h-1 vs. 0.21 h-1. It was

reported that the ratio of butyrate to acetate (B/A ratio) increased with decreased cell

growth rate (Michel-Savin, et al., 1990). The different biomass yield and specific growth

rate of PAK-Em indicated that the carbon and energy flux were redistributed in its

metabolic pathways, which would result in significant changes in the production of

various fermentation products. In the FBB, the C. tyrobutyricum cells were attached to the matrix surfaces and entrapped as large cell clumps within the matrix, so the cell density in the fibrous bed was very high (>70 g/L) (Zhu, et al., 2002), which resulted that the cell biomass yield was greatly decreased in the fermentor broth, only 0.04 g/g. The low biomass yield indicated that more carbon and energy flowed into the metabolic pathway that produces butyric acid.

Butyric acid production by PAK-Em in free cell fermentation was greatly improved in terms of final product concentration and yield. The final concentration of butyric acid produced in the fed-batch fermentation using glucose was increased from

130

19.98 g/L by the wild type to 41.65 g/L by the mutant, suggesting that the butyric acid tolerance of PAK-Em was enhanced. As noted in Table 5.1, the butyric acid yield was increased by 23%, from 0.34 g/g for wild type to 0.42 g/g for PAK-Em. Furthermore, the

B/A ratio increased from 4.52 withwild type to 5.41 with PAK-Em, suggesting that the metabolic pathway in the mutant had been shifted to favor butyric acid production over acetic acid formation. It is obvious that the immobilized PAK-Em mutant produced a much higher concentration of butyric acid (50.11 g/L) than the free cell at 37 oC and pH

6.0. The butyric acid yield was also increased from 0.42 g/g by free cells to 0.447 g/g by

immobilized cells. Therefore, the butyric acid tolerance of the immobilized PAK-Em

cells was enhanced, due to the adaptation benefit of the FBB. The butyrate/acetate ratio

(B/A ratio) also increased from 5.41 for free-cell PAK-Em and to 5.99 for immobilized-

cell PAK-Em. The enhanced B/A ratio in the mutant indicated that the selectivity of

butyric acid over acetic acid was increased by the inactivation of the PTA-AK acetic acid

forming pathway.

It is found that hydrogen production was significantly increased by PAK-Em with

gene manipulation. The hydrogen yield of the PAK-Em mutant was improved from 0.016

g/g to 0.024 g/g from glucose, indicating that more electrons were transferred to H+, which produced more hydrogen and NADH. It is obvious that PAK-Em had a higher ability to produce butyric acid and hydrogen as compared with the wild type. As shown in Table 5.1, the production of hydrogen was similar to that of carbon dioxide in the wild type, which has a H2/CO2 ratio of about 1, but the H2/CO2 ratio was increased to 1.44 in

free-cell PAK-Em fermentation and 1.59 in immobilized-cell fermentation using FBB. As

shown by the enzyme assay and protein expression sessions, the hydrogenase activity and

131

expression were improved in the mutant, which led to the high production of hydrogen.

The high production of hydrogen by the immobilized PAK-Em cell was probably due to the cell adaptation under high pressure of hydrogen product.

5.3.6 Metabolic Shift Analysis

As shown in Table 5.2, the wild type C. tyrobutyricum produced 14.4 g/L of butyric acid with a yield of 0.365 g/g, 5.43 g/L acetic acid with a yield of 0.14 g/g, and no lactic acid from glucose at pH 5.0. With xylose as the sugar source at pH 5.0, the main product was 33.5 g/L of lactic acid and 25.5 g/L of acetic acid, while the concentration of butyric acid was very low (5.3 g/L). It is found that most of the carbon source and energy from butyric acid pathway using glucose was shifted to the lactic acid and acetic acid pathways in wild type C. tyrobutyricum at pH 5.0 (Zhu and Yang, 2004).

PAK-Em was applied to fermentation at pH 5.0 using both glucose and xylose to characterize the metabolic shift and metabolic pathway changes with ack gene inactivation. The fermentation kinetics of mutant PAK-Em are shown in Figure 5.7, and all the fermentation data are summarized in Table 5.2. PAK-Em produced butyric acid as its main product, with a concentration of 14.79 g/L and a yield of 0.424 g/g. Acetic acid was produced as byproduct, with a concentration of 2.19 g/L and a yield of 0.06 g/g, and no lactic acid was produced from glucose fermentation. This glucose fermentation result by PAK-Em was similar to that of the wild type except for the lowered acetic acid production. However, the fermentation using xylose by PAK-Em showed that the butyric acid was produced as the only product, with a concentration of 13.635 g/L and a yield of

132

0.44 g/g. There was no byproduct, acetic acid nor lactic acid, which was distinct from the wild type. The lactic acid and acetic acid formation metabolic pathways were blocked when using xylose as sugar source, indicating that there was no metabolic shift to lactic acid and acetic acid at pH 5.0 in PAK-Em. This result indirectly proved that the PTA-AK pathway was inactivated completely by the ack gene disruption. The main cost of butyric acid fermentation comes from the separation and purification during the downstream process. If the production of byproducts is inactivated, then the butyric acid separation cost would be significantly decreased. In addition, the fermentation was operated at pH

5.0, close to the pKa value of butyric acid (4.89), so most of the butyric acid product was free acid.

5.3.7 Effects of Gene Manipulation and Cell Adaptation

In this work, the engineered mutant PAK-Em with inactivated ack was created using insertional inactivation of the gene on the chromosome by homologous recombination with an integrational plasmid containing the partial DNA sequence of ack gene. The improved tolerance to butyrate inhibition and the reduced flux through the

PTA-AK pathway resulted in the high butyric acid concentration produced by PAK-Em.

However, acetic acid production by the PAK-Em mutant was not significantly affected in the fermentation at pH 6.0, although the mutant had a much lower AK activity and the

PTA-AK pathway should have been impaired. Apparently, in addition to PTA and AK, there are other enzymes in C. tyrobutyricum that can also produce acetate from acetyl-

CoA and perhaps other substrates as well, which has been discussed in a previous study

133

(Zhu, et al. 2005). But the metabolic shift study using glucose and xylose at pH 5.0 proved that the PTA-AK pathway and the lactic acid forming pathway were blocked by ack gene deletion.

The hydrogen production of the mutant was also increased by the enhanced hydrogenase activity and expression. This is the first time the hydrogen production of C. tyrobutyricum has been improved by gene manipulation. The energy efficiency was decreased due to reduced flux through the PTA-AK pathway and lactic acid formation pathway, which resulted in more hydrogen production due to energy balance. Therefore, the gene manipulation in the acetic acid formation metabolic pathway resulted in global metabolic flux changes, including the acetic acid formation, lactic acid formation, butyric acid formation and hydrogen producing metabolic pathway due to the redistribution of carbon and energy.

5.4 Conclusion

In this work, the ack gene fragment was cloned from C. tyrobutyricum and was

used to inactivate the parental gene to develop the metabolically engineered mutant strain

of C. tyrobutyricum for butyrate and hydrogen production from glucose. The improved

tolerance to butyrate inhibition and the reduced flux through the PTA-AK pathway

resulted in the high butyric acid concentration produced by the metabolically engineered

mutant, PAK-Em. The hydrogen production of PAK-Em was also increased by the

enhanced hydrogenase expression, thus producing more energy. Cell adaptation in FBB

was very important to improving the fermentation production of butyric acid and 134

hydrogen. All these results suggest that the PAK-Em has a metabolic pattern distinct from the wild type.

135

5.5 References

Andersch, W., Bahl, H., Gottschalk, G. Levels of enzymes involved in acetate, butyrate, acetone and butanol formation by Clostridium acetobutylicum. Eur. J. Appl. Microbiol. Biotechnol. 1983, 17, 327-332.

Das, D., and Veziroglu, T. N. Hydrogen production by biological processes: a survey of literature. Int. J. Hydrogen Energy. 2001, 26, 13-28.

Drake, H. L. Demonstration of hydrogenase in extracts of the homoacetate-fermenting bacterium Clostridium thermoaceticum. J. Bacteriol. 1982, 150, 702-9.

Dunn, S. Hydrogen futures: toward a sustainable energy system. Int. J. Hydrogen Energy. 2002, 27, 235-64.

Green, E. M., Boynton, Z. L., Harris, L. M., Rudolph, F. B., Papoutsakis, E.T., and Bennett, G.N. Genetic manipulation of acid formation pathways by gene inactivation in Clostridium acetobutylicum ATCC 824. Microbiology 1996, 142, 2079-2086.

Guérout-Fleury, A. M., Shazand, K., Frandsen, N., and Stragier, P. Antibiotic-resistance cassettes fro Bacillus subtilis. Gene 1995, 167, 335-336.

Huang, Y. L., Mann, K., Novak, J. M., and Yang, S. T. Acetic acid production from fructose by Clostridium formicoaceticum immobilized in a fibrous-bed bioreactor. Biotechnol. Prog. 1998, 14, 800-806.

Michel-Savin, D., Marchal, R., and Vandecasteele, J. P. Control of the selectivity of butyric acid production and improvement of fermentation performance with Clostridium tyrobutyricum. Appl. Microbiol. Biotechnol. 1990a, 32, 387-392.

Momirlan, M., and Veziroglu, T. Recent directions of world hydrogen production. Energy Rev. 1999, 3, 219-31. 136

Papousakis, E. T., and Meyer, C. L. Equations and calculations of product yields and preferred pathways for butanediol and mixed-acid fermentations. Biotechnol. Bioeng. 1985, 27, 50-56.

Richards, D. F., Linnett, P. E., Oultram, J. D., and Young, M., Restriction endonucleases in Clostridium pasteurianum ATCC 6013 and C. thermohydrosulfuricum DSM 568. J. Gen. Microbiol. 1988, 134, 3151-3157.

Rose, I. A. Acetate kinase of bacteria (acetokinase). Methods Enzymol. 1955, 1, 591-595.

Silva, E. M., and Yang, S. T. Kinetics and stability of a fibrous bed bioreactor for continous production of lactic from unsupplemented acid whey. J. Biotechnol. 1995, 41, 59-70.

Vandak, D., Zigova, J., Sturdik, E., and Schlosser, S. Evaluation of solvent and pH for extractive fermentation of butyric acid. Process Biochemistry. 1997, 32, 245-251.

Williams, E. A., Coxhead, J. M., and Mathers, J. C. Anti-cancer effects of butyrate: use of micro-array technology to investigate mechanisms. Proceed Nutrition Society. 2003, 62, 107-115.

Wu, Z., and Yang, S. T. Extractive fermentation for butyric acid production from glucose by Clostridium tyrobutyricum. Biotechnol. Bioeng. 2003, 82, 93-102.

Zhu, Y., Wu, Z., and Yang, S. T. Butyric acid production from acid hydrolysate of corn fiber by Clostridium tyrobutyricum in a fibrous-bed bioreactor. Process Biochem. 2002, 38, 657-666.

Zhu, Y., and Yang, S. T. Adaptation of Clostridium tyrobutyricum for enhanced tolerance to butyric acid in a fibrous-bed bioreactor. Biotechnol. Prog. 2003, 19, 365-372.

Zigova, J., Sturdik, E., Vandak, D., and Schlosser, S. Butyric acid production by Clostridium butyricum with integrated extraction and pertraction. Proc. Biochem. 1999, 34, 835-843.

137

Strains Wild type PAK-Em PAK-Em Free Free Immobilized Cell Growth µ (h-1) 0.21 ± 0.03 0.14 ± 0.007 0.14 ± 0.01

Biomass yield (g/g) 0.10 ± 0.01 0.06 ± 0.002 0.04 ± 0.08 Acid Production Butyric acid conc. (g/L) 19.98 ± 3.07 41.65 ± 0.63 50.11 ± 2.42

Butyric acid yield (g/g) 0.34 ± 0.02 0.42 ± 0.006 0.45 ± 0.02 Acetic acid conc. (g/L) 4.42 ± 0.55 7.75 ± 0.76 8.38 ± 0.04

Acetic acid yield (g/g) 0.07 ± 0.01 0.07 ± 0.01 0.08 ± 0.01 B/A ratio (g/g) 4.52 ± 0.85 5.41 ± 0.61 5.99 ± 0.55 Gas Production

H2 yield (g/g) 0.016 ± 0.001 0.024 ± 0.001 0.023 ± 0.004

CO2 yield (g/g) 0.32 ± 0.02 0.37 ± 0.02 0.34 ± 0.01

H2/CO2 ratio (mole/mole) 1.04 ± 0.001 1.44 ± 0.06 1.59 ± 0.01 Carbon Balance (%) 0.95 ± 0.04 0.95 ± 0.01 0.96 ± 0.02

Table 5.1 Comparison of fed-batch fermentations by C. tyrobutyricum wild type and ack deleted mutant at 37 oC and pH 6.0.

138

Wild type PAK-Em Strains Glucose Xylose Glucose Xylose

Cell Growth µ (h-1) NA 0.04 ± 0.006 0.08 ±0.004 0.05 ±0.01 Biomass yield (g/g) NA NA 0.11 ±0.03 0.04 ±0.003 Acid concentration Butyric acid (g/L) 14.4 5.3 14.79 ±0.99 13.635 ±1.223 Acetic acid (g/L) 5.43 25.5 2.19 ±1.30 0 Lactic acid (g/L) 0 33.5 0 0 Acid yield Butyric yield (g/g) 0.365 0.05 ± 0.01 0.42 ±0.03 0.44 ±0.01 Acetic yield (g/g) 0.14 0.43 ± 0.03 0.06 ±0.004 0 Lactic yield (g/g) 0 0.61 ± 0.03 0 0

Gas yield Hydrogen (g/g) NA NA 0.023±0.001 0.021±0.001 Carbon dioxide (g/g) NA NA 0.037±0.007 0.30±0.002

Metabolic shift Yes No

Table 5.2 Metabolic shift analysis by immobilized cell of wild type and PAK-Em mutant using fibrou-bed bioreactor from glucose and xylsoe at 37oC and pH 5.0.

139

EcoRI EcoRI ack pCR-AK (4.5kb) HindIII 0.56 kb SphI

r Em (1.6kb) HindIII HindIII DELETE

f1 pCR 2.1 pDG647 ColE1 4.3 kb 3.9kb

Kmr Apr

SphI

EcoRI EcoRI HindIII ack SphI

Emr pAK-Em 4.6kb Apr

ColE1 HindIII

Figure 5.1 Construction of integrational plasmid pAK-Em with one 0.56-kb ack fragment cloned from C. tyrobutyricum ATCC 25755. Some important restriction sites used in this work are indicated. Abbreviations: ack, partial cka gene; f1, f1 filamentous phage origin of replication with helper phage; Apr, ampicillin resistance gene; Emr, erythromycin resistance gene; ColE1, compatibility group origin of replication in E. coli.

140

Bsu 1 DTAFHQTMPEQSYLYSLPYEYYEKFGIRKYGFHGTSHKYVTERAAELLGRPLKDLRLISC Mes 1 DTAFHQTMPPYAYMYALPYDLYEKHGVRKYGFHGTSHKYVAERAALMLGKPAEETKIITC Cac 1 DTAFHQTIPDYAYMYAIPYEYYDKYKIRKYGFHGTSHKYVSRTAAEFIGKKVEDLKMVVC Cty 1 DTAFHQTMPAHAYRYALPKFLYTQHNVRRYGFHGTSHAYVSERGSELAG-SYKHGGWLTA Eco 1 DTAFHQTMPEESYLYALPYNLYKEHGIRRYGAHGTSHFYVTQEAAKMLNKPVEELNIITC Hae 1 DTAFHQTMPEEAFLYALPYSLYKEHGVRRYGAHGTSHYFVSREVAKYVGKPADQVNAIIC Myc 1 DTTFHTTIPRENYLYAVPENWEKNNLVRRYGFHGTSYKYINEFLEKKFN--KKPLNLIVC

Bsu 61 HLGNGASIAAVEGGKSIDTSMGFTPLAGVAMGTRSGNIDPALIPYIMEKTGQTADEVLNT Mes 61 HLGNGSSITAVEGGKSVETSMGFTPLEGLAMGTRCGSIDPAIVPFLMEKEGLTTREIDTL Cac 61 HMGNGASITAVENGKSVDTSMGFTPPGGLAMGARSGDMDPAVVTFLMDKLNINASEVNNL Cty 60 HLGNGSSTCAIWNGQSVDTSMGLTPLEGVVMGTRSGDVDPSIHSFLASNLGWDIYKIDKM Eco 61 HLGNGGSVSAIRNGKCVDTSMGLTPLEGLVMGTRSGDIDPAIIFHLHDTLGMSVDAINKL Hae 61 HLGNGGSVSVVRNGQCIDTSMGLTPLEGLVMGTRCGDIDPAIVFYLYKTLGMSMDQIEET Myc 59 HLGNGASVCAIKQGKSLNTSMGFTPLEGLIMGTRSGDIDPAIVSYIAEQQKLSCNDVVNE

Bsu 121 LNKKSGLLGISGFSSDLRDIVEATKEG-NERAETALEVFASRIHKYIGSYAARMSG--VD Mes 121 MNKKSGVLGVSGLSNDFRDLDEAASKG-NRKAELALEIFAYKVKKFIGEYSAVLNG--AD Cac 121 LNKKSGIEGLSGISSDMRDIKKGNYVDKDPKAMLAYSVFTYKIKQFIGSYTAVMNG--LD Cty 120 LNSESGLLGLSDLSNDMRTLIEASEQG-NEDATLAIEVFCYRLAKSLAALSCGLPR--ID Eco 121 LTKESGLLGLTEVTSDCRYVEDNY--ATKEDAKRAMDVYCHRLAKYIGAYTALMDG-RLD Hae 121 LVKKSGLLGLTEVTSDCRYAEDNYDDESKPETRRALNVYSYRLAKYIGAYMAVLGDDHLD Myc 119 LNKKSGMFAITGSS-DMRDIFDKPEIN-----DIAIKMYVNRVADYIAKYLNQLSG-EID

Bsu 178 AIIFTAGIGENS- Mes 178 AVVFTAGIGENS- Cac 179 CLVFTGGIGENS- Cty 177 GLFFTGGIGENS- Eco 178 AVVFTGGIGENA- Hae 181 AIAFTGGIGENS- Myc 172 SLVFTGGVGENAS

Figure 5.2 Alignment of partial amino acid sequences of AK from B. subtilis (Bsu; GBAN L17320), Methanosarcina thermophila (Mes; GBAN L23147), C. acetobutylicum (Cac; GBAN U38234), C. tyrobutyricum (Cty, GBAN AY706093), E. coli (Eco; GBAN M22956), H. influenzae (Hae; GBAN L45839), and Mycoplasma genitalium (Myc; GBAN L43967).

141

160

140

120

100

Relative activity (%) 40

0 PTA AK Hydrogenase

Figure 5.3 Relative activities of key enzymes in acetate and butyrate-forming pathways in the PPTA-Em mutant as compared with the wild type.

142

1 2 3 KDa

116 80 70

51.8

34.7

32 30

Figure 5.4 SDS polyacrylamide gel electrophoresis of cellular proteins from C. tyrobutyricum (Lane 1: Molecular Marker; Lane 2: Wild Type; Lane 3: PAK-Em mutant).

143

100

60 PAK-Em

40 Relative Growth Rate (%) Growth Relative 20 Wild Type

0 0 5 10 15 20 Butyrate Concentration (g/L)

Figure 5.5 Noncompetitive inhibition of butyric acid on cell growth of C. tyrobutyricum wild type (∆) and PAK-Em (□).

144

35 25 Wild Type, Free, Glucose 30 H60 20

15 20

OD Glucose 15 Acetate 10 Butyrate Series6 10 H2 (L) production Gas CO2 Acid concentration (g/L); OD 5 5

0 0 A 0 1020304050607080 Time (h)

50 60

PAK-Em, Free, Glucose pH 6.0 50 40

40 30

OD 30 Glucose 20 Acetate Butyrate 20 H2 (L) production Gas CO2 Concentration (g/L); OD 10 10

0 0 B 0 20 40 60 80 100 120 140 160 180 Time (h)

145

60 70 PAK-Em, FBB,Glucose pH 6.0 60 50

OD 50 Glucose 40 Acetate Butyrate 40 H2 30 CO2 30

20 Gas production (L) production Gas 20 Concentration (g/L); OD

10 10

0 0 C 0 20 40 60 80 100 120 140 160 180 200 220 Time (h)

Figure 5.6 Fermentation kinetics of the C. tyrobutyricum from glucose by free cell of wild type (A), free cell of PAK-Em (B), Immobilized cell of PAK-Em (C) at pH 6.0 o and 37 C. OD600 (×), sugar concentration (■), butyrate concentration (○), acetate concentration (∆), hydrogen (●) and carbon dioxide (▲).

146

20 70 PAK-Em, FBB, Glucose, pH 5.0 60

15 50 OD Acetate Butyrate 40 Glucose 10 H2 CO2 30

20 5 Acid concentration (g/L); OD 10 Gas production (L) and glucose (g/L)

0 0 A 0 20 40 60 80 100 120 140 160 180 200 Time (h)

20 40 PAK-Em, FBB, Xylose, pH 5.0

15 30

OD Butyrate Xylose 10 H2 20 CO2

5 10 Acid concentration (g/L); OD (g/L); concentration Acid Gas production (L) and Xylose (g/L) Xylose and (L) production Gas

0 0 B 0 20406080100120140 Time (h)

Figure 5.7 Fermentation kinetics of the C. tyrobutyricum from glucose by immobilized cell of PAK-Em using glucose (A) and xylsoe (B) at pH 5.0 and 37oC. OD600 (×), sugar concentration (■), butyrate concentration (○), acetate concentration (∆), hydrogen (●) and carbon dioxide (▲).

147

CHAPTER 6

BUTYRIC ACID AND HYDROGEN PRODUCTION BY PAK-EM MUTANT OF CLOSTRIDIUM TYROBUTYRICUM

Summary

Fed-batch fermentations with free cells of the metabolically engineered mutant

PAK-Em produced a high concentration of butyric acid and high yield of hydrogen at pH

6.0 and 37oC, including 41.65 g/L of butyric acid and 0.024 g/g of hydrogen using

glucose, 39.03 g/L of butyric acid and 0.024 g/g of hydrogen with xylose, and 36.26 g/L

of butyric acid and 0.23 g/g of hydrogen grew on low-value waste grape juice from wine

manufacture. A fibrous-bed bioreactor (FBB) was used to immobilize PAK-Em mutant cells to further improve fermentation production. The butyric acid production by the

immobilized mutant was increased to a concentration of 50.1 g/L and a yield of 0.45 g/g

from glucose, as compared to the free-cell fermentation by PAK-Em (concentration: 41.6

g/L, yield: 0.42 g/g). It was found that the carbon source affected butyric acid selectivity,

with a butyrate/acetate ratio of 6.2 g/g-xylose vs. 5.4 g/g-glucose in free-cell fermentation

and 8.0 g/g-xylose vs. 6.0 g/g-glucose in immobilized-cell fermentation. Through

adaptation in the FBB, a high butyric acid concentration of 80 g/L was obtained at pH 6.3

with PAK-Em. This concentration is the highest ever attained in butyric acid 148

fermentation to date. Because of reduced cell growth in the immobilized-cell fermentation, the butyric acid yield was also increased to 0.45 g/g. In addition, a new mutant HydEm that produced even more hydrogen, with yield of 0.04 g/g, was discovered from the FBB adaptation. Finally, the protein expression in PAK-Em changed with gene manipulation and different substrates in the SDS-PAGE experiment.

149

6.1 Introduction

Clostridium tyrobutyricum is a gram-positive, rod-shaped, spore-forming, obligate anaerobic bacterium capable of producing butyrate and acetate from a wide variety of carbohydrates. Historically, butyric acid fermentation in cheese (late blowing) caused by the outgrowth of Clostridial spores present in raw milk, most commonly originating from silage, can result in considerable product loss (Klijin et al., 1995). On the other hand, butyric acid has many applications in chemical, food, and pharmaceutical industries

(Zigova, et al. 1999; Vandak et al. 1997; Williams et al., 2003). It is used a pure acid to enhance butter-like notes in food flavors. The esters are used as additives for increasing fruit fragrance and as aromatic compounds for the production of perfumes. Butyric acid is one of the short-chain fatty acids generated by bacterial fermentation of dietary fibers in the colon, and has been shown to have anticancer effect and can be used as neutraceuticals or even as a drug to cure colo-rectal cancers (Williams et al., 2003).

The production of butyric acid from renewable resources has become an increasingly attractive alternative to the current petroleum-based chemical synthesis because of public concerns about the environmental pollution caused by the petrochemical industry and consumers’ preference for bio-based natural ingredients in foods, cosmetics, and pharmaceuticals. Recently, our group developed a fibrous-bed immobilized-cell bioreactor (FBB) for several organic acid fermentations with significantly improved productivity, yield, and product concentration (Lewis and Yang,

1992; Huang and Yang, 1998; Silva and Yang, 1995). With the FBB, a much higher final product concentration that is 2 to 3-fold higher than those previously attained can be 150

achieved, not only because of the higher cell density in the FBB, but also through the adaptation of the culture to become more tolerant of the fermentation product (Yang et al., 1996). It was also found that the butyrate-forming enzyme activities of the adapted cells were increased, resulting in significantly improved tolerance to butyric acid (Zhu and Yang, 2003). For economical production of butyric acid from biomass, it is desirable to further improve the final product concentration and productivity of fermentation process. It is also desirable to reduce acetate production in the fermentation in order to facilitate the separation and purification of the final product, butyric acid.

Hydrogen, with a high energy content per unit weight (141.86 kJ/g or 61,000

Btu/lb), is considered the most promising future fuel. Hydrogen can be generated through catalytic fuel reforming, electrolysis of water, photosynthesis with algae and photosynthetic bacteria, and fermentation with anaerobic bacteria (Momirlan and

Veziroglu, 1999; Das and Veziroglu, 2001). It is thus possible to produce hydrogen by anaerobic fermentation from low-cost renewable biomass using C. tyrobutyricum, which can add value to the butyric acid fermentation process.

Concerns about future scarcity, cost, and environmental impact of fossil fuel have stimulated interest in the exploitation of cheap renewable biomass (Lynd et al., 1999).

Plant biomass represents a useful and valuable resource. Plant biomass consist of lignin, cellulose, and hemicellulose. The cellulose and hemicellulose are heteropolymers of hexose and pentose sugars, with glucose and xylose as two major constituents. While fermentation has been widely used to produce various fuels and chemicals, little study has been done on using pentose in industrial fermentation processes. Some low-value wastes, such as grape juice from wine manufacture, contains rich sugar sources, including

151

glucose and fructose. Previous work has shown that both glucose and xylose can be fermented by wild type C. tyrobutyricum with similar butyric acid yield (Zhu, et al.,

2002), but the fructose fermentation ability has not been studied and data on gas production from these carbon sources by C. tyrobutyricum are not available.

Recently, our group developed an ack-deleted mutant, PAK-Em, by integrational mutagensis. This mutant showed reduced acetate kinase activity and improved, and improved butyric acid production as compared with the wild type C. tyrobutyricum (Liu et al., 2005; Zhu, et al., 2005). The objective of this work was to evaluate the potential to produce butyric acid and hydrogen from various sugar sources by C. tyrobutyricum mutant, PAK-Em. The kinetics of the glucose, xylose, and fructose fermentations using free cells of the mutant were studied to evaluate its fermentation ability from different carbon sources. The FBB was applied to immobilize PAK-Em cells to further improve butyric acid and hydrogen production from glucose and xylose fermentation. One adaptated mutant (HydEm) was screened from the FBB cotton fibrous matrix and the glucose fermentation kinetics was studied. The fermentation results using the engineered mutant and the fibrous-bed bioreactor were compared and discussed in this paper.

6.2 Material and Methods

6.2.1. Culture and Media

The acidogenic bacterium C. tyrobutyricum ATCC 25755 was cultivated at 37°C

in a previously described synthetic medium (CGM) (Huang, et al., 1998) with glucose,

152

xylose, or grape juice as the substrates. The C. tyrobutyricum colonies were maintained on Reinforced Clostridial Medium (RCM; Difco) plates in an anaerobic chamber. The

PAK-Em and HydEm mutants were selected and stored on RCM plates containing 40

µg/mL of antibiotics erythromycin (Em).

6.2.2 Fermentation Kinetic Studies

Fed-batch fermentations of C. tyrobutyricum were performed in a 5-L stirred-tank fermentor (Marubishi MD-300) containing 2 L of the CGM medium and substrate agitated at 150 rpm with pH and temperature controls. For the repeated fed-batch fermentation with immobilized cells, a fibrous-bed bioreactor (FBB) with a working volume of ~500 mL was made of a glass column packed with spiral wound cotton towel, as described elsewhere (Silva and Yang, 1995). The anaerobiosis was reached by sparging the fermentor medium and the FBB with N2. About 100 mL of cell suspension

prepared in the serum bottle was inoculated into the fermentor. After 3 days of growth,

the cells with medium in the fermentor were circulated through the fibrous bed and

immobilized in the fibrous matrix. Most of the cells were immobilized and there was no

change in cell density in the fermentor broth after 2 days; the spent medium was then replaced with fresh medium. The fed-batch mode was operated by pulse feeding concentrated substrate solution when the sugar level in the fermentation broth was close to zero. The Micro-oxymax system (Columbus Instrument, Columbus, OH) connected to the fermentor was used to measure the production of H2 and CO2. Samples were taken at

regular intervals from the fermentor for the analyses of cell density, substrate, and

153

product concentration. Three different carbon substrates, including glucose, xylose and grape juice, were studied. The fermentations were stopped when the sugar substrate was no longer consumed by the cells due to inhibition by butyric acid. The sugar content of the grape juice, measured with HPLC, was about 50% glucose and 50% fructose.

6.2.3 Adapted Mutant Screening

At the end of the FBB fermentation study, cells immobilized in the FBB were washed off from the fibrous matrix, collected, and maintained as suspended cultures in the serum bottles with glucose as the carbon source. The fresh culture of the adapted cells was spread on the RCM plates with 40 µg/ml of antibiotics Em for the engineered mutant. The isolated colonies with good growth were chosen and purified on RCM plates. These purified colonies were cultivated with 15 mL CGM medium supplied with carbon source. Samples were taken from the 15 mL of culture for product analysis and the cell density was measured to monitor the bacterial growth. The purified culture of the adapted mutant was stored in -80oC freezer.

6.2.4 Preparation of Cell Extract and SDS-PAGE

Bacterial cells were cultivated until the exponential phase (OD600 = ~1.5) in serum

bottle at 37°C. After being harvested and washed, the cell pellets were resuspended in 10

mL of 25 mM Tris-HCl buffer (pH 7.4) and sonicated to obtain the cell extract. The

protein extract was centrifuged to remove cell debris and concentrated using four

volumes of acetone (40 mL) to precipitate protein at -20oC overnight. The protein pellet 154

was re-dissolved in 2 mL of 25 mM Tris-HCl buffer (pH 7.4). The protein content in the cell extract sample was determined following standard Bradford protocol (Bio-Rad,

Hercules, CA). Protein samples, 24 µg per well, were loaded into 12.5% SDS-PAGE gel and run at 100 V for 2.5 h with PROTEAN II xi Cell (Bio-Rad). The gel was stained and distained following the instruction of the manufacturer.

6.2.5 Analytical Methods

Cell density was analyzed by measuring the optical density of the cell suspension at a wavelength of 600 nm (OD600) with a spectrophotometer (Sequoia-turner, model

340). One unit of OD600 corresponded to 0.68 g/L cell dry weight for cells grown in the xylose medium. A high performance liquid chromatography (HPLC) system was used to analyze the organic compounds, including xylose, lactate, butyrate, and acetate in the fermentation broth. Gas (H2 and CO2) production in the fermentation was monitored using an on-line respirometer Micro-oxymax system (Columbus Instrument, Columbus,

OH).

6.3 Results and Discussions

6.3.1 Free Cell Fermentation Using Different Sugar Sources

Glucose and xylose are two most abundant sugars in plant biomass, and fructose

is a fermentable sugar commonly found in corn steep liquor and many other food

155

processing wastes (Huang, et al., 1998). In this work, different sugar sources were used to study butyric acid and hydrogen production by PAK-Em.

6.3.1.1 Fermentation Kinetics

The kinetics of fed-batch fermentations by free cell of C. tyrobutyricum mutant

PPTA-Em grown on glucose, xylose and grape juice (containing glucose and fructose) are shown in Figure 6.1. The PAK-Em cells began growing shortly after inoculation and reached the stationary phaseby the end of the first fed-batch. The acid (acetic acid and butyric acid) and gas (H2 and CO2) production increased significantly during the

exponential phase. Acetic acid production stopped in the stationary phase; however,

butyric acid and gas production increased to a much higher level. Figure 1C shows that

the PAK-Em mutant consumed glucose and fructose at similarly paces and that it had no

preference for either of these two sugars.

6.3.1.2 Fermentation Kinetics

Table 6.1 compares the fermentation results using the mutant from glucose, xylose, and grape juice waste in terms of cell growth, biomass yield, and acid and gas production. The acid and gas fermentation production by wild type is also listed for comparison. As can be seen in Table 6.1, the mutant’s specific growth rate was reduced by 30% (0.14 h-1 vs. 0.21 h-1) on glucose and 16% (0.10 h-1 vs. 0.12 h-1) on xylose as

compared with wild type. The specific growth rate of the mutant on grape juice (0.17 h-1)

156

was higher than the growth rate on glucose and xylose, but it was still lower than that of the wild type on glucose. More ATP is generated from pyruvate oxidation to acetate than its oxidation to butyrate, so ack gene inactivation led to the lower specific growth rate for the mutant. The lower specific growth rate with xylose as the substrate is because xylose fermentation gives a lower energy efficiency than glucose and fructose fermentation.

6.3.1.3 Butyric Acid Production

The butyric acid concentration and yield by mutant PAK-Em on glucose, xylose, and grape juice were improved greatly as compared to the wild type. As can be seen in

Table 6.1, PAK-Em produced 23% more butyric acid from both glucose and xylose (0.42 g/g), and 15% more butyric acid from glucose/fructose mixture (0.39 g/g) than the wild type. The increased butyrate production in the mutant can be attributed to more carbon and flux to the butyrate formation pathway, the reduced specific growth rate, and the lowered biomass yield. Although the mutant still produced a significant amount of acetate grew on different sugars, inactivating the ack gene enhanced butyric acid selectivity and increased the butyrate/acetate (B/A) ratio by 19% - 25%, 5.4 g/g-glucose vs. 4.5 g/g for glucose, and 6.2 g/g vs. 4.9 g/g for xylose. It was reported that the B/A ratio increased with a decreased cell growth rate (Michel-Savin, et al., 1990a).

The final butyric acid concentrations produced in the fed-batch fermentations were also increased to 41.7 g/L from glucose, 39.0 g/L from xylose, and 36.3 g/L from glucose/fructose by PAK-Em, which were about 100% higher as compared to wild type

(butyric acid concentration of 19.9 g/L from glucose and 17.9 g/L from xylose). Previous

157

study has shown that the acetic acid-forming enzymes (phosphotransacetylase and acetate kinase) are more sensitive than the butyric acid-forming enzymes

(phosphotransbutyrylase and butyrate kinase) to butyric acid inhibition (Zhu and Yang,

2003). The mutant’s reduced sensitivity to butyric acid inhibition thus can be attributed to the reduced carbon flux through the PTA-AK pathway. However, the final acetic acid concentration produced in the fermentation by PAK-Em was not significantly affected because of other enzymes or pathways present in C. tyrobutyricum that can also produce acetate from pyruvate or acetyl CoA (Liu et al. 2005; Zhu, et al., 2005).

6.3.1.4 Hydrogen Production

Both hydrogen and carbon dioxide were produced throughout the fermentation.

The PAK-Em mutant’s hydrogen yield from glucose, xylose and flucose/fructose was improved to ~0.024 g/g from ~0.017 g/g for the wild type using (PAK-Em mutant). The mole ratio of hydrogen to carbon dioxide (H2/CO2) was also improved by about 40%

(~1.4 mole/mole) as compared with wild type (1.0 mole/mole). It is obvious that the ack- eliminated mutation in PAK-Em significantly changed hydrogen production.

6.3.1.5 Effect of Carbon Source

Previous work has shown that both glucose and xylose canbe fermented by wild type C. tyrobutyricum with similar butyric acid yield (Zhu, et al., 2002), but the fructose fermentation ability and gas production from glucose, xylose and fructose by C.

158

tyrobutyricum have not been studied. Fed-batch fermentations using glucose, xylose, and grape juice by free cells of PAK-Em mutant were carried out at pH 6.0 and 37oC. As

observed in this work, all the sugar sources can be efficiently used in butyric acid

fermentation and there is no obvious preference for either glucose, xylose or fructose as

the carbon source, although cell growth on xylose was significantly lower than those on

glucose and fructose. The lower growth rate from xylose should not be a concern for

immobilized cell fermentation using the FBB is discussed below.

6.3.2 Fermentation Using Fibrous-bed Bioreactor

The novel fibrous-bed bioreactor (FBB) with bacterial cells immobilized in the

fibrous matrix was developed for the production of various organic acids, including

lactic, propionic and butyric acids. It has been shown that the FBB has high active cell

densities (>70 g/L) of biomass and high viability (> 84%) immobilized in the fibrous

matrix (Zhu and Yang, 2003). In this work, the FBB bioreactor was used to immobilize

the PAK-Em mutant cells to further improve the fermentation product from glucose and

xylose due to these advantages.

6.3.2.1 Acid and Gas Production

The repeated fed-batch fermentation kinetics using immobilized PAK-Em on both

glucose and xylose at 37oC and pH 6.0 are shown in Figure 6.2, and the fermentation data

are summarized in Table 6.2. Most cells were attached to fibrous matrix surfaces and

159

entrapped as large cell clumps. Therefore, the biomass yield in fermentor broth was only

~0.04 g/g. The immobilized PAK-Em produced a higher concentration of butyric acid

(~50 g/L) from glucose and xylose than the free cell fermentation (~40 g/L). It is obvious that the butyric acid tolerance of immobilized PAK-Em was enhanced due to gradual cell adaptation in the fibrous matrix of the FBB bioreactor. The butyric acid yield also increased from 0.42 g/g by the free cells to 0.45 g/g by the immobilized cells on both glucose and xylose. As compared to the free-cell fermentation, the B/A ratio produced by immobilization fermentation using PAK-Em were improved, with 5.99 g/g vs. 5.41 g/g from glucose and 7.99 g/g vs. 6.19 g/g from xylose. Also the H2/CO2 ratio was increased

to 1.59 from glucose and 1.61 from xylose. Compared to the wild type, the butyric acid

concentration, butyrate yield, and B/A ratio from immobilized PAK-Em cells were also higher although the effect was not as large as in the free cell fermentation.

6.3.2.2 Effect of Cell Adaptation and Carbon Source

This work shows that the FBB increased the final product concentration and yield

as compared to conventional free-cell fermentations because of the adaptation of the

immobilized cells in the FBB, which was consistent with previous study (Zhu and Yang,

2003). It is found that the acetic acid concentration produced by PAK-Em from xylose

was 6.24 g/L with a yield of 0.06 g/g, which was much lower than that from glucose with

concentration of 8.38 g/L and yield of 0.08 g/g. Thus the butyric acid selectivity from

xylose was much higher than the selectivity using glucose as carbon source.

160

6.3.3 Fermentation by Mutant at Different pHs

Previous results show that C. tyrobutyricum can grow between pH 5.0 and pH 7.0, but a higher pH favors butyrate production and gave a higher butyrate yield and concentration (Zhu and Yang, 2004). The FBB can give good long-term performance for cell adaptation, and it is easier to optimize for fermentation conditions such as pH and nutrient supplement. In this study, different pH values (pH 5.0, 6.0, 6.3, and 7.0) were applied to the immobilized fermentation using FBB by PAK-Em to improve its fermentation production. Figure 6.3 describes the fermentation kinetics of pAK-Em using glucose as sugar source at 37oC, pH 5.0, pH 6.3 and 7.0. The effects of pH on growth

rate, biomass yield, acid concentration and yield, and gas production are shown in Table

6.3.

6.3.3.1 Acid Production

As shown in Figure 6.3, most butyric acid and hydrogen was produced after cells

grew to exponential phase and before the glucose concentration leveled off. The butyric

acid production was changed greatly by the fermentation media pH. The concentration of

butyric acid increased to 50.1 g/L at pH 6.0, 80.2 g/L at pH 6.3, and 61.5 g/L at pH 7.0

from 14.79 g/L at pH 5.0. As can be seen in Table 6.3, the butyric acid yield was similar

for all the pH values, about 0.44 g/g, which is close to the maximum butyric acid yield in

C. tyrobutyricum. The acetic acid concentration was still high because of the coenzyme

and other pathways, and the acetic acid production increased with increasing butyric acid production. The B/A ratio was also affected by the media pH, with 5.99 g/g at pH 6.0,

161

5.26 g/g at pH 6.3, and 11.46 g/g at pH 7.0 using glucose, but with high B/A ratio of

12.37 g/g at ph 5.0. High and low pH values were not the best conditions for high concentration of butyric acid production, but the B/A ratio at high and low pH were higher than those at other pH. It is obvious that pH 6.3 was favored for high concentration of butyric acid, but butyrate selectivity was decreased.

6.3.3.2 Gas Production

Table 6.3 shows that the H2/CO2 molar ratio was 1.53 at pH 5.0, 1.59 at pH 6.0,

1.85 at pH 6.3, and 2.02 at pH 7.0. It is obvious that the hydrogen production increased at

higher media pH. Probably, the solubility of CO2 increased with increasing fermentation

broth’s pH value.

6.3.3.3 Effect of pH on Fermentation

In this study, we obtained the highest butyric acid concentration (81.9 g/L) and

yield (0.45 g/g) by immobilized PAK-Em mutant in the FBB at pH 6.3 and 37oC. It is the

best result for butyric acid production from fermentation yet reported: the previous best

results were a concentration of 62.8 g/L with yield of 0.45 g/g from sucrose (Fayolle, et

al., 1990) and butyric acid concentration of 42.5 g/L with yield of 0.36 g/g from glucose

(Michel-Savin et al., 1990b) by C. tyrobutyricum.

162

6.3.4 Fermentation by the Adapted Mutant

It has been reported that culturing in the FBB facilitates the adaptation and selection of mutants (Huang and Yang, 1998). The immobilized PAK-Em cells in the

FBB were harvested after six FBB fermentations by washing off the cells from the fibrous matrix under high a pressure in the fermentor. The cells of the adapted mutant,

HydEm, were then used in free-cell fermentation with glucose at pH 6.0 and 37oC, and the results are shown in Table 6.1 and Figure 6.4. The growth rate of this mutant (0.21 h-

1) was much faster than that of the parent strain PAK-Em (0.14 h-1) and similar to that of

the wild type (0.21 h-1). Previous study of wild type C. tyrobutyricum in the FBB also

showed that the specific growth rate of the adapted culture was improved (Zhu and Yang,

2003). As shown in Figure 6.4, the maximum cell density in the free-cell fermentation of

HydEm (OD600 = 11.53), was much higher than that of PAK-Em (OD600 = 4.42) and even

higher than the wild type (OD600 = 7.05). The biomass yield of HydEm (0.14 g/g) was

also higher than that of the wild type (0.10 g/g) and PAK-Em (0.064 g/g) from glucose. It

is found that the gas production of this mutant was improved significantly, including

hydrogen yield of 0.04 g/g, and H2/CO2 ratio of 2.69. It is clear that the FBB adaptation

improved both butyric acid and hydrogen production of C. tyrobutyricum.

6.3.5 Protein Expression

To better understand the fermentation kinetics changes that resulted from

mutations generated by gene manipulation and sugar sources, protein expression was

studied by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

163

Figure 6.5 shows that the protein expressions of the wild type and PAK-Em using glucose and xylose were different.

6.3.5.1 The Protein with 32 KDa

Compared to the wild type, PAK-Em cells expressed less 32-kDa protein, especially for growth on glucose. This protein is likely the deleted acetate kinase (AK), although it is not possible to identify AK from SDS-PAGE gel due to limited proteomics information for C. tyrobutyricum and similar species. As compared with the growth on glucose, the PAK-Em cells grew on xylose had a higher expression level of this protein.

This result was consistent with previous work, which showed that the metabolic flux using xylose as the substrate was distinct from the flux using glucose as the carbon source. The expression of the protein with ~32 kDa was changed by gene manipulation, cell adaptation, and carbon source, resulting in the redistribution of carbon and energy flux.

6.3.5.2 The Protein with 70 KDa

The expression of the protein with ~70 kDa molecular mass was higher in PAK-

Em from both glucose and xylose than in wild type. This 70 kDa protein is probably the hydrogenase which has a molecular mass ~67 kDa in C. tyrobutyricum (Watrous, 2003).

Because more ATP was produced from the acetate-producing (PTA-AK) pathway than the butyrate-producing (PTB-BK) pathway, to compensate for the lost energy efficiency

164

due to reduced flux through the PTA-AK pathway, the PAK-Em mutant increased its butyrate production and hydrogen production. The increased hydrogenase expression not only resulted in more hydrogen production, but might also have increased the amount of

NADH and energy produced.

6.4 Conclusion

The mutant C. tyrobutyricum obtained using integrational mutagenesis to selectively inactivate ack gene was used to study the effects of different sugar sources and cell adaptation in the FBB. As compared with the wild type, butyric acid and

hydrogen production by PAK-Em was improved greatly, with higher final product

concentration and yield. This work demonstrated that butyric acid and hydrogen can be

produced from various sugar sources with high final concentration and yield. It should be

noted that the adapted culture from FBB was more resistant to butyrate inhibition and that

the butyric acid yield was improved further by applying FBB. Increasing the final butyric

acid concentration and hydrogen yield from fermentation using different sugar sources

should reduce the production cost for bio-based butyric acid and hydrogen and allow

bioproduction to compete more favorably in the marketplace.

165

6.5 References

Das, D., and Veziroglu, T. N. Hydrogen production by biological processes: a survey of literature. Int. J. Hydrogen Energy. 2001, 26, 13-28.

Huang, Y., and Yang, S. T. Acetate production from whey lactose using co-immobilized cells of homolactic and homoacetic bacteria in a fibrous-bed bioreactor. Biotechnol. Bioeng. 1998, 60, 499-507.

Huang, Y. L., Mann, K., Novak, J. M., and Yang, S. T. Acetic acid production from fructose by Clostridium formicoaceticum immobilized in a fibrous-bed bioreactor. Biotechnol. Prog. 1998, 14, 800-806.

Klijn, N., Bovie, C., Dommes, J., Hoolwerf, J. D., van der Walls, C. B., Weerkamp, A. H., and Nieuwenhof, F. F. J. Identification of Clostridium tyrobutyricum and related species using sugar fermentation, organic acid formation and DNA probes based on specific 16S sRNA sequences. System. Appl. Microbiol. 1994, 17, 249-256.

Klijin, N., Nieuwenhof, F. F. J., Hoolwerf, J. D., van Der Waals, C. B., and Weerkamp, A. H. Identification of Clostridium tyrobutyricum as the causative agent of late blowing in cheese by species-specific PCR amplification. Appl. Environmental Microbiol. 1995, 61, 2919-2924.

Lewis, V. P. and Yang, S. T. Continuous propionic acid fermentation by immobilized Propionibacterium acidipropionici in a novel packed-bed bioreactor. Biotechnol. Bioeng. 1992, 40, 465-474.

Liu, X., Zhu, Y., Yang, S.T. Butyric acid and hydrogen production by Clostridium tyrobutyricum ATCC 25755 and mutants. Enzyme Microbial. Technology. 2005, In press.

Lynd, L.R., Wyman, C. E., and Gerngross, T. U. Biocommodity engineering. Biotechnol. Prog. 1999, 15, 777-793.

Michel-Savin, D., Marchal, R., and Vandecasteele, J. P. Control of the selectivity of butyric acid production and improvement of fermentation performance with Clostridium tyrobutyricum. Appl. Microbiol. Biotechnol. 1990a, 32, 387-392.

166

Momirlan, M., and Veziroglu, T. Recent directions of world hydrogen production. Energy Rev. 1999, 3, 219-31.

Silva, E. M., and Yang, S. T. Kinetics and stability of a fibrous bed bioreactor for continous production of lactic from unsupplemented acid whey. J. Biotechnol. 1995, 41, 59-70.

Vandak, D., Zigova, J., Sturdik, E., and Schlosser, S. Evaluation of solvent and pH for extractive fermentation of butyric acid. Process Biochemistr. 1997, 32, 245-251.

Watrous, M. M., Clark, S., Kutty, R., Huang, S., Rudolph, F. B., Hughers, J. B., Bennett, G.N., 2003. 2,4,6-Trinitrotoluene reduction by an Fe-only hydrogenase in Clostridium acetobutylicum. Appl. Environ. Microbiol. 26, 542-547.

Williams, E. A., Coxhead, J. M., and Mathers, J. C. Anti-cancer effects of butyrate: use of micro-array technology to investigate mechanisms. Proceed. Nutrition Society. 2003, 62, 107-115.

Yang, S. T. Extractive fermentation using convoluted fibrous bed bioreactor. U. S. Patent No. 5563069. 1996.

Zhu, Y., Wu, Z., and Yang, S. T. Butyric acid production from acid hydrolysate of corn fiber by Clostridium tyrobutyricum in a fibrous-bed bioreactor. Process Biochem. 2002, 38, 657-666.

Zhu, Y., and Yang, S. T. Adaptation of Clostridium tyrobutyricum for enhanced tolerance to butyric acid in a fibrous-bed bioreactor. Biotechnol. Prog. 2003, 19, 365-372.

Zhu, Y., and Yang, S. T. Effect of pH on metabolic pathway shift in fermentation of xylose by Clotridium tyrobutyricum. J. Biotechnol. 2004, 110, 143-157.

Zigova, J., Sturdik, E., Vandak, D., and Schlosser, S. Butyric acid production by Clostridium butyricum with integrated extraction and pertraction. Proc. Biochem. 1999, 34, 835-843. 167

Strains Wild type PAK-Em HydEm Sugar Sources Glucose Xylose Glucose Xylose Grape Juice Glucose

Cell Growth

µ (h-1) 0.21 ± 0.03 0.12 ± 0.01 0.14 ± 0.007 0.10 ± 0.001 0.17 ± 0.05 0.21 ± 0.002 Biomass yield (g/g) 0.10 ± 0.01 0.09 ± 0.003 0.06 ± 0.002 0.06 ± 0.0001 0.075 0.14 ± 0.007 Acid Production Butyric acid conc. (g/L) 19.98 ± 3.07 17.86 ± 2.21 41.65 ± 0.63 39.03 ± 3.46 36.26 41.38 ± 0.24 Butyric acid yield (g/g) 0.34 ± 0.02 0.35±0.02 0.42 ± 0.006 0.42 ± 0.01 0.39 ± 0.02 0.40 ± 0.004

Acetic acid conc. (g/L) 4.42 ± 0.55 3.63 ± 0.04 7.75 ± 0.76 7.14 ± 0.97 7.34 10.45 ± 1.62

Acetic acid yield (g/g) 0.07 ± 0.01 0.06 ± 0.001 0.07 ± 0.007 0.08 ± 0.005 0.09 ± 0.01 0.11 ± 0.004 B/A ratio (g/g) 4.52 ± 0.85 4.93 ± 0.58 5.41 ± 0.61 6.19 ± 0.40 4.94 4.01 ± 0.64 Gas Production

H yield (g/g) 0.016 ± 0.001 0.018 ± 0.001 0.024 ± 0.001 0.024 ± 0.001 0.023 ± 0.002 0.04 ± 0.005 2 0.32 ± 0.02 0.33 ± 0.03 0.37 ± 0.002 0.37 ± 0.01 0.363 ± 0.026 0.344 ± 0.006 CO2 yield (g/g)

H2/CO2 ratio (mole/mole) 1.04 ± 0.001 1.05 ± 0.04 1.44 ± 0.06 1.44 ± 0.07 1.45 ± 0.03 2.69 ± 0.27 Carbon Balance 0.95 ± 0.04 0.92 ± 0.015 0.95 ± 0.01 0.95 ± 0.06 0.946 0.93 ± 0.02

Table 6.1 Comparison of fed-batch fermentation of C. tyrobutyricum wild type and mutant grew on different sugar sources at 37 oC and pH 6.0.

168

Strains Wild Type PAK-Em

Sugar Sources Glucose* Xylose* Glucose Xylose Cell Growth µ (h-1) 0.11 ± 0.01 0.06 ± 0.002 0.14 ± 0.01 0.11 ± 0.06 Biomass yield (g/g) 0.06 ± 0.001 0.09 ± 0.001 0.04 ± 0.008 0.04 ± 0.004 Acid Production Butyric acid conc. (g/L) 44.1 ± 0.1 37.3 50.1 ± 2.4 48.6 ± 2.9

Butyric acid yield (g/g) 0.44 ± 0.004 0.42 ± 0.01 0.45 ± 0.02 0.45 ± 0.004

Acetic acid conc. (g/L) 8.6 ± 0.9 4.9 8.4 ± 0.1 6.2 ± 1.6

Acetic acid yield (g/g) 0.09 ± 0.001 0.05 ± 0.004 0.08 ± 0.01 0.06 ± 0.01

B/A ratio (g/g) 5.1 ± 0.5 7.6 ± 0.2 6.0 ± 0.6 8.0 ± 1.6 Gas Production

H2 yield (g/g) NA NA 0.023 ± 0.004 0.026 ± 0.001

CO2 yield (g/g) NA NA 0.34 ± 0.001 0.35 ± 0.005

H2/CO2 ratio (mole/mole) NA NA 1.59 ± 0.01 1.61 ± 0.01 Carbon Balance (%) NA NA 0.96 ± 0.02 0.98 ± 0.03

* These data come from previous study (Zhu, 2003).

Table 6.2 Comparison of fed-batch fermentations by immobilized cells of C. tyrobutyricum wild type and mutant on glucose and xylose at 37oC and pH 6.0.

169

Strains PAK-Em

pH Values 5.0 6.0 6.3 7.0

Cell Growth

µ (h-1) 0.08 ± 0.04 0.14 ± 0.01 0.06 ± 0.02 0.09 ± 0.01

Biomass yield (g/g) 0.11 ± 0.03 0.04 ± 0.01 0.06 ± 0.01 0.07 ± 0.01

Acid Production

Butyric acid conc. (g/L) 14.79 ± 0.99 50.11 ± 2.42 80.15 ± 2.54 61.52 ± 0.78

Butyric acid yield (g/g) 0.42 ± 0.03 0.45 ± 0.02 0.44 ± 0.002 0.44 ± 0.002 Acetic acid conc. (g/L) 2.19 ± 1.32 8.38 ± 0.37 13.10 ± 3.50 6.14 ± 1.16

Acetic acid yield (g/g) 0.06 ± 0.44 0.08 ± 0.01 0.07 ± 0.006 0.06 ± 0.01

B/A ratio (g/g) 8.44 ± 5.56 5.99 ± 0.55 6.31 ± 1.49 10.19 ± 1.80 Gas Production

H2 yield (g/g) 0.022 ± 0.002 0.025 ± 0.001 0.026 ± 0.001 0.027 ± 0.002

CO2 yield (g/g) 0.34 ± 0.04 0.34 ± 0.001 0.32 ± 0.03 0.30 ± 0.01

H2/CO2 ratio (mole/mole) 1.53 ± 0.18 1.59 ± 0.014 1.85 ± 0.092 2.02 ± 0.20

Carbon Balance 0.96 ± 0.05 0.96 ± 0.02 0.93 ± 0.03 0.94 ± 0.01

Table 6.3 Effects of pH on fed-batch fermentation kinetics by immobilized cells of C. tyrobutyricum mutant PAK-Em grew on glucose at 37oC, and various pH value.

170

45 60 PAK-Em, Free Cell OD Glucose 40 Glucose, pH 6.0 Acetate Butyrate 50 35 H2 CO2

30 40

25 30 20

15 20 Gas Production (L) Production Gas Concentration (g/L); OD 10 10 5

0 0

A 0 25 50 75 100 125 150 175 200 225 Time (h)

40 60 PAK-Em, Free Cell OD 35 Xylose, pH 6.0 xylose acetate 50 butyrate 30 H2 CO2 40 25

20 30

15 20 Gas Production (L) Production Gas

Concentration (g/L); OD 10

10 5

0 0 B 0 20 40 60 80 100 120 140 160 180 200 Time (h)

171

40 60 PAK-Em, Free Cell 35 Grape juice, pH 6.0 50 OD600 30 Glucose Fructose Acetate 40 25 Butyrate H2 CO2 20 30

15 20 Gas Production (L) Production Gas

Concentration (g/L); OD 10

10 5

0 0 C 0 20 40 60 80 100 120 140 160 180 200 220 Time (h)

Figure 6.1 Kinetics of fed-batch fermentations by free cells of the C. tyrobutyricum mutant PAK-Em with glucose (A), xylose (B), grape juice (C) at 37oC and pH 6.0. OD600 (×), sugar (■), butyrate (○), acetate (∆), hydrogen (●), and carbon dioxide (▲).

172

60 70 PAK-Em, FBB OD600 Glucose Glucose, pH 6.0 60 50 Acetate Butyrate H2 CO2 50 40

40 30 30

20 Concentration (g/L) Concentration 20 (L) Production Gas

10 10

0 0 A 0 20 40 60 80 100 120 140 160 180 200 220 240 Time (h)

60 70 PAK-Em, FBB OD Glucose Xylose, pH 6.0 60 50 Acetate Butyrate H2 CO2 50 40

40 30 30

20 (L) Production Gas Concentration (g/L) Concentration 20

10 10

0 0 B 0 20406080100120140160180200220 Time (h)

Figure 6.2 Kinetics of fed-batch fermentations by C. tyrobutyricum mutant PAK-Em o immobilized in the FBB with glucose (A) and xylose (B) at 37 C and pH 6.0. OD600 (×), sugar (■), butyrate (○), acetate (∆), hydrogen (●), and carbon dioxide (▲). 173

20 70 PAK-Em, FBB OD Glucose, pH 5.0 Acetate Butyrate 60 Glucose 15 H2 CO2 50

40 10 30

20 Concentration (g/L); OD 5

10 glucose(g/L) and (L) production Gas

0 0 A 0 20 40 60 80 100 120 140 160 Time (h)

90 140 PAK-Em, FBB OD 80 Glucose, pH 6.3 Glucose Acetate 120 70 Butyrate H2 100 CO2 60

50 80

40 60

30 40 (L) Production Gas Concnetration (g/L); OD (g/L); Concnetration 20

20 10

0 0 B 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 Time (h)

174

70 100 PAK-Em, FBB OD Glucose, pH 7.0 Glucose 60 Acetate Butyrate 80 H2 50 CO2

60 40

30 40 Gas Production (L) Production Gas 20 Concentration (g/L); OD 20 10

0 0 C 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 Time (h)

Figure 6.3 Kinetics of fed-batch fermentations by C. tyrobutyricum mutant PAK-Em o immobilized in the FBB with glucose at 37 C and pH 6.3 (A) and pH 7.0 (B). OD600 (×), sugar (■), butyrate (○), acetate (∆), hydrogen (●), and carbon dioxide (▲).

175

50 120 HydEm, Free OD600 Glucose, pH 6.0 Glucose Acetate 100 40 Butyrate H2 CO2 80 30

20 40 Gas Production (L) Production Gas Concentration (g/L); OD

10 20

0 0 0 20406080100120 Time (h)

Figure 6.4 Kinetics of fed-batch fermentations by free cells of the adapted mutant o HydEm of C. tyrobutyricum with glucose at 37 C and pH 6.0. OD600 (×), sugar (■), butyrate (○), acetate (∆), hydrogen (●), and carbon dioxide (▲).

176

1 2 3 4 5 kDa

97 70 66

Figure 6.5 SDS polyacrylamide gel electrophoresis of cellular proteins from immobilized C. tyrobutyricum. (Lane 1: Molecular marker; Lane 2: wild type with glucose; Lane 3: wild type with xylose; Lane 3: PAK-Em with glucose; Lane 4: PAK-Em with xylose.)

177

CHAPTER 7

CONCLUSIONS AND RECOMMENDATIONS

7.1 Conclusions

This research demonstrated the feasibility of using metabolically engineered mutant of Clostridium tyrobutyricum for the production of butyric acid and hydrogen, the advantages of using fibrous-bed bioreactor (FBB) for the immobilization fermentation, potential applications of shotgun microarray to study genome function and protein expression, and re-engineering of the metabolically engineered mutant by high throughput screening from mutant library. The important results and conclusions obtained in this study are summarized below.

7.1.1 Butyric Acid and Hydrogen Fermentation by pta-Deleted Mutant

The gene integrational mutagenesis can be used to modify the metabolic pathway in

C. tyrobutyricum by gene knockout. A pta gene inactivated mutant, PPTA-Em, was

thus obtained with improved biomass yield (0.16 g/g), slower specific cell growth rate

(0.12 h-1), two-fold higher fermentation productivity (0.63 g/L·h), and much higher

final butyric acid tolerance with inhibition rate constant Kp 1.59 g/L. 178

Free fermentation by PTA-Em mutant using glucose produced 15% more butyric acid

(32.5 g/L) with yield of 0.38 g/g and 14% less acetic acid (4.28 g/L) as compared to

the wild type strain. However, the acetic acid byproduct was not completely

eliminated in the mutant. The gas production was not significantly changed by gene

manipulation.

A high butyric acid concentration of 50 g/L from glucose and 51.55 g/L from xylose

by PPTA-Em were obtained using FBB fermentation. The acetic acid production

from xylose was lower than that from glucose in FBB fermentation, 6.424 g/L vs.

8.740 g/L for the concentration and 0.045 g/g vs. 0.081 g/g for the yield.

The mutant cell growth rate using glucose as carbon source was higher than that using

xylose. The butyric acid concentration and yield had no obvious changes from

different sugar sources, but the butyric acid selectivity was increased greatly with

higher butyric acid/acetic acid (B/A) ratio by PPTA-Em.

7.1.2 Butyric Acid and Hydrogen Fermentation by ack-Deleted Mutant

Integrational plasmid containing the ack fragment gene was constructed and

integrated into the chromosomal DNA by homologous recombination, producing an

ack-deleted mutant PAK-Em.

-1 -1 The specific cell growth of the mutant decreased by 42%, from 0.24 h to 0.14 h ,

due to the impaired PTA-AK pathway. The butyric acid production from fed-batch

fermentations was improved greatly with higher butyric acid yield of 0.42 g/g, final

concentration of 41.65 g/L, and improved final product tolerance. Butyric acid yield

179

was increased by 23% and the B/A ratio increased from 3.68 by wild type to 5.41 by

the mutant. The hydrogen production by PAK-Em mutant increased significantly,

with higher hydrogen yield of 0.024 g/g and H2/CO2 ratio of 1.44. The

immobilization fermentation using FBB produced high butyric acid concentration

(50.11 g/L) and yield (0.45 g/g) at 37 oC and pH 6.0.

Different sugar sources, including glucose, xylose and fructose were applied to the

fed-batch fermentation using PAK-Em mutant. Although the butyric acid and

hydrogen production from different sugar sources were similar, the cell growth rate,

biomass yield, and butyric acid selectivity were greatly changed. This study provided

directions for the application of low value biomass in the biotechnology industry.

One adaptation mutant screened from FBB was obtained with similar specific cell

growth rate (0.21 h-1) to that of wild type, higher hydrogen productivity, and higher

hydrogen yield (0.40 g/g).

It was found that high concentration of butyric acid (81.94 g/L) with high yield (0.45

g/g) was obtained at pH 6.0 using FBB fermentation by PAK-Em, which is the best

butyric acid fermentation ever attained from fermentation to date. Higher B/A ratio

were found at low and high pH conditions.

The metabolic shift was studied by fermentation with glucose and xylose using PAK-

Em under pH 5.0. It was found that there was no acetic acid and lactic acid

production, and butyric acid was the only product at pH 5.0 using xylose, which

could decrease the butyric acid production cost using fermentation.

180

7.1.3 Protein Expression with Gene Manipulation

The enzyme activities in the PPTA-Em mutant were reduced by more than 60% for

PTA and 80% for AK and increased about 40% for BK. The SDS-PAGE and two-

dimensional protein electrophoresis showed that at least two highly expressed protein

spots with PI 6.2 and PI 6.8 around ~32 kDa disappeared completely, and one protein

was dramatically down regulated in the mutant. Different sugar sources and the cell

adaptation in the fibrous-bed matrix also greatly affected the global protein

expression of the PPTA-Em mutant.

The enzyme assay showed that the AK activity in PAK-Em decreased by ~50%,

meanwhile, the PTA and hydrogenase activities increased by ~40%, respectively. The

expression of protein with 70 kDa decreased significantly, and the protein with 70

kDa had much higher expression in PAK-Em than that in the wild type. Protein

expression varied when using glucose and xylose was changed indicating that

different pathways or enzymes were used for the consumption of different carbon

sources.

7.1.4 Metabolic Flux and Pathway Analysis

Southern hybridization analysis using mutant indicated that the original pta gene in

PPTA-Em mutant was completely inactivated, but the acetic acid production was not

zero. A possible reason is that some other enzymes or metabolic pathways exist for

the production of acetic acid in C. tyrobutyricum. The slow cell growth rate and

higher final product tolerance were due to the inactivation of higher ATP producing

181

acetic acid pathway. The metabolic flux to butyric acid was increased but the flux to

acetic acid production was decreased greatly with higher ATP accumulation in the

PPTA-Em mutant, indicating the global metabolism changes in the pathway.

The reduced flux through PTA-AK pathway resulted in higher butyric acid

concentration from PAK-Em with improved final product tolerance. The hydrogen

production by PAK-Em mutant was increased; this was the first time to improve

hydrogen production of C. tyrobutyricum by gene manipulation. The metabolic flux

analysis based on the fermentation using different sugar sources showed that the C.

tyrobutyricum can used different pathway for the butyric acid and hydrogen

production.

The metabolic shift analysis indicated that the PTA-AK pathway in PAK-Em was

deleted completely because of no acetic acid production at pH 5.0. The blocking of

lactic acid formation pathway indicated that the gene manipulation resulted in

significant energy changes in the metabolic pathway of PAK-Em, which led to higher

yield of hydrogen.

7.2 Recommendations

This research focused on the improvement of butyric acid and hydrogen production by metabolically engineered mutants using free and immobilization fermentation from various sugar sources. The butyric acid and hydrogen production were significantly improved. The protein expression changes, the metabolic pathway and flux in the mutants were also studied. However, there is very little genomic and proteomic 182

information available for this bacterium. Thus the initial effort will be on developing and utilizing functional genomic and proteomic using shotgun DNA microarrays and 2D-

PAGE to analyze and identify genes expressed and proteins present. The results from functional genomics studies will be used in genetic and metabolic engineering of mutants for industrial applications. The major tasks for future work include: the construction of

DNA microarrays based on random genomic library of the organisms; studying the proteomics via 2D-PAGE and tandem MS to identify proteins on the gel maps; using the functional genomic data to develop metabolic engineering strategies for reconstruction of acid-tolerant mutants with high product yield; developing mutant library and screening for metabolically superior mutants for further genome shuffling.

7.2.1 Functional Genomics by DNA Microarray

Shotgun DNA microarray and proteomics will be used to analyze and identify genes expressed and the proteins present in wild type and mutant strains of C. tyrobutyricum under different environmental conditions. The focus of the study will be on the mechanisms of cell survival or response to environmental changes in, e.g., pH, temperature, and salt and product concentrations. By comparing different gene expression and fermentation kinetics between wild type and different mutants, it is possible to identify critical genes in the metabolic pathway as targets for rational metabolic engineering. The results from the functional genomic studies will then be used in genetic and metabolic engineering of mutants for industrial applications.

183

1. A random genomic library for C. tyrobutyricum has been constructed. The white

colonies were picked from the agar plates and stored in 96-well plates at -80oC

with 15% glycerol for further study. The library was created as follows. Some

randomly picked DNA insert was checked by minipreps and the results showed

that the redundancy of the C. tyrobutyricum gene library was low, indicating that

the gene library should cover the entire genome.

2. Colonies will be cultivated, processed, and the insert DNA in each clone was

amplified using 96-well PCR. Genomic DNA microarray with spots representing

the C. tyrobutyricum genome will be printed on silylated slides (TeleChem

International) with 75 µm spot size and 30 µm spacing using an arrayer at 45-50%

humidity; the slides will be processed following the standard procedure online

(http://derisilab.ucsf.edu/arraymaker.shtml).

3. Cells cultured under different conditions will be harvested and the total RNA will

be extracted using Qiagen RNA extraction kit. The RNA samples from control

and test conditions will be labeled by cDNA synthesis with reverse transcriptase

and Cy3- and Cy5-dUTP primed with random hexamers. Chip hybridization,

DNA sequencing and analysis will then be carried out following the procedure

described in the appendix.

7.2.2 Protein Electrophoresis and Proteomics

1. Bacterial cells grown under various fermentation conditions will be harvested,

and the cell extract will be applied to run two-dimensional protein electrophoresis.

184

The 2D gel will be analyzed using Phoretix 2D AdvancedTM software, and the

targeted proteins on the 2D gels will be isolated, sequenced, and identified using

protein sequencing by mass spectrometer and analysis following the protocol in

the appendix.

2. The interested protein spots on the gel will be examined by Western blotting

when the corresponding protein antibodies are available. Enzyme activity assays

also will be performed to correlate with the observed protein expression levels.

7.2.3 Re-Engineer Metabolically Engineered Mutant

With the high throughput DNA microarray data (transcriptomics) and two- dimensional database (proteomics), new molecular pathways may be discovered from protein interaction and gene expression data. Proteome analysis can provide insightful information on cellular metabolism that is difficult to obtain by traditional approaches.

Based on the results of functional genomics and proteomics, new metabolic pathways under different conditions or stress can be identified, enabling us to develop rational strategies for engineering metabolic pathways and cellular properties useful for fermentation, which relies extensively on strain improvement for commercialization.

1. With the study of physiology changes, gene expression, and proteome changes

using DNA microarrays and proteomic analysis, and the metabolic flux analysis

based on the fermentation kinetics results, we can define the key enzymes or the

limiting control pathway in the metabolic network for further design of new gene

manipulation strategies for a high-yield mutant. The sequences in rational

185

metabolic engineering design are as follows: (i) identify potentially limiting

enzymes in the metabolic pathway, (ii) theoretically and/or experientially examine

the possible flux changes that can be achieved by gene manipulations of the

enzymes identified, (iii) select the final candidate enzymes to be changed, (iv)

examine the consequences of this metabolic and cellular engineering and (v)

repeat (ii) to (iv) until the objective is accomplished. However, the process in

rational metabolic engineering is a long exercise and may not be successful until a

full knowledge of the genome, proteome, and metabolome is obtained.

2. Besides the mutants we have already created previously, we will generate more

mutants with random gene knockout and gene knockin. The basic idea is to

randomly knock out the genes on the chromosome using integrational

mutagenesis, develop the high-throughput screening method to obtain the mutant

that we want, and to further study the functional genomics and proteomics of the

mutants. The random DNA inserts are the fragments amplified from gene library

using PCR. The constructed plasmids will be used to transform the C.

tyrobutyricum cells and inactivate the homologous genes on the chromosome. An

efficient mutant screening method is very important in obtaining the desirable

mutants. In this regard, 96 well plates will be used to cultivate all the candidate

mutant strains generated in this study by applying gradient pressure such as pH,

nutrient, and final produc. Since CO2 gas is produced in butyric acid fermentation,

cell growth and metabolic activities may be screened by directly measuring gas

pressure in the microwells with deformable diaphragm bottom attached to a small

piece of stainless steel metal connected to a pressure/strain transducer (probe).

186

When the gas is produced in fermentation, pressure is exerted on the fixed bottom

metal. The strength of pressure will be converted via the pressure sensor array and

monitored in a central computer by LabVIEW Control System. Gas pressure is

proportional to the total volume of gas produced in the microwell. To further

evaluate acid production, we will add pH-sensitive fluorescence dye in each well

and monitor the fluorescence intensity as the pH change due to acid production.

The robust mutants under different environmental conditions will be screened by

this high throughput screening method.

3. With comparative genomic analysis to identify beneficial mutations, we will be

afforded the opportunity to do “genome breeding” for the reconstruction of a

robust production strain, a minimal mutation strain with useful mutations and no

unnecessary mutations (Ikeda and Nakagawa, 2003). Genome shuffling is a

process that combines the advantages of multi-parental crossing allowed by DNA

shuffling with the recombination of entire genomes normally associated with

conventional breeding. This method has been applied with several bacteria, and

will be studied once we have identified and obtained mutants with desirable traits.

The success of genome shuffling approach depends on the initial selection of

variants, the efficiency of the genetic recombination process, and the power of the

selection method. Because the key feature of this technology is the formation of a

library of candidate cells by shuffling the genomes of mutants with improved

phenotypes (Stephanopoulos, 2002), the classical strain-improvement methods

also can be used to generate populations with subtle improvement in acid

tolerance or acid production. The improvement in the distribution of complex

187

progeny can be achieved by the recursive fusion of a mixed protoplast population.

Each pooled fusion mimics each thermal cycle of a DNA shuffling reaction, and recursive fusions should result in the efficient shuffling of the population (Patnaik, et al., 2002). In this project, we will explore the genome shuffling approach after obtaining mutants with desirable processing traits. We can use genome shuffling to improve both acid tolerance and product formation at the same time. The advance made here will be a departure from the paradigm of examining one mutation at a time. This approach is used in recognition of the distribution of phenotype control throughout the genome and the need to simultaneously modify several genes in order to improve it.

188

BIBLIOGRAPHY

Abbad-Andaloussi, S., Durr, C., Raval, G., and Petitdemange, H. Carbon and electron flow in Clostridium butyricum grown in chemostat culture on glycerol and on glucose. Microbiology. 1996, 142, 1149-1158.

Allen, S. P., and Blaschek, H. P. Factors involved in the electroporation-induced transformation of Clostridium perfringens. FEMS Microbiol. Lett. 1990, 70, 217-220.

Bailey, J. E. Toward a science of metabolic engineering. Science 1991, 252, 1668-1675.

Borch, E., Berg, H., and Holst, O. Heterolactic fermentation by a homofermentative Lactobacillus sp. during glucose limitation in anaerobic continuous culture with complete cell recycle. J. Appl. Bacteriol. 1991, 71, 265-269.

189

Campbell, A. M. Episomes. Adv. Genet. 1962, 11, 101-146.

Chin, H. L., Chen, Z. S., and Chou, P. Fedbatch operation using Clostridium acetobutylicum suspension culture as biocatalyst for enhancing hydrogen production. Biotechnol. Prog. 2003, 19, 383-388.

Das D, Veziroglu T. N. Hydrogen production by biological processes: a survey of literature. Int. J. Hydrogen Energy. 2001, 26, 13-28.

Danyluk, B., and Kijowski, J. Influence of the lysozyme monomer on the growth of Clostridium tyrobutyricum bacteria. Pol. Przemysl Spozywczy. 2001, 55, 16-19. de Vos, W. M., Bron, P. A., Kleerebezem, M. Post-genomics of lactic acid bacteria and other food-grade bacteria to discover gut functionality. Current Opinion in Biotechnol. 2004, 15, 86-93. de Vos, W. M., and Hugenholtz, J. Engineering metabolic highways in Lactococci and other lactic acid bacteria. Trends Biotechnol. 2004, 22, 72-79.

Desai, R. P., Nielsen, L. K. and Papoutsakis, E. T. Stoichiometric modeling of Clostridium acetobutylicum fermentation with non-linear constraints. J. Biotechnol. 1999b, 71, 191-205.

Drake, H. L. Demonstration of hydrogenase in extracts of the homoacetate-fermenting bacterium Clostridium thermoaceticum. J. Bacteriol. 1982, 150, 702-709.

Dunn, S. Hydrogen futures: toward a sustainable energy system. Int. J. Hydrogen Energy. 2002, 27, 235-264.

190

Elsser, D., Morgan, A. J., Thomas, L. V., and Yu, S. Anhydrofructose derivative antimicrobial agents for food use. U.S. Pat. Appl. Publ. 2003. 31 pp.

Flory, M. R., Griffin, T. J., Martin, D., and Aebersold, R. Advances in quantitative proteomics using stable isotope tags. Trends in Botechnology. 2002, 20, 23-29.

Fayolle, F., Marchal, R., and Ballerini, D. Effect of controlled substrate feeding on butyric acid production by Clostridium tyrobutyricum. J. Ind. Microbiol. 1990, 6, 179-183.

Gaertner, A. L., Antelmann, H., Sapolsky, R., Ferrari, E., Chotani, G., Miller, B., Hecker, M. Qantitative proteomic and transcriptomic profiling during fermentation of pleiotropic Bacillus subtilis mutants. Abstracts of Papers, 227th ACS National Meeting, Anaheim, CA, March 28-April 1, 2004.

Girbal, L. Croux, C. Vasconcelos, I., and Soucaille, P. Regulation of metabolic shifts in Clostridium acetobutylicum ATCC 824. FEMS Microbiol. Rev. 1995, 17(3), 287-297.

Goodwin, S. and Zeikus, J. G. Physiological adaptations of anaerobic bacteria to low pH: metabolic control of protonmotive force in Sarcina ventriculi. J. Bacteriol. 1987, 169, 2150-2157.

Green, E. M. and Bennett, G. N. Inactivation of an aldehyde/alcohol dehydrogenase genes from Clostridium acetobutylicum ATCC 824. Appl. Biochem. Biotechnol. 1996, 57-58, 213-221.

Green, E. M. and Bennett, G. N. Genetic manipulation of acid and solvent formation in Clostridium acetobutylicum ATCC 824. Biotechnol. Bioeng. 1998, 58, 215-221.

Grobben, N. C., Egging, G., Cuperus, F. P., and Huizing, H. J. Production of ABE from potato wastes: fermentation with integrated membrane extraction. Appl. Microbiol. Biotechnol. 1993, 39, 494-498.

Guérout-Fleury, A. M., Shazand, K., Frandsen, N., and Stragier, P. Antibiotic-resistance cassettes for Bacillus subtilis. Gene 1995, 167, 335-336.

Han, M. J., Lee, S. Y. Proteome profiling and its use in metabolic and cellular engineering. Proteomics. 2003, 3, 2317-2324. 191

Hara, H. Physiological effects of short-chain fatty acid produced from prebiotics in the colon. Bioscience Microflora 2002, 21, 35-42.

Hartmanis, M. G. N. Butyrate kinase from Clostridium acetobutylicum. J. Biol. Chem. 1987, 262, 617-621.

Harwood, C. S., and Caale-Parola, E. Properties of acetate kinase isozymes and a branched-chain fatty kinase from a spirochete. J. Bacteriol. 1982, 52, 246-254.

Haynes, P. A., Gygi, S. P., Figeys, D., and Aebersold, R. Electrophoresis 1998, 19, 1862

Herman, L. M., De Block, J. H. G. E., and Waes, G. M. A. V. J. A direct PCR detection method for Clostridium tyrobutyricum spores in up to 100 milliliters of raw milk. Appl. Environ. Microbiol. 1995, 61, 4141-4146.

Heyndrickx, M., DeVos, P., and DeLey, J. Fermentation of D-xylose by Clostridium butyricum LMG 1213t1 in chemostats. Enzyme Microb. Technol. 1991, 13, 893-897.

Hofvendahl, K., van Niel, E. W. J., and Hahn-Hägerdal, B. Effect of temperature and pH on growth and product formation of Lactococcus lactis ssp. lactis ATCC 19435 growing on maltose. Appl. Microbiol. Biotechnol. 1999, 51, 669-672.

Hsu, S.-T., and Yang, S.-T. Propionic acid fermentation of lactose by Propionibacterium acidipropionici: effects of pH. Biotechnol. Bioeng. 1991, 38, 572-578.

Huang, Y., and Yang, S.-T., Acetate production from whey lactose using co-immobilized cells of homolactic and homoacetic bacteria in a fibrous-bed bioreactor. Biotechnol. Bioeng. 1998, 60, 499-507.

192

Ikeda, M., and Nakagawa, S. The Corynebacterium glutamicum genome: features and impacts on biotechnological processes. Appl. Microbiol. Biotechnol. 2003, 62, 99-109.

Jungblut, P., and Wittmann-Liebold, B. J. Biotechnol. 1995, 41, 111-120.

Kataoka, N., Miya, A., and Kiriyama, K. Studies on hydrogen production by continuous culture system of hydrogen-producing anaerobic bacteria. Water Sci. Technol. 1997, 36, 41-47.

Kirkpatrick, C., Maurer, L. M., Oyelakin, N. E., Yoncheva, Y. N., Mauer, R., Slonczewski, J. L. Acetate and formate stress: opposite responses in the proteome of Escherichia coli. J. Bacterial. 2001, 183, 6466-6477.

Klose, J., and Kobalz, U. Electrophoresis 1995. 16, 1034-1059.

Koffas, M., Roberge, C., Lee, K., and Stephanopoulos, G. Metabolic engineering. Annu. Rev. Biomed. Eng. 1999, 1, 535-557.

193

Kuipers, O. P. Genomics for food biotechnology: prospects of the use of high-throughput technologies for the improvement of food microorganisms. Current Opinion in Biotechnol. 1999, 10, 511-516

Kumar, A., Jain, S. R., Sharma, C. B., Joshi, A. P., and Kalia, V. C.. Increased H2 production by immobilized microorganisms. World J. Microbiol. Biotechnol. 1995, 11, 156-159.

Lee, S. Y. and Papoutsakis, E. T. The challenges and promise of metabolic engineering. In: Metabolic engineering, eds., S. Y. Lee and E. T. Papoutsakis, Marcel Dekker Inc., NY, 1999, pp 1-12.

Lefranc, L. and Cie, E. A process for the manufacture of butyric acid and other fatty acids with recovery of the gases of fermentation. Br. Patent 1923, No. 186572.

Lewis, V. P. and Yang, S. T. A novel extractive fermentation process for propionic acid production from whey lactose. Biotechnol. Prog. 1992, 8(2), 104-110.

Lindmark, D. G. Acetate production by Tritrichomonas foetus. Biochem. Parasites Host- Parasite Relat. Proc. Int. Sym. 1976, 10, 15-21.

Loessner, M. J., Maier, S. K., Schiwek, P., Scherer, S. Long-chain polyphosphates inhibit growth of Clostridium tyrobutyricum in processed cheese spreads. J. Food Protection. 1997, 60, 493-498.

Lynd, L. R., Wyman, C. E., and Gerngross, T. U. Biocommodity engineering. Biotechnol. Prog. 1999, 15, 777-793.

Maharbiz, M. M., Holtz, W. J., Howe, R. T., and Keasling, J. D. Microbioreactor arrays with parametric control for high-throughput experimentation. Biotechnol. Bioeng. 2004, 85, 376-381.

Majewski, R. A. and Domach, M. M. Simple constrained optimization view of acetate overflow in E. coli. Biotechnol. Bioeng. 1990, 35, 732-738.

Matsuyama, A. H., Yamamoto, H., and Nakano, E. Cloning, expression, and nucleotide sequences of the Escherichia coli K-12 ackA gene. J. Bacteriol. 1989, 171, 577-580.

194

Mermelstein, L. D., Welker, N. E., Bennett, G. N., and Papoutsakis, E. T. Expression of cloned homologous fermentative genes in Clostridium acetobutylicum ATCC 824. Bio/Technology 1992, 10, 190-195.

Mclaren, J. S., and Jeffries, T. W. Genetics and genomics in bioenergy and bioproducts. Appl. Biotechnol. Biochem. 2003. 107, 631-632.

Michel-Savin, D., Marchal, R., and Vandecasteele, J. P. Control of the selectivity of butyric acid production and improvement of fermentation performance with Clostridium tyrobutyricum. Appl. Microb. Biotechnol. 1990a, 32, 387-392.

Mitchell, W. J. Biology and physiology. In: Clostridia, Biotechnology and medical applications, eds., H. Bahl and P. Dürre, Wiley-VCH, Weinheim, 2001, pp 49-104.

Mizuno, O., Dinsdale, R., Hawkes, F. R., Hawkes, D. L. and Noike, T. Enhancement of hydrogen production from glucose by nitrogen gas sparging. Bioresour. Technol. 2000, 73, 59-65.

Momirlan, M, and Veziroglu, T. Recent directions of world hydrogen production. Energy Rev. 1999, 3, 219-31.

O’Farrell, P. H. J. Biol. Chem. 1975, 250, 4007-4021.

Papoutsakis, E. T. A useful equation for fermentation of butyric acid bacteria. Biotechnol. Lett. 1983, 5(4), 253-258.

195

Papoutsakis, E. T. Equations and calculations for fermentations of butyric acid bacteria. Biotechnol. Bioeng. 1984, 26, 174-187.

Papoutsakis, E. T. and Meyer, C. L. Equations and calculations of product yields and preferred pathways for butanediol and mixed-acid fermentations. Biotechnol. Bioeng. 1985a, 27, 50-66.

Papoutsakis, E. T. and Meyer, C. L. Fermentation equations for propionic acid bacteria and production of assorted oxychemicals from various sugars. Biotechnol. Bioeng. 1985b, 27, 67-80.

Patnaik, R., Louie, S. Gavrilovic, V. Perry, K. Stemmer, W. P. C. Ryan, C. M. and del Cardayre, S. Genome shuffling of Lactobacillus for improved acid tolerance. Nat. Biotechnol. 2002, 20, 707-712.

Phillips-Jones, M. K. Plasmid transformation of Clostridium perfringens by electroporation methods. FEMS Microbiol. Lett. 1990, 66, 221-226.

Playne, M. J. Propionic acid and butyric acids. In: Comprehensive biotechnology, vol. 3, eds., M. Moo-Young, Pergamon Press, Oxford, 1985, pp 731-759.

Pouillart, P. R. Role of butyric acid and its derivatives in the treatment of colorectal cancer and hemoglobinopathies. Life Sci. 1998, 63(20), 1739-1760.

Reardon, K., Scheper, T.-H., and Bailey, J. Metabolic pathway rates and culture fluorescence in batch fermentations of Clostridium acetobutylicum. Biotechnol. Prog. 1987, 3, 153-167.

196

Reid, C. A., Hillman, K., Henderson, C., and Glass, H. Fermentation of native and processed starches by the porcine caecal anaerobe Clostridium butyricum (NCIMB 7423). J. Appl. Bacteriol. 1996, 80, 191-198,

Rogers, P. and Gottschalk, G. Biochemistry and regulation of acid and solvent production in Clostridia. In: The clostridia and biotechnology, eds. D. R. Woods, Butterworth- Heinemann, MA, 1993, pp 25-50.

Rose, I. A. Acetate kinase of bacteria (acetokinase). Methods Enzymol. 1955, 1, 591-595.

Schulz, C. M., and Ruzicka, J. Real-time determination of glucose consumption by live cells using a lab-on-valve system with an integrated microbioreactor. Analyst 2002, 127, 1293-1298.

Sharpell, F. H. J. Microbial flavors and fragrances. In: Comprehensive biotechnology, eds., M. Moo-Young. Pergamon Press, New York, 1985, pp 965-981.

Shimizu, H. Metabolic engineering integrating methodologies of molecular breeding and bioprocess systems engineering. J. Biosci. Bioeng. 2002, 94, 563-573.

Silva, E. M. and Yang, S. T. Kinetics and stability of a fibrous-bed bioreactor for continuous production of lactic from unsupplemented acid whey. J. Biotechnol. 1995, 41(1), 59-70.

Sneath, P. H. A. Bergey’s manual of systematic bacteriology 2, Waverly Press, Baltimore, 1986, pp 1141-1200.

Sridhar, J. and Eiteman, M. A. Metabolic flux analysis of Clostridium thermosuccinogenes, effects of pH and culture redox potential. Appl. Biochem. Biotechnol. 2001, 94, 51-69.

Stephanopoulos, G. Metabolic engineering by genome shuffling. Nat. Biotechnol. 2002, 20, 66-667.

197

Stephanopoulos, G. and Sinskey, A. J. Metabolic engineering – methodologies and future prospects. Trends Biotech. 1993, 11, 392-396.

Stinson, E. E. and Naftulin, K. A. Effect of pH on organic acid production by Clostridium propionicum in test tube and fermentor cultures. J. Ind. Microbiol. 1991, 8, 59-63.

Taguchi, F., Yamada, K., Hasegawa, K., Taki-saito, T. and Hara, K. Continuous hydrogen production by Clostridium sp. stain no. 2 from cellulose hydrolysate in an aqueous two-phase system. J. Ferment. Bioeng. 1996, 82, 80-83.

Thomas, L. V., Ingram, R. E., Bevis, H. E., Davis, E. A., Milne, C. F., and Delves- Broughton, J. Effective use of nisin to control Bacillus and Clostridium spoilage of a pasteurized mashed potato product. J. Food Protection. 2002, 65, 1580-1585.

Tomas, C. A., Alsaker, K. V., Bonarius, H. P. J., Hendriksen, W. T., Yang, H., Beamish, J.A., Paredes, C.J., and Papoutsakis, E.T. DNA array-based transcriptional analysis of asporogenous, nonsolventogenic Clostridium acetobutylicum strain SKO1 and M5. J. Bacteriol. 2003, 185, 4539-4547.

Torino, M. I., Taranto, M. P., Sesma, F., and Font de Valdez, G. Heterofermentative pattern and exopolysaccharide production by Lactobacillus helveticus ATCC 15807 in response to environmental pH. J. Appl. Microbiol. 2001, 91, 846-852. van Andel, J. G., Zouttberg, G. R. Crabbendam P. M., and Breure, A. M. Glucose fermentation by Clostridium butyricum grown under a self generated gas atmosphere in chemostat culture. Appl. Microbiol. Biotechnol. 1985, 23, 21-26. van der Lelie, D., van der Vossen, J. M. B. M., and Venema, G. Effect of plasmid incompatibility on DNA transfer to Streptococcus cremoris. Appl. Environ. Microbiol. 1988, 54, 865-871. van der Meer, J. R., Ludwig, W., and De Vos W. M. Characterization of a ribosomal RNA gene cluster from Clostridium tyrobutyricum: phylogenetic positioning based on the 16S and 23S nucleotide sequences. System. Appl. Microbiol. 1993, 16, 201-207.

Vandak, D., Telgarsky, M., and Sturdik, E. Influence of growth factor components on butyrate production from sucrose by Clostridium butyricum. Folia Microbiol. 1995b, 40(6), 669-672. 198

Vandak, D., Zigová, J., Šturdík, E., and Schlosser S. Evaluation of solvent and pH for extractive fermentation of butyric acid. Proc. Biochem. 1997, 32, 245-251.

Varma, A. and Palsson, B. O. Metabolic flux balancing: basic concepts, scientific and practical use. Bio/Technology 1994a, 12, 994-998.

Venkatesh, K. V. Metabolic flux analysis of lactic acid fermentation: effects of pH and lactate ion concentration. Proc. Biochem. 1997, 32(8), 651-655.

Walter, K. A., Mermelstein, L. D., and Papoutsakis, E. T. Studies of recombinant Clostridium acetobutylicum with increased dosage of butyrate formation genes. Ann. N. Y. Acad. Sci. 1994, 721, 69-72.

Watrous, M. M., Clark, S., Kutty, R., Huang, S., Rudolph, F. B., Hughers, J. B., and Bennett, G. N. 2,4,6-Trinitrotoluene reduction by an Fe-only hydrogenase in Clostridium acetobutylicum. Appl. Environ. Microbiol. 2003, 26, 542-547.

Weimer, P. J. Effects of dilution rate and pH on the ruminal cellulolytic bacterium Fibrobacter succinogenes S85 in cellulose-fed continuous culture. Arch. Microbiol. 1993, 160, 288-294.

Williams, E. A., Coxhead, J. M., and Mathers, J. C. Anti-cancer effects of butyrate: use of micro-array technology to investigate mechanisms. Proceeding of the Nutrition Society. 2003, 62, 107-115.

Wiseman, A. Novel food-proteomics by control of bioprocess-enzymology outcomes? Trends Food Science & Technol. 2004, 15, 44-45.

Wiesenborn, D. P., Rudolph, F. B., and Papoutsakis, E. T. Phosphotransbutyrylase from Clostridium acetobutylicum ATCC 824 and its role in acidogenesis. Appl. Environ. Microbiol. 1989, 55, 317-322.

Wong, T. K. K., Philp, R., Nissom, P. M., Zheng, L., Lau, L. Y. A., Wong, C. L., and Yap, M. G. S. The study of cell adaptation to different growth conditions using quantitative proteomics and DNA microarray. Abstracts of Papers, 227th ACS National Meeting, Anaheim, CA, March 28-April 1, 2004.

199

Woods, D. R. and Santangelo, J. D. Molecular analysis and expression of nitrogen metabolism and electron transport genes of Clostridium. In: The clostridia and biotechnology, eds. D. R. Woods, Butterworth-Heinemann, MA, 1993, pp 201-226.

Willims, E. A., Coxhead, J. M., Mathers, J. C. Anti-cancer effects of butyrate: use of microarray technology to investigate mechanisms. Proc. Nutr. Soc. 2003, 62, 107-115.

Wu, Z., and Yang, S. T. Extractive fermentation for butyric acid production from glucose by Clostridium tyrobutyricum. Biotechnol. Bioen. 2003, 82, 93-02.

Yang, S.-T. Extractive fermentation using convoluted fibrous bed bioreactor. U.S. Patent 1996, No. 5563069.

Yang, S.-T., Huang, Y., and Hong, G. A novel recycle batch immobilized cell bioreactor for propionate production from whey lactose. Biotechnol. Bioeng. 1995, 45, 379-386.

Zeng, A.-P. Pathway and kinetic analysis of 1,3-propanediol production from glycerol fermentation by Clostridium butyricum. Bioprocess Engineering 1996, 14, 169-175.

Zhang, C., Zhang, M., Ju, J., Nietfeldt, J., Wise, J., Terry, P. M., Olson, M., Kachman, S. D., Wiedmann, M., Samadpour, M., Benson, A. K. Genome diversification in phylogenetic lineages I and II of Listeria monocytogenes: Identification of segments unique to lineage II populations. J. Bacteriol. 2003, 185, 5573-5584.

Zhang, Y-X, Perry, K., Vinci, V. A., Powell, K., Stemmer, W. P. C., and Cardayre, S. B. Genome shuffling leads to rapid phenotypic improvement in bacteria. Nature 2002, 415, 644-646.

Zhu, Y., Wu, Z., and Yang, S.-T. Butyric acid production from acid hydrolysate of corn fiber by Clostridium tyrobutyricum in a fibrous bed bioreactor. Proc. Biochem. 2002, 38, 657-666.

Zhu, Y., and Yang, S. T. Adaptation of Clostridium tyrobutyricum for enhanced tolerance to butyric acid in a fibrous-bed bioreactor. Biotechnol. Prog. 2003, 19, 365-372. 200

Zhu, Y., and Yang, S. T. Effect of pH on metabolic pathway shift in fermentation of xylose by Clotridium tyrobutyricum. J. Biotechnol. 2004, 110, 143-157.

Zigová, J., and Šturdík, E. Advances in biotechnological production of butyric acid. J. Ind. Microbiol. Biotechnol. 2000, 24, 153-160

Zigová, J., Šturdík, E., Vandak, D., and Schlosser, S. Butyric acid production by Clostridium butyricum with integrated extraction and pertraction. Proc. Biochem. 1999, 34, 835-843.

201

APPENDIX A

MEDIUM COMPOSITIONS

The basal medium contained (per liter of distilled water): 40 ml of mineral #1

solution, 40 ml of mineral #2 solution, 10 ml of trace metals solution, 10 ml of vitamin

solution, 10 ml of 0.005% NiCl·6H2O, 2 ml of 0.2% FeSO4·7H2O, 5 g of trypticase, 5 g

of yeast extract, 6 g of NaHCO3. The mineral #1 solution contained 7.86 g/L

K2HPO4·3H2O. Mineral #2 solution consisted of the following (per liter): 6 g of KH2PO4;

6 g of (NH4)2SO4; 12 g of NaCl; 2.5 g of MgSO4·7H2O; 0.16 g of CaCl2·2H2O. The

composition of trace metal solution was as follows (per liter): 1.5 g of nitrilotriacetic acid;

0.1 g of FeSO4·7H2O; 0.5 g of MnSO4·2H2O; 1.0 g of NaCl; 0.1 g of CoCl2; 0.1 g of

CaCl2·2H2O; 0.1 g of ZnSO4·5H2O; 0.01 g of CuSO4·5H2O; 0.01 g of AlK(SO4)2; 0.01 g of H3BO3; 0.01 g of Na2MoO4·3H2O. The vitamin solution contained the following (per

liter): 5 mg of thiamine-HCl; 5 ml of riboflavin; 5 mg of nicotinic acid; 5 mg of

Capantothenate; 0.1 mg of vitamin B12; 5 mg of p-aminobenzoic acid; 5 mg of lipoic acid.

Glucose or xylose was added to the basal medium for the fermentation study. The pH of

medium was adjusted to 6.0 except for the pH experiments.

202

Appendix B Analysis Methods

B.1 High Performance Liquid Chromatography

The high performance liquid chromatography (HPLC) system consisted of an

automatic injector (Shimadzu SIL-10Ai), a pump (Shimadzu LC-10Ai), an organic acid

analysis column (Bio-Rad HPX-87H), a column oven at 45°C (Shimadzu CTO-10A), and

a reflective index detector (Shimadzu RID-10A). The eluent was 0.01 M H2SO4 at a flow

rate of 0.6 ml/min. Samples with a volume of 15 µl were injected by the auto-sampler.

The running time was 25 min. The AUX RANGE parameter of RI detector was set at 2.

Peak height was used to calculate the concentration of each component based on the

analysis of the standard mixture containing all the compounds at 2 g/L.

Samples of cell-free fermentation broth were diluted with distilled water at

different ratios from 1/6 to 1/20 depending on the concentration of the compounds to be

analyzed. A typical chromatogram of the fermentation sample with xylose as the

substrate is shown below.

B.2 Gas Production by Micro-oxymax

An on-line respirometer Micro-oxymax system equipped with both H2 and CO2 sensors (Columbus Instrument, OH) was used for automatic measurement of gas (H2 and

CO2) production during the fermentation of C. tyrobutyricum. The fermentor was

connected to a tightly sealed bottle with a volume of 5400 ml used for the collection of

203

the gas produced, which was diluted in this bottle with air. The change of gas concentration inside the bottle was measured periodically by the Micro-oxymax system.

Before starting each experiment, calibration with standard gas containing 1.8% of

H2 and 4.0% of CO2 was carried out. The parameters of Micro-oxymax were set as:

refresh volume 6; refresh flowrate 2 L/min; refresh threshold% 0; interval 1; window

auto. The total volume of gas produced in the fermentation was calculated from the data

of the pressure and gas composition in the bottle.

B.3 Enzyme Assay and SDS-PAGE

Cells cultivated in 100 mL of CGM at 37 °C were allowed to grow to the

exponential phase (OD600 = ~1.5), and then harvested and washed. The cell pellets

suspended in 10 mL of 25 mM Tris-HCl buffer (pH 7.4) were sonicated, and the protein

extract was collected by centrifugation. The protein extract was then used in PTA and

AK activity assays and all these operations were done under ambient conditions. For

hydrogenase activity assay, cells were suspended in 1 mL of TE buffer (10 mM Tris-HCl,

1 mM EDTA, pH 8.0) and lysed at 37 oC for 30 min with mutanolysin (100 µg/mL;

Sigma). The cell debris was then removed by centrifugation. All procedures for the

hydrogenase assay were carried out in the anaerobic chamber. The protein content in the

cell extract sample was determined following standard Bradford protocol (Bio-Rad,

Hercules, CA).

204

The activity of PTA was measured spectrophotometrically at 405 nm by detecting the liberation of CoA from acetyl-CoA, following the protocol of Andersch (Andersch, et al., 1983). One unit of PTA activity is defined as the amount of enzyme converting 1

µmol of acetyl-CoA per minute. The activity of AK was assayed using potassium acetate as substrate, by the method of Rose (Rose, 1955). One unit of AK is defined as the amount of enzyme producing 1 µmol of hydroxamic acid per minute. Hydrogenase activity was detected using the procedure developed by Drake (Drake, 1982). One unit of hydrogenase activity is defined as 2 µmol of methyl viologen reduced (equivalent to 1

µmol of H2 oxidized) per minute. Specific enzyme activity was calculated as the units of

activity per mg of protein. The specific enzyme activities in the mutant as compared with

the corresponding specific enzyme activities of the wild type were reported as the relative

enzyme activity (%) in this work.

Protein samples, 24 µg per well, were loaded into 12.5% SDS-PAGE gel and run

at 100 V for 2.5 h with PROTEAN II xi Cell (Bio-Rad). The gel was stained and

distained following the instruction of the manufacturer.

B.4 Two-dimensional Protein Electrophoresis

For two-dimensional protein electrophoresis (2DE) analysis, cell extract was

concentrated by acetone and then dissolved in rehydration buffer (8 M urea, 4% CHAPS,

10 mM DTT, 0.2% (w/v) Bio-Lytes 3/10) for sample preparation. The first dimension

was performed on a 7 cm IPG strip with a nonlinear immobilized pH 3-10 gradient

205

(Amersham). The IPG strip was rehydrated in rehydration buffer with 6 µg protein sample at 50 V for 12 h using PROTEAN IEF Cell (Bio-Rad). After rehydration, the protein was focused on IPG strip by preset method, at 250 V for 15 min to remove excess salts, then ramped linearly from 250 V to 4000 V for 2 h, and finally maintained at 4000

V for 5 h for focusing purpose. After isoelectric focusing (IEF), the strip was equilibrated in equilibrated buffer I (6 M urea, 2% SDS, 0.375 M Tris-HCl, pH 8.8, 20% glycerol and

130 mM DTT) for 10-15 min and in equilibrate buffer II (6 M urea, 2% SDS, 0.375 M

Tris-HCl, pH 8.8, 20% glycerol and 135 mM iodoacetamide) for 10-15 min. The equilibrated strip was applied to a polyacrylamide/PDA SDS gel to run the second dimension electrophoresis at 100 V for 90-120 min with Mini-PROTEAN 3 Cell (Bio-

Rad). The protein spots were developed using silver staining kit (Amersham). The two- dimensional protein electrophoresis gels were analyzed using Phoretix 2D AdvancedTM software (Nonlinear Dynamics Ltd, Newcastle upon Tyne, UK).

B.5 Southern Hybridization

Restriction enzyme SmaI was used to digest the chromosomal DNA of both wild type and mutant completely at 30oC. After being separated on a 1% agarose gel with low voltage, all digested DNA fragments were transferred from the agarose gel to a Hybond-

N+ nylon membrane (Amersham, Piscataway, NJ) by upward Southern capillary transfer.

Pre-hybridization of blotted nylon membrane was performed at 50oC for 1 h. Two probes

were used separately for the hybridization of Emr gene and pta gene. The Emr probe was

206

prepared from HindIII-digested pPTA-Em followed with SacI digestion, resulting in a partial Emr gene fragment of ~345 bp. The gene probe was the same as the cloned pta

fragment from PCR. Both probes and the HindIII-digested λ DNA, which was used as

DNA-size marker, were labeled with alkaline phosphatase (Amersham). The

hybridization with the probes was carried out with gently shaking at 62oC overnight.

HyperfilmTM ECL (Amersham) was then used for the detection.

B.6 Metabolic Flux Analysis

1. Xylose or Glucose utilization (a) 3 Xylose + 5 NAD+ + 5 ADP→ 5 Pyruvate + 5 NADH + 5 H+ + 5 ATP

(a)’ glucose → 3 ATP + 1 butyric acid + 2 H2 + 2 CO2 2. Production of AcCoA

(b) Pyruvate + CoA + Fd → AcCoA + FdH2 + CO2

3. Production of H2

(c) FdH2 ⇔ H2 + Fd 4. NAD reduction + + (d) FdH2 + NAD ⇔ NADH +Fd + H 5. Production of acetate (e) AcCoA + Pi + ADP → Acetate + CoA + ATP 6. Production of butyrate (f) 2 AcCoA + 2 NADH + 2 H+ + Pi + ADP → Butyrate + CoA + 2 NAD+ + ATP

7. Production and utilization of lactate (g) Pyruvate + NADH + H+ → Lactate + NAD+ (g′) Lactate → Pyruvate

207

8. Biomass formation + + 4 Pyruvate + 5.75 NADH + 5.75 H + 33.7 ATP → 3 C4H4pO4nN4q + 5.75 NAD + 33.7 ADP

208

Appendix C Bioreactor Construction and Operation

C.1 Construction of Immobilized Cell Bioreactor

The immobilized cell bioreactor was made of a packed glass column with a water

jacket. A piece of cotton towel laid with a stainless steel mesh was spirally wound and

put inside the column for cell immobilization. The gap between the layers was about 5

mm. A 1″ to 1.5″ layer of rasching rings was filled in the bottom of column to support the

spirally wound matrix and create a homologous flow distribution. The packed glass

column was sealed with rubber stoppers at both ends and connected to a 5-L fermentor

(Marubishi MD-300). The fibrous bed bioreactor was maintained at the optimal growth

temperature for the specific microorganism by circulating water with temperature control

through the water jacket in the glass column.

C.2 Bioreactor Start-up and Operation

Fed-batch fermentations of C. tyrobutyricum were performed in a 5-L stirred-tank

fermentor (Marubishi MD-300) containing 2 L of the CGM medium and substrate agitated at 150 rpm with pH and temperature controls. For the repeated-batch fermentations with immobilized cells, a 0.5-L fibrous-bed bioreactor (FBB) with a working volume of ~500 mL was made of a glass column packed with spiral wound cotton towel, which was described elsewhere (Silva and Yang, 1995). The anaerobiosis was reached by sparging the fermentor medium and the FBB with N2. The 100 mL of cell

209

suspension prepared in serum bottle was inoculated into the fermentor. After 3 days growth, the cells with medium in the fermentor were circulated through the fibrous bed and immobilized in the fibrous matrix. Most of the cells were immobilized and there was no change in cell density in the fermentor broth after 2 days and the spent medium was then replaced with fresh medium. The fed-batch mode was operated by pulse feeding concentrated substrate solution when the sugar level in the fermentation broth was close to zero. The Micro-oxymax system (Columbus Instrument, Columbus, OH) connected to the fermentor was used to measure the production of H2 and CO2. Samples were taken at

regular intervals from the fermentor for the analyses of cell density, substrate and

products concentration.

210

Appendix D Protocols of Functional Genomic and Proteomics

D.1 Genome Library Construction

A random genomic library for C. tyrobutyricum has been constructed. More than

20,000 white colonies in one standard transformation reaction using E. coli DH5α

competent cells have been obtained. The white colonies were picked from the agar plates

and stored in 96-well plates at -80oC with 15% glycerol for further study. The library was

created as follows. Bacterial chromosomal DNA was partially digested with Sau 3AI, and

agarose gel-purified fragments from 2-5 kb was ligated to BamHI completely digested

plasmid pXL with dephophosphorylation. The ligation mixture was used to transform

Max E. coli DH5α competent cells (Invitrogen), and the clones were plated on LB with

50 µg/ml Kanamycin. Since most of Clostridia have genomic size between 3 Mb to 4 Mb, we can assume a chromosomal size of 4 Mb. The total number of clones (n) needed to construct the gene library can be estimated as n = ln(1-p)/ln(1-f), where f = 0.00075

(fractional portion of genome present in a single DNA fragment or the average size of the insert (3 kb) divided by the haploid genomic size, 4 Mb) and p is the probability of genes of interest present in the gene library. Depending on the size and probability of each

DNA fragment inserted into the plasmid, the estimated number of clones is between 4000

(for p = 0.95) and 9200 (for p = 0.999). More clones (white colonies) would be needed since redundancy factor and the partial digestion factor need to be multiplied to ensure the complete gene library. Therefore, totally 12,000 white color colonies should be enough to cover the whole genomic DNA. Some randomly picked DNA insert was

211

checked by minipreps and the results showed that the redundancy of the C. tyrobutyricum gene library was low, indicating that the gene library should cover the entire genome.

A. Gene library construction B. Microarray chip construction C. Hybridization and analysis

C. tyrobutyricum cell Chromosomal DNA isolation Sample mRNA extraction Reference Insert amplification by PCR Colony Chromosomal DNA Reverse transcription (partial) digestion mRNA mRNA mRNA labeling

DNA fragment (insert) Insert Insert purification Ligation

Cloning vector Green Cy3 Red Cy5 Plasmids Array printing Probe Transformation Competitive E. coli DH5α hybridization

White/blue screen Arrayer Colony transfer to 96 format Store arrayer slides Scan and analyze

Store 96 well plates

Sequence 2D/MS G A T C Analyze Slide storage cassette Gene library

Figure A.1 Shotgun DNA microarray. A. Construction of gene library; B. Construction of microarray gene chips; C. RNA labeling, chip hybridization, and data analysis.

D.2 DNA Microarray Construction

Colonies were cultivated, processed, and the insert DNA in each clone was amplified using 96-well PCR. Two oligonucleotide pairs were used as the universal primers for the PCR: M13 forward primer (5’-GTAAAACGACGGCCAG-3’) and M13

212

reverse primer (5’-CAGGAAACAGCTATGAC-3’). These two primers can bind to vector sequences bordering each probe sequence, and thus allow for amplification of all target DNAs with a single primer pair. The PCR products were checked by agarose gel electrophoresis in a mini-ready-to-run system (Amersham Biosciences) and purified by using a 100 µl PCR purification kit (TeleChem International, Sunnyvale, CA). Genomic

DNA microarray with spots representing 3,000 inserts (1,500 per slide) from the genomic library, approximately one-fourth of the C. tyrobutyricum genome, will be printed on silylated slides (TeleChem International) with (75 µm spot size with 30 µm spacing) using an arrayer at 45-50% humidity, and the slides will be processed following the standard procedure online (http://derisilab.ucsf.edu/arraymaker.shtml). This will be done with a high-precision fast motion (3000 status per second) arrayer with 48 printing heads in our lab. Each slide containing at least 1,500 spots will be constructed with controls located on different parts of the chip. Some of these control spots include PCR-amplified

DNA fragments codifying the 16S and part of the 23S rRNA as positive controls and eukaryotic DNA (e.g., herring sperm DNA) with no known homologies to the C. tyrobutyricum genome sequence as negative control. On-chip internal DNA markers will be chosen from several known genes involved in acetic acid metabolic pathway, such as pta encoding phosphotransacetylase, ack encoding acetate kinase, monophosphate dehydrogenase gene, arginine kinase gene, and enoate reductase gene (Rohdich et al.,

2001). Complete or partial DNA sequences of these genes are available from previous research. The whole PCR-amplified DNA fragments containing these genes will be spotted at different concentrations (from 50 to 250 ng/µl) in different places on the chip

213

and used as internal markers for primary metabolism. After spotting, the slides will be

UV cross-linked and baked in an oven at 80oC for 2 to 4 h (Tomas, et al., 2003).

D.3 RNA Extraction, Labeling, and Chip Hybridization

Standard procedures available in the literature and public websites will be

followed. Briefly, cells cultured under different conditions will be harvested and the total

RNA will be extracted using a Qiagen (Valencia, CA) RNA extraction kit. RNA will be

treated with DNase I, RNase-free, followed by phenol and chloroform extraction. Total

RNA quality will be checked by spectrometry. The RNA samples from control and test

conditions will be labeled by cDNA synthesis with reverse transcriptase (SuperScript II,

Invitrogen) and Cy3- and Cy5-dUTP primed with random hexamers

(http://cmgm.stanford.edu/pbrowmn/protocols/4_Ecoli_RNA.txt;

http://www.microarrays.org/pdfs/amino-allyl-protocol.pdf). Chip hybridization (55oC

from 6 to 12 h) will then be carried out (www.arrayit.com). Printed microarrays will be washed first to remove unbound materials and the double-stranded DNAs will be denatured by boiling for 3 min in distilled water. The microarrays will be hybridized in hybridization cassettes (TeleChem) for 5 h at 42°C under 18 mm x 18 mm optically flat cover slips with 5.0 µl probe solution containing 5X SSC + 0.2% SDS + 0.2 mg/ml bovine serum albumin (BSA) + 2 µM 15-mer oligonucleotide containing either a Cy3 or

Cy5 label on the 5' end. Hybridized microarrays will be washed twice for 5 min each in

2X SSC + 0.2% SDS at 25°C, once for 1 min in 2X SSC at 25°C and spun dry for 1 min at 500 x g. Probe pre-heating and BSA addition can both improve microarray data.

214

D.4 Scanning and Data Analysis

Microarrays will be scanned with Affymetrix 428 scanner, which is a confocal laser microscope with a flying head, i.e. the objective lens scans over the slide. The 428 scanner can analyze any microarray on a glass slide (25 x 75 x 1mm) that has been probed with DNA labeled with Cy3 or Cy5 as well as other common fluorescence dyes.

The slide is inserted into the scanner, and the resulting fluorescence is measured after excitation at two different wavelengths: 532nm and 635nm. Subsequently, the files can be analyzed with additional software to compare the signal from the two different dyes and therefore differences between the samples. The Affymetrix software (Microarray

Suite 5.0, MicroDB 3.0 and Data Mining Tool 3.0) will be used to analyze the spotted microarray images and data (Plant-Microbe Genomics Facility, The Ohio State

University, http://www.biosci.ohio-state.edu/~pmgf/428_scanner.htm). Images also can be overlapped and analyzed with SCANALYZE2 software

(http://rana.lbl.gov/EisenSoftware.htm) or other free software. Because this is the first

time to apply the DNA microarray to C. tyrobutyricum and there is no previous data

knowledge to be input as a teacher signal, we have to use the following unsupervised

method for data mining:

a) Principle component analysis: a data reduction technique used to identify

uniquely expressing genes (Rnudsen, 2002);

b) Clustering: to identify clusters of coexpressed genes, such that gene in one cluster

relate to a common biological phenomenon, while gene in different clusters relate to

different phenomena (Amaratunga and Dhammika, 2004).

215

D.5 DNA Sequencing and Analysis

The clones with large expression difference analyzed by shotgun microarray will be identified by further sequencing. The Sequence reactions will be made from plasmid minipreps using dye terminator cycle sequencing reactions and run with automated 3730

DNA Analyzer (Applied Biosystems) and BigDye Terminator Cycle Sequencing chemistry. This instrument utilizes fluorescent dye labeled nucleotides and provides high volume capabilities, e.g. 384 sequencing reactions in 12 h, and typically yields 650 to 750 bases per reaction. Sequences will be analyzed by using software Sequencing Analysis version 5.1 and Blast analysis will be run using NCBI to search the possible function

(Plant-Microbe Genomics Facility, The Ohio State University, http://www.biosci.ohio- state.edu).

D.6 2D-PAGE Data Analysis

The protein electrophoresis gels will be analyzed using Phoretix 2D AdvancedTM software (Nonlinear Dynamics Ltd). The molecular mass and PI value of all protein spots will be identified. The protein expression level detected from the 2D gels will be analyzed and a database of all the proteins present in the cell extracts will be constructed.

By comparing the protein maps between different samples derived from various treatments, we can identify the difference in protein expression. These proteomic data will be compared with the Microarray (mRNA) data, and the results can be used to find interested genes.

216

D.7 Protein Separation and Purification

Targeted proteins on the 2D gels will be isolated, sequenced, and identified. For protein identification, 200 µg of cellular protein will be loaded onto each IPG strip. The slab gel proteins will be transferred to an Immobilon-P polyvinylidene difluoride membrane (Millipore) using the ESA Investigator graphite electroblotter, type II

(Genomic Solutions). The membrane will be stained with Coomassie blue, and protein spots excised for sequence determination. Protein spots cut from the transfer membrane will be washed three times in 10% methanol and then dried and stored at -70oC.

D.8 Protein Sequencing by Mass Spectrometer and Analysis

The protein analysis procedure will be done at the Mass Specrometry &

Proeomics Facility at the Ohio State University (http://www.ccic.ohio- state.edu/MS/services.htm). Proteomics is a service offered at OSU through collaboration

between the CCIC-MS Facility and the Plant-Microbe Genomics Facility. The PMGF

services include separation of protein mixtures on 2D gels, robotized analysis of spots

and specific coring of the gels using the BioRad Proteome Works Station. The Coomassie

stained gel cores are sent to the CCIC Mass Spectrometry Facility for subsequent

digestion, extraction and peptide fingerprint analysis using a Bruker Reflex III MALDI-

TOF-MS or a Micromass Q-TOF II with capillary LC/MS/MS capabilities.

Mass spectral data will be obtained using a Micromass Tof-Spec 2E instrument

equipped with a 337-nm N2 laser at 20-35% power in the positive ion reflectron mode.

Spectral data will be obtained by averaging 10 spectra each of which is the composite of

217

10 laser firings. The mass axis will be calibrated using known peaks from tryptic autolysis. Peptide mass fingerprinting will be used for protein identification from tryptic fragment sizes by using the MASCOT search engine (www.matrixscience.com) based on

the entire NVBInr protein data base Mass tolerance of 150 ppm is the window of error to

be allowed for matching the peptide mass values. Probability-based MOWSE scores will

be estimated by comparison of search results against estimated random match population

(Thongboonkerd et al., 2002).

D.9 Mutant Library Construction

Figure A.2 shows the construction of a non-replicative integrational plasmid used in creating the gene-knock out mutants in our lab. In this project, we will modify the plasmid to posses the capability to accept (ligate with) any DNA sequence with a high efficiency similar to the plasmid cassettes used in TA cloning or PCR. The basic idea is to randomly knock out the genes on the chromosome using integrational mutagenesis, develop the high-throughput screening method to obtain the mutant that we want, further study the functional genomics and proteomics of the mutants. The random DNA inserts are the fragments amplified from gene library using PCR. The constructed plasmids will be used to transform the C. tyrobutyricum cells and inactivate the homologous genes on the chromosome.

218

Not I

HindIII EcoR V insert SphI

Sac I

Emr Apr

ColE1 SmaI HindIII

Figure A.2 Mutant library construction.

219

Appendix E Reagents and Buffers

EDTA (pH 8.0), 0.5 M

Dissolve 18.61 g EDTA in 80 ml distilled water and adjust pH to 8.0 with solid

NaOH. Mix and add distilled water to 100 ml.

Ethidium Bromide (1000x)

Dissolve 0.05 g ethidium bromide in 100 ml distilled water. Mix well and store it in the dark.

TE Buffer

10 mM Tris/HCl with adjusted pH of 8.0 and 1 mM EDTA (pH 8.0).

TAE Electrophoresis Buffer, 50x

Dissolve 242.28 g Tris base, 18.61 g EDTA, 34.02 g CH3COONa·3H2O in 50 ml

distilled water and adjust pH to 7.6 with concentrated HCl. Mix and add distilled water to

1 liter.

Tris/HCl, 1M

Dissolve 12.114 g Tris base in 80 ml distilled water and adjust to the desired pH

with concentrated HCl. Mix and add distilled water to 100 ml.

220