2,3-Butanediol production using

acetogenic bacteria

Dissertation

Submitted for the fulfillment of the requirements for the doctoral degree Dr. rer. nat. at the Faculty of Natural Sciences,

Ulm University

Catarina Erz

from

Stuttgart – Bad Cannstatt

2017

The present study was performed at the Institute of Microbiology and Biotechnology, Ulm University, under the direction of Prof. Dr. Peter Dürre.

Faculty dean in office: Prof. Dr. Peter Dürre First reviewer: Prof. Dr. Peter Dürre Second reviewer: Prof. Dr. Bernhard J. Eikmanns

Date of the doctoral examination: October 11 th , 2017

Content

Content

Abbreviations ______I

1. Introduction ______1

2. Material and Methods ______10

2.1 Microorganisms, plasmids, and primers ______10

2.1.1 Bacterial strains ______10

2.1.2 Vectors and plasmids ______11

2.1.3 Primers ______14

2.2 Chemicals, enzymes, and gases ______17

2.3 Bacterial cultivation ______19

2.3.1 Cultivation media ______19

2.3.1.1 Modified 1.0 ATCC 1612 medium ______19

2.3.1.2 Modified ATCC 1612 medium ______22

2.3.1.3 Lyophilization medium ______22

2.3.1.4 Lysogeny broth medium -modified- ______23

2.3.1.5 Rajagopalan medium -modified- ______23

2.3.1.6 Super optimal broth medium -modified- ______24

2.3.1.7 Tanner medium -modified- ______25

2.3.1.8 Yeast tryptone fructose medium -modified-______25

2.3.2 Antibiotic selection ______26

2.4 Growth conditions ______27

2.4.1 Aerobic cultivation ______27

2.4.2 Anaerobic cultivation ______27

2.4.3 Strain conservation ______28 Content

2.4.3.1 Glycerol ______28

2.4.3.2 SMP buffer/DMSO ______28

2.4.3.3 Lyophilization ______29

2.4.4 Determination of growth and metabolic production parameters ______29

2.4.4.1 Determination of optical density ______29

2.4.4.2 Quantification of substrate and microbial metabolic products by high performance liquid chromatography ______30

2.4.4.3 Quantification of microbial metabolic products by gas chromatography _ 31

2.5 Nucleic acid methods______32

2.5.1 Nucleic acid isolation ______32

2.5.1.1 Plasmid isolation from E. coli using the Zyppy TM Plasmid Miniprep Kit ___ 32

2.5.1.2 Plasmid isolation from E. coli using QIAGEN Plasmid Midi Kit ______32

2.5.1.3 Genomic DNA isolation from Gram positive bacteria using MasterPure TM Gram Positive DNA Purification Kit ______33

2.5.1.4 RNA isolation from clostridia ______33

2.5.2 Polymerase chain reaction ______35

2.5.2.1 Composition of PCR reaction mixtures with different DNA polymerases _ 35

2.5.2.2 Standard PCR program ______36

2.5.2.3 Insertion of restriction recognition sequences via PCR ______37

2.5.3 DNA modification ______37

2.5.3.1 Restriction enzyme digestion of DNA ______37

2.5.3.2 Dephosphorylation of digested plasmids ______38

2.5.3.3 DNA ligation ______38

2.5.3.4 In-Fusion® HD Cloning Kit ______39

2.5.3.5 Radioactive labelling of DNA fragments ______39 Content

2.5.4 Purification of nucleic acids ______40

2.5.4.1 Purification of amplified and digested DNA ______40

2.5.4.2 Purification of radioactively labelled DNA ______41

2.5.4.3 Purification of ligated DNA by microdialysis ______41

2.5.5 Gel electrophoresis ______41

2.5.5.1 Non-denaturating gel electrophoresis ______41

2.5.5.2 Denaturating gel electrophoresis ______42

2.5.6 Determination of nucleic acid concentration ______43

2.6 DNA sequencing and data analysis ______43

2.6.1 Sequencing of vectors and DNA fragments ______43

2.6.2 Genome sequencing ______44

2.6.3 Mapping for single nucleotide polymorphism (SNP) analysis ______44

2.6.4 Comparison of locus tags of different acetogens ______44

2.7 DNA transfer in bacteria ______45

2.7.1 DNA transfer in Escherichia coli ______45

2.7.1.1 Preparation of electrocompetent E. coli cells ______45

2.7.1.2 Preparation of fast competent E. coli cells ______45

2.7.1.3 Transformation of electrocompetent E. coli cells ______45

2.7.1.4 Preparation of chemically competent E. coli cells ______46

2.7.1.5 Transformation of chemically competent E. coli cells ______47

2.7.2 DNA transfer in A. woodii and Clostridium by electroporation______47

2.7.2.1 Preparation of electrocompetent A. woodii and Clostridium cells ______47

2.7.2.2 Transformation of electrocompetent A. woodii and Clostridium ______47

2.7.3 DNA transfer in Clostridium by conjugation ______48

2.7.4 Northern blot ______49 Content

2.7.4.1 Transfer of RNA onto a nylon membrane ______49

2.7.4.2 Hybridization of radioactively labelled probe with RNA ______50

3. Results ______53

3.1 Improvement of 2,3-BD production in acetogenic bacteria ______53

3.1.1 Natural 2,3-BD production in acetogenic bacteria ______53

3.1.2 Construction of plasmids for enhancement of natural 2,3-BD production __ 54

3.1.3 Homologous 2,3-BD overproduction in C. ljungdahlii ______56

3.1.4 Homologous 2,3-BD overproduction in C. coskatii ______60

3.1.5 Modifications of 2,3-BD production plasmid pJIR750_23BD ______63

3.1.5.1 Removal of a potential terminator structure ______64

3.1.5.2 Exchange of bdh with adh (primary-secondary alcohol dehydrogenase) _ 66

3.1.5.3 BudABC operon from Raoultella terrigena ______68

3.1.5.4 Growth experiments of different C. ljungdahlii strains harboring the modified 2,3-BD plasmids ______70

3.1.6 Sequencing of C. ljungdahlii [pMTL82151] and C. ljungdahlii WT with SNP analysis ______75

3.1.7 Detection and quantification of transcripts responsible for 2,3-BD production ______77

3.1.8 Heterologous 2,3-BD production in A. woodii ______79

3.1.9 Enhancement of 2,3-BD production by overexpressing nifJ ______82

3.2 Heterologous butanol production ______90

4. Discussion ______95

4.1 Natural microbial 2,3-BD production ______95

4.2 Enhancement of 2,3-BD production using acetogens ______103 Content

4.2.1 Enhancement of 2,3-BD production via overexpression of alsS , budA , and bdh ______103

4.2.2 Modifications of 2,3-BD production plasmid to optimize 2,3-BD production 107

4.2.3 PFOR enzyme as potential bottle-neck in 2,3-BD production? ______112

4.3 Detection of transcripts from synthetic 2,3-BD operons in C. ljungdahlii 116

4.4 Investigation of mutated C. ljungdahlii [pMTL82151] showing increased flux

towards ethanol and 2,3-BD ______118

4.5 A. woodii as suitable host for 2,3-BD production from H 2 + CO 2? ______122

4.6 Heterologous butanol production in C. ljungdahlii - the conflict of large

plasmid transformation ______124

5. Summary ______127

6. Zusammenfassung ______129

7. References ______132

8. Supplement ______159

Contribution to this work ______165

Danksagung ______167

Curriculum vitae ______171

Statement______173

Abbreviations I

Abbreviations

A adenosine A. Acetobacterium ; Aeromonas ; Aspergillus AA amino acid abbr. abbreviation ad . addiere (fill up to) ADP adenosine diphosphate AG Aktiengesellschaft (joint-stock company) AG & Co. Aktiengesellschaft (joint-stock company) & Compagnie (trading company) ATCC American Type Culture Collection ATP adenosine triphosphate B. Bacillus ; Butyribacterium BLAST Basic Local Alignement Search Tool Bp base pair(s) 2,3-BD 2,3-butanediol C cytosine c centi (10 -2) C. Clostridium , Corynebacterium °C degree(s) Celsius

C1 compound with one carbon

C4 compound with four carbons CA California cal calory CGMCC China General Microbiological Culture Collection Center CICC China Center of Industrial Culture Collection CO carbon monoxide

CO 2 carbon dioxide D D-isomer Δ delta; gene deletion DAD diode array detector [α-32 P]-dATP deoxyadenosine triphosphate, labeled on the alpha phosphate group with 32 P II Abbreviations dATP deoxyadenosine triphosphate DMSO dimethylsulfoxide DSMZ Deutsche Sammlung von MIkroorganismen und Zellkulturen (German Collection of Microorganisms and Cell Culture) DNA deoxyribonucleic acid dNTP deoxyribonucleotide triphosphate E. Escherichia ; Enterobacter EDTA ethylenediaminetetraacetic acid et al. et alii (and others) E-value expect value Fd ferredoxin Fd 2- reduced ferredoxin FID flame ionization detector fwd forward G guanine g gram GC gas chromatography GmbH Gesellschaft mit beschränkter Haftung (limited liability company) h hour(s) H+ proton(s)

H2 hydrogen

H2O water HPLC high-performance liquid chromatography ID inner diameter Inc. Incorporated IPTG isopropyl β-D-1-thiogalactopyranoside K kilo (10 3) K. Klebsiella L L-isomer l liter λ lambda; wavelength LB Luria-Bertani; lysogeny broth Abbreviations III

Ltd. Limited M molar m milli (10 -3); meter µ micro (10 -6) MA Massachusetts max. maximal MES 2-(N-morpholino)ethanesulfonic acid min minute mod. modified n nano (10 -9)

N2 nitrogen Na + sodium ion NAD + oxidized nicotinamide adenine dinucleotide NADH reduced nicotinamide adenine dinucleotide NADP + oxidized nicotinamide adenine dinucleotidephosphate NADPH reduced nicotinamide adenine dinucleotidephosphate NC North Carolina NH New Hampshire NJ New Jersey

O2 oxygen

OD 600nm optical density at 600 nm P promoter P. Paenibacillus PCR polymerase chain reaction pH negative decade logarithm of the proton concentration ( potentia hydrogenii )

Pi inorganic phosphate PIPES 1,4-piperazinediethanesulfonic acid R. Raoultella rev reverse RID refraction index detector rpm rounds per minute s second IV Abbreviations

S. Saccharomyces ; Serratia ; Synechococcus ; Synechocystis SMP sucrose-magnesium-phosphate SNP single nucleotide polymorphism SOB super optimal broth sp. Species SSF simultaneous saccharification and fermentation subsp. subspecies Syngas synthesis gas T terminator; thymine TE tris-EDTA TM trademark TSE tris-sucrose-EDTA U uracile UV ultraviolet V volume w weight WI Wisconsin WT wild type x g times gravity (unit of relative centrifugal force)

1. Introduction 1

1. Introduction

There are four stable isomers of butanediol (BD): 1,2-BD, 1,3-BD, 2,3-BD, and 1,4-BD. Two of these four different isomers are important for commercial use. On the one hand, 1,4-BD is a significant bulk chemical, with production of 1.3 million tons per year from fossil resources (Zeng and Sabra, 2011). On the other hand, 2,3-BD is considered as an important fine and potential platform chemical, with impact on the specialty chemical market. It can be easily converted to butanes, butenes, and butadienes, which are building blocks used for production of specialty chemicals. The key downstream products of 2,3-BD have a potential global market of approximately 32 million tons per year with a value of around 43 billion dollars on the sales market (Köpke et al ., 2011b). Transparancy market research company reports that the global market size of 2,3-BD was estimated at over 61.8 kilo tons in 2012 and is predicted to expend to 74 kilo tons by 2018. There are three stereoisomers of 2,3-BD: the optically active forms 2S,3S-BD (dextrorotatory/L(+)-form) and 2R,3R-BD (levorotatory/D(-)-form), as well as the optically inactive form 2R,3S-BD (meso -form) (Figure 1). Since the levorotatory form has a low freezing point of -60 °C, it is considered to be used commercially as an anti-freeze agent (Kuhz et al ., 2017). Furthermore, 2R,3R -BD as well as 2S,3S -BD are excellent chiral compounds for asymmetric synthesis of liquid crystals (Xiao et al ., 2010; Liu et al ., 2011; Zeng and Sabra, 2011). Apart from applications of the optical forms, 2,3-BD can be converted to convenient derivatives via certain chemical reactions (Ji et al ., 2011a; Nie et al ., 2011; Gubbels et al ., 2013). The dehydration of 2,3-BD leads to the excellent organic solvent methyl ethyl ketone (MEK), which is used for resins, fuels, paints, and laquers (Tran and Chambers, 1987; Ji et al ., 2012; Zhang et al ., 2012b). Further dehydrating leads to 1,3-butadiene, finding applications in manufacturing synthetic rubbers, polyesters, and polyurethanes (Haveren et al ., 2008).

Figure 1: Stereoisomers of 2,3-butanediol 2 1. Introduction

Moreover, dehydrogenation of 2,3-BD can form acetoin and diacetyl, which are used for flavoring dairy products and margarines, giving a buttery taste (Bartowsky and Henschke, 2004; Faveri et al ., 2003; Ji et al ., 2013). Additionally, diacetyl can be applied as a bacteriostatic food additive, since it inhibits growth of some microorganisms (Celińska and Grajek, 2009). As a further chemical reaction, esterification of 2,3-BD can form a polyimide precursor, which finds application in lotions, drugs, and cosmetics. Other products formed by esterification of 2,3-BD with maleic acid are polyurethane-melamides (PUMAs), which are useful in cardiovascular applications (Petrini et al ., 1999). Not only dehydration and esterification are possible, but also polymerization of 2,3-BD. The polymerization leads to polyesters with high potential for industry, if they are bio-based and even bio-degradable (Aeschelmann and Carus, 2015; Hu et al ., 2016; Debuissy et al ., 2016). Furthermore, due to its high octane rating 2,3-BD might be a potential aviation fuel (Celińska and Grajek, 2009; Ji et al ., 2012) and it is useful as raw material in the manufacture of printing inks, pesticides, plasticizers, moisturizing and softening agents, explosives, fumigants, etc. (Magee and Kosaric, 1987; Garg and Jain, 1995; Syu, 2001; Kuhz et al ., 2017).

The chemical production of 2,3-BD is performed by removal of butadiene and isobutene from crack gases (product from steam cracking petroleum refining), whereby a C 4-hydrocarbon fraction is obtained, consisting of approximately 77 % butenes and 23 % of a mixture containing butane and isobutane (Gräfje et al ., 2000). The chlorohydrination of this fraction with a chlorine/water solution and subsequent cyclization of the chlorohydrins with sodium hydroxide leads to a butene oxide mixture. This mixture contains 55 % trans -2,3-butene oxide, 30 % cis-2,3-butene oxide, and 15 % 1,2-butene oxide (Gräfje et al ., 2000). By subsequent hydrolysis of the butene oxides, a mixture of butanediols is obtained and separation is performed by vacuum fractionation. By using an excess of water during hydrolysis, formation of polyethers is avoided. Thereby, the fractionation of butanediols is easier than of the butene oxide mixture (Gräfje et al ., 2000). The optically inactive meso -2,3-BD is obtained from trans - 2-butene via trans -2,3-butene oxide, whereas the racemic mixture of D(-)-2,3-BD and L(+)-2,3-BD is formed in an analogous manner from cis -2-butene via cis -2,3-butene oxide

(Gräfje et al ., 2000). Furthermore, MEK is also formed as a by-product (Myszkowski and Zielinski, 1965). 1. Introduction 3

The chemical chiral synthesis of 2,3-BD as well as its separation represent expensive steps. Thus, application of bacteria for biotechnological production of 2,3-BD with a high enantiomeric purity reveals an alternative, sustainable, and competitive approach. Moreover, bio-based 2,3-BD production would be independent from oil supply. This economic aspect has boosted the overall interest in the biotechnological process recently. The microbial 2,3-BD production has a history of more than 100 years. In 1906, Harden and Walpole reported for the first time microbial 2,3-BD production in Klebsiella pneumoniae , formerly known as Aerobacter aerogenes and Klebsiella aerogenes (Harden and Walpole, 1906). At that time, their method was very cost-effective and less expensive than chemical synthesis. In 1926, 2,3- BD accumulation was also detected in Paenibacillus polymyxa (formerly Bacillus polymyxa ; reclassified by Ash et al ., 1993) by Garg and Jain (Garg and Jain, 1995). The first microbial industrial-scale production of 2,3-BD was proposed in 1933 (Fulmer et al ., 1933). During World War II there was a lack of 1,3-butadiene (pre-curser for synthetic rubber), which highly promoted interest in 2,3-BD fermentation. Thus, the peak of pilot-scale development for manufacturing 2,3-BD and its conversion to 1,3-butadiene was reached by this time (Ji et al ., 2011a). However, it suddenly ended when less expensive ways for 1,3-butadiene production by chemical synthesis with petroleum as feed-stock were available (Ji et al ., 2011a; Białkowska et al ., 2016). The advantage of the chemical method over biotechnological production lasted until the mid-1970s, since crude oil prices highly increased due to gradual depletion of that feedstock. This again promoted interest in 2,3-BD production from biomass (Ji et al ., 2011a; Białkowska et al ., 2016). Nowadays, fossil fuel prices remain very unstable, chemical compounds catalyzing the unique diol structure during chemical synthesis become more and more expensive, and the petrochemical synthesis requires high energy input. This shows that the focus needs to shift towards a sustainable microbial 2,3-BD production. Until today, the best 2,3-BD production strains are sugar- or citrate-fermenting bacteria with the majority belonging to risk group 2 organisms, e. g. Klebsiella sp. and Serratia marcescens with 150 g/l and 152 g/l, respectively (Ma et al ., 2009; Zhang et al ., 2010a). However, application of these organisms in industrial processes is unfavorable due to safety requirements of risk group 2 organisms (Celińska and Grajek, 2009; Ji et al ., 2011a). One sustainable alternative is represented by acetogenic bacteria (acetogens), which are obligate anaerobe organisms capable of using CO or/and CO 2 + H2 as sole carbon and energy source to produce the central intermediate acetyl-CoA via the reductive acetyl-CoA pathway, which is also known as Wood- 4 1. Introduction

Figure 2: Overview of the Wood -Ljungdahl pathway of acet ogens with enzymes and natural products starting from the central intermediate acetyl-CoA with focus on 2,3-BD production. Rnf, Rhodobacter nitrogen fixation; ATPase, adenosine triphosphatase; CODH, carbon monoxide dehydrogenase; Acs/CODH, complex of acetyl-CoA synthase with carbon monoxide dehydrogenase; THF, tetrahydrofolate; [CO], enzyme-bound CO; Fdh, formate dehydrogenase; Fhs, formyl-THF synthetase; Mtc, methenyl-THF cyclohydrolase; Mtd, methylene-THF deyhdrogenase; Mtr, methylene-THF reductase; Met, methyl transferase; AlsS, acetolactate synthase; BudA, acetolactate decarboxylase; Bdh, 2,3-BD dehydrogenase; sAdh, primary-secondary alcohol dehydrogenase; Co- FeS-P, corrinoid iron-sulfur protein; HS-CoA, coenzyme A; [H], reducing equivalent.

Ljungdahl pathway (Wood, 1991; Ragsdale and Pierce, 2008; Figure 2). This pathway is considered to be one of the oldest biochemical pathways in life and was first described in Moorella thermoacetica (formerly known as Clostridium thermoaceticum ) (Fontaine et al ., 1942; Daniel et al ., 1990) by Harland Goff Wood and Lars Gerhard Ljungdahl (Ljungdahl, 1986; Wood, 1991; Drake et al ., 2008). The Wood-Ljungdahl pathway is one of six pathways capable 1. Introduction 5

of fixing CO 2, but instead of being circular it follows a linear manner (Fuchs, 2011). It consists of the methyl and carbonyl branch (Figure 2), leading to the central intermediate acetyl-CoA.

In the methyl branch, CO 2 is reduced stepwise to methyl-tetrahydrofolate (THF) via the enzymes formate dehydrogenase (Fdh), formyl-THF synthetase (Fhs), methenyl-THF cyclohydrolase (Mtc), methylene-THF dehydrogenase (Mtd), and methylene-THF reductase (Mtr) (Ljungdahl, 1986; Ragsdale and Pierce, 2008). Activation of formate to formate-THF by the enzyme formyl-THF synthetase requires one mol of ATP. After binding the methyl-group of methyl-THF to a corrinoid iron-sulfur protein (CoFeS-P) in the last step of the methyl branch, the enzyme complex Acs/CODH (acetyl-CoA synthase/carbon monoxide dehydrogenase) merges the enzyme-bound methyl group (methyl branch), CO (carbonyl branch), and coenzyme A (CoA) to the central intermediate acetyl-CoA. Afterwards, it is either used for anabolism (production of biomass) or converted to other products (Ragsdale, 2007). The Rnf complex (Rhodobacter nitrogen fixation; ferredoxin:NAD + oxidoreductase) plays an important role for energy conservation in acetogens during autotrophic growth. It couples the transfer of electrons from reduced ferredoxin to NAD + with simultaneous translocation of cations across the cell membrane, leading to a cation gradient (Imkamp et al ., 2007; Müller et al ., 2008; Biegel and Müller, 2010; Tremblay et al ., 2012; Hess et al ., 2016). Afterwards, this gradient is used by an ATPase for generation of additional ATP (Reidlinger and Müller, 1994). Thereby, this system is also coupled to a flavin-based electron-bifurcating hydrogenase providing cations from oxidation of hydrogen with concomitant reduction of oxidized ferredoxin (Schuchmann and Müller 2012, Wang et al ., 2013a, Buckel and Thauer, 2013). Other acetogens such as Moorella thermoacetica harbour other systems involved in energy conservation, including a so-called energy-converting hydrogenase (Ech) complex, cytochromes and quinones (Schuchmann and Müller, 2014). Over 100 acetogenic species were identified to date, of which 90 % produce acetate as sole end product (Köpke et al ., 2011a). In addition, ethanol, butanol, butyrate, lactate, hexanol, and hexanoate can be by- products depending on the strain (Bruant et al ., 2010; Schiel-Bengelsdorf and Dürre, 2012; Phillips et al ., 2015; Ramió-Pujol et al ., 2015). Recently, 2,3-BD production was shown for C. ljungdahlii (Figure 3A), Clostridium autoethanogenum (Figure 3B), and Clostridium ragsdalei (1.4-2 mM) from steel mill waste gas (Köpke et al ., 2011b). These organisms mainly produce 6 1. Introduction the D(-)-form with 94 %, while only 6 % of meso -2,3-BD is formed. The production of 2,3-BD in C. autoethanogenum from pyruvate is catalyzed by the enzymes acetolactate synthase (AlsS), acetolactate decarboxylase (BudA), and 2,3-BD dehydrogenase (Bdh) or primary- secondary alcohol dehydrogenase (Adh) (Köpke et al ., 2011b; Köpke et al ., 2014). Acetogens being typically used for commercial syngas fermentation are C. ljungdahlii (Figure 3A) and C. autoethanogenum (Figure 3B), C. ragsdalei , Clostridium coskatii , Clostridium carboxidivorans , Clostridium aceticum , M. thermoacetica (formerly known as Clostridium thermoaceticum ), Acetobacterium woodii , and Butyribacterium methylotrophicum . They can use pure CO, or syngas, or other waste gases originating from e. g. steel mills or chemical production lines (Daniell et al ., 2016; Dürre, 2016a). Gas fermentation is a process showing several advantages such as the use of waste gases diminishing environmental pollution by reduction of industrial gas emissions. Reducing CO 2-emissions is a major topic of today’s society, due to increasing global warming. Furthermore, gas fermentation represents a non- cellulosic process not only saving food resources, but also agricultural land. Aerobic as well as anaerobic gas fermentation have both reached commercial level. Especially acetogens have attracted attention in gas fermentation, including industrial waste gases and synthesis gas

(syngas; a mixture of mainly CO and H 2), since these organisms are less sensitive to variations and contaminants in the composition of the gas as well as leading to a high product specifity

(Köpke et al ., 2011a; Schiel-Bengelsdorf and Dürre, 2012; Dürre and Eikmanns, 2015).

In addition to microbial 2,3-BD production, biotechnologically produced 1-butanol (butanol) is a valuable product for industry. Due to its high heating value, low corrosiveness compared to ethanol, low volatility, and energy density similar to gasoline, it is a more interesting biofuel

A B

2 µm 2 µm

Figure 3: Scanning electron microscope image of C. ljungdahlii (A) and C. autoethanogenum (B) growing with fructose as energy and carbon source. The white bar represents the scale. 1. Introduction 7 than ethanol. The four-carbon alcohol is commonly used as a chemical and solvent in paints, coatings, sealants, textiles, printing inks, glues, and plastics (Hahn et al ., 2000). In 2013, a study presented by the company Informa Economics revealed that the annual global market of butanol exceeded 3.6 million tons/year and was valued over 6 billion dollars at that time. Butanol was produced petrochemically in the past decades. In this process called oxo- synthesis, propylene is converted to butyraldehyde by using a homogenous catalyst via hydroformylation with CO. Subsequently, butyraldehyde is hydrogenated over a heterogenous catalyst to butanol. The Guerbet reaction represents a second pathway for chemical synthesis of butanol (O’Lenick, 2001; Kozlowski and Davis, 2013). This process consists of three steps: dehydrogenation of ethanol to acetaldehyde, acetaldehyde aldol- condensation to form crotonaldehyde, and hydrogenation of crotonaldehyde to form 1- butanol (Kozlowski and Davis, 2013; Sun and Wang, 2014). In order to achieve high activity and selectivity of the Guerbet reaction, the catalyst used in this process needs to have certain characteristics. On the one hand, it can be a basic agent like an alkali metal hydroxide, a salt dissolved in the reaction medium (homogenous catalyst), or a solid base (heterogenous catalyst) (Kuhz et al ., 2017). On the other hand, it needs to facilitate the dehydrogenation of ethanol at the respective reaction temperature. Examples for typical dehydrogenating agents are metals including platinum, nickel, and copper (Veibel and Nielsen, 1967; Kozlowski and Davis, 2013). If a homogenous catalyst is used in the formation of butanol, a precious metal catalyst such as an organometallic complex needs to be applied in order to perform the de/hydrogenation steps and an inorganic base aids in the aldol coupling step (Koda et al ., 2009; Chakraborty et al ., 2015). A different alternative to this three-step mechanism represents a coupling step without obtaining two molecules of acetaldehyde (Faba et al ., 2016). This process involves a direct surface coupling between the α-carbon of an aldehyde and one alcohol (Yang and Meng, 1993; Ndou et al ., 2003).

Instead of chemical synthesis, 66 % of butanol used worldwide was produced by acetone- butanol-ethanol (ABE) fermentation until 1950 (Dürre, 2008), which is one of the oldest bioprocesses in history. The first discovery of biological butanol production was found in Vibrion butyrique (Pasteur, 1862), but it propably was not a pure culture. Most likely it contained Clostridium butyricum , which is also able to produce butanol under certain conditions (Dürre, 2005). The most commonly used bacteria for biotechnological butanol 8 1. Introduction production are clostridia such as Clostridium acetobutylicum , Clostridium beijerinckii , Clostridium pasteurianum , Clostridium sporogenes , Clostridium saccharobutylicum , and Clostridium saccharoperbutylacetonicum (Visioli et al ., 2014). The metabolism of these strains shows an acidogenic phase in the exponential growth phase converting the substrate to acids (acetate and butyrate) followed by the solventogenic phase, in which the substrate and produced acids are transformed into solvents (acetone, butanol, and ethanol) (Dürre, 2005). It is characterized by a typical ABE ratio of 3:6:1 (Jones and Woods, 1986; Formanek et al ., 1997). Due to rising fossil oil production after the World War II, butanol production by ABE fermentation was too expensive compared to oxo-synthesis and the last ABE-plant closed in the late 1980s (Li et al ., 2014a). Recently, interest in biotechnological butanol production dramatically increased again, since acetogens were reported to produce butanol from the renewable and sustainable feedstock syngas. The organism Butyribacterium methylotrophicum was the first acetogen reported to naturally produce butanol autotrophically (Grethlein et al ., 1991). For the formation of butanol two mol of acetyl-CoA are condensed to acetoacetyl-CoA, which is stepwise reduced to butanol (Fast and Papoutsakis, 2012). Recently, C. ljungdahlii was genetically modified to produce butanol growing on syngas by introduction of the plasmid pSOB ptb (Köpke et al ., 2010). By overexpression of the six genes thlA (thiolase), hbd (3-hydroxybutyryl-CoA dehydrogenase), crt (crotonase), bcd (butyryl-CoA dehydrogenase), adhE (bifunctional butyraldehyde/alcohol dehydrogenase), and bdhA (butanol dehydrogenase) production of 2 mM butanol with C. ljungdahlii was detected (Köpke et al ., 2010). This work demonstrated that production of butanol by syngas fermentation is possible.

The main focus of this study was to enhance 2,3-BD production of different acetogens by overexpressing the genes alsS (acetolactate synthase), budA (acetolactate decarboxylase), and bdh (2,3-BD dehydrogenase) responsible for 2,3-BD formation starting from pyruvate in C. ljungdahlii . For this purpose, the respective genes were cloned into two different shuttle vectors containing different replicons from Gram-positive bacteria (Clostridium perfringens , Clostridium botulinum ) enabling investigation of their influence on 2,3-BD formation. Furthermore, examination of additional overexpression of nifJ (pyruvate:ferredoxin oxidoreductase) was performed to elucidate, if this modification can direct the metabolic flux towards 2,3-BD production. 1. Introduction 9

The second aim of this study was the modification of the above-mentioned butanol plasmid pSOB ptb (Köpke et al ., 2010). The main disadvantage of this plasmid is that it does not contain the genes etfA and etfB (electron-transferring flavoproteins), which were shown to be essential for reduction of crotonyl-CoA to butyryl-CoA e. g. in C. kluyveri (Li et al ., 2008). These proteins are also considered to be needed for butanol formation in C. acetobutylicum (Lütke- Eversloh and Bahl, 2011). In this study, the plasmid pSB3C5-UUMKS 3 containing the genes etfA and etfB from C. acetobutylicum (Schuster, 2011) should be equipped with two different replicons from Gram-positive bacteria ( C. perfringens , Clostridium butyricum ). This aims at construction of a recombinant acetogenic strain capable of efficient butanol formation.

10 2. Material and Methods

2. Material and Methods

2.1 Microorganisms, plasmids, and primers

2.1.1 Bacterial strains Bacterial strains used in this work are shown in Table 1.

Table 1: Bacterial strains

Strain Relevant geno- or phenotype* Origin/Reference

Acetobacterium woodii Type strain DSMZ ** GmbH, DSM 1030 Brunswick, Germany Clostridium aceticum Type strain DSMZ GmbH, DSM 1496 Brunswick, Germany Clostridium autoethanogenum Ty pe strain DSMZ GmbH, DSM 10061 Brunswick, Germany Clostridium carboxidivorans Type strain DSMZ GmbH, DSM 15243 Brunswick, Germany Clostridium coskatii Type strain ATCC *** , Manassas, PTA-10522 VA, USA Clostridium ljungdahlii Type strain DSMZ GmbH, DSM 13528 Brunswick, Germany Clostridium ragsdalei Type strain DSMZ GmbH, DSM 15248 Brunswick, Germany Escherichia coli

CA434 hsdS20 (r -B, m -B), supE44 , thi-1, Purdy et al. , 2002 recAB , ara-14 , leuB5proA2 , lacY1 , galK , rpsL20 (Str R), xyl-5, mtl-1 including the conjugative plasmid R702 (Tet R, Sm R, Su R, Hg R, Tra +, Mob +) *Standard genotype abbreviations for E. coli (Berlyn, 1998) **Deutsche Sammlung von Mikroorganismen und Zellkulturen (German Collection of Microorganisms and Cell Culture) ***American Type Culture Collection

2. Material and Methods 11

Table 1: Bacterial strains (continued)

Strain Relevant geno- or phenotype* Origin/Reference

ER2275 trp31 his1 tonA2 rpsL104 Mermelstein and supE44 xyl -7 mtl -2 metB1 Papoutsakis, 1993 e14 - Δ(lac )U169 endA1 recA1 R( zgbZ10 ::Tn 10 ) Tc s Δ(mcr-hsd-mrr) 114::1510 (F´ proAB traD36 lacI q Δ M15 zzf ::mini Tn 10 (Km R))

XL1 -Blue MRF' Δ( mcrA )183 Δ( mcrCB - Agilent hsdSMR -mrr)173 endA1 Technologies, Santa supE44 thi -1 recA1 gyrA96 Clara, CA, USA

relA1 lac (F’proAB lacI q ZΔM15 Tn 10 (Tet R))

*Standard genotype abbreviations for E. coli (Berlyn, 1998)

2.1.2 Vectors and plasmids Vectors and plasmids constructed and used in this work are listed in Table 2.

Table 2: Vectors and plasmids

Vector/ Plasmid Size [bp] Characteristics Origin/Reference

pACYC184 4,245 p15A ori (rep ), Cm r (catP ), Tc r DSMZ GmbH, (tetR ) Brunswick, Germany

pACYC184_MCljI 7,414 pACYC184 carrying Matthias Beck, CLJU_c03310 and unpublished CLJU_c03320 from C. ljungdahlii under control

of Pptb -buk from C. acetobutylicum

pACYC184_MCljI_pSC101ori 8,327 pACYC184_MCljI carrying This study pSC101 ori (rep101 ) 12 2. Material and Methods

Table 2: Vectors and plasmids (continued)

Vector/Plasmid Size [bp] Characteristics Origin/Reference

pCE2 14 ,692 pSB3C5 -UUMKS 2 carrying This study pIP404 ori (rep ) pCE3 14 ,686 pSB3C5 -UUMKS 3 carrying This study pCB102 ori (repH) pCE3_traJ 15 ,452 pCE3 carrying traJ of This study pMTL82151 pJIR750 6,568 pMB1 ori (rep ), pIP404 ori Bannam and Rood, (rep ), Cm r (catP ), lacZ 1993 pJIR750_23BD 10 ,960 pJIR750 carrying alsS , budA , This study and bdh from C. ljungdahlii under control of

Ppta-ack from C. ljungdahlii pJIR750_23BD_budAshort 10 ,867 pJIR750_23BD carrying a 3' - This study end shortened sequence of budA from C. ljungdahlii pJIR750_23BD_budAshort_adh 10 ,798 pJIR750_23BD_budA short This study carrying CLJU_c24860 from C. ljungdahlii instead of bdh pJIR750_budABC 9,771 pJIR750 carrying budA , budB , This study and budC from Raoultella terrigena pJIR750_budABC_Ppta -ack 10 ,153 pJIR750 _budABC carrying This study

Ppta-ack from C. ljungdahlii pJIR750_budABCoperon 10 ,342 pJIR750 _budABC_Ppta -ack This study

carrying Tsol-adc from C. acetobutylicum pJIR750_23BD_PFOR 14 ,510 pJIR750_23BD carrying nifJ This study from C. ljungdahlii pJIR750_23BD_PFOR_traJ 15 ,274 pJIR750_23BD_PFOR carry ing This study traJ of pMTL82151 2. Material and Methods 13

Table 2: Vectors and plasmids (continued)

Vector/Plasmid Size [bp] Characteristics Origin/Reference pJIR750_oBDH_PFOR 13 ,212 pJIR750_23BD without bdh This study carrying nifJ from C. ljungdahlii pJIR750_oBDH_PFOR_traJ 13,976 pJIR750_oBDH_PFOR carrying This study traJ of pMTL82151

pJR3 14 ,199 pSB3C5 -UUMKS 3 carrying This study pCB102 ori (repH )

pJR3_traJ 14 ,965 pJR3 carrying traJ of This study pMTL82151

pJIR750_oBDH_PFOR_traJ 13 ,976 pJIR750_oBDH_PFOR carrying This study traJ of pMTL82151

pMTL82151 5,254 colE1 ori (rep ), pBP1 ori (repA , Heap et al. , 2009 orf2 ), Cm r (catP ), traJ

pMTL82151_23BD 9,646 pMTL82151 carrying alsS , This study budA , and bdh from C. ljungdahlii under control of

Ppta-ack from C. ljungdahlii

pMTL82151_23BD_PFOR 13 ,196 pMTL82151_23BD carrying This study nifJ from C. ljungdahlii

pMTL82151_oBDH_PFOR 11 ,898 pMTL82151_23BD without This study bdh carrying nifJ from C. ljungdahlii

pMTL83151 4,476 colE1 ori (rep ), pCB102 ori Heap et al. , 2009 (repH ), Cm r (catP ), traJ

pMK 2,393 colE1 ori (rep ), Km r (kanR ) Thermo Fisher Scientific Inc., Waltham, MA, USA

pMK_budABCoperon 5,537 colE1 ori (rep ), budA , budB , Biopolis S. L., budC from R. terrigena , Km r Valencia, Spain (kan ) 14 2. Material and Methods

Table 2: Vectors and plasmids (continued)

Vector/Plasmid Size [bp] Characteristics Origin/Reference pSB3C5 3,828 p15A ori (rep ), Cm r (catP), ATG:biosynthetics ccdB GmbH, Freiburg, Germany pSB3C5 -UUMKS 3 12 ,515 pSB3C5 carrying crt , bcd , ATG:biosynthetics etfA , etfB , hbd , thlA , adhE2 GmbH, Freiburg, from C. acetobutylicum under Germany

control of P ptb -buk from C. acetobutylicum

pSC101 9,263 pSC101 ori (rep101 ), Tc r (tetR ) DSMZ GmbH, Brunswick, Germany/Cohen and Chang, 1977

pUC57 2,710 ColE1 ori (rep ), Ap r (bla ), lacZ GenScript USA Inc., Piscataway, NJ, USA pUC57_23BD 10 ,960 pUC carrying carrying alsS , GenScript USA Inc., budA and bdh from Piscataway, NJ, USA C. ljungdahlii under control of

Ppta-ack

2.1.3 Primers Primers used in this study are presented in Table 3. They were synthesized by biomers.net GmbH (Ulm, Germany) and dissolved in sterile, ultrapure water to get a concentration of 100 pmol/µl. Primers were used for either amplification of DNA fragments or sequencing. For storage purposes, primers were kept at -20°C. When required, restriction sites added at the 5'-end are underlined in bold and primers that contain overlapping regions to the utilized cloning vector for “In-Fusion® HD Cloning” are italicized.

2. Material and Methods 15

Table 3: Primers

Primer Sequence (5' → 3') Restriction enzyme

fD1 * AGAGTTTGATCCTGGCTCAG rP2 * ACGGCTACCTTGTTACGACTT pMTL82151catP_fwd CATACCGGGAATATGTAG pMTL82151catP_rev GTTTGCAAGCAGCAGATTAC pMTL82151_23BDSeq1 GCAGAACGAGCACTTTCAATATG pMTL821 51_23BDSeq2 TGCCCGTAGGTGCTTTAG pMTL82151_23BDSeq3 GTTGGTGATGGCGGTTTC pMTL82151_23BDSeq4 AGCCTATGGCTGAAGTTG pMTL82151_23BDSeq5 CAAGAGTACTGCACCAGTAG pJIR750catP_fwd CTTTCGGCTCGATTTCAC pJIR750catP_rev CGGCAAGTGTCCAAGAAG pIP404PstI_fwd ACA CTGCAG ATC CCGCTTTAATCCCAC Pst I pIP404SalI_rev ACA GTCGAC GGTTTACTTCAACGGC Sal I CljbudAKpnI_fwd ACA GGTACC GAGGTGAATGTAATATGG Kpn I Clj -budABamHI_rev ACA GGATCC CTATCAATACTTTATTTTTC Bam HI CljADHBamHI_fwd ACA GGATCC GGAGGTTGTATTATGAAAGG Bam HI CljADHSalI_rev ACA GTCGAC GTTCT TGGCATCAGGTTGAG Sal I budABC_SEQ CTGAATGCACCTGCCAGGAG BudABC_Seq1 CATATGCAGGGCCTTAAC BudABC_Seq2 GCGCCTGCACGCCAATATTC BudABC_Seq3 TGGTCTCCAACGCCTTCCG BudABC_Seq4 GGCCGCTGTATCCATATGAC pIP404Seq1 CTGTCGACGGTTTACTTC pIP404Seq2 GGAGCAGATAGACAAGCCTT AG pIP404Seq3 TCTAATGTTAGAAGTACC CPECPpta -ackEcoRI_fwd CACACAGGAAACAGCTATGACCATGATTAC Eco RI GAATTC CCATGTCAAGAACTCTGTTTATTTC CPECPpta -ackEcoRI_rev CCTGGCAGGTGCATTCAGGATAATGATTCAC Eco RI GAATTC GGTGGTGGCTTTAAATTTAACACAAA ATTAC

*(Weisburg et al ., 1991) 16 2. Material and Methods

Table 3: Primers (continued)

Primer Sequence (5' → 3') Restriction enzyme

TsoladcXbaI_fwd ACA TCTAGA CTAGCAAAGAAATACTATG Xba I TsoladcHindIII_rev ACA AAGCTT CCTTAGAATCCATTAC Hin dIII pJIRSEQ_fwd GATAACCGTATTACCGCCTTTG pJIR_budABC_P ptaack CAATCAGTTCGCCATCGAGTTC SEQ_rev pCB102SalI_fwd ACA GTCGAC AGCCTGAATGGCGAATGG Sal I pCB102PstI_rev ACA CTGCAG CCGGCCGC TTATAATCCATAAC Pst I pJR3Seq1 TCGACAGCCTGAATGGCGAATG pJR3Seq2 GGGTGAGCAAAGTGACAGAG pJR3Seq3 AATATTGAGAGTGCCGACACAG pJR3Seq4 CAAAGCCGTTTCCAAATC pSC101SacII_fwd ACA CCGCGG AGCGCAGCGAACTGAATGTC Sac II pSC101ClaI_rev ACA ATCGAT GAACAGCTTTAAATGCAC Cla I Clj alsS -sonde_fwd GTTTAGAAGCCGAAGGAG CljalsS -sonde_rev TTATAGCACCGCTTCCAG CljBudA -sonde_fwd GAGGTGAAAGTCCCAAAC CljBudA -sonde_rev CCAGTTCAGTGACTACAG Clj23bdh -sonde_fwd GTGGTTCTGACTTGCATGAG Clj23bdh -sonde_rev GAGAATGAAGGGCAACTG PFORInfusionBamHI_ fw d TAAGTAAGTCCACCTGTCC GGATCC TGAAAAG Bam HI GAGAGGAATTTTTATG PFORInfusionBamHI_ rev CTATGTGCCATTATGACAC GGATCC AATTTACT Bam HI TGATTACTGTTCATTATCAAG PFORohneBDHBamHI_ fwd TAAGTAAGTCCACCTGTCC GGATCC TGAAAAG Bam HI GAGAGGAATTTTTATG PFORohneBDHSalI_ rev CAAGCTTGCATGCC TGCAG GTCGAC AATTAAA Sal I AAAGTACCAGACAGC PFOR -Seq1 TCCAGTAACTCGTGGTACAG PFOR -Seq1 PFOR -Seq2 AACGCTGCTATAGACAAGGG PFOR -Seq2 PFOR -Seq3 GCTGAAGGTGAAGGAACAAGAG PFOR -Seq3 PFOR -Seq4 CACCAGAAGGAACTACAG PFOR -Seq4 2. Material and Methods 17

Table 3: Primers (continued)

Primer Sequence (5' → 3') Restriction enzyme

PFOR -Seq5 CTACAAGCAGGAGCATTGAC

traJInfusionPstI_fwd AACCCTTAGTGACTC CTGCAG CGATCGGTCTTG Pst I CCTTGC traJInfusionPstI_rev AACCCTTAGTGACTC CTGCAG CGATCGGTCTTG Pst I CCTTGC traJseq1 GATTAAAGCGGGATCTGC

traJSeq_rev CGCCTCCATCCAGTCTATTC

InfusionTraJEheI_fwd CAGCCTGAATGGCGAAT GGCGCC TGCTTGCGG Ehe I GTCATTATAG InfusionTraJEheI_rev GAGAAAATACCGCATCA GGCGCC GATCGGTCT Ehe I TGCCTTGC pJR3 -traJ Seq CCTGGCGTTACCCAACTTAATC

M13 -FP * TGTAAAACGACGGCCAGT

M13 -RP * CAGGAAACAGCTATGACC

*GATC Biotech AG, Constance, Germany

2.2 Chemicals, enzymes, and gases The chemicals and enzymes used in this work were all purchased from the following companies:

 AppliChem GmbH, Darmstadt, Germany  Biozym Scientific GmbH, Oldenburg, Germany  Carl Roth GmbH & Co. KG, Karlsruhe, Germany  Difco Laboratories, Augsburg, Germany  Epicentre Technologies Corp., an Illumina company, Madison, WI, USA  GE Healthcare EUROPE GmbH, Munich, Germany  Genaxxon Bioscience GmbH, Ulm, Germany  Invitrogen GmbH, Karlsruhe, Germany  MACHEREY-NAGEL GmbH & Co. KG, Dueren, Germany  Merck KGaA, Darmstadt, Germany  Qiagen GmbH, Hilden, Germany  SERVA Electrophoresis GmbH, Heidelberg, Germany 18 2. Material and Methods

 Sigma-Aldrich Chemie GmbH, Steinheim, Germany  Takara Bio USA Inc., Mountain View, CA, USA  Thermo Fisher Scientific, Waltham, MA, USA  VWR International GmbH, Darmstadt, Germany  ZYMO Research Europe GmbH, Freiburg, Germany

All gases and gas mixtures used during this study are listed in Table 4. Gases, excluding syngas and H2 + CO 2, were purchased from the company MTI IndustrieGase AG (Neu-Ulm, Germany).

Syngas and H2 + CO 2 were ordered at Westfalen AG (Münster, Germany).

Table 4: Utilized gases and gas mixtures

Gas Gas composition Purity Application

Forming gas 095 .0 % N 2 3.0 Anaerobic chamber

005.0 % H 2 3.0

N2 100 .0 % N 2 5.0 Anaerobic media preparation, gas chromatography (carrier gas)

Synthetic air 079.5 % N 2 5.0 Gas chromatography

020.5 % O 2 5.0 (detector gas)

H2 + CO 2 067 .0 % H 2 5.0 Autotrophic growth

033.0 % CO 2 3.0 experiments Synthesis gas (s yngas ) 040 .0 % CO 2.0 Autotrophic growth

040.0 % H 2 5.0 experiments

010.0 % CO 2 3.0

010.0 % N 2 5.0

N2 + CO 2 080 .0 % N 2 5.0 Anaerobic media preparation

020.0 % CO 2 3.0

H2 100 .0 % H 2 5.0 Gas chromatography (detector gas)

2. Material and Methods 19

2.3 Bacterial cultivation

2.3.1 Cultivation media In general, for preparing culture media (see chapter 2.3.1.1 - 2.3.1.8), chemicals were weighed to corresponding composition into beakers and dissolved in water of the ultrapure water system “18.2 MΩ-cm, Type I water” (PURELAB Classic, ELGA LabWater, Celle). After adjusting pH and end volume, media were autoclaved at 121 °C and 1.2 bar for 15 to 20 min. Heat-liable components such as antibiotics were filter-sterilized (“Filtropur S 2.0”, pore size 0.2 µm, Sarstedt AG & Co, Nümbrecht, Germany) and were added to the media directly prior to use. In order to prepare solid media, 1.5 % agar were added to liquid culture media before autoclaving. Afterwards, media were cooled down to approximately 55 °C and antibiotics were supplemented. The prepared media were poured into sterile petri dishes and solidified overnight. Solid agar plates were sealed in sterile plastic bags and stored at 4 °C.

Growth of clostridial strains requires anaerobic culture media. Resazurin was supplemented to anaerobic media as an indicator for oxygen. Under anaerobic conditions, resazurin is reduced to the colourless dihydroresorufin, while under aerobic conditions the reoxidized resorufin is formed appearing as a pink colour. Media were prepared aerobically, filled in Hungate tubes (Ochs GmbH, Bovenden, Germany) or glass flasks (Müller Krempel AG, Bülach, Switzerland, or Thermo Fisher Scientific Inc., Waltham, MA, USA), closed with butyl rubber stoppers (Ochs GmbH, Bovenden or Maag Technic GmbH, Göppingen) and screw caps (Ochs GmbH, Bovenden, Germany or Thermo Fisher Scientific Inc., Waltham, MA, USA). Finally, vacuum was applied which was followed by purging with N 2 + CO 2 to get an anaerobic atmosphere. These last two steps were repeated 5 times. Media which contained agar (1.5 % [w/v]) were poured into sterile petri dishes in the anaerobic chamber after cooling down to 55 °C. Agar plates were enabled to solidify overnight, sealed in sterile plastic bags, and were stored in the anaerobic chamber at room temperature.

2.3.1.1 Modified 1.0 ATCC 1612 medium (Hoffmeister, unpublished) Cultivation of C. aceticum was performed in modified 1.0 ATCC 1612 medium.

Modified 1.0 ATCC 16121 medium

NH 4Cl 00.20 g 03.7 mM

KH 2PO 4 00.33 g 02.0 mM 20 2. Material and Methods

K2HPO 4 00.45 g 03.0 mM

MgSO 4 x 7 H 2O* 00.10 g 00.4 mM Vitamin solution 02.00 ml 00.2 % [v/v] SL -9 T race element solution 02.00 ml 00.2 % [v/v] Bacto ® yeast extract 03.00 g 00.3 % [w/v]

NaHCO 3 10.00 g 00.1 M 000000000

Cysteine HCl x H 2O 00.30 g 01.7 mM

Na 2S x 9 H 2O 00.30 g 01.2 mM Resazurin 01.00 mg 04.4 µM

H2O ad 1,000 ml 0000 *Sterile, anaerobic stock solution (750 mM) was supplemented after autoclaving.

As carbon source, a sterile and anaerobic fructose stock solution (1.11 M) was added after autoclaving. The pH was adjusted to 8.5 by adding Na 2CO 3 (sterile, anaerobic stock solution; 5 % [w/v]). For cultivation of C. aceticum on syngas, additional trace elements (sterile, anaerobic stock solutions) were supplemented to 50 ml medium prior to use (0.2 ml FeCl 2 stock solution, 0.1 ml Na 2SeO 3 + Na 2WO 4 stock solution, 0.4 ml NiCl 2 stock solution, and 0.4 ml of ZnSO 4 stock solution).

FeCl 2 Stock solution

FeCl 2 x 4 H 2O 0.12 g 8.10 mM

H2O ad 50 ml 0000

Na 2SeO 3 + Na 2WO 4 Stock solution

NaOH 0.5 g 15.5 mM

Na 2SeO 3 x 5 H 2O 3.0 mg 11.4 µM

Na 2WO 4 x 2 H 2O 4.0 mg 12.1 µM

H2O ad 1000 ml 0000

NiCl 2 Stock solution

NiCl 2 x 6 H 2O 0.03 g 0.36 mM

H2O ad 50 ml 0000

ZnSO 4 Stock solution

ZnSO 4 x 7 H 2O 0. 06 g 3.91 mM

H2O ad 50 ml 0000 2. Material and Methods 21

Vitamin solution -modified- (Wolin et al. , 1963)

Pyridoxine HCl 50 mg 205 µM Thiamine HCl 50 mg 148 µM Riboflavin 50 mg 130 µM Calcium pantothenate 50 mg 105 µM α-Lipoic acid 25 mg 120 µM 4-Aminobenzoic acid 50 mg 365 µM Nicotinic acid 50 mg 405 µM

Vitamin B 12 25 mg 020 µM D-Biotin 25 mg 100 µM Folic acid 25 mg 055 µM

H2O ad 1,000 ml 0000 SL-9 Trace element solution (Tschech and Pfennig, 1984)

Nitrilotriacetic acid 12.8 g 67.0 mM

MnCl 2 x 4 H 2O 00.1 g 00.5 mM

FeCl 2 x 4 H 2O 02.0 g 10.1 mM

CoCl 2 x 6 H 2O 00.2 g 00.8 mM

ZnCl 2 70.0 mg 00.5 mM

CuCl 2 x 2 H 2O 02.0 mg 11.7 µM

NaMoO 4 x 2 H 2O 36.0 mg 00.2 mM

NiCl 2 x 6 H 2O 24.0 mg 00.1 mM

H3BO 3 06.0 mg 97.0 µM NaCl 05.0 g 85.6 mM

H2O ad 1,000 ml 0000 For preparation of trace element stock solution, nitrilotriacetic acid was dissolved in water by adjusting pH to 6 using KOH. Afterwards, the other components were added and the final pH was adjusted to 7.0 using KOH.

22 2. Material and Methods

2.3.1.2 Modified ATCC 1612 medium (Hoffmeister et al ., 2016) Cultivation of A. woodii was performed in modified ATCC 1612 medium.

Modified ATCC 1612 medium

NH 4Cl 00.20 g 03.7 mM

KH 2PO 4 01.76 g 12.9 mM

K2HPO 4 08.44 g 48.5 mM

MgSO 4 x 7 H2O* 00. 33 g 01.3 mM Vitamin stock solution (see above) 02.00 ml 00.2 % [v/v] SL -9 t race element stock solution (see above) 02.00 ml 00.2 % [v/v] Bacto ® yeast extract 02.00 g 00.2 % [w/v]

NaHCO 3 10.00 g 00.1 M000000000

Cystein e HCl x H 2O 00.30 g 01.7 mM

Na 2S x 9 H 2O 00.30 g 01.2 mM Resazurin 01.00 mg 04.4 µM

H2O ad 1,000 ml 0000 *Sterile, anaerobic solution was supplemented after autoclaving.

As carbon source, a sterile and anaerobic stock solution of fructose (1.11 M) was added after autoclaving.

2.3.1.3 Lyophilization medium (Flüchter, unpublished) All anaerobic strains were stored by lyophilization, using lyophilization medium.

Lyophilization medium

Meat extract 00 2 g 00 0.2 % [w/v] Sucrose 120 g 350.6 mM Resazurin 00 1 mg 00 4.4 µM

H2O (anaerobi c) ad 1,000 ml Prior to use, 0.3 ml of anaerobic ferrous (II) sulfide stock solution (20 mg/ml) were added to 5 ml medium.

2. Material and Methods 23

2.3.1.4 Lysogeny broth medium -modified- (Bertani, 1951) Cultivation of E. coli cells was done in lysogeny broth (LB) medium.

LB medium -modified-

Bacto ® tryptone 10 g 00 1.0 % [v/v] NaCl 10 g 171.1 mM Bacto ® yeast extract 05 g 00 0.5 % [v/v]

H2O ad 1,000 ml 0000 2.3.1.5 Rajagopalan medium -modified- (Rajagopalan et al ., 2002) Cultivation of C. ragsdalei was performed in Rajagopalan medium.

Rajagopalan medium -modified-

MES 20 .00 g 102 .0 mM Bacto ® yeast extract 02.00 g 00 0.2 % [v/v]

NH 4Cl 01.00 g 019 .0 mM NaCl 01.00 g 017 .0 mM KCl 00.1 0 g 00 1.3 mM

KH 2PO 4 00.1 0 g 00 0.7 mM

MgSO 4 x 7 H 2O 00.20 g 00 0.8 mM

CaCl 2 00.02 g 00 0.2 mM

L-Cysteine -HCl x H 2O 01.00 g 00 5.0 mM Vitamin solution 10.00 ml 00 1.0 % [v/v] PE TC 1754 trace element solution 10.00 ml 00 1.0 % [v/v] Resazurin 01.00 mg 00 4.4 µM

H2O ad 1,000 ml 0000 All components were mixed and pH was adjusted to 6.1 using NaOH . As carbon source, a sterile and anaerobic stock solution of fructose (1.11 M) was added after autoclaving.

Vitamin solution (Tanner, 2007)

Pyridoxin e HCl 10 mg 49 .0 µM Thiamine HCl 05 mg 15 .0 µM Riboflavin 05 mg 13 .0 µM Calcium p antothenate 05 mg 21 .0 µM

α-Lipoic acid 05 mg 24 .0 µM 24 2. Material and Methods

4-Aminobenzoic acid 05 mg 36 .0 µM Nicotinic acid 05 mg 41 .0 µM

Vitamin B 12 05 mg 03.7 µM D-Biotin 02 mg 08.2 µM Folic acid 02 mg 04.5 µM 2-Mercaptoethanesulfonic acid (MESNA) 02 mg 14 .0 µM

H2O ad 1,000 ml 0000 PETC 1754 trace element solution

Nitrilotriacetic acid 2.00 g 105 µM

MnSO 4 x H 2O 1.00 g 059 µM

Fe(NH 4)2 (SO 4)2 x 6 H 2O 0.8 0 g 020 µM

CoCl 2 x 6 H 2O 0.2 0 g 712 µM

ZnSO 4 x 7 H 2O 1.00 mg 00 3 µM

CuCl 2 x 2 H 2O 0.02 g 117 µM

NaMoO 4 x 2 H 2O 0.02 g 097 µM

Na 2SeO 3 x 5 H 2O 0.02 g 102 µM

NiCl 2 x 6 H 2O 0.02 g 084 µM

Na 2WO 4 x 2 H2O 0.02 g 064 µM

H2O ad 1,000 ml 0000 For preparation of trace element solution, nitrilotriacetic acid was dissolved in water by adjusting pH to 6 using KOH. Afterwards, the other components were added.

2.3.1.6 Super optimal broth medium -modified- (Sun et al. , 2009b) Chemically competent cells of E. coli were prepared in super optimal broth (SOB).

SOB medium -modified-

Bacto ® tryptone 020.0 g 02.0 % [v/v] NaCl 00 0.5 g 08.6 mM Bacto ® yeast extract 00 5.0 g 00.5 % [v/v] KCl 190 .0 g 02.6 M

Mg Cl 2 (2M) 00 5.0 ml 10 .0 mM

H2O ad 1,000 ml 0000

*Sterile solution was added after autoclaving . 2. Material and Methods 25

2.3.1.7 Tanner medium -modified- (Tanner, 2007) Cultivation of C. ljungdahlii , C. autoethanogenum , and C. coskatii was achieved in Tanner medium. For cultivation of C. carboxidivorans , 3 g of yeast extract and for C. coskatii 2 g of yeast extract was used.

Tanner medium -modified-

MES 20 .0 g 102 .00 mM Bacto ® yeast extract 00.5 g 00 0.05 % [v/v] Mineral solution 25 .0 ml 00 0.5 0 % [v/v] PETC 1754 T race element solution (see above) 10 .0 ml 00 1.00 % [v/v] Vitamin solution ( see above) 10.0 ml 00 1.00 % [v/v]

L-Cysteine HCl x H 2O 01.0 g 00 5.00 mM Resazurin 01.0 mg 00 4.40 µM

H2O ad 1,000 ml 0000 All components were mixed and pH was adjusted to 5.9 using NaOH. As carbon source, a sterile and anaerobic fructose stock solution (1.11 M) was added after autoclaving.

Mineral solution (Tanner, 2007)

NaCl 080 g 00 1.4 M

NH 4Cl 100 g 00 1.9 M KCl 010 g 134.0 mM

KH 2PO 4 010 g 073.0 mM

MgSO 4 x 7 H 2O 020 g 081.0 mM

CaCl 2 x 2 H 2O 00 4 g 027.0 mM

H2O ad 1,000 ml 0000 2.3.1.8 Yeast tryptone fructose medium -modified- (Leang et al., 2013) Cultivation of C. ljungdahlii after transformation via electroporation (see chapter 2.7.2) or conjugation (see chapter 2.7.3) was performed on solid yeast tryptone fructose (YTF) medium.

YTF medium -modified-

Bacto ® yeast extract 10 g 1.0 % [w/v] Bacto ® tryptone 16 g 1.6 % [w/v] Fructose 05 g 27.8 mM NaCl 04 g 68.4 mM

L-Cysteine -HCl x H 2O 01 g 05.7 mM 26 2. Material and Methods

Resazurin 01 mg 00 4.4 µM

H2O ad 1,000 ml 0000 2.3.2 Antibiotic selection The selection of recombinant strains was achieved by using antibiotics which are listed in Table 5. Antibiotic stock solutions were 1000-fold concentrated, sterile-filtrated (sterile filter “Filtropur S 2.0”, pore size 0.2 µm, Sarstedt AG & Co, Nümbrecht, Germany) and stored at -20 °C. Addition of antibiotics to the media was done prior to use.

Table 5: Antibiotics used in this work

Antibiotic Stock solution Final Organism Supplier [mg/ml] (solvent) concentration [μg/ml]

Ampicillin 100 (in H 2O) 100 E. coli Carl Roth GmbH Chloramphenicol 30 (in 96 % [v/v] ethanol) 30 E. coli & Co.KG, Karlsruhe, Germany Tetracycline HCl 10 (in 50 % [v/v] ethanol) 10 E. coli Fluka Chemie GmbH, Buchs, Switzerland

Kanamycine 50 (in H 2O) 50 E. coli Carl Roth GmbH & Co.KG, Karlsruhe, Germany

Colistin sodium 10 (in H 2O) 10 E. coli Sigma -Aldrich methanesulfonate Chemie GmbH, Thiamphenicol* 15 (in 7.5 A. woodii , Steinheim, dimethylformamide or C. ljungdahlii, Germany acetonitrile) C. coskatii

*light-sensitive solution

2. Material and Methods 27

2.4 Growth conditions

2.4.1 Aerobic cultivation In general, E. coli was grown at a temperature of 37 °C. For preparation of chemically competent cells (see chapter 2.7.1.4), cells were grown at 18 °C. In order to cultivate E. coli for e.g. plasmid isolation (see chapter 2.5.1.1), 5 ml LB medium (see chapter 2.3.1.4) in test tubes, closed with loose aluminium caps were used. Inoculation was performed with 10 µl of E. coli stock (see chapter 2.4.3.1). For isolating higher amounts of plasmid DNA (see chapter 2.5.1.2), 500 ml medium in baffled 2-l Erlenmeyer flasks were inoculated with 5 ml of growing culture. Optimal growth of E. coli strains was maintained by continuous oxygenation on a rotary shaker at 120 to 180 rpm at 37 °C. Agar plates with E. coli strains were incubated upside down at 37 °C.

2.4.2 Anaerobic cultivation For 5 ml culture volume, Acetobacterium and Clostridium strains were grown anaerobically in Hungate tubes (Ochs GmbH, Bovenden, Germany), which were closed with airtight synthetic rubber stoppers (Ochs GmbH, Bovenden, Germany) and plastic screw caps (Ochs GmbH, Bovenden, Germany). For larger volumes (50 ml and 100 ml), glass flasks (Müller & Krempel AG, Bülach, Switzerland or Thermo Fisher Scientific Inc., Waltham, MA, USA), sealed with natural rubber stoppers (Maag Technic GmbH, Göppingen, Germany) and stainless-steel screw caps (Müller & Krempel AG, Bülach, Switzerland or Thermo Fisher Scientific Inc., Waltham, MA, USA) were used. Cultures of A. woodii , C. aceticum , and C. ragsdalei were grown at 30 °C, while C. ljungdahlii , C. carboxidivorans , C. autoethanogenum , and C. coskatii were cultivated at 37 °C. Incubation of agar plates took place upside down in an incubator within the anaerobic chamber at 37 °C. Heterotrophic growth experiments were performed in 50 ml medium in 125-ml anaerobic glass flasks (see above) and for autotrophic growth experiments either 50 ml medium in 500-ml glass flasks or 100 ml in 1-l glass flasks were used. The respective medium was inoculated with the corresponding strain to an optical density at

600 nm (OD 600nm ) of 0.1 for heterotrophic growth experiments and to OD 600nm of 0.05 for autotrophic growth experiments, respectively. Thereafter, OD600nm was monitored and samples for product analysis were withdrawn. After each growth experiment, genomic DNA was isolated and the identity of the strain was verified by amplification and sequencing of the 16S rDNA using the primers fD1 and rP2 (Table 3). For plasmid verification either primers 28 2. Material and Methods pMTL82151catP_fwd and pMTL82151catP_rev for pMTL82151 backbone or pJIR750catP_fwd and pJIR750catP_rev for pJIR750 based vectors were used. Reactivation of sucrose- magnesium-phosphate (SMP) buffer/dimethylsulfoxide (DMSO) cultures (see chapter 2.4.3.2) was achieved by inoculation of 5 ml medium with 10 µl of culture. Subsequent passages of the strains into medium were performed with approximately 10 % [v/v] of the final culture volume. Lyophilized cells (see chapter 2.4.3.3) were reactivated by adding 2 ml of respective medium to the cells, the pellet was allowed to dissolve and afterwards returned to the Hungate tube. Three days of incubation at the respective temperature were sufficient for reactivation.

2.4.3 Strain conservation

2.4.3.1 Glycerol For conservation of E. coli strains, cells were grown over night at 37 °C in 5 ml LB medium (see chapter 2.3.1.4) with respective antibiotics added (Table 5), if needed. 500 µl of this culture were mixed with 500 µl of 50 % [v/v] glycerol and afterwards stored in cryogenic vials at -80 °C.

2.4.3.2 SMP buffer/DMSO For short-time conservation of anaerobic strains, the (sucrose-magnesium-phosphate) SMP buffer/DMSO method was used (Hoffmeister, 2017). 5 ml medium in a Hungate tube were inoculated with 500 µl of growing cells. When late exponential growth phase was reached, cells were harvested in a “Hungate tube centrifuge EBA20” (Andreas Hettich GmbH & Co. KG, Tuttlingen, Germany) at 2,400 x g for 10 min at 4 °C. Afterwards, supernatant was discarded with a 5-ml syringe and cell pellet was dissolved in 400 µl SMP buffer and 200 µl anti-freezing buffer. The cell suspension was stored at -80 °C.

SMP buffer

Sucrose 92.40 g 270 mM

MgCl 2 x 6 H 2O 00.20 g 00 1 mM

NaH 2PO 4 00.84 g 00 7 mM

H2O ad 1,000 ml 0000 2. Material and Methods 29

The pH was adjusted to 6 by adding HCl, buffer was boiled for 20 min and cooled on ice for 20 min, while sparging with N 2. Buffer was transferred in an anaerobic chamber and filled in airtight glass flasks (see chapter 2.4.2).

Anti-freezing buffer

SMP buffer 20 ml 40 % [v/v] DMSO 30 ml 60 % [v/v] Anti-freezing buffer was prepared with anaerobic SMP buffer and DMSO in an anaerobic chamber and filled in airtight glass flasks (see chapter 2.4.2).

2.4.3.3 Lyophilization In order to preserve anaerobic strains for a long time, lyophilization was performed. Acetobacterium or Clostridium were grown in a Hungate tube to late exponential growth phase. Thereupon, cells were sedimented in a “Hungate tube centrifuge EBA20” (Andreas Hettich GmbH & Co. KG, Tuttlingen, Germany) at 2,400 x g for 10 min at 4 °C, supernatant was removed with a 5-ml syringe and resuspended in 1.5 ml of anaerobic lyophilization medium (see chapter 2.3.1.3). Aliquots of 0.4 ml were transferred under sterile conditions into sterile 8-ml glass vials, closed instantly with sterile cotton wool plugs (Paul Hartmann AG,

Heidenheim, Germany) and flash-freezed in liquid N 2. For keeping cells in this frozen state, a cooling block (-80 °C) was used to take the cells to the pre-cooled lyophilisator (-54 °C) (Christ Alpha 1-4 freeze-dryer; Martin Christ Gefriertrocknungsanlagen GmbH, Osterode am Harz, Germany). Then, vacuum was applied for 19 h, which represents the first drying step. Afterwards, cotton wool plugs were exchanged under sterile conditions with butyl rubber stoppers and plastic screw caps (Ochs GmbH, Bovenden, Germany). Thereafter, a second vacuum step took place by piercing each butyl rubber stopper with a sterile cannula which was linked to the lyophilisator by a sterilised manifold. 24 hours later, cannulas were removed and the freeze-dried cells were stored at 4 °C.

2.4.4 Determination of growth and metabolic production parameters

2.4.4.1 Determination of optical density

For determination of bacterial growth, optical density at 600 nm (OD 600nm ) of 1 ml sample was determined in 1.6-ml semi-micro cuvettes with 10 mm optical pathway (Sarstedt AG & Co., Nümbrecht, Germany) by using “Ultrospec® 3100 pro” (Amersham Bioscience Europe GmbH, 30 2. Material and Methods

Freiburg, Germany). If OD 600nm was exceeding a value of 0.3, samples were diluted by with water to obtain values in the linear range of the photometer. In this case, the sterile medium being used as blank was also diluted in the same ratio with water. For calculation of OD 600nm the respective dilution factor was taken into account.

2.4.4.2 Quantification of substrate and microbial metabolic products by high performance liquid chromatography The method and parameters for high performance liquid chromatography (HPLC) analysis were described previously (Nielsen et al ., 2010). The following method was used to determine fructose, acetate, lactate, and stereoisomers of 2,3-BD in 300 µl cell culture supernatant (centrifugation 15,000 x g, 30 min, 4 °C), which were transferred in 2-ml vials (CS- Chromatographie Service GmbH, Langerwehe, Germany) and sealed with aluminium caps (CS- Chromatographie Service GmbH, Langerwehe, Germany). Analysis was performed with a “Infinity 1260” HPLC (Agilent Technologies, Waldbronn, Germany), equipped with an autosampler and pre-column followed by a CS-Organic acid column (CS-Chromatographie Service GmbH, Langerwehe, Germany). For detection of organic acids (e.g. acetate, lactate) a diode array UV detector (DAD) was used. Analysis of fructose and solvents (e.g. 2R,3R -BD) was performed with a refraction index detector (RID). Calibration was achieved by using 1-ml samples with different concentrations of the investigated compounds (1 mM-200 mM) which served as external standards. All data analysis was done with “Agilent OPEN-Lab CDS Version A.01.03.” (Agilent Technologies, Waldbronn, Germany).

The following parameter setup was used for HPLC analysis:

Column: CS -Organic acid ( 150 mm x 8 mm) Pre -column: CS -Organic acid ( 40 mm x 8 mm) Packing material: Poly(styrene -divinylbenzene)

Mobile phase: 5 mM H 2SO 4 (0.7 ml/min) Injection temperature: 25 °C Column temperature: 40 °C Detector temperature: 35 °C Detection wavelength (DAD) : 210 nm Running time : 24 min Sample volume: 20 µl at 15 °C autosampler 2. Material and Methods 31

2.4.4.3 Quantification of microbial metabolic products by gas chromatography The method used for quantification of metabolic products by gas chromatography (GC) was modified from a previously described protocol (Playne, 1985). For this purpose, 1 ml of culture supernatant (centrifugation at 15,000 x g, 30 min, 4 °C) in 2-ml vials (CS-Chromatographie Service GmbH, Langerwehe, Germany), sealed with aluminium caps (CS-Chromatographie Service GmbH, Langerwehe, Germany) was needed to measure ethanol as well as 2,3-BD in Clostridium and A. woodii by using a “Clarus 600” gas chromatograph (PerkinElmer LAS GmbH, Rodgau, Germany). It was equipped with a flame ionization detector (FID), an autosampler, and a column packed with Porapak P. Isobutanol solution (110 mM isobutanol in 2 M HCl) served as internal standard (100 µl for 1 ml sample). For calibration, one vial containing internal standard and 5 mM of each analyzed compound was used. All data analysis was executed with “TotalChrom Version 6.3.1” (PerkinElmer, LAS GmbH, Rodgau, Germany).

The GC analysis was performed with the following conditions for ethanol analysis:

Column: Metal (ID 2 mm x 2,000 mm) Packing material : Porapak P (80/100 mesh)

Carrier gas : N2 (2.17 bar ) Injection temperature : 200 °C Detector temperature: 300 °C Detector type: Flame ionization detector (FID)

Detect or gases: H2 (45 ml/min flow rate), synthetic air (450 ml/min flow rate) Temperature setup: 130 °C for 1 min, 130°C to 140 °C at 5 °C/min, 140 °C for 10 min 140 °C to 170 °C at 25 °C/min, 170 °C for 1 min Sample volume: 1 µl For 2,3-BD analysis along with acetonitrile (solvent of thiamphenicol for 2,3-BD analysis in A. woodii ) the following parameters were used:

Column: Metal (ID 2 mm x 2,000 mm) Packing material: Porapak P (80/100 mesh)

Carrier gas: N2 (1.5 bar ) 32 2. Material and Methods

Injection temperature: 200 °C Dete ctor temperature: 300 °C Detector type: Flame ionization detector (FID)

Detector gases: H2 (45 ml/min flow rate), synthetic air (450 ml/min flow rate) Temperature setup: 160 °C for 1 min, 160°C to 170 °C at 1 °C/min, Sample volume: 1 µl 2.5 Nucleic acid methods

2.5.1 Nucleic acid isolation

2.5.1.1 Plasmid isolation from E. coli using the Zyppy TM Plasmid Miniprep Kit For isolating small amounts of plasmid DNA, the “Zyppy TM Plasmid Miniprep Kit” (ZYMO Research Europe GmbH, Freiburg, Germany) was used. This kit contained all necessary reagents and columns. The instructions were slightly modified in the following way. 3 ml of an overnight grown E. coli culture were harvested at 13,000 x g for 1 min in the same 1.5-ml reaction tube. All steps were performed at room temperature. The cell pellet was dissolved in 600 µl sterile water. Then, a washing step with 400 µl Zymo-Spin TM wash buffer was conducted (13,000 x g for 30 s). After the last washing step, drying of the column was performed at 10,000 x g for 3 min. Finally, plasmid DNA was eluted with 35 µl sterile water at 13,000 x g for 15 s, instead of tris-EDTA (TE) buffer. Plasmid DNA was stored at -20 °C.

2.5.1.2 Plasmid isolation from E. coli using QIAGEN Plasmid Midi Kit For isolating larger volumes of plasmid DNA with a high concentration, which is needed for transformation of plasmid DNA in anaerobic bacteria, the “QIAGEN Plasmid Midi Kit” (QIAGEN-tip 100; Qiagen GmbH, Hilden, Germany) was used. The protocol which was applied is for isolating very low-copy plasmids and cosmids with slight modifications. E. coli was grown in 500 ml LB medium in baffled 2-l Erlenmeyer flasks with shaking at 300 rpm overnight. Cells were harvested in a centrifugation vessel at 6,000 x g for 15 min at 4 °C. Then, the bacterial pellet was resuspended in 20 ml buffer P1 containing RNase A. Afterwards, 20 ml of buffer P2 were added to the sample and mixed by vigorously inverting the sealed vessel 4-6 times. After incubation at room temperature for 5 min, 20 ml of chilled buffer P3 were added and the solution was mixed immediately and thoroughly by inverting 4-6 times. Thereafter, incubation 2. Material and Methods 33 for 30 min on ice followed. The sample was centrifuged at 20,000 x g for 30 min at 4 °C to sediment the cell debris. Before centrifugation the sample was mixed again. After adding buffer P1, P2, and P3 and centrifugation at 20,000 x g for 30 min at 4 °C, the supernatant containing plasmid DNA was directly filtered over a pre-wetted, folded filter paper (ø 125 mm; Schleicher & Schuell Bioscience GmbH, Dassel, Germany) to get rid of cell debris. In the end, DNA pellet was air-dried for 15 min at 60 °C and resuspended in 150 µl Tris-EDTA (TE) buffer (pH 8). Plasmid DNA was stored at -20 °C.

TE buffer

Tris 1.20 g 10 mM EDTA, disodium salt 0.37 g 01 mM

H2O ad 1,000 ml 0000 The pH value was adjusted to 8, using 1 M HCl.

2.5.1.3 Genomic DNA isolation from Gram positive bacteria using MasterPure TM Gram Positive DNA Purification Kit In order to isolate genomic DNA from Gram-positive bacteria for 16S rDNA PCR or retransformation into E. coli cells, the “MasterPure TM Gram Positive DNA Purification Kit” (Epicentre Technologies Corp., an Illumina company, Madison, WI, USA) was used. Harvesting of cells was done with 4 ml of a stationary growth phase culture at 13,000 x g for 3 min at room temperature and isolation was handled according to the kit instructions with slight modifications. Incubation of cells with lysozyme was performed overnight at 37 °C. After the washing step with 70 % [v/v] ethanol and centrifugation at 10,000 x g for 2 min, DNA was allowed to air-dry and resuspended in 35 µl TE buffer. Genomic DNA was stored at -20 °C.

2.5.1.4 RNA isolation from clostridia For isolation of RNA, all material and reagents were autoclaved twice to eliminate RNases. While working with RNA, wearing nitrile gloves was obligatory to minimize the possibility of RNA degradation by RNases of human skin. Furthermore, all work surfaces were cleaned with “RNase-Exitus Plus TM “(AppliChem GmbH, Darmstadt, Germany), which destroys RNases. All reagents were treated with RNase-free SafeSeal-Tips® (Biozym Scientific GmbH, Oldenburg). Clostridium culture for RNA isolation was grown in 50 ml of the respective medium for 65 h and harvested in 50-ml falcon tubes at 6,000 x g for 10 min at 4 °C. Cell pellet was stored at -80 °C overnight. After thawing the pellet on ice, 600 µl lysis buffer (20 mg/ml lysozyme in tris- 34 2. Material and Methods sucrose-EDTA (TSE) buffer) were added to the sample and vortexed. The following steps were performed with 700 µl of this sample in 2-ml reaction tubes. Cells were incubated at 37 °C for 5 min. After adding 1 ml chilled TRI Reagent® (Sigma-Aldrich Chemie GmbH, Steinheim), the sample was pipetted up and down. Thereafter, 0.3 ml of chloroform/isoamyl alcohol (24:1 v/v) were applied and the solution was mixed vigorously. After an incubation at room temperature for 3 min, sample was centrifuged at 12,000 x g for 15 min at 4 °C. The upper aqueous phase was transferred into a 1.5-ml reaction tube and 0.5 ml cold isopropanol were added. After inverting the tube several times, sample was centrifuged at 12,000 x g for 10 min at 4 °C. Supernatant was removed and pellet was washed with 1 ml of chilled 70 % ethanol [v/v]. Then, sample was centrifuged at 12,000 x g for 4 min at 4 °C and supernatant was discarded. Finally, RNA pellet was air-dried for 20 min and resuspended in 44 µl of RNase-free water. If the pellet could not be completely resuspended, sample was incubated at 55 °C for 15 min.

The RNA isolation protocol was completed with a DNA digest method by Invitrogen™ Ambion™ TURBO DNA-free Kit (Thermo Fisher Scientific Inc., Waltham, MA, USA). For standard DNA digestion, the following reaction mixture was used:

RNA 44 µl ~ 20 µg/ µl 10 x TURBO DNase buffer 05 µl 10 % [v/v] TU RBO DNase 01 µl 02 % [v/v]

H2O ad 50 µl After 30 min incubation at 37 °C, another 1 µl of TURBO DNase was added and sample was incubated for 30 min at 37 °C. Thereafter, 5 µl DNase inactivation reagent were applied and sample was vortexed. After incubation at room temperature for 5 min, RNA was centrifuged at 10,000 x g for 90 s. DNA-free RNA was transferred into a new 1.5-ml reaction tube. Finally, RNA concentration was determined by using the spectral photometer NanoDrop 2000c (Thermo Fisher Scientific Inc., Waltham, MA, USA) (see chapter 2.5.6).

TSE buffer

Sucrose 62.50 g 730 mM Tris -HCl 01.50 g 050 mM EDTA, disodium salt 04.65 g 050 mM

H2O ad 250 ml 0000 The pH value was adjusted to 7.6, using 1 M NaOH. 2. Material and Methods 35

2.5.2 Polymerase chain reaction Polymerase chain reaction (PCR) is a method, which was used in this study for amplification of specific DNA fragments, e.g. for cloning procedures or strain verification (Mullis et al ., 1986). PCR requires a three-step cycling process. In the first step, denaturation at 95 °C separates double-stranded DNA. In the second step, single-stranded primers attach to the dissociated DNA, while each primer is complementary to either 5' end or 3' end of the sequence of interest. In the third step, DNA polymerase synthesizes the new DNA strands starting from the respective primer (Schochetman et al ., 1988). Primer design was performed with the software “Clone Manager 7.11“ (Scientific & Educational Software, Cary, NC USA) or NEBuilder® Assembly Tool (New England Biolabs® Inc., Ipswich, MA, USA). For different DNA polymerases, the polymerase chain reaction mixtures changed slightly (see chapter 2.5.2.1).

2.5.2.1 Composition of PCR reaction mixtures with different DNA polymerases In this study, the amplification of DNA fragments for proof of plasmid presence in recombinant strains was done with Taq DNA polymerase (Genaxxon Bioscience GmbH, Ulm, Germany). In contrast, amplification of genes, promoters, or terminators for cloning procedures, for which proof reading is obligatory, was performed with ReproFast DNA polymerase (Genaxxon Bioscience GmbH, Ulm, Germany). If problems occurred with DNA amplification (e.g. no PCR product), the ReproFast DNA polymerase was used with a slightly modified method including “FailSafe TM PCR Premix E” (Epicentre Technologies Corp., an Illumina company, Madison, WI, USA). This premix contains a PCR enhancer with betaine which is helpful to amplify DNA fragments. DNA fragments to be used in the cloning procedure “In-Fusion® HD Cloning Kit” (Takara Bio USA Inc., Mountain View, CA, USA), were amplified with the “CloneAmp TM HiFi polymerase” (Takara Bio USA Inc., Mountain View, CA, USA).

PCR with Taq DNA polymerase

DNA template 3.00 µl ~ 10 .00 ng/ µl dNTPs (10 mM each) 1.00 µl ~ 1 0.2 0 mM each

10 x Buffer with MgCl 2 5.00 µl ~ 10 .00 % [v/v] Primers (100 µM each) 0.50 µl each ~ 1 1.0 0 µM each Taq polymerase 0.25 µl ~ 1 0.04 U/ µl

H2O ad 50 µl 36 2. Material and Methods

PCR with ReproFast polymerase

DNA template 3.00 µl ~ 10 .00 ng/ µl dNTPs (10 mM each) 1.00 µl ~ 1 0.2 0 mM e ach

10 x ReproFast buffer with MgSO 4 5.00 µl ~ 10 .00 % [v/v] Primers (100 µM each) 0.50 µl each ~ 1 1.0 0 µM each ReproFast polymerase (5 U/ µl) 0.25 µl ~ 1 0.0 3 U/ µl

H2O ad 50 µl PCR with ReproFast polymerase and FailSafe TM PCR Premix E

DNA template 03.00 µl ~ 10 .00 ng/ µl FailSafe TM PCR Prem ix E 25.00 µl ~ 50 .00 % [v/v] Primers ( 100 µM each ) 00.50 µl each ~ 1 1.0 0 µM each ReproFast polymerase (5 U/ µl) 00. 5 µl ~ 1 0.05 U/ µl

H2O ad 50 µl PCR with CloneAmp TM HiFi PCR Premix

DNA template 01.00 µl <100 .0 ng 2 x CloneAmp TM HiFi PCR Premix 12.50 µl <0 50 .0 % [v/v] Primers ( 10 µM each) 01.00 µl each <00 0.2 µM

H2O ad 25 µl 2.5.2.2 Standard PCR program The standard PCR program was performed with the following conditions:

1. Initial denaturation 95 °C 30 0 s 2. Denaturati on 95 °C 045 s 3. Annealing Variable* 045 s 4. Elongation 72 °C Variable ** 5. Final elongation 72 °C 600 s 6. Cooling 12 °C For ever *The annealing temperature was either calculated with the software Tm calculator (Thermo Fisher Scientific Inc., Waltham, MA, USA) or NEBuilder® Assembly Tool (New England Biolabs® Inc., Ipswich, MA,MA,USA) USA ) **Elongation time was calculated by size of the amplified DNA fragment and speed of the utilized DNA polymerase: Taq polymerase (1,000 bps/60s) and ReproFast polymerase (600 bps/60s) 2. Material and Methods 37

Steps 2-4, which represent one complete PCR cycle, were repeated 32 times before going to step 5. For PCRs with “CloneAmp TM HiFi PCR Premix” (Takara Bio USA Inc., Mountain View, CA, USA) a slightly modified PCR program was used:

1. Denaturation 95 °C 010 s 2. Annealing 55 °C 015 s 3. Elongation 72 °C Variable* 4. Cooling 04 °C For ever *Elongation time was calculated by size of the amplified DNA fragment and speed of the CloneAmp TM HiFi polymerase (1,000 bps/30-60 s)

Steps 1-3, which are one complete PCR cycle, were repeated 30 times before going to step 4.

2.5.2.3 Insertion of restriction recognition sequences via PCR By means of PCR, restriction recognition sequences can be attached to the DNA fragment of interest. The desired restriction recognition sequence was linked to the 5'-end of the utilized primer. Additionally, the bases ACA were set upstream of the restriction recognition sequence, since restriction enzymes require several bases to bind and cut DNA.

2.5.3 DNA modification

2.5.3.1 Restriction enzyme digestion of DNA Restriction enzyme digestion was used for analytical and preparative digest of plasmids and DNA fragments for cloning procedures and recombinant strain analysis, respectively. Composition of the reaction mixture, when using one restriction enzyme, was prepared following the instructions of the manufacturer, whereas the setting for using two restriction enzymes was planned with the “DoubleDigest TM tool” (Thermo Fisher Scientific Inc., Waltham, MA, USA). If vector was digested with only one enzyme for linearization, dephosphorylation (see chapter 2.5.3.2) was mandatory after inactivation of restriction enzyme.

Analytical restriction enzyme digestion

DNA (plasmid) 3 µl ~ 150 .0 ng/ µl 10 x FastDigest green buffer 2 µl ~ 0 10.0 % [v/v] FastDigest restriction enzyme 1 µl 0~ 0 0.5 U/ µl

H2O ad 20 µl The analytical digestion was performed in 1.5-ml reaction tubes at 37 °C for 30 min. 38 2. Material and Methods

Preparative restriction enzyme digestion

DNA (plasmid or DNA fragment) 5 µl ~ 150 .0 ng/ µl 10 x F ast Digest green buffer 5 µl ~ 0 10.0 % [v/v] FastDigest restriction enzyme 2 µl ~ 0 00.5 U/ µl

H2O ad 50 µl The preparative digestion was performed in 1.5-ml reaction tubes at 37 °C for 30 min.

2.5.3.2 Dephosphorylation of digested plasmids To avoid religation of linearized plasmids, which were constructed by cutting with one restriction enzyme, the 5'-phosphate groups of the sticky ends must be removed prior to ligation (see chapter 2.5.3.3). For this purpose, the “FastAP Thermosensitive Alkaline Phosphatase” (Thermo Fisher Scientific Inc., Waltham, MA, USA) was the enzyme of choice. After inactivation of restriction enzymes used in the digestion reaction, 1 µl of FastAP Thermosensitive Alkaline Phosphatase (1 U/l) were added to the preparative digestion, vortexed and incubated at 37 °C for 30 min. After an inactivation step at 65 °C for 15 min, the dephosphorylated DNA was purified via “DNA Clean & Concentrator TM -5 Kit” (ZYMO Research Europe GmbH, Freiburg, Germany) (see chapter 2.5.4.1). This purified vector was ready to be used in DNA ligation.

2.5.3.3 DNA ligation In order to construct plasmids of choice, purified DNA fragments (e.g. digested PCR product or restriction enzyme digested DNA fragment; see chapter 2.5.3.1) and purified, digested vectors were linked by DNA ligation (Weiss et al ., 1968). The enzyme needed for DNA ligations is “T4 DNA ligase” (Thermo Fisher Scientific Inc., Waltham, MA, USA), which is capable to form phosphodiester bonds of exposed 5'-phosphate and adjacent 3'-hydroxyl terminal groups in double-stranded DNA molecules. This reaction requires ATP as cofactor, which is present in the DNA ligase buffer. ATP is a molecule, which is very unstable, since it contains three negatively charged phosphate groups. For this reason, ligase buffer is stored in 10-µl aliquots for not thawing and freezing too often. The ratio between insert DNA fragment and vector backbone was set in between 1:1 or 7:1. There was always a control for each ligation experiment, which did not contain any insert DNA. This reaction served as control for recircularization of vector backbone. Ligations were carried out by using the following reaction composition: 2. Material and Methods 39

Insert DNA 1-7 µl ~ 10 .00 -70 .00 ng/ µl Lin earised vector 0-1 µl 000 00 ~ 10 .00 ng/ µl 10 x T4 DNA ligase buffer 0-2 µl 00000 ~ 10 .00 % [v/v] T4 DNA ligase (5 U/ µl) 0-1 µl 00 000 0~ 0.25 U/ µl

H2O ad 20 µl Ligation reaction mixture was pipetted into a 0.2-ml microtube and was incubated on a thermal cycler for 1.5 h at 22 °C. After inactivation of T4 DNA ligase, ligation reaction mixture was transformed into chemically competent E. coli cells (see chapter 2.7.1.4). If ligation mixture was transformed into electrocompetent E. coli cells, ligation mixture was purified by dialysis (see chapter 2.5.4.3).

2.5.3.4 In-Fusion® HD Cloning Kit By using “In-Fusion HD Cloning Kit” (Takara Bio USA Inc., Mountain View, CA, USA) one or more DNA fragments can be cloned directly into any vector. The “In-Fusion Enzyme” fuses DNA fragments (e.g. PCR products and linearized vectors) by recognizing a 15-bp overlap at their ends. This overlap is engineered by primer design for amplification of desired DNA fragments. Designing of required primers (see chapter 2.1.3) was accomplished by using the web-based “NEBuilder® Assembly Tool” (New England Biolabs® Inc., Ipswich, MA, USA). “In-Fusion HD Cloning” belongs to the next generation cloning techniques like circular polymerase extension cloning (CPEC) or Gibson Assembly (Quan and Tian, 2011; Gibson et al ., 2009). DNA fragment and linearized vector were cut from agarose gel and purified (see chapter 2.5.5.1). The “In- Fusion cloning” reaction was performed by using the following set-up:

Purified PCR fragment 0.5 µl 50 -100 ng Purified linearized vector 4.0 µl 50 -200 ng 5 x In -Fusion HD enzyme premix 2.0 µl 20 % [v/v]

H2O ad 10 µl The reaction mixture was incubated at 50 °C for 15 min in 0.25-ml microtubes. Subsequently, sample was placed on ice and 5 µl of “In-Fusion cloning” reaction mixture was transformed into chemically competent E. coli cells (see chapter 2.7.1.5).

2.5.3.5 Radioactive labelling of DNA fragments In order to prepare [α-32 P]-labelled DNA probes (10mCi/ml) for Northern blot experiments “Random Primers DNA Labelling System Kit” (Thermo Fisher Scientific Inc., Waltham, MA, 40 2. Material and Methods

USA) was performed according to manufacturer’s instructions. For radioactive labelling, 25 ng (dissolved in 20 µl sterile water) of the desired DNA fragment were used. After addition of 5 µl stop buffer, non-incorporated radioactive bases were removed by using “illustra™ MicroSpin™ G-25 Columns” (GE Healthcare, Buckinghamshire, GB) (see chapter 2.5.4.2). For measurement of total radioactivity, 3 µl sample (1:1000 dilution) were transferred to a scintillation vessel and 4 ml of scintillation solution were added. The radioactivity of the sample was measured in a “Tri-Carb® 2800TR Liquid Scintillation Analyzer” (PerkinElmer LAS GmbH, Rodgau, Germany).

2.5.4 Purification of nucleic acids

2.5.4.1 Purification of amplified and digested DNA In order to purify DNA fragments after PCR (see chapter 2.5.2) or after DNA digest (see chapter 2.5.3.1) either “Zymoclean TM Gel DNA Recovery Kit” or “DNA Clean & ConcentratorTM -5 Kit” (ZYMO Research Europe GmbH, Freiburg, Germany) were used. If no unspecific PCR fragments were present or for purification of digested DNA fragments, the “Clean & Concentrator TM -5 Kit” (ZYMO Research Europe GmbH, Freiburg) was used to remove all undesired reaction components. The kit was used according to the manufacturer’s instructions with slight modifications. After washing the column, it was dried by centrifugation at 10,000 x g for 3 min. Finally, it was placed into a sterile 1.5-ml reaction tube and DNA elution was achieved by adding 10 µl sterile water to the column, and centrifugation at 10,000 x g for 30 s. If PCR reactions contained DNA fragments, caused by unspecific binding of primers or for purification of preparative digests (see chapter 2.5.3.1) the “Zymoclean TM Gel DNA Recovery Kit” (ZYMO Research Europe GmbH, Freiburg, Germany) was applied. At first, the DNA fragment of interest was cut with a scalpel from stained agarose gel (see chapter 2.5.5.1) by transferring the gel on a “UST-30L-8E” UV-transilluminator (λ = 365 nm; biostep GmbH, Burkhardtsdorf, Germany)). The gel slice was placed into a 1.5-ml reaction tube and weighed. Subsequently, the kit was used according to manufacturer’s instructions with the following modifications. As mentioned above, a drying step was performed at 10,000 x g for 3 min after the washing step and DNA was eluted with 10 µl of sterile water (10,000 x g, 30 s). 2. Material and Methods 41

2.5.4.2 Purification of radioactively labelled DNA For the purpose of separating radioactive labelled DNA from non-incorporated [α-32 P]-dATP, the “illustra™ MicroSpin™ G-25 Columns” (GE Healthcare, Buckinghamshire, GB) were used according to the manufacturer’s instructions.

2.5.4.3 Purification of ligated DNA by microdialysis If plasmid DNA was transformed after ligation into electrocompetent E. coli cells (see chapter 2.7.1.3), purification via microdialysis was mandatory. This step is required, since T4 DNA ligase buffer (Thermo Fisher Scientific Inc., Waltham, MA, USA) contains a high concentration of MgCl 2 (100 mM). This high salt concentration in ligation reaction mixture would cause arcing during electroporation. Microdialysis was performed with the complete ligation reaction mixture by using “MF TM -Membrane Filters” (0.025 µm; Merck KGaA, Darmstadt, Germany). The sample was pipetted onto the filter, which was placed in an ultra-pure water containing petri dish. After 15 min incubation, the scaled-down volume of ligation sample was ready to be used in transformation of electrocompetent E. coli cells.

2.5.5 Gel electrophoresis In order to separate nucleic acids according to their size, the method gel electrophoresis was applied. The phosphate groups of nucleic acids are loaded negatively and so they move to the anode in an electric field. Due to the molecular structure of agarose gels, nucleic acids move at different speeds through the gel. Thus, smaller fragments move faster than larger nucleic acids (Green and Sambrook, 2012).

2.5.5.1 Non-denaturating gel electrophoresis Agarose gel electrophoresis can be used to separate DNA fragments according to their size. In this manner, restriction digests and PCR DNA fragments were analyzed, using the size indicator “GeneRuler TM DNA ladder Mix” (Thermo Fisher Scientific Inc., Waltham, MA, USA). DNA with a size over 1000 bp was separated by 0.8 % agarose [w/v] and for smaller fragments 2.0 % agarose [w/v] was used. Agarose was dissolved in 1 x TAE buffer by heating in a microwave and then poured in an electrophoresis chamber equipped with a plastic comb. After polymerization, the gel was overlaid with 1 x tris-acetic acid-EDTA buffer (TAE) buffer and DNA sample was mixed with “6 x DNA loading dye” (Thermo Fisher Scientific Inc., Waltham, MA, USA) in a ratio of 5:1. This dye contains tracking components to monitor electrophoresis by migration of xylene cyanol FF (4160 bp) and bromphenol blue (370 bp). Electrophoresis was 42 2. Material and Methods performed at 120 V covering a distance of 7 cm. For visualizing the DNA at a wavelength of 312 nm with a “TFP-M/WL” UV-transilluminator (biostep GmbH, Burkhardtsdorf, Germany), the gel was stained with 0.5 % [w/v] ethidium bromide. In order to save the results of DNA staining, the gel was photographed by using a photo documentation facility (Decon Science Tec GmbH, Hohengandern, Germany).

50 x TAE buffer (Green and Sambrook, 2012)

Tris 242.3 g 02 M Glacial acetic acid 057.1 g 01 M EDTA 014.6 g 50 mM

H2O ad 1,000 ml The pH was adjusted to 7.5 with HCl.

2.5.5.2 Denaturating gel electrophoresis In order to separate RNA, denaturating gel electrophoresis was the method of choice (Lehrach et al ., 1977). For this purpose, 2.4 g agarose was boiled in 180 ml water (autoclaved twice) and afterwards 20 ml 10 x MOPS-EDTA-sodium acetate (MESA) buffer, 3.6 ml formaldehyde, and 2 µl ethidium bromide were added. After cooling down to 50 °C, the solution was poured in an electrophoresis chamber equipped with a plastic comb. Then, the gel was allowed to polymerize for 1 h. Afterwards, it was overlaid with 1 x MAE running buffer and pre- electrophoresis was performed at 100 V for 30 min. RNA samples were diluted 1:2 with “2 x RNA loading dye” (Thermo Fisher Scientific Inc., Waltham, MA, USA) and pipetted into the gel pockets. 3 µl of “RiboRuler TM High Range DNA Ladder” (Thermo Fisher Scientific Inc., Waltham, MA, USA) served as size indicator. The electrophoresis of RNA molecules was performed at 100 V for 3 h and 45 min. RNA was made visible by an “TFP-M/WL” UV transilluminator (λ = 312 nm; biostep GmbH, Burkhardtsdorf, Germany) and gel was photographed by using a photo documentation facility (Decon Science Tec GmbH, Hohengandern, Germany). Bands of the size standard were marked with a scalpel in the gel for later labelling by probe hybridization.

2. Material and Methods 43

10 x MESA buffer

MOPS 41.85 g 200 mM

Sodium acetate x 3 H 2O 06.80 g 050 mM EDTA, disodium salt 01.86 g 00 5 mM

H2O ad 1,000 ml The pH was adjusted to 7 with HCl.

1 x MESA running buffer

10 x MESA buffer 150 ml 10 % [v/v] Formald ehyde 030 ml 03 % [v/v]

H2O ad 1,500 ml 2.5.6 Determination of nucleic acid concentration The absorption of DNA was measured at wavelengths of 260 nm and 280 nm with a spectral photometer “NanoDrop 2000c” (Thermo Fisher Scientific Inc., Waltham, MA, USA). Ratio of absorptions (A260/A280) informs about contamination of DNA with proteins. The ratio for DNA should be in between 1.7 and 2.0 and for RNA between 2.0 and 2.2. If the absorption accounts for 1, the concentration of double stranded DNA is 50 µg/ml and for RNA 40 µg/ml, respectively (Green and Sambrook, 2012). The solvent of DNA and RNA served as blank value.

2.6 DNA sequencing and data analysis

2.6.1 Sequencing of vectors and DNA fragments Sequencing of vectors or DNA fragments was done by GATC Biotech AG (Constance, Germany) using custom primers or universal primers. The method “SUPREMErun” is based on Sanger sequencing. It requires purified DNA (30-100 ng/µl for plasmids and 10-50 ng/µl for DNA fragments in a volume of 20 µl) in 1.5-ml reaction tubes. Resulting sequences were analyzed using “Clone Manager 7.11“ (Scientific & Educational Software, Cary, NC, USA) for plasmid validation after cloning and transformation processes. Validation of bacterial strains via 16S rDNA sequencing was achieved with Basic Local Alignment Search Tool (BLAST) by using the method “Nucleotide BLAST” (BLASTN). This browser-based tool of the National Center for Biotechnological Information (NCBI) (http://blast.ncbi.nlm.nih.gov) aligns different nucleotide databases using a nucleotide query (Altschul et al ., 1990). If any mismatches of bases appeared, chromatograms of the original sequencing data were further analyzed. 44 2. Material and Methods

2.6.2 Genome sequencing Chromosomal DNA of C. ljungdahlii DSM 13528 and C. ljungdahlii [pMTL82151] was isolated using the “MasterPure TM Gram Positive DNA Purification Kit” (Epicentre, Madison, WI, USA)

(see chapter 2.5.1.3). Sequencing of respective strains was performed by the Göttingen Genomics Laboratory (Göttingen, Germany). Resequencing of C. ljungdahlii wildtype was done to search for possible sequence errors in the original sequence in order to exclude false positive single nucleotide polymorphisms (SNPs). Illumina shotgun libraries were generated from the extracted DNA according to the manufacturer’s protocol. Sequencing was performed using a MiSeq instrument and the MiSeq reagent kit version 3, as recommended by the manufacturer (Illumina Inc., San Diego, CA, USA) resulting in 3,480,880 paired end reads for C. ljungdahlii [pMTL82151] and 3,092,112 paired end reads for DSM 13528 with a size of 301 bp. Quality filtering using Trimmomatic version 0.36 (Bolger et al ., 2014) resulted in 3,231,581 reads for the recombinant and 3,016,568 reads for the wild type strain, respectively.

2.6.3 Mapping for single nucleotide polymorphism (SNP) analysis Bowtie2 (Langmead and Salzberg, 2012) and SAMtools (Li, 2011) were used to map the trimmed reads of C. ljungdahlii [pMTL82151] on the reference genome (CP001666.1). Variant calling was performed using VarScan v2.3.9 (Koboldt et al ., 2012) with the pileup2snp option. Finally, changes in genome sequence and resulting changes in amino acid translation were edited with an unpublished in-house tool of Göttingen Genomics Laboratory (Göttingen, Germany).

2.6.4 Comparison of locus tags of different acetogens The orthology detection tool “Proteinortho” (Lechner et al ., 2011) version 4.26 (default specification with following cut off settings: blast = blastp v2.2.24, E-value = 1 -6, algorithm- connection = 0.1, coverage = 0.3, percent_identity = 25, adaptive_similarity = 0.95, incorporated_pairs = 1, incorporated_singles = 1, selfblast = 1, unambiguous = 0;) was used by the Göttingen Genomics Laboratory (Göttingen, Germany) to find orthologous genes within genome sequences of different acetogenic species. 2. Material and Methods 45

2.7 DNA transfer in bacteria

2.7.1 DNA transfer in Escherichia coli

2.7.1.1 Preparation of electrocompetent E. coli cells The modified method of choice for preparing electrocompetent E. coli cells was adapted from a published protocol (Dower et al ., 1988). Respective E. coli strain was grown overnight in 5 ml LB medium (see chapter 2.3.1.4). This sample was used to inoculate 250 ml LB medium in a baffled 2-l Erlenmeyer flask to an OD 600nm of 0.1. Cells were incubated at 37 °C and

150 rpm to an OD 600nm of 0.5-0.8. Afterwards, culture was transferred to a centrifugation vessel and incubated on ice for 20 min. Cells were harvested at 4,000 x g for 10 min at 4 °C. After two washing steps with 250 ml chilled and sterile water (4,000 x g, 10 min, 4°C), cells were resuspended in 50 ml cold 10 % [v/v] glycerol. Thereupon, centrifugation at 4,000 x g for 10 min at 4 °C followed and culture was suspended in 30 ml cold 10 % [v/v] glycerol. After a final centrifugation step at 4,000 x g for 10 min at 4 °C, cells were resuspended in 1 ml cold

10 % [v/v] glycerol and 50-µl aliquots were first flash-freezed in liquid N2 and then stored in 1.5-ml reaction tubes at -80 °C or directly used for electroporation procedure.

2.7.1.2 Preparation of fast competent E. coli cells For preparing smaller amounts of electrocompetent E. coli cells, 5 ml of an overnight grown culture were transferred into two 2-ml reaction tubes. All centrifugation steps were performed at room temperature. After centrifugation at 5,040 x g for 1 min, cells were washed twice with 1 ml chilled, sterile water (5,040 x g, 1 min). Subsequently, culture was resuspended in cold 10 % glycerol [v/v] and after a final centrifugation at 5,040 x g for 3 min, suspended in 100 µl cold 10 % glycerol [v/v]. For electroporation 50-µl aliquots were used. Short-term storage of fast competent cells was achieved at -80 °C.

2.7.1.3 Transformation of electrocompetent E. coli cells According to the described protocol for transformation of electrocompetent or fast competent E. coli cells (Dower et al ., 1988), 50 µl of cells (see chapter 2.7.1.1 and 2.7.1.2) were thawed on ice and mixed with approx. 500 ng of purified plasmid DNA (see chapter 2.5.4) or dialyzed ligation (see chapter 2.5.4.3). The sample was transferred to pre-cooled electroporation cuvettes with a gap width of 2 mm (Biozym Scientific GmbH, Hessisch Oldendorf, Germany). The electric pulse (2500 V, 25 µF, and 200 Ω; retention time 4.6-5.0 ms) was performed with “Gene Pulser Xcell TM pulse generator” (Bio-Rad Laboratories GmbH, 46 2. Material and Methods

Munich, Germany). Afterwards, 800 µl sterile LB medium (see chapter 2.3.1.4) were added to the pulsed cells and transferred to a sterile 1.5-ml reaction tube. After incubation at 37 °C and 1350 rpm, the sample was centrifuged at 6,000 x g for 1 min at room temperature, 800 µl of supernatant were discarded, cells were suspended in the remaining suspension, and plated on selective LB agar plates (see chapter 2.3.1.4). Plates were incubated upside down at 37 °C until colonies appeared.

2.7.1.4 Preparation of chemically competent E. coli cells Chemically competent E. coli cells were prepared according to a previously described protocol (Inoue et al ., 1990). 5 ml of an overnight grown E. coli culture were used to inoculate 250 ml

SOB medium (see chapter 2.3.1.6) in a baffled 2-l Erlenmeyer flask to an OD 600nm of 0.1. Cells were incubated at 18 °C, until they reached an OD 600nm of 0.6-0.8. Thereafter, the culture was transferred to centrifugation vessels and incubated on ice for 30 min. After cells were harvested at 3,900 x g for 10 min at 4 °C, they were suspended in 40 ml tris-borate (TB) buffer and incubated on ice for another 10 min. Subsequently, cells were centrifuged again at 3,900 x g for 10 min at 4 °C, the sediment was suspended in 10 ml TB buffer and 1.5 ml of sterile DMSO were added. Finally, 200-µl aliquots were flash-freezed in liquid N2 and stored at -80 °C or used directly for transformation.

TB buffer

Solution I

PIPES 756 mg 10 mM

CaCl 2 420 mg 15 mM

H2O ad 125 ml The pH was adjusted to 6.7 with KOH.

Solution II

KCl 4.66 g 250.0 mM

MnCl 2 x 4 H 2O 1.72 g 034.7 mM

H2O ad 125 ml Both solutions were autoclaved separately and combined afterwards. 2. Material and Methods 47

2.7.1.5 Transformation of chemically competent E. coli cells In order to transform chemically competent E. coli cells, the following modified protocol was used (Inoue et al ., 1990). Cells were thawed on ice, then 20 µl of ligation mixture (see chapter 2.5.3.3), 5 µl of “In-Fusion® HD Cloning Kit” sample (see chapter 2.5.3.4), or 500 ng plasmid DNA were added, and incubated on ice for 10 min. Heat shock was performed at 42 °C for 1 min. Afterwards, cells were incubated on ice for another 10 min, 800 µl pre-warmed (37 °C) LB medium (see chapter 2.3.1.4) were added, and further incubated at 1350 rpm and 37 °C for 1 h. Thereafter, cells were centrifuged at 4,000 x g for 1 min and 800 µl of supernatant were discarded. Cell pellet was resuspended in the remaining LB medium and plated on selective agar plates, which were incubated upside down at 37 °C, until colonies appeared.

2.7.2 DNA transfer in A. woodii and Clostridium by electroporation

2.7.2.1 Preparation of electrocompetent A. woodii and Clostridium cells The protocol used for preparing electrocompetent A. woodii and Clostridium cells was slightly modified from a previously described method (Leang et al ., 2013). All plastic material (e.g. syringes, 5 µg plasmid DNA in 1.5-ml reaction tubes, electroporation cuvettes) was placed in an anaerobic chamber the day before transformation. An early stationary growth phase culture of the respective strain was used to inoculate 100 ml modified ATCC 1612 medium (see chapter 2.3.1.2) or Tanner medium (see chapter 2.3.1.7), which contained

DL-threonine (150 mM) and fructose (40 mM), to an OD 600nm of 0.09. DL-Threonine is used to permeabilize the cell wall, but the exact mechanism is not known (Zhang et al. , 2011). A. woodii and Clostridium were incubated over night at 30 °C and 37 °C, respectively, until they reached an OD 600nm of 0.3-0.7. Cells were harvested in 50-ml falcon tubes in an anaerobic chamber by centrifugation at 6,000 x g for 10 min at 4 °C. Afterwards, cells were washed twice with 30 ml chilled SMP buffer (see chapter 2.4.3.2) (6,000 x g, 10 min, 4°C) and suspended in 600 µl SMP buffer and 120 µl anti-freezing buffer (see chapter 2.4.3.2). Cells were used directly for transformation or stored as 500-µl aliquots in cryogenic vials at -80 °C for further use.

2.7.2.2 Transformation of electrocompetent A. woodii and Clostridium In an anaerobic chamber, electrocompetent cells were first transferred to a 1.5-ml reaction tube containing 5 µg of plasmid DNA and incubated on ice for 5 min. Then, sample was placed to a pre-cooled electroporation cuvette with a gap width of 1 mm (Biozym Scientific GmbH, Hessisch Oldendorf, Germany) and electroporation was performed with “Gene Pulser Xcell TM 48 2. Material and Methods pulse generator” (Bio-Rad Laboratories GmbH, Munich, Germany) at 0.625 kV, 25 µF, and 600 Ω with a retention time of 9-11 ms. Afterwards, 800 µl from a Hungate tube with respective medium were transferred to the cuvette and the mixture was placed back into the

Hungate tube. This tube was incubated at 37 °C, until an increase in OD 600nm was observed. In case of Clostridium , 600 µl cells were plated in an anaerobic chamber on selective YTF agar plates (see chapter 2.3.1.8) until colonies appeared. Single colonies were selected in the anaerobic chamber and transferred to Hungate tubes using a syringe. In contrast, A. woodii was mixed with 5 µl of the respective antibiotic directly in the Hungate tube. When cells were growing in presence of antibiotic, they were inoculated into fresh medium containing the respective antibiotic, genomic DNA was isolated (see chapter 2.5.3.1) and recombinant strain was verified via PCR targeting 16S rDNA (fD1 and rP2) and vector backbone (see chapter 2.4.2), respectively.

2.7.3 DNA transfer in Clostridium by conjugation Conjugation of plasmid DNA in Clostridium was performed with a previously described, slightly modified protocol (Mock et al ., 2015). This protocol is based on methods for conjugation of Clostridium acetobutylicum and Clostridium difficile (Purdy et al ., 2002). As conjugation donor strain, E. coli CA434 was used to transfer large plasmid DNA in Clostridium. By conjugation, restriction modification (R-M) systems of bacteria can be partially eluded. At first, the plasmid to be conjugated into Clostridium was transformed in fast competent E. coli CA434 cells (see chapter 2.7.1.3). The resulting recombinant E. coli CA434 strain was inoculated the day before conjugation into 5 ml LB medium (see chapter 2.3.1.4), containing tetracycline and the respective antibiotic depending on the plasmid used, and was incubated overnight at 37 °C. In contrast, Clostridium was inoculated in Tanner medium (40 mM fructose) 9 h earlier than E. coli CA434, since it grows more slowly. On the day of conjugation, 2 ml LB medium containing respective antibiotics were inoculated with 100 µl of the overnight grown E. coli

CA434. When both organisms reached an OD 600nm of about 0.5, both cultures were transferred into an anaerobic chamber. E. coli CA434 was transferred to a sterile 2-ml reaction tube, and centrifuged at 3,000 x g for 3 min at room temperature. Supernatant was discarded, cell pellet was carefully washed with sterile phosphate-buffered saline (PBS) and centrifuged again as mentioned above. Thereafter, cells were carefully mixed with 200 µl of Clostridium culture and spotted onto YTF agar plates (see chapter 2.3.1.8) without any antiobiotic selection. After 2. Material and Methods 49 incubation of plates at 37 °C for 24 h without inverting, cells were harvested by flooding the agar plates with 500 µl PBS. Before flooding the plates, a small piece of agar at the lower part of the plate was removed by using a sterile inoculation loop. This way, the PBS/cell mixture aggregated in the gap, remaining cells were scraped off with the same inoculation loop and pipetted on selective YTF agar plates supplemented with 10 μg/ml colistin (to counterselect the E. coli CA434 donor strain) and 7.5 μg/ml thiamphenicol (to select plasmid uptake). Plates were incubated upside down for 3-4 days at 37 °C, until colonies appeared and conjugants were re-streaked on a new selective YTF agar plate, incubated again under the same conditions to further eliminate E. coli CA434 cells and to get single colonies which were inoculated in 5 ml modified Tanner medium. When cells were growing in liquid medium, genomic DNA was isolated (see chapter 2.5.1.3) and the recombinant strain was verified via PCR of 16S rDNA (fD1 and rP2) (see chapter 2.4.2) and PCR of vector backbone (see chapter 2.4.2), respectively.

PBS

NaCl 8.0 g 0.14 M

Na 2HPO 4 x 2 H 2O 1.4 g 7.86 mm KCl 0.2 g 2.68 mm

KH 2PO 4 0.2 g 1.47 mM

H2O ad 1,000 ml 2.7.4 Northern blot RNA of Clostridium ljungdahlii was investigated via Northern blot analysis. By using this method, RNA transcripts were analyzed according to their size and amount (Alwine et al ., 1977).

2.7.4.1 Transfer of RNA onto a nylon membrane Separation of 10 µg RNA was achieved in a denaturating agarose gel (see chapter 2.5.5.1) according to its size and amount. Afterwards, RNA was transferred to a nitrocellulose membrane using capillary blot. The setup of the Northern blot is depicted in Figure 4. One piece of Whatman paper (20 x 12 cm; Whatman International Ltd., Maidstone, GB ) was placed on a blotting device (Wissenschaftliche Werkstatt Feinwerktechnik, Ulm University, Germany), which was filled with 10 x saline-sodium citrate (SSC) buffer. The ends of this Whatman paper were dipped in 10 x SSC buffer to ensure capillary force. Three more layers of Whatman paper 50 2. Material and Methods

Figure 4: Scheme of Northern blot setup

(14.5 x 20 cm), the agarose gel (see chapter 2.5.5.2), one piece of nitrocellulose membrane (MACHEREY-NAGEL GmbH & Co. KG, Düren, Germany), one more Whatman paper, a paper towel stack, and a weight (~500 g) were placed onto the block (Figure 4). The capillary force originates from the paper towel stack and 10 x SSC buffer. SSC buffer migrates to the paper towel stack and transfers negatively loaded RNA from agarose gel to the positively loaded nitrocellulose membrane. Blotting was performed for 15 h at room temperature and thereafter the nitrocellulose membrane was dried at 37 °C. Finally, RNA was covalently fixed to the membrane by using “Ultraviolet Crosslinker” (GE Healthcare EUROPE GmbH, Munich, Germany) at a wavelength of 254 nm for 90 s and 120,000 µJ/cm 2.

10 x SSC buffer

NaCl 87.7 g 3.0 M

Tri sodium citrate x H 2O 44.1 g 0.3 M

H2O ad 1,000 ml 2.7.4.2 Hybridization of radioactively labelled probe with RNA In order to hybridize the radioactively labelled probe with RNA, nitrocellulose membrane was washed initially with RNase-free water and then overlaid with hybridization buffer. Afterwards, the membrane was transferred without any bubbles and overlapping to a hybridization tube and 20 ml hybridization buffer were added. Then, pre-hybridization was performed in a “rotary oven OV2” (Biometra GmbH, Göttingen, Germany) at 65 °C for 1 h. Denaturation of the radioactively labelled DNA probe was achieved by boiling at 95 °C for 5 min and afterwards it was added to the pre-hybridized membrane. Hybridization was 2. Material and Methods 51 finished after incubation for 18 h at 65 °C in the rotary oven. After discarding hybridization buffer, the membrane was washed twice with 50 ml wash buffer I. At first, buffer was discarded immediately after washing, while the second washing step was performed for 15 min at 65 °C in the rotary oven. Thereupon, two further washing steps with 50 ml wash buffer II and wash buffer III followed under the same conditions as mentioned above. Before transferring the nitrocellulose membrane to a film cassette, it was wrapped in cling film, a phosphor imaging plate was placed on the wrapped membrane for 5 h and 24 h, and finally analyzed by a “Fujifilm BAS-1000 Bio Imaging Analyzer IPR 1000” (Fuji Photo Film Co., Ltd., Tokyo, Japan) with the software “Image Reader V 1.2” (Fuji Photo Film Co., Ltd., Tokyo. Japan). Phosphor imaging plates allow computed radiography and were always cleaned with an eraser (raytest Isotopenmessgeräte GmbH, Straubenhardt, Germany) before and after use.

Hybridization buffer

5 x Hybridization buffer 10.0 ml 20 % [v/v] 50 x Denhardt solution 10.0 ml 20 % [v/v] 10 % SDS 02.5 ml 05 % [v/v]

H2O ad 50 ml Incubation at 65 °C for 15 min, then addition of:

NaCl 02.9 g 990 mM Herring sperm DNA 00.5 ml 010 mg/ml [w/v] 5 x Hybridization solution

Tris 3.03 g 250.0 mM Sodium pyrophosphate 0.50 g 018.8 mM

H2O ad 100 ml The pH was adjusted to 7.5 with HCl

50 x Denhardt solution

Bovine serum albumin 1 g 1 % [ w/v] (BSA; albumin fraction V) Polyvinylpyrrolidone - K 30 1 g 1 % [w /v] Ficoll 400 1 g 1 % [w /v]

H2O ad 100 ml

52 2. Material and Methods

Wash buffer I

10 x SSC buffer 60 ml 20 % [v/v] 10 % SDS [w/v] 03 ml 01 % [v/v]

H2O ad 300 ml Wash buffer II

10 x SSC buffer 10 ml 10 % [v/v] 10 % SDS [w/v] 01 ml 01 % [v/v]

H2O ad 100 ml Wash buffer III

10 x SSC buffer 1 ml 1 % [v/v] 10 % SDS [w/v] 1 ml 1 % [v/v]

H2O ad 100 ml 3. Results 53

3. Results

3.1 Improvement of 2,3-BD production in acetogenic bacteria

3.1.1 Natural 2,3-BD production in acetogenic bacteria At first, a suitable host for improvement of natural 2,3-BD production using syngas as energy and carbon source was characterized. It was already reported that C. autoethanogenum ,

A 160 5 B 140 6

140 120 4 120 1 100 4 1 100 3 80 80 60 60 2 0.1 2 2,3-BD [mM] 2,3-BD [mM] 40 Growth [OD600nm] 40

Growth [OD600nm] 1

20 Acetate, Ethanol [mM] 20 Acetate, Ethanol [mM] 0.1 0 0 0.01 0 0 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 Time [h] Time [h]

C 60 4 D 80 1 1 70 50 3 60 40

600nm] 50

30 2 40 0.1 0.1 30 20 2,3-BD [mM] 1 20 Acetate [mM] Growth [OD

Growth [OD600nm] 10 Acetate, Ethanol [mM] 10

0.01 0 0 0.01 0 0 100 200 300 400 500 600 700 0 50 100 150 200 250 300 Time [h] Time [h] E 50

1 3 40

600nm] 30 2

0.1 20 1 Butyrate [mM] 10 Growth [OD Acetate, Ethanol [mM]

0.01 0 0 0 50 100 150 200 250 300 350 Time [h]

Figure 5: Growth and production profile of C. autoethanogenum (A), C. ljungdahlii (B) in 50 ml Tanner medium, C. ragsdalei (C) in 50 ml Rajagopalan medium, C. aceticum (D) in 50 ml m odified 1.0 ATCC 1612 medium, and C. carboxidivorans (E) in 50 ml Tanner medium (3 g yeast extract), growing with syngas (1.8 bar overpressure) as energy and carbon source. Growth ( ), growth control ( ), acetate production ( ) , ethanol production ( ), butyrate production ( ), and 2,3-BD production production ( ). 54 3. Results

C. ragsdalei , and C. ljungdahlii naturally produce 1.4-2 mM 2,3-BD from steel mill waste gas (Köpke et al., 2011b). In order to determine the best acetogenic natural 2,3-BD producer, the above-mentioned organisms as well as C. aceticum and C. carboxidivorans were examined in autotrophic growth experiments. All organisms were cultivated in 500-ml glass flasks containing 50 ml of the respective medium using syngas as energy and carbon source (overpressure 1.8 bar). All growth experiments were performed as biological duplicates. Growth and production profile of C. autoethanogenum , C. ljungdahlii, C. ragsdalei , C. aceticum , and C. carboxidivorans with syngas is depicted in Figure 5. The organisms C. autoethanogenum , C. ljungdahlii, and C. ragsdalei produced mainly acetate (52.6 mM-138.7 mM) and ethanol (21.8 mM-49.1 mM). Only small amounts of 2,3-BD were detected with a maximal concentration of 3.7 mM, 4.8 mM, and 3.1 mM, respectively (Figure 5A and B and C). In contrast, C. aceticum produced only acetate (71.4 mM) without any by-products, and C. carboxidivorans produced acetate (42.6 mM) and ethanol (26.7 mM) in addition to small amounts of butyrate (2.4 mM) (Figure 5D and E). This experiment demonstrated that C. ljungdahlii showed the highest natural 2,3-BD production with a concentration of 4.8 mM. For further genetic engineering experiments, C. ljungdahlii was the organism of choice in order to enhance 2,3-BD production. Furthermore, there is an efficient and reproducible transformation protocol available (Leang et al ., 2013), which facilitates genetic engineering of this organism. Moreover, the acetogen C. coskatii represents a further interesting organism for enhancement of natural 2,3-BD overproduction showing 98.3 % average nucleotide identity (ANI) compared to C. ljungdahlii (Bengelsdorf et al ., 2016). Analysis with “Proteinortho” detection tool revealed that C. coskatii harbors the homologous genes alsS (CCOS_39110), budA (CCOS_38400), and bdh (CCOS_06940) for 2,3-BD production. So, C. coskatii was also taken into account to overproduce 2,3-BD.

3.1.2 Construction of plasmids for enhancement of natural 2,3-BD production For homologous overproduction of 2,3-BD, the three genes alsS (acetolactate synthase; CLJU_c38920), budA (acetolactate decarboxylase; CLJU_c08380), and bdh (2,3-BD dehydrogenase; CLJU_c23220) from C. ljungdahlii (Köpke et al ., 2011b; 2014) were synthesized in form of an operon (GenScript USA Inc., Piscataway, NJ, USA) under control of promoter P pta-ack from C. ljungdahlii . This native promoter is constitutive and originates from pta-ack operon of C. ljungdahlii (Hoffmeister et al ., 2016; Köpke and Mueller, 2016). The 3. Results 55

Figure 6: 2,3-BD production in C. ljungdahlii and C. coskatii originating from pyruvate via enzymes AlsS (acetolactate synthase), BudA (acetolactate decarboxylase), Bdh (2,3-BD dehydrogenase), and LdhA (lactate dehydrogenase). enzymes acetolactate synthase (AlsS), acetolactate decarboxylase (BudA), and 2,3-BD dehydrogenase (Bdh) promote 2,3-BD production starting from two molecules of pyruvate (Figure 6). The synthetic 2,3-BD operon was received in form of the plasmid pUC57_23BD (GenScript USA Inc., Piscataway, NJ, USA (Figure 7)). In order to clone the synthetic 2,3-BD operon into vectors pMTL82151 and pJIR750, the respective plasmid and vectors were digested using restriction enzymes Sac I and Sal I. Subsequently, ligation of the purified 2,3-BD

Figure 7: Cloning strategy for construction of plasmids pMTL82151_23BD and pJIR750_23BD. Plasmid pUC57_ 23BD, as well as vectors pMTL82151 and pJIR750 were digested using restriction enzymes Sac I and Sal I. 2,3-BD operon (P pta-ack , alsS , budA , and bdh ) was ligated with linearized pMTL82151 and pJIR750, resulting in plasmids pMTL82151_23BD and pJIR750_23BD. P pta-ack , (promoter region upstream of pta-ack operon from C. ljungdahlii ); alsS , acetolactate synthase from C. ljungdahlii ; budA , acetolactate decarboxylase from C. ljungdahlii ; bdh , 2,3-BD dehydrogenase from C. ljungdahlii ; bla , ampicillin resistance gene; rep , ColE1, replicon for Gram-negative bacteria; repA /orf2 , pBP1-replicon for Gram-positive bacteria from C. botulinum ; rep , pMB1-replicon for Gram-negative bacteria; rep , pIP404-replicon for Gram-positive bacteria from C. perfringens ; catP , chloramphenic ol resistance gene; traJ , gene for conjugal transfer function. 56 3. Results operon into the linearized vectors was performed. The resulting plasmids were named pMTL82151_23BD and pJIR750_23BD (Figure 7). Successful cloning was confirmed by restriction enzyme digestion of these plasmids using Sac I and Sal I as well as sequencing (M13- FP, M13-RP, pMTL82151_23BDSeq1, pMTL82151_23BDSeq2, pMTL82151_23BDSeq3, pMTL82151_23BDSeq4, pMTL82151_23BDSeq5) by GATC Biotech AG (Constance, Germany). 3.1.3 Homologous 2,3-BD overproduction in C. ljungdahlii The empty vectors pMTL82151 and pJIR750, as well as the two different 2,3-BD overproduction plasmids pMTL82151_23BD and pJIR750_23BD (Figure 7) were transformed into C. ljungdahlii . The transformed strains growing in presence of thiamphenicol were verified using two different methods. On the one hand, isolation of genomic DNA from transformed C. ljungdahlii strains was performed and catP of pMTL82151 backbone (expected DNA fragment 1,121 bp) or pJIR750 backbone (expected DNA fragment 1,075 bp) was amplified via PCR (pMTL82151catP_fwd, pMTL82151catP_rev, pJIR750catP_fwd, pJIR750catP_rev) (Figure 8A). On the other hand, isolated genomic DNA from recombinant C. ljungdahlii strains was transformed into chemically competent E. coli XL-1 Blue MRF' cells and single colonies growing on agar plates containing chloramphenicol were picked for plasmid DNA isolation.

A B C M 1 2 3 4 M 1 2 3 4 5 M 6 7 1 2 3 4 5 M bp bp bp bp 10,000----- 6,000 ----- 5,0 00 -- 3,000 ------1,500 1,500 --- 2,0 00 --- 1,000-- 1,000 --- --1,000 700 -- 1,5 00 ---

Figure 8: Analysis of DNA fragments resulting from amplification of catP (A), a nalytical restriction enzyme digestion of retransformed plasmids (B), and amplification of 16S rDNA gene of genomic DNA from recombinant C. ljungdahlii strains (C) using 0.8 % agarose gels. GeneRuler DNA Ladder Mix (M). (A) 1,121-bp (pMTL82151 backbone) and 1,075-bp (pJIR750 backbone) DNA fragments resulting from PCR of catP from C. ljungdahlii [pMTL82151] (1), C. ljungdahlii [pMTL82151_23BD] (2), C. ljungdahlii [pJIR750] (3), and C. ljungdahlii [pJIR750_23BD] (4). (B) Analytical restriction enzyme digestion using Apa I of retransformed pMTL82151 with expected DNA fragments 4,488 bp and 766 bp (1), using Bam HI and Bsu 15I of retransformed pMTL82151_23BD with expected DNA fragments 7,983 bp and 1,663 bp (2-4), using Bam HI and Sac I of retransformed pJIR750_23BD with expected DNA fragments 7,890 bp and 3,070 bp (5), using Nde I of retransformed pJIR750 with expected DNA fragments 4,773 bp and 1,825 bp (7), and undigested control (6). (C) 1,500-bp DNA fragment resulting from amplification of 16S rDNA gene of C. ljungdahlii WT as control (1), C. ljungdahlii [pMTL82151] (2), C. ljungdahlii [pMTL82151_23BD] (3), C. ljungdahlii [pJIR750] (4), and C. ljungdahlii [pJIR750_23BD] (5). 3. Results 57

This procedure is called retransformation and plasmid DNA isolated from recombinant E. coli XL-1 Blue MRF' is called retransformed plasmid. Figure 8B and C show the analytical restriction enzyme digestions using Apa I of retransformed pMTL82151 (expected DNA fragments 4,488 bp, 766 bp), Nde I of retransformed pJIR750 (expected DNA fragments 4,773 bp, 1,825 bp), Bam HI and Bsu 15I of retransformed pMTL82151_23BD (expected DNA fragments 7,983 bp, 1,663 bp), and Bam HI and Sac I of retransformed pJIR750_23BD (expected DNA fragments 7,890 bp, 3,070 bp). Additionally, amplification of 16S rDNA gene (1,500 bp; fD1, rP2) and concomitant sequencing of the resulting DNA fragment by GATC Biotech AG (Constance, Germany) supported correct identity of the recombinant C. ljungdahlii strains (Figure 8D). Using these methods, recombinant strains C. ljungdahlii [pMTL82151], C. ljungdahlii [pJIR750], C. ljungdahlii [pMTL82151_23BD], and C. ljungdahlii [pJIR750_23BD] were verified. The recombinant C. ljungdahlii strains were analyzed under both, heterotrophic and autotrophic growth conditions.

Figure 9 presents growth, fructose consumption, and production pattern of C. ljungdahlii wildtype (WT), C. ljungdahlii [pMTL82151], C. ljungdahlii [pJIR750], C. ljungdahlii [pMTL82151_23BD] and C. ljungdahlii [pJIR750_23BD] under heterotrophic conditions. Growth experiments were performed as biological triplicates in 125-ml glass flasks containing 50 ml Tanner medium with 40 mM fructose as energy and carbon source. All strains showed a similar growth behaviour. C. ljungdahlii WT and C. ljungdahlii [pMTL82151] reached a max.

OD 600nm of 2.2 after 72 h while the max. OD 600nm of C. ljungdahlii [pJIR750] and C. ljungdahlii [pJIR750_23BD] was 2.0 after 72 h. The strain C. ljungdahlii [pMTL82151] displayed the lowest max. OD 600nm of 1.7 after 48 h. In the medium of C. ljungdahlii WT and C. ljungdahlii [pJIR750_23BD] fructose concentration in the beginning was a little lower with 33.0 mM and 30.2 mM, respectively. Furthermore, the highest acetate concentration was detected for C. ljungdahlii WT with 102.0 mM, while for the other strains acetate concentration reached values between 87.9 mM and 96.8 mM. Regarding ethanol production, C. ljungdahlii [pMTL82151] showed a slightly higher ethanol production with 11.2 mM compared to the other strains ranging from 4.9 mM to 8.3 mM. In contrast, C. ljungdahlii [pJIR750_23BD] and C. ljungdahlii [pMTL82151_23BD] showed slightly higher 2,3-BD concentrations with 1.2 mM and 1.1 mM, respectively, in comparison to the strains with the empty vectors and the WT strain (0.8 mM-1.0 mM). This shows that C. ljungdahlii [pJIR750_23BD] and C. ljungdahlii 58 3. Results

] 50 1 40 600nm 30 0.1 20 10 Fructose [mM]

Growth [OD 0.01 0 120 100 80 60 40 20

Acetate [mM] 0 12 10 8 6 4 2 Ethanol [mM] 0 1.4 1.2 1.0 0.8 0.6 0.4 0.2 2,3-BD [mM] 0.0 0 25 50 75 100 125 150 175 200 225 Time [h]

Figure 9: Growth (solid line), fructose consumption (dashed line), and production pattern (solid lines) of C. ljungdahlii WT ( ), C. ljungdahlii [pMTL82151] ( ), C. ljungdahlii [pJIR750] ( ), C. ljungdahlii [pMTL82151_23BD] ( ), and C. ljungdahlii [pJIR750_23BD] ( ) growing in 50 ml Tanner medium containing 40 mM fructose as carbon source.

[pMTL82151_23BD] produced slightly higher amounts of 2,3-BD with fructose as carbon source compared to C. ljungdahlii WT, C. ljungdahlii [pMTL82151], and C. ljungdahlii [pJIR750].

Afterwards, the 2,3-BD production with syngas was examined. The growth experiment was performed in 1-l glass flasks containing 100 ml Tanner medium and syngas as energy and carbon source with 1.8 bar overpressure. Figure 10 shows growth and production pattern of C. ljungdahlii WT, C. ljungdahlii [pMTL82151], C. ljungdahlii [pJIR750], C. ljungdahlii [pMTL82151_23BD], and C. ljungdahlii [pJIR750_23BD] under autotrophic growth conditions. Regarding growth, it is noticeable that C. ljungdahlii [pJIR750] showed a more slowly and decreased growth with a max. OD 600nm of 0.7 after 1,344 h in comparison to the other strains. 3. Results 59 ] 1 600nm

0.1

Growth [OD 0.01 350 300 250 200 150 100

Acetate [mM] 50 0 120 100 80 60 40

Ethanol [mM] 20 0 8

6

4

2 2,3-BD [mM] 0 0 500 1000 1500 2000 2500 Time [h] Figure 10 : Growth and production patte rn of C. ljungdahlii WT ( ), C. ljungdahlii [pMTL82151] ( ), C. ljungdahlii [pJIR750] ( ), C. ljungdahlii [pMTL82151_23BD] ( ), and C. ljungdahlii [pJIR750_23BD] ( ) growing in 100 ml Tanner medium with syngas (1.8 bar overpressure).

The highest OD 600nm of 1.8 after 1,008 h was detected for C. ljungdahlii WT compared to C. ljungdahlii [pMTL82151] with 1.5 after 840 h, C. ljungdahlii [pJIR750_23BD] with 1.4 after 432 h, and C. ljungdahlii [pMTL82151_23BD] with 1.2 after 840 h. With respect to acetate production, all recombinant C. ljungdahlii strains showed decreased acetate concentrations (260.4 mM-241.6 mM) compared to C. ljungdahlii WT (287.8 mM). The stain C. ljungdahlii [pJIR750_23BD] produced least acetate with a concentration of 219.6 mM. In contrast, all recombinant C. ljungdahlii strains except of C. ljungdahlii [pJIR750] (60.4 mM) showed an increased ethanol production compared to C. ljungdahlii WT (71.9 mM) with concentrations of 82.0 mM to 83.3 mM. The highest ethanol concentration of 92.8 mM showed C. ljungdahlii [pMTL82151]. Regarding 2,3-BD production, only C. ljungdahlii [pJIR750_23BD] 60 3. Results produced more 2,3-BD (6.4 mM) than C. ljungdahlii WT (4.4 mM). In contrast, C. ljungdahlii [pMTL82151_23BD] did not produce more 2,3-BD (3.8 mM) than C. ljungdahlii WT.

In summary, heterotrophic (Figure 9) as well as autotrophic (Figure 10) growth experiments with C. ljungdahlii WT and the recombinant C. ljungdahlii strains revealed that C. ljungdahlii [pJIR750_23BD] produced higher amounts of 2,3-BD compared to C. ljungdahlii WT. In contrast, C. ljungdahlii [pMTL82151_23BD] did only produce higher amounts of 2,3-BD compared to C. ljungdahlii WT under heterotrophic growth conditions.

3.1.4 Homologous 2,3-BD overproduction in C. coskatii For investigation of 2,3-BD overproduction in C. coskatii , the vectors pMTL82151 and pJIR750 as well as the two 2,3-BD production plasmids pMTL82151_23BD and pJIR750_23BD (Figure 7) were transformed into C. coskatii . The respective recombinant C. coskatii strains were verified either by isolation of genomic DNA and amplification (pMTL82151catP_fwd, pMTL82151catP_rev, pJIR750catP_fwd, pJIR750catP_rev) of catP originating from pMTL82151 backbone (expected DNA fragment 1,121 bp; Figure 11A) and pJIR750 backbone (expected DNA fragment 1,075 bp; Figure 11A), respectively or via retransformation of genomic DNA from recombinant strains into E. coli XL1-Blue MRF' and subsequent plasmid isolation and analytical restriction enzyme digestion. Figure 11B depicts the digestion of

Figure 11 : Analysis of DNA fragments resulting from amplification of catP (A), analytical restriction enzyme digestion (B), and amplification of 16S rDNA gene of genomic DNA from recombinant C. coskatii strains (C) using 0.8 % agarose gels. GeneRuler DNA Ladder Mix (M). (A) 1,121-bp (pMTL82151 backbone) and 1,075-bp (pJIR750 backbone) DNA fragments resulting from PCR of catP from C. coskatii [pMTL82151] (3), C. coskatii [pMTL82151_23BD] (4), C. coskatii [pJI R750] (7), C. coskatii [pJIR750_23BD] (8), positive control with empty vector pMTL82151 (2) and pJIR750 (6) as template, and negative control with sterile water as template (1, 5). (B) Analytical restriction enzyme digestion using Sal I and Sac I of retran sformed pJIR750_23BD with expected DNA fragments 6,545 bp and 4,415 bp (1), using Nde I of retransformed pJIR750 with expected DNA fragments 4,773 bp and 1,825 bp (2), using Eco RI of retransformed pMTL82151 with expected DNA fragment 5,254 bp (3) and of ret ransformed pMTL82151_23BD with expected DNA fragment 5,284 bp, 4,338 bp, and 22 bp (4). (C) 1,500-bp DNA fragment resulting from amplification of 16S rDNA gene of C. coskat ii [pJIR750_23BD] (3), C. coskatii [pMTL82151] (6), C. coskatii [pMTL82151_23BD] (7), C. coskat ii [pJIR750] (4), and C. coskatii [pJIR750] (8), negative control with sterile water as template (1, 4), and positive control with genomic DNA from C. coskatii WT (2, 5). 3. Results 61 retransformed pJIR750_23BD using Sal I and Sac I (expected DNA fragments 6,545 bp, 4,415 bp), retransformed pJIR750 using Nde I (expected DNA fragments 4,773 bp, 1,825 bp), retransformed pMTL82151 using Eco RI (expected DNA fragment 5,254 bp), and retransformed pMTL82151_23BD using Eco RI (expected DNA fragments 5,284 bp, 4,338 bp, 22 bp). Furthermore, amplification of 16S rDNA gene (1,500 bp; fD1, rP2), sequencing of DNA fragment by GATC Biotech AG (Constance, Germany) with subsequent BLAST analysis supported correct identity of the recombinant strains (Figure 11C). This way, recombinant strains C. coskatii [pMTL82151], C. coskatii [pJIR750], C. coskatii [pMTL82151_23BD], and C. coskatii [pJIR750_23BD] were verified and subsequently growth and production patterns were investigated under heterotrophic, as well as autotrophic growth conditions. Heterotrophic growth experiments were performed as biological triplicates in 125-ml glass flasks containing 50 ml Tanner medium with 30 mM fructose as energy and carbon source.

Figure 12 presents growth, fructose consumption, and production pattern of C. coskatii WT , C. coskatii [pMTL82151], C. coskatii [pJIR750], C. coskatii [pMTL82151_23BD], and C. coskatii [pJIR750_23BD]. In terms of growth behaviour, no significant difference between C. coskatii WT and the recombinant strains occurred. Although this heterotrophic growth experiment was conducted with less fructose (30 mM) compared to C. ljungdahlii (Figure 9), none of the C. coskatii strains were able to consume fructose completely. The C. coskatii strains reached max. OD 600nm values of 2.2-1.9 after 219 h. Furthermore, C. coskatii [pJIR750_23BD] and C. coskatii [pMTL82151] produced more acetate (92.0 mM and 78.0 mM) than C. coskatii WT (67.8 mM). Moreover, it is noticeable that C. coskatii [pMTL82151_23BD] produced more ethanol (15.7 mM) compared to the other tested strains (6.1 mM-7.7 mM), but it did not produce more 2,3-BD (0.3 mM) than C. coskatii WT (1.1 mM). Also, C. coskatii [pJIR750_23BD] produced a similar 2,3-BD concentration (1.0 mM) compared to C. coskatii WT.

All C. coskatii strains were also investigated under autotrophic growth conditions with syngas (Figure 13). The growth experiment was performed in 500-ml glass flasks containing 50 ml Tanner medium and syngas as energy and carbon source with 1.0 bar overpressure. Overpressure of syngas in anaerobic glass flasks was reduced from 1.8 bar to 1 bar, since the problem of bursting glass flasks occurred, when overpressure was too high and moreover 500-ml glass flasks were more stable than 1-l glass flasks. Regarding growth of the different 62 3. Results ]

1 30 600nm 20 0.1 10 Fructose [mM] Growth [OD 0.01 0 100 80 60 40 20 Acetate [mM] 0 20

15

10

5 Ethanol [mM] 0 1.2 1.0 0.8 0.6 0.4 0.2 2,3-BD [mM] 0.0 0 25 50 75 100 125 150 175 200 225 Time [h]

Figure 12 : Growth (solid line), fructose consu mption (dashed line), and production pattern (solid lines) of C. coskatii WT ( ), C. coskatii [pMTL82151] ( ), C. coskatii [pJIR750] ( ), C. coskatii [pMTL82151_23BD]( ), and C. coskatii [pJIR750_23BD] ( ) growing in 50 ml Tanner medium containing 30 mM fructose. C. coskatii strains under autotrophic growth conditions (Figure 13), it was striking that all strains grew more slowly and to decreased OD 600nm values of 0.8. The strain C. coskatii

[pJIR750] showed lowest growth (OD 600nm 0.5) compared to C. ljungdahlii (OD 600nm 1.8; Figure 10). In terms of acetate production, all C. coskatii strains showed very similar acetate concentrations (113.7 mM-117.3 mM) with C. coskatii [pJIR750] producing least acetate (101.4 mM), which can be ascribed to decreased growth. According to the results of the growth experiment using fructose as carbon source (Figure 12), C. coskatii [pMTL82151_23BD] produced more ethanol (7.6 mM) than the other C. coskatii strains (2.3-3.0 mM) and regarding 2,3-BD concentration, it is noticeable that C. coskatii WT produced less 2,3-BD than 3. Results 63

] 1 600nm

0.1

Growth [OD Growth 0.01 125 100 75 50 25 Acetate [mM] Acetate 0 8

6

4

2 Ethanol Ethanol [mM] 0 2.0

1.5

1.0

0.5 2,3-BD [mM] 2,3-BD

0.0 0 250 500 750 1000 1250 1500 1750 2000 2250 Time [h]

Figure 13: Growth and production pattern of C. coskatii WT ( ), C. coskatii [pMTL82151] ( ), C. coskatii [pJIR750] ( ), C. coskatii [pMTL82151_23BD] ( ), and C. coskatii [pJIR750_23BD] ( ) growing in 50 ml Tanner medium with syngas (1.0 bar overpressure). C. ljungdahlii WT. Moreover, the recombinant C. coskatii strains did not show enhancement in 2,3-BD production compared to C. coskatii WT.

3.1.5 Modifications of 2,3-BD production plasmid pJIR750_23BD An optimization of the 2,3-BD production plasmid pJIR750_23BD was obligatory to further enhance the homologous 2,3-BD production in C. ljungdahlii . In autotrophic growth experiments with syngas, C. ljungdahlii [pJIR750_23BD] produced higher amounts of 2,3-BD (6.4 mM) than C. ljungdahlii [pMTL82151_23BD] (3.8 mM) (Figure 10). For this reason, further modification steps of 2,3-BD plasmids were only performed with pJIR750_23BD. 64 3. Results

3.1.5.1 Removal of a potential terminator structure The DNA analytical program “Clone Manager 7.11“ (Scientific & Educational Software, Cary, NC, USA) was applied to analyze the DNA sequence of the synthetic 2,3-BD operon. Due to this analysis, a hairpin loop structure was identified downstream of budA gene (Figure 14). Although a poly-A tail is missing, it could not be excluded that transcription of bdh from plasmid pJIR750_23BD is aborted at this terminator structure. For removal of the potential terminator structure downstream of budA , a shortened sequence of budA was amplified (CljbudAKpnI_fwd, Clj-budABamHI_rev) using genomic DNA from C. ljungdahlii WT. The amplified DNA fragment, as well as pJIR750_23BD were digested with restriction enzymes Kpn I and Bam HI and afterwards ligated. Successful cloning was confirmed by analytical restriction digestion with Kpn I and Sal I, as well as by sequencing (pMTL82151_23BDSeq3, pMTL82151_23BDSeq4) by GATC Biotech AG (Constance, Germany). The resulting plasmid was named pJIR750_23BD_budAshort and has a size of 10,867 bp, which is 93 bp shorter than pJIR750_23BD with 10,960 bp (Figure 15). The modified plasmid pJIR750_23BD_budAshort was transformed into C. ljungdahlii WT. Transformation was only successful after in vivo methylation using E. coli ER2275 [pACYC184_MCljI] and isolation of plasmid DNA using “QIAGEN Plasmid Midi Kit” (QIAGEN-tip 100; Qiagen GmbH, Hilden, Germany). Afterwards,the transformed strain growing in presence of thiamphenicol was verified by PCR of 16S rDNA gene (fD1, rP2; expected DNA fragment 1,500 bp; Figure 16C) with genomic DNA as template followed by sequencing (fD1, rP2) by GATC Biotech AG (Constance, Germany). The presence

Tbdh [bp]

free energy: -9.3 kcal/mol

Figure 14 : 2,3 -BD operon containing P pta-ack (promoter region upstream of pta -ack operon from C. ljungdahlii), alsS (acetolactate synthase from C. ljungdahlii ), budA (acetolactate decarboxylase from C. ljungdahlii ), bdh (2,3-BD dehydrogenase from C. ljungdahlii ), and T bdh (terminator of bdh from C. ljungdahlii ) of plasmid pJIR750_23BD showing hairpin loop structure after budA gene with respective sequence and energy content. 3. Results 65

Figure 15 : Cloning strategy for construction of pJIR750_23BD_budAshort. Plasmid pJIR750_23BD as well as DNA fragment budA (short) were digested with restriction enzymes Kpn I and Bam HI. DNA fragment budAshort was ligated with linearized pJIR750_23BD, resulting in plasmid pJIR750_23BD_budAshort. Ppta-ack , promoter region upstream of pta-ack operon from C. ljungdahlii ; alsS , acetolactate synthase from C. ljungdahlii ; budA , acetolactate decarboxylase from C. ljungdahlii ; bdh , 2,3-BD dehydrogenase from C. ljungdahlii ; rep , pMB1-replicon for Gram-negative bacteria; rep , pIP404-replicon for Gram-positive bacteria from C. perfringens ; catP , chloramphenicol resistance gene. of plasmid was confirmed by PCR of catP (pJIR750catP_fwd, pJIR750catP_rev; expected DNA fragment 1,075 bp; Figure 16A) with genomic DNA as template as well as analytical restriction enzyme digestion with Kpn I and Bam HI (expected DNA fragments 10,120 bp, 747 bp; Figure 16B) after retransformation of genomic DNA from the recombinant C. ljungdahlii strain into E. coli XL1-Blue MRF' and subsequent plasmid isolation.

Figure 16 : Analysis of DNA fragments resulting from amplification of catP (A), analytical restriction enzyme digestion (B), and amplification of 16S rDNA gene of genomic DNA from recombinant C. ljungdahlii strain (C) using 0.8 % agarose gels. GeneRuler DNA Ladder Mix (M). (A) 1,075-bp DNA fragments resulting from PCR of catP from C. ljungdahlii [pJIR750_23BD_budAshort] (3), positive control with empty vector pJIR750 as template (2), and negative control with sterile water as template (1). (B) Analytical restriction enzyme digestion using Kpn I and Bam HI of retransformed pJIR750_23BD_budAshort with expected DNA fragments 10,120 bp and 747 bp (1). (C) 1,500-bp DNA fragment resulting from amplification of 16S rDNA gene of C. ljungdahlii [pJIR750_23BD_budAshort] (2), and water as template for negative control (1). 66 3. Results

3.1.5.2 Exchange of bdh with adh (primary-secondary alcohol dehydrogenase) As a second approach to improve the pJIR750_23BD production plasmid, the gene bdh , which encodes 2,3-BD dehydrogenase, was exchanged with the adh gene (primary-secondary alcohol dehydrogenase). Recently, the primary-secondary alcohol dehydrogenase (sAdh) from C. autoethanogenum (CAETHG_0553) was identified, which is able to convert acetoin to 2,3- BD and also acetone to 1,3-propanediol (Köpke et al ., 2014). The identical adh gene (CLJU_c24860) is also present in the genome of C. ljungdahlii . Instead of using NADH as reducing equivalent for reduction of acetoin to 2,3-BD, sAdh requires NADPH (Figure 17). To test whether the primary-secondary alcohol dehydrogenase shows a higher performance for 2,3-BD production in C. ljungdahlii , plasmid pJIR750_23BD_budAshort_adh was constructed (Figure 18). At first, adh was amplified (CljADHBamHI_fwd, CljADHSalI_rev) using genomic DNA from C. ljungdahlii WT. Subsequently, purified adh gene and plasmid pJIR750_23BD_budAshort were linearized using restriction enzymes Bam HI and Sal I and ligated with DNA fragment adh . Confirmation of successful cloning was achieved by analytical restriction enzyme digestion using Bam HI and Sal I and sequencing (M13-FP) by GATC Biotech AG (Constance, Germany). The resulting plasmid was named pJIR750_23BD_budAshort_adh (Figure 18). It was also necessary to methylate the plasmid in vivo , using E. coli ER2275 [pACYC184_MCljI] prior to transformation into C. ljungdahlii and isolation of plasmid DNA was performed with “QIAGEN Plasmid Midi Kit” (QIAGEN-tip 100; Qiagen GmbH, Hilden, Germany) to receive high plasmid DNA concentrations. Verification of the C. ljungdahlii

Figure 17 : 2,3 -BD production in C. ljungdahlii originating from pyruvate via enzymes AlsS (acetolactate synthase), BudA (acetolactate decarboxylase), and sAdh (primary-secondary alcohol dehydrogenase). 3. Results 67

Figure 18 : Cloning strategy for construction of pJIR750_23BD_budAshort_adh. Plasmid pJIR750_23BD_budAshort, as well as DNA fragment adh wer e digested with restriction enzymes Bam HI and Sal I. DNA fragment adh was ligated with linearized pJIR750_23BD_budAshort, resulting in plasmid pJIR750_23BD_budAshort_adh. Ppta-ack , promoter region upstream of pta-ack operon from C. ljungdahlii ; alsS , acetolactate synthase from C. ljungdahlii ; budA , acetolactate decarboxylase from C. ljungdahlii ; bdh , 2,3-BD dehydrogenase from C. ljungdahlii ; adh , primary-secondary alcohol dehydrogenase from C. ljungdahlii ; rep , pMB1-replicon for Gram-negative bacteria; rep , pIP404- replicon for Gram-positive bacteria from C. perfringens ; catP , chloramphenicol resistance gene.

[pJIR750_23BD_budAshort_adh] strain was done by PCR of catP (pJIR750catP_fwd, pJIR750catP_rev; expected DNA fragment 1,075 bp; Figure 19A) using genomic DNA as template as well as analytical restriction enzyme digestion using Bam HI and Sal I (expected DNA fragments 9,615 bp, 1,183 bp; Figure 19B) after retransformation of genomic DNA from recombinant C. ljungdahlii strain into E. coli XL1-Blue MRF' and subsequent plasmid isolation.

Figure 19 : Analysis of DNA fragments resulting from amplification of catP (A), analytical restriction enzyme digestion (B), an d amplification of 16S rDNA gene of genomic DNA from recombinant C. ljungdahlii strain (C) using 0.8 % agarose gels. GeneRuler DNA Ladder Mix (M). (A) 1,075-bp DNA fragments resulting from PCR of catP from C. ljungdahlii [pJIR750_23BD_budAshort_adh] (3), positive control with empty vector pJIR750 as template (2), and negative control with sterile water as template (1). (B) Analytical restriction enzyme digestion using Bam HI and Sal I of retransformed pJIR750_23BD_budAshort_adh with expected DNA fragments 9,615 bp and 1,183 bp (1). (C) 1,500- bp DNA fragment resulting from amplification of 16S rDNA gene of C. ljungdahlii [pJIR750_23BD_budAshort_adh] (2), and water as template for negative control (1). 68 3. Results

Furthermore, 16S rDNA gene was amplified (fD1, rP2; expected DNA fragment 1,500 bp; Figure 19C) using genomic DNA from C. ljungdahlii [pJIR750_23BD_budAshort_adh] as template and sequenced (fD1, rP2) by GATC Biotech AG (Constance, Germany) to verify the recombinant strain.

3.1.5.3 BudABC operon from Raoultella terrigena As a third possibility for modification of pJIR750_23BD to enhance 2,3-BD production, the 2,3- BD operon budABC from Raoultella terrigena (formerly known as Klebsiella terrigena ; Blomqvist et al ., 1993) was cloned into vector pJIR750 under control of a promoter from

C. ljungdahlii (P pta-ack ) and a terminator from C. acetobutylicum (T sol-adc ). Tsol-adc is a strong rho- independent terminator, which is functional in sense as well as antisense direction. Figure 20 shows the 2,3-BD production pathway of R. terrigena , in which acetolactate decarboxylase (BudA), which converts S-acetolactate to acetoin, is encoded by budA , and the gene budB encodes acetolactate synthase (AlsS), which metabolizes pyruvate to S-acetolactate. The final reduction of R-acetoin or S-acetoin to meso -2,3-BD and 2S,3S -BD, respectively, is catalyzed by acetoin reductase (Acr), encoded by budC , which can also operate as 2,3-BD dehydrogenase and oxidizes 2,3-BD to acetoin with concomitant reduction of NAD +. In order to test heterologous 2,3-BD production with budABC operon from R. terrigena in C. ljungdahlii , plasmid pJIR750_budABCoperon was constructed. Firstly, plasmid pMK_budABCoperon was digested with Eco RI and Bam HI to obtain the budABC operon. The operon was ligated with linearized pJIR750 resulting in plasmid pJIR750_budABC (Figure 21). Afterwards, Ppta-ack was

Figure 20 : 2,3 -BD production in R. terrigena originating from pyruvate via enzymes AlsS (acetolactate synthase), BudA (acetolactate decarboxylase), and Acr (acetoin reductase). 3. Results 69

Figure 21 : Cloning strategy for construction of pJIR750_budABCoperon. Vector pJIR750 a s well as DNA fragment budABC were digested with restriction enzymes Eco RI and Bam HI . DNA fragment budABC was ligated with linearized pJIR750, resulting in plasmid pJIR750_budABC. Ppta-ack was cloned into linearized plasmid pJIR750_budABC via “In-Fusion HD cloning”, resulting in plasmid pJIR750_budABC_Ppta-ack. T sol-adc was digested with Xba I and Hin dIII and ligated with linearized plasmid pJIR750_budABC_Ppta-ack, resulting in plasmid pJIR750_budABCoperon. rep (pMB1) , pMB1-replicon for Gram-negative bacteria; rep (pIP404) , pIP404-replicon for Gram-positive bacteria from C. perfringens ; catP , chloramphenicol resistance gene; lacZ , β-galactosidase gene of lac -operon from E. coli ; budABC , budABC operon from R. terrigena ; Ppta-ack , promoter region upstream of pta- ack operon from C. ljungdahlii ; T sol-adc , terminator region downstream of sol-adc operon from C. acetobutylicum . amplified (CPECPpta-ackEcoRI_fwd, CPECPpta-ackEcoRI_rev) and cloned into linearized (with Eco RI digested) plasmid pJIR750_budABC using “In-Fusion HD Cloning” resulting in plasmid pJIR750_ budABC_Ppta-ack (Figure 21). To finish plasmid construction, Tsol-adc was amplified (TsoladcXbaI_fwd, TsoladcHindIII_rev ), DNA fragment was digested with restriction enzymes Xba I and Hin dIII, and ligated with linearized pJIR750_budABC_Ppta-ack. The resulting plasmid was named pJIR750_budABCoperon (Figure 21). Successful cloning of pJIR750_budABCoperon was confirmed by analytical restriction enzyme digestion using Nde I (expected fragments 5,425 bp, 3,092 bp, 1,825 bp) and sequencing (M13-RP, budABC_SEQ, BudABC_Seq1, BudABC_Seq2, BudABC_Seq3, BudABC_Seq4, M13_FP) by GATC Biotech AG (Constance, Germany). Plasmid pJIR750_budABCoperon was transformed into C. ljungdahlii , but again in vivo methylation and plasmid isolation with “QIAGEN Plasmid Midi Kit” (QIAGEN-tip 100; Qiagen GmbH, Hilden, Germany) was mandatory prior to the transformation process. Recombinant strain C. ljungdahlii [pJIR750_budABCoperon] was verified by both PCR of catP (pJIR750catP_fwd, pJIR750catP_rev; expected DNA fragment 1,075 bp; Figure 22A) with genomic DNA as template and analytical restriction enzyme digestion using Eco RI (expected 70 3. Results

Figure 22 : Analysis of DNA fragments resulting from amplification of catP (A), analytical restriction enzyme dige stion (B), and amplification of 16S rDNA gene of genomic DNA from recombinant C. ljungdahlii strain (C) using 0.8 % agarose gels. GeneRuler DNA Ladder Mix (M). (A) 1,075-bp DNA fragments resulting from PCR of catP from C. ljungdahlii [pJIR750_budABCoperon ] (2), positive control with empty vector pJIR750 as template (1). (B) Analytical restriction enzyme digestion using Eco RI of retransformed pJIR750_budABCoperon with expected DNA fragments 9,960 bp and 382 bp (1). (C) 1,500-bp DNA fragment resulting from amplification of 16S rDNA gene of C. ljungdahlii [pJIR750_budABCoperon] (2), and water as template for negative control (1).

DNA fragments 9,960 bp, 382 bp; Figure 22B) after retransformation of genomic DNA into E. coli XL1-Blue MRF' and subsequent plasmid isolation. Afterwards, identity of recombinant C. ljungdahlii strain growing in the presence of thiamphenicol was verified by PCR of 16S rDNA gene (fD1, rP2; expected DNA fragment 1,500 bp; Figure 22C) with genomic DNA as template followed by sequencing (fD1, rP2) by GATC Biotech AG (Constance, Germany).

3.1.5.4 Growth experiments of different C. ljungdahlii strains harboring the modified 2,3-BD plasmids The recombinant strains C. ljungdahlii [pJIR750_23BD_budAshort], C. ljungdahlii [pJIR750_23BD_budAshort_adh], and C. ljungdahlii [pJIR750_budABCoperon] were investigated under both heterotrophic and autotrophic growth conditions. Heterotrophic growth experiments were performed as biological triplicates in 125-ml glass flasks containing 50 ml Tanner medium as well as 40 mM fructose as energy and carbon source.

In Figure 23, growth, pH, fructose consumption, and production pattern of C. ljungdahlii WT, C. ljungdahlii [pMTL82151], C. ljungdahlii [pJIR750], C. ljungdahlii [pMTL82151_23BD], C. ljungdahlii [pJIR750_23BD], C. ljungdahlii [pJIR750_23BD_budAshort], C. ljungdahlii [pJIR750_23BD_budAshort_adh], and C. ljungdahlii [pJIR750_budABCoperon] under heterotrophic conditions is presented. In terms of growth behaviour, it was noticeable that

C. ljungdahlii [pJIR750_23BD], C. ljungdahlii [pMTL82151_23BD], as well as C. ljungdahlii 3. Results 71

6.5

6.0 ] 1 5.5 600nm 5.0 OD pH [

4.5

Growth 0.1 4.0

3.5 45 120 40 35 100 30 80 25 20 60

15 40 Acetate [mM] Acetate Fructose [mM] Fructose 10 20 5 0 0 40 8 35 7 30 6 25 5 20 4 15 3 2,3-BD[mM] Ethanol [mM] Ethanol 10 2 5 1 0 0 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 Time [h] Time [h]

Figure 23: Growth, pH, fructose consumption, and production pattern of C. ljungdahlii WT ( ), C. ljungdahlii [pMTL82151] ( ), C. ljungdahlii [pJIR750] ( ), C. ljungdahlii [pMTL82151_23BD] ( ), C. ljungdahlii [pJIR750_23BD] ( ), C. ljungdahlii [pJIR750_23BD_budAshort] ( ), C. ljungdahlii [pJIR750_23BD_budAshort_adh] ( ), and C. ljungdahlii [pJIR750_budABCoperon] ( ) growing in in 50 ml Tanner medium containing 40 mM fructose.

[pJIR750] showed decreased max. OD 600nm values compared to the other strains of 0.9, 1.7, and 1.6, respectively (Figure 23). Moreover, C. ljungdahlii [pMTL82151_23BD] and C. ljungdahlii [pJIR750] did not consume fructose completely. All other strains reached max.

OD 600nm values above 2. The pH decreased concomitantly with increasing acetate concentrations, but it is striking that C. ljungdahlii [pMTL82151] produced with 52.1 mM much less acetate than the other C. ljungdahlii strains (96.1 mM to 110.4 mM). Due to less acetate 72 3. Results production, pH of C. ljungdahlii [pMTL82151] decreased only to a value of 4.5, while all other strains showed pH values of 4.25 after 356 h. Regarding ethanol and 2,3-BD production, C. ljungdahlii [pMTL82151] showed even more differences compared to the other strains. After 66 h C. ljungdahlii [pMTL82151] produced 31.5 mM ethanol, but after 74 h ethanol concentration decreased to 19.3 mM with a subsequent increase to 24.1 mM. Also, C. ljungdahlii [pJIR750] produced higher ethanol amounts of 29.5 mM after 212 h. In terms of 2,3-BD production, C. ljungdahlii [pMTL82151] produced the highest 2,3-BD concentration of 6.5 mM, which is 3.6 times higher than C. ljungdahlii WT (1.8 mM). From the newly constructed strains C. ljungdahlii [pJIR750_23BD_budAshort], C. ljungdahlii [pJIR750_23BD_budAshort_adh], and C. ljungdahlii [pJIR750_budABCoperon], only C. ljungdahlii [pJIR750_23BD_budAshort_adh] produced higher amounts of 2,3-BD with 2.1 mM compared to C. ljungdahlii WT, but still lower concentrations than C. ljungdahlii [pJIR750_23BD] with 2.7 mM.

Figure 24 presents growth, pH, and production pattern of C. ljungdahlii WT, C. ljungdahlii [pMTL82151], C. ljungdahlii [pJIR750], C. ljungdahlii [pMTL82151_23BD], C. ljungdahlii [pJIR750_23BD], C. ljungdahlii [pJIR750_23BD_budAshort], C. ljungdahlii [pJIR750_23BD_budAshort_adh], and C. ljungdahlii [pJIR750_budABCoperon] under autotrophic conditions. Growth experiments were performed as biological triplicates in 500-ml glass flasks containing 50 ml Tanner medium with syngas (1 bar overpressure) as energy and carbon source. All strains showed a very similar growth behaviour with a max.

OD 600nm from 1.7 for C. ljungdahlii [pJIR750_23BD_budAshort] to 2.2 for C. ljungdahlii [pJIR750_23BD_budAshort_adh]. The pH dropped to 4 within 384 h for all C. ljungdahlii strains except for C. ljungdahlii [pJIR750], which decreased the pH within 816 h (Figure 24). Concerning max. acetate concentrations, C. ljungdahlii [pMTL82151] produced 120.1 mM followed by C. ljungdahlii WT and C. ljungdahlii [pJIR750_23BD_budAshort_adh] with 119.8 mM and 110.9 mM, respectively. In contrast, C. ljungdahlii [pJIR750_23BD], C. ljungdahlii [pJIR750_23BD_budAshort], and C. ljungdahlii [pJIR750_budABCoperon] produced less than 100 mM acetate (98 mM, 99 mM, and 98.6 mM, respectively). Regarding ethanol production, C. ljungdahlii [pMTL82151] produced higher ethanol concentrations (68.1 mM) compared to the other C. ljungdahlii strains (57.1 mM-31.6 mM), which was in accordance with the growth experiment under heterotrophic conditions (Figure 23). However, 3. Results 73

] 6 1 600nm

5 pH

0.1 Growth [OD 4

80 120 70 100 60

80 50 40 60 30 40 Ethanol [mM] Ethanol Acetate [mM] Acetate 20 20 10 0 0 4.5 0 500 1000 1500 2000 2500 3000 4.0 Time [h] 3.5 3.0 2.5 2.0 1.5 2,3-BD [mM] 2,3-BD 1.0 0.5 0.0 0 500 1000 1500 2000 2500 3000 Time [h]

Figure 24 : Growth, pH, and production pattern of C. ljung dahlii WT ( ), C. ljungdahlii [pMTL82151]( ) C. ljungdahlii [pJIR750] ( ), C. ljungdahlii [pMTL82151_23BD] ( ), C. ljungdahlii [pJIR750_23BD] ( ), C. ljungdahlii [pJIR750_23BD_budAshort] ( ), C. ljungdahlii [pJIR750_23BD_budAshort_adh] ( ), and C. ljungdahlii [pJIR750_budABCoperon] ( ) growing in 50 ml Tanner medium with syngas (1 bar overpressure). in terms of 2,3-BD production, C. ljungdahlii [pMTL82151] did not produce higher 2,3-BD concentrations (1.4 mM) than the other C. ljungdahlii strains (Figure 24), which does not reflect the results of the heterotrophic growth experiment (Figure 23). The highest 2,3-BD production is shown for C. ljungdahlii [pJIR750_23BD_budAshort_adh] (3.3 mM), which is very similar compared to C. ljungdahlii [pMTL82151_23BD], C. ljungdahlii [pJIR750_23BD], and 74 3. Results

C. ljungdahlii [pJIR750_23BD_budAshort] (3.1 mM, 2.9 mM, and 2.8 mM, respectively) and about two times higher than C. ljungdahlii WT (1.7 mM).

To sum it up, modifications of plasmid pJIR750_23BD did not lead to much higher 2,3-BD concentrations with 3.3 mM for C. ljungdahlii [pJIR750_23BD_budAshort_adh] compared to 2.9 mM for C. ljungdahlii [pJIR750_23BD] (Figure 24). Furthermore, the strains C. ljungdahlii [pJIR750_23BD_budAshort] and C. ljungdahlii [pJIR750_budABCoperon] did not produce more 2,3-BD than C. ljungdahlii [pJIR750_23BD] under autotrophic growth conditions (Figure 24). Nevertheless, a higher ethanol and 2,3-BD production was shown for C. ljungdahlii [pMTL82151] under heterotrophic growth conditions (Figure 23). Since this strain contains the vector pMTL82151 without any genes for 2,3-BD production, it was interesting to find the reason for higher ethanol and 2,3-BD concentrations in C. ljungdahlii [pMTL82151]. Table 6 and Table 7 present the summarized growth and production pattern of the investigated C. ljungdahlii strains under heterotrophic and autotrophic growth conditions.

Table 6: Summary of max. OD 600nm , max. acetate production, max. ethanol production, and max. 2,3-BD production of recombinant C. ljungdahlii strains in comparison to C. ljungdahlii WT under heterotrophic growth conditions.

Strain Max. Max. acetate Max. ethanol Max. 2,3 -BD

OD 600nm production [mM] production production [mM] [mM]/[mg/l] Heterotrophic growth condition s C. ljungdahlii WT 2.7 105.0 8.5 1.8 /162.2 C. ljungdahlii [pMTL82151] 2.0 52.1 31.5 6.5 /585.8 C. ljungdahlii [pJIR750] 1.6 96.1 7.2 1.0 /90.1 C. ljungdahlii [pMTL82151_23BD] 1.7 119.5 7.5 1.3 /117.2 C. ljungdahlii [pJIR750_23BD] 0.9 98.4 29.5 2.7 /243.3 C. ljungdahlii 2.3 95.1 6.2 1.5 /135.1 [pJIR750_23BD_budAshort] C. ljungdahlii 2.8 103.1 7.1 2.1 /189.3 [pJIR750_23BD_budAshort_adh] C. ljungdahlii 2.6 110.4 9.0 1.9 /171.2 [pJIR750_budABCoperon] 3. Results 75

Table 7: Summary of max. OD 600nm , max. acetate production, max. ethanol production, and max. 2,3-BD production of recombinant C. ljungdahlii strains in comparison to C. ljungdahlii WT under autotrophic growth conditions.

Strain Max. Max. acetate Max. ethanol Max. 2,3 -BD

OD 600nm production [mM] production production [mM] [mM]/[mg/l] Autotrophic growth conditions C. ljungdahlii WT 2.0 119.8 41.9 1.7 /153.2 C. ljungdahlii [pMTL82151] 1.9 120.1 68.1 1.5 /135.2 C. ljungdahlii [pJIR750] 1.5 105.4 40.6 1.1 /099.1 C. ljungdahlii [pMTL82151_23BD] 1.8 109.0 52.5 3.1 /27 9.4 C. ljungdahlii [pJIR750_23BD] 1.9 98.0 57.1 2.9 /261.3 C. ljungdahlii 1.7 99.0 42.2 2.8 /252.3 [pJIR750_23BD_budAshort] C. ljungdahlii 2.2 110.9 50.3 3.3 /297.4 [pJIR750_23BD_budAshort_adh] C. ljungdahlii 1.6 98.6 31.6 1.8 /162.2 [pJIR750_budABCoperon]

3.1.6 Sequencing of C. ljungdahlii [pMTL82151] and C . ljungdahlii WT with SNP analysis Growth experiments using fructose or syngas as energy and carbon source (Figure 23 and Figure 24) demonstrated that C. ljungdahlii [pMTL82151] produced decreased acetate concentrations accompanied by increased ethanol and 2,3-BD production. Thus, the question evoked why this production pattern occurred. Since C. ljungdahlii [pMTL82151] harbors the vector pMTL82151 without any genes for 2,3-BD production, there might be one or more mutations in genes on the chromosome that affect ethanol and 2,3-BD production. For this reason, sequencing of C. ljungdahlii [pMTL82151] and C. ljungdahlii WT was performed in cooperation with the Göttingen Genomics Laboratory (Göttingen, Germany). SNP analysis of C. ljungdahlii [pMTL82151] compared to C. ljungdahlii WT revealed many SNPs in different locus tags, which are presented in Table 12 (chapter 7. Supplement). Table 8 summarizes locus tags CLJU_c07590, CLJU_c07600, CLJU_c16510, CLJU_c16520, CLJU_c24850, and

76 3. Results

Table 8: Results of SNP analysis of C. ljungdahlii [pMTL82151] compared to C. ljungdahlii WT

Locus tag* Annotat ion AA position changed Original AA Changed AA

CLJU_c07590 putative secretion 1123, 1126, 1144, 1146, V, Y, H, H, A, G D, F, R, R, T, D protein HlyD 1148, 1171 CLJU_c07600 putative transporter 2794, 2797, 2821, 2836, D, G, M, Q, P, I G, V, K , P, H, R protein 2848, 2854

CLJU_c16510 bifunctional 544, 568, 607 M, A, S R, V , no AA aldehyde/alcohol dehydrogenase CLJU_c16520 bifunctional 520, 532, 1768, 1769 D, I, E, E A, S, D, D aldehyde/alcohol dehydrogenase

CLJU_c24850 signal -transduction 802, 1342 E, R V, T and transcriptional- control protein CLJU_c24870 signal -transduction 1817, 1818 M, M I, I and transcriptional- control protein

*according to IMG /MER (https://img.jgi.doe.gov/cgi-bin/mer/main.cgi), 25 th of april, 2017.

CLJU_c24870 showing one or more changed amino acids. The locus tag CLJU_c07590 (putative secretion protein HylD) showed six changed AA and CLJU_c07600 (putative transporter protein) displayed five changed AA. The locus tags CLJU_c16510 (gene adh1 ) and CLJU_c16520 (gene adh2 ), which are both annotated as bifunctional aldehyde/alcohol dehydrogenases, showed three and four changed AA, respectively. Moreover, locus tag CLJU_c16510 contained a stop codon at AA position 607 (length of total gene 2613 bp). This means that a shortened protein with 668 less amino acids was translated in C. ljungdahlii [pMTL82151] compared to C. ljungdahlii WT. Furthermore, locus tags CLJU_c24850 (gene stc1) and CLJU_c24870 (gene stc2 ), which are both annotated as signal-transduction and transcriptional-control proteins showed two changed AA. Therefore, strain C. ljungdahlii [pMTL82151] harboring these SNPs is designated as mutated C. ljungdahlii [pMTL82151] in further experiments. 3. Results 77

3.1.7 Detection and quantification of transcripts responsible for 2,3-BD production In a further experiment, mRNAs of the genes alsS , budA , and bdh were detected by Northern blot analysis to show that the transcript alsS -budA -bdh is completely expressed in recombinant strains C. ljungdahlii [pMTL82151_23BD], C. ljungdahlii [pJIR750_23BD], C. ljungdahlii [pJIR750_23BD_budAshort], and C. ljungdahlii [pJIR750_23BD_adh]. All C. ljungdahlii strains for Northern blot analysis were grown with 40 mM fructose and RNA was isolated after 65 h during early stationary growth phase, when 2,3-BD production starts, which was checked by HPLC analysis (data not shown). Figure 25 depicts the gene organisation of the synthetic 2,3-BD operon (alsS -budA -bdh ) as well as the genes alsS , budA , and bdh present on the chromosome of C. ljungdahlii with respective sizes of mRNA. For Northern blot experiment, mRNAs originating from the chromosome were detected by using RNA of C. ljungdahlii WT, C. ljungdahlii [pMTL82151], and C. ljungdahlii [pJIR750] and as negative control, RNA from E. coli XL1-Blue MRF' was applied. Figure 26 shows Northern blot analysis with probes alsS (Figure 26A), budA (Figure 26B), and bdh (Figure 26C), which were amplified using CljalsS-sonde_fwd, CljalsS-sonde_rev, CljBudA-sonde_fwd, CljBudA-sonde_rev, Clj23bdh-sonde_fwd, and Clj23bdh-sonde_rev, respectively. The probes were hybridized with RNA from C. ljungdahlii WT, C. ljungdahlii [pMTL82151], C. ljungdahlii [pJIR750], C. ljungdahlii

Figure 25: Schematic organisation of 2,3-BD genes on chromosome of C. ljungdahlii with size of transcript alsS (A), budA (B), and bdh (C) as well as sizes of synthetic 2,3-BD operons of overproduction plasmids pMTL81151_23BD as well as pJIR750_23BD (D), pJIR750_23BD_budAshort (E), and pJIR750_23BD_budAshort_adh (F). 78 3. Results

Figure 26 : No rthern blot analysis with alsS probe (A), budA probe (B), and bdh probe (C) hybridized with RNA of C. ljungdahlii WT (1), C. ljungdahlii [pMTL82151] (2), C. ljungdahlii [pJIR750] (3), C. ljungdahlii [pMTL82151_23BD] (4), C. ljungdahlii [pJIR750_23BD] (5), C. ljungdahlii [pJIR750_23BD_budAshort] (6), C. ljungdahlii [pJIR750_23BD_budAshort_adh] (7), and E. coli XL1- Blue MRF' as negative control (8); RiboRuler High Range RNA Ladder (M).

[pMTL82151_23BD], C. ljungdahlii [pJIR750_23BD], C. ljungdahlii [pJIR750_23BD_budAshort], C. ljungdahlii [pJIR750_23BD_budAshort_adh], and E. coli XL1-Blue MRF' as negative control. The transcript alsS from C. ljungdahlii chromosome was successfully detected in C. ljungdahlii WT at a size of 1,670 bases (Figure 26A, lane 1) and the transcript alsS-budA-bdh could be identified in C. ljungdahlii [pJIR750_23BD] at a size of 3,802 bases (Figure 26A, lane 5). In contrast, no signal at all was detected for C. ljungdahlii [pMTL82151] (Figure 26A, lane2), C. ljungdahlii [pMTL82151_23BD] (Figure 26A, lane 4), and C. ljungdahlii [pJIR750_23BD_budAshort_adh] (Figure 26A, lane 7). Moreover, a signal with a smaller size of about 1,000 bases was detected in the strain C. ljungdahlii [pJIR750] (Figure 26A, lane 3) and in C. ljungdahlii [pJIR750_23BD_budAshort] an additional signal at 1,500 bases occurred (Figure 26A, lane 6). Analysis of transcript budA of C. ljungdahlii WT (Figure 26B, lane 1) revealed a signal at size of about 1,250 bases, but it actually should show a length of 720 bases. Moreover, this transcript was also present in C. ljungdahlii [pMTL82151] and C. ljungdahlii [pJIR750_23BD_budAshort_adh] (Figure 26B, lane 2,7), but not in the other recombinant strains. A shorter transcript at a size of 750 bases was again existing in C. ljungdahlii [pMTL82151], C. ljungdahlii [pJIR750], C. ljungdahlii [pJIR750_23BD], and C. ljungdahlii [pJIR750_23BD_budAshort_adh]. Furthermore, the transcript alsS-budA-bdh is again detected solely in C. ljungdahlii [pJIR750_23BD] (Figure 26B, lane 5), but in C. ljungdahlii [pMTL82151_23BD] a very weak signal occured at a size of about 2,500 bases, which is shorter than the transcript alsS-budA-bdh. Figure 26C shows Northern blot analysis with bdh probe and transcript bdh with a size of 1,074 bases is present in C. ljungdahlii WT as well as C. ljungdahlii [pMTL82151] (Figure 26C, lane 1 and 2). The recombinant strain C. ljungdahlii [pJIR750_23BD_budAshort_adh] was not analyzed, since the plasmid contains the adh gene 3. Results 79 instead of the bdh gene and so detection of the transcript alsS -budAshort -adh is not possible by bdh probe in this strain. Besides, in accordance with results of alsS probe and budA probe, transcript alsS-budA-bdh was only detected in C. ljungdahlii [pJIR750_23BD] (Figure 26C, lane 5). Moreover, no transcript at all was detected in C. ljungdahlii [pMTL82151_23BD] and C. ljungdahlii [pJIR750_23BD_budAshort] and as mentioned above, a shorter signal at a size of 500 bases was present.

Summarizing the results of Northern blot analysis with probes alsS , budA , and bdh , it can be stated that transcript originating from C. ljungdahlii chromosome was always present in C. ljungdahlii WT but not in each recombinant strain. Furthermore, transcript for 2,3-BD overproduction was only detected in C. ljungdahlii [pJIR750_23BD].

3.1.8 Heterologous 2,3-BD production in A. woodii The acetogenic organism A. woodii is a promising organism for heterologous 2,3-BD production, since it was used previously for heterologous production of acetone with H 2 + CO 2 (Hoffmeister et al ., 2016). For that reason, A. woodii was also taken into account to overproduce 2,3-BD with the above mentioned 2,3-BD production plasmids (Figure 7, Figure 15, Figure 18, Figure 21). For investigation of 2,3-BD production in A. woodii , the five different unmethylated 2,3-BD production plasmids pMTL82151_23BD, pJIR750_23BD, pJIR750_23BD_budAshort, pJIR750_23BD_budAshort_adh, and pJIR750_budABCoperon were transformed into A. woodii. The recombinant strains A. woodii [pMTL82151] and A. woodii [pJIR750] already existed (Hoffmeister, 2017), but were verified after reactivation of lyophilized cultures by amplification (pMTL82151catP_fwd, pMTL82151catP_rev, pJIR750catP_fwd, pJIR750catP_rev) of catP from pMTL82151 backbone (expected DNA fragment 1,121 bp; Figure 27A) or pJIR750 backbone (expected DNA fragment 1,075 bp; Figure 27A). Furthermore, 16S rDNA was amplified (fD1, rP2; expected DNA fragment 1,500 bp; Figure 27C) with genomic DNA as template followed by sequencing (fD1, rP2) by GATC Biotech AG (Constance, Germany). New recombinant strains were verified either by isolation of genomic DNA and amplification of catP (expected DNA fragment 1,500 bp; Figure 27A) or via retransformation of genomic DNA from recombinant strains into E. coli XL1- Blue MRF', subsequent plasmid isolation, and analytical restriction enzyme digestion (Figure 27B). 80 3. Results

Figure 27 : Analysis of DNA fragments resulting from amplification of catP (A), analytical restriction enzyme digestion (B), and amplification of 16S rDNA gene of genomic DNA from recombinant A. woodii strains (C) using 0.8 % agarose gels. GeneRuler DNA Ladder Mix (M). (A) 1,121-bp (pMTL82151 backbone) and 1,075-bp (pJIR750 backbone) DNA fragments resulting from PCR of catP from A. woodii [pMTL82151] (3), A. woodii [pMTL82151_23BD] (4), A. woodii [pJIR750] (7), A. woodii [pJIR750_23BD] (8), A. woodii [pJIR750_23BD_budAshort] (9), A. woodii [pJIR750_23BD_budAshort_adh] (10), A. woodii [pJIR750_budABCoperon] (11), positive control with empty vector pMTL82151 (2) and pJIR750 (6) as template, and negative control with sterile water as template (1, 5). (B) Analytical restriction enzyme digestion using Eco 32I of retransformed pJIR750_23BD_budAshort with expected DNA fragments 6,233 bp, 3,875 bp and 759 bp (1), using Bam HI and Sac I of retransformed pJIR750_23BD with expected DNA fragments 7,890 bp and 3,070 bp (2), using Bam HI and Bsu 15I of retransformed pMTL821 51_23BD with expected DNA fragments 7,983 bp and 1,663 bp (3), using Bam HI and Sal I of retransformed pJIR750_23BD_budAshort_adh with expected DNA fragments 9,522 bp and 1,218 bp (4), and using Nde I of retransformed pJIR750_budABCoperon with expected DNA fr agment 5,425 bp, 3,092 bp and 1,825 bp (4). (C) 1,500-bp DNA fragment resulting from amplification of 16S rDNA gene of A. woodii [pMTL82151] (2), A. woodii [pJIR750] (3), A. woodii [pMTL82151_23BD] (4), A. woodii [pJIR750_23BD] (5), A. woodii [pJIR750_23BD_budAshort] (6), A. woodii [pJIR750_23BD_budAshort_adh] (7), A. woodii [pJIR750_budABCoperon] (8), and negative control with sterile water as template (1). Figure 27B shows the analytical restriction enzyme digestions using Eco32 I of retransformed plasmid pJIR750_23BD_budAshort (expected DNA fragments 6,233 bp, 3,875 bp, 759 bp), using Bam HI and Sac I of retransformed pJIR750_23BD (expected DNA fragments 7,890 bp, 3,070 bp), using Bam HI and Bsu 15I of retransformed pMTL82151_23BD (expected DNA fragments 7,983 bp, 1,663 bp), using Bam HI and Sal I of retransformed pJIR750_23BD_budAshort_adh (expected DNA fragments 9,522 bp, 1,218 bp), and using Nde I of retransformed pJIR750_budABCoperon (expected DNA fragments 5,425 bp, 3,092 bp, 1,825 bp). Additionally, amplification of 16S rDNA gene (1,500 bp; fD1, rP2) and subsequent sequencing of the resulting DNA fragment supported correct identity of the recombinant A. woodii strains (Figure 27C). Using these methods, recombinant strains A. woodii [pMTL82151_23BD], A. woodii [pJIR750_23BD], A. woodii [pJIR750_23BD_budAshort], A. woodii [pJIR750_23BD_budAshort_adh], and A. woodii [pJIR750_budABCoperon] were verified. 3. Results 81

Figure 28 presents growth, pH, and production pattern of A. woodii WT, A. woodii [pMTL82151], A. woodii [pJIR750], A. woodii [pMTL82151_23BD], A. woodii [pJIR750_23BD], A. woodii [pJIR750_23BD_budAshort], A. woodii [pJIR750_23BD_budAshort_adh], and

A. woodii [pJIR750_budABCoperon] under autotrophic conditions with H2 + CO 2 as energy and carbon source. Growth experiments were performed as biological triplicates in 500-ml glass flasks containing 50 ml modified ATCC 1612 medium and H2 + CO2 as energy and carbon source (1 bar overpressure). In terms of growth behaviour, A. woodii WT and A. woodii [pJIR750] showed higher max. OD 600nm values with 1.0 and 1.1, respectively, after 120 h compared to the other strains (0.3-0.5). After this maximum OD 600nm (120 h), growth of all strains decreased. Furthermore, the pH started at 7.5 and decreased to 5.5 after 984 h in accordance

] 1 600nm

0.1 Growth [OD 0.01 8

7 pH 6

5 180 160 140 120 100 80 60

Acetate [mM] 40 20 0 0 100 200 300 400 500 600 700 800 900 1000 Time [h]

Figure 28 : Growth, pH, and production pattern of A. woodii WT ( ), A. woodii [pMTL82151] ( ), A. woodii [pJIR750] ( ), A. woodii [pMTL82151_23BD] ( ), A. woodii [pJIR750_23BD] ( ), A. woodii [pJIR750_23BD_budAshort] ( ), A. woodii [pJIR750_23BD_budAshort_adh] ( ), and A. woodii [pJ IR750_budABCoperon] ( ) growing in 50 ml modified ATCC 1612 medium with H 2 + CO 2 (1 bar overpressure). 82 3. Results with increasing acetate concentration. Acetate was the only detectable product in all strains with A. woodii [pJIR750_23BD_budAshort] producing most (166.0 mM). Also, A. woodii [pJIR750_23BD] and A. woodii [pJIR750_23BD_budAshort_adh] produced slightly higher amounts of acetate with 163.9 mM and 156.3 mM, respectively, compared to A. woodii WT (148.5 mM). All in all, none of the recombinant A. woodii strains was able to produce 2,3- BD heterologously.

3.1.9 Enhancement of 2,3-BD production by overexpressing nifJ The enzyme pyruvate:ferredoxin oxidoreductase (PFOR) converts acetyl-CoA to pyruvate by oxidizing reduced ferredoxin. Additionally, two mol of pyruvate are needed to build one mol of 2,3-BD (Figure 29). The company LanzaTech Inc. (Skokie, IL, USA) recently published a patent (Köpke and Mueller, 2016) in which they describe C. autoethanogenum strains that overexpress the gene nifJ , which encodes PFOR enzyme, in addition to the genes alsS and budA . The enzymes capable of reducing acetoin to 2,3-BD were not overproduced, since investigation of enzyme activities revealed that 2,3-BD dehydrogenase (bdh ) and primary- secondary alcohol dehydrogenase (adh ) from C. autoethanogenum WT showed a high enzyme activity with 1.2 U/mg (0.8 U/mg with NADH and 0.4 U/mg with NADPH) compared to PFOR with 0.11 U/mg (Köpke and Mueller, 2016). According to the results of LanzaTech Inc. (Skokie, IL, USA), it seems not necessary to overproduce 2,3-BD dehydrogenase or primary-secondary

Figure 29 : 2,3 -BD production in C. ljungdahlii originating from acetyl -CoA via enzymes PFOR (pyruvate:ferredoxin oxi doreductase), AlsS (acetolactate synthase), BudA (acetolactate decarboxylase), BDH (2,3-BD dehydrogenase). 3. Results 83 alcohol dehydrogenase for 2,3-BD overproduction. Therefore, new plasmids were constructed starting from the plasmids pMTL82151_23BD and pJIR750_23BD. In one attempt, the nifJ gene was cloned in addition to alsS , budA , and bdh into pMTL82151_23BD and pJIR750_23BD and in another attempt, the gene bdh was exchanged against nifJ . The sequence of nifJ was amplified (PFORInfusionBamHI_fwd, PFORInfusionBamHI_rev, PFORohneBDHBamHI_fwd, PFORohneBDHBamHI_rev) using genomic DNA from C. ljungdahlii WT as template, and plasmids pMTL82151_23BD as well as pJIR750_23BD were digested either using Bam HI for linearization or Bam HI and Sal I for cutting out bdh (Figure 30 and Figure 31). Subsequently, amplified DNA fragments were cloned into the linearized plasmids by “In-Fusion HD Cloning”. Successful cloning was confirmed by analytical restriction digestion using restriction enzymes Bam HI or Bam HI and Sal I, as well as sequencing (PFOR-Seq1, PFOR-Seq2, PFOR-Seq3, PFOR-

Figure 30 : Cloning strategy for construction of plasmids pMTL82151_23BD_PFOR (A) and pMTL82151_23BD_oBDH_PFOR (B). Plasmid pMTL82151_23BD was digested either with restriction enzymes Bam HI for linearization or Bam HI and Sal I for cutting out bdh . Gene nifJ was coned into linearized pMTL82151_23BD by “In-Fusion HD cloning”, resulting in plasmids pMTL82151_23BD_PFOR and pMTL82151_23BD_oBDH_PFOR. Ppta-ack , promoter region upstream of pta -ack operon from C. ljungdahlii ; alsS , acetolactate synthase from C. ljungdahlii ; budA , acetolactate decarboxylase from C. ljungdahlii ; bdh , 2,3-BD dehydrogenase from C. ljungdahlii ; nifJ , PFOR from C. ljungdahlii ; rep (ColE1) , ColE1-replicon for Gram-negative bacteria; repA (pBP1) /orf2 , pBP1-replicon for Gram-positive bacteria from C. botulinum ; rep (pMB1, replicon for Gram-negative bacteria); rep(pIP404) , pIP404-replicon for Gram-positive bacteria from C. perfringens ; catP , chloramphenicol resistance gene; traJ , gene for conjugal transfer function. 84 3. Results

Figure 31 : Cloning strategy for construction of plasmids pJIR750_23BD_PFOR (A), pJIR750_23BD_oBDH_PFOR (C), pJIR750_23BD_PFOR_ traJ (B), and pJIR750_23BD_oBDH_PFOR_traJ (D). Plasmid pJIR750_23BD was digested either using restriction enzymes Bam HI (A) for linearization or Bam HI and Sal I for cutting out bdh (C). Gene nifJ was cloned into linearized pJIR750_23BD by In- Fusion HD cloni ng, resulting in plasmids pJIR750_23BD_PFOR (A) and pJIR750_23BD_oBDH_PFOR (C). For construction of plasmids pJIR750_23BD_PFOR_traJ (B) and pJIR750_23BD_oBDH_PFOR_traJ (D), digestion of pJIR750_23BD_PFOR and pJIR750_23BD_oBDH_PFOR with Ehe I was performed. DNA fragment oriT /traJ was cloned into linearized pJIR750_23BD_PFOR and pJIR750_23BD_oBDH_PFOR by In-Fusion HD cloning. P pta-ack , promoter region upstream of pta-ack operon from C. ljungdahlii ; alsS , acetolactate synthase from C. ljungdahlii ; budA , acetola ctate decarboxylase from C. ljungdahlii ; bdh , 2,3-BD dehydrogenase from C. ljungdahlii ; nifJ , PFOR from C. ljungdahlii ; rep (ColE1) , replicon for Gram-negative bacteria; repA /orf2 (pBP1) , pBP1-replicon for Gram-positive bacteria from C. botulinum ; rep (pMB1) , pMB1-replicon for Gram-negative bacteria; rep (pIP404) , pIP404-replicon for Gram-positive bacteria from C. perfringens ; catP , chloramphenicol resistance gene; traJ , gene for conjugal transfer function; oriT , origin of conjugal transfer function. Seq4, PFOR-Seq5) by GATC Biotech AG (Constance, Germany). The resulting plasmids were named pMTL82151_23BD_PFOR, pMTL82151_23BD_oBDH_PFOR, pJIR750_23BD_PFOR, and pJIR750_23BD_oBDH_PFOR (Figure 30A and B as well as Figure 31A and C). All constructed plasmids were in vivo methylated using E. coli ER2275 [pACYC184_MCljI] and subsequently isolated using “QIAGEN Plasmid Midi Kit” (QIAGEN-tip 100; Qiagen GmbH, Hilden, Germany). Attempts of transforming the methylated plasmids pMTL82151_23BD_PFOR, pMTL82151_23BD_oBDH_PFOR, pJIR750_23BD_PFOR, and pJIR750_23BD_oBDH_PFOR into C. ljungdahlii by electroporation were not successful and since pMTL82151 vector harbors already genes for conjugal transfer ( oriT and traJ ), conjugation experiments were conducted in order to transform the newly constructed plasmids into C. ljungdahlii . Since vector pJIR750 does not contain any genes for conjugation, the oriT /traJ region of pMTL82151 was amplified 3. Results 85

(InfusionTraJEheI_fwd, InfusionTraJEheI_rev). Afterwards, plasmids pJIR750_23BD_PFOR and pJIR750_23BD_oBDH_PFOR were linearized by restriction enzyme digestion using Ehe I and subsequently oriT /traJ was inserted using “In-Fusion HD cloning” (Figure 31D). Successful cloning was confirmed by analytical restriction digestion using Ehe I, as well as sequencing (traJseq1, traJSeq_rev) by GATC Biotech AG (Constance, Germany). The resulting plasmids were named pJIR750_23BD_PFOR_traJ and pJIR750_23BD_oBDH_PFOR_traJ (Figure 31B and D). Subsequent conjugation experiments were only successful for plasmids pMTL82151_23BD_PFOR and pMTL82151_23BD_oBDH_PFOR. The transformed strains growing in presence of thiamphenicol and colistin were verified by two different methods. On the one hand, isolation of genomic DNA from recombinant C. ljungdahlii strains was performed and catP (expected DNA fragment 1,121 bp) was amplified via PCR (pMTL82151catP_fwd, pMTL82151catP_rev) to verify presence of the plasmid (Figure 32A). On the other hand, isolated genomic DNA was retransformed into chemically competent E. coli XL-1 Blue MRF' cells and single colonies growing on agar plates containing chloramphenicol were picked for plasmid DNA isolation. Figure 32B and C show the analytical restriction enzyme digestions using Nde I of retransformed plasmids pMTL82151_23BD_PFOR (expected DNA fragments 8,719 bp, 4,477 bp) and pMTL82151_23BD_oBDH_PFOR (expected

Figur e 32 : Analysis of DNA fragments resulting from amplification of catP (A), analytical restriction enzyme digestion (B), and amplification of 16S rDNA gene of genomic DNA from recombinant C. ljungdahlii strains (C) using 0.8 % agaro se gels. GeneRuler DNA Ladder Mix (M). (A) 1,121-bp (pMTL82151 backbone) DNA fragment resulting from PCR of catP from C. ljungdahlii [pMTL82151_23BD_PFOR] (3), C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] (4), positive control with empty vector pMTL82151 (2) as template, and negative control with sterile water as template (1). (B) Analytical restriction enzyme digestion using Nde I of retransformed pMTL82151_23BD_oBDH_PFOR with expected DNA fragments 7,421 bp and 4,477 bp (2), and undigested control (1), using Nde I of retransformed pMTL82151_23BD_PFOR with expected DNA fragments 8,719 bp and 4,477 bp (4), and undigested control (3). (C) 1,500-bp DNA fragment resulting from amplification of 16S rDNA gene of C. ljungdahlii [pMTL82151_23BD_PFOR] (3), C. ljungdahli i [pMTL82151_23BD_oBDH_PFOR] (4), positive control C. ljungdahlii WT (2), and negative control sterile water (1). 86 3. Results

DNA fragments 7,421 bp, 4,477 bp). Additionally, amplification of 16S rDNA gene (1,500 bp; fD1, rP2) and concomitant sequencing of the resulting DNA fragment by GATC Biotech AG (Constance, Germany) supported correct identity of the recombinant strains (Figure 32C). Using these methods, recombinant strains C. ljungdahlii [pMTL82151_23BD_PFOR] and C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] were verified. These recombinant C. ljungdahlii strains were analyzed under both heterotrophic and autotrophic growth conditions. Figure 33 presents growth, fructose consumption, and production pattern of C. ljungdahlii WT, mutated C. ljungdahlii [pMTL82151], C. ljungdahlii [pMTL82151_23BD_PFOR], and C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] under heterotrophic conditions. Growth experiments were performed as biological triplicates in 125-ml glass flasks containing 50 ml Tanner medium with 30 mM fructose as energy and carbon source. In terms of growth,

C. ljungdahlii WT reached a higher max. OD 600nm (2.4) after 50 h compared to the other strains with 1.9-1.8 after 50 h. Regarding fructose concentration it was striking that C. ljungdahlii [pMTL82151_23BD_PFOR] was the only strain, which was not completely consuming fructose. The pH decreased to 4.0 after 72 h for all strains and mutated C. ljungdahlii [pMTL82151] showed a decreased acetate production of 46.9 mM after 218 h compared to C. ljungdahlii WT and C. ljungdahlii [pMTL82151_23BD_PFOR] of 62.9 mM and 63.0 mM, respectively. In contrast, C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] produced the highest acetate concentration with 71.4 mM. Comparing ethanol production of the different C. ljungdahlii strains, it was noticeable that in accordance with previous results (Figure 23) mutated C. ljungdahlii [pMTL82151] produced higher ethanol concentrations (23.1 mM) in contrast to the other strains (5.8 mM-7.1 mM). In terms of 2,3-BD production, C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] exposed the highest 2,3-BD concentration after 218 h with 2.2 mM, which is very similar to mutated C. ljungdahlii [pMTL82151] with 1.9 mM. However, C. ljungdahlii [pMTL82151_23BD_PFOR] produced comparable concentrations of 2,3-BD as C. ljungdahlii WT with 0.7 mM after 218 h.

Figure 34 depicts growth, pH, and production pattern of C. ljungdahlii wildtype (WT), mutated C. ljungdahlii [pMTL82151], C. ljungdahlii [pMTL82151_23BD_PFOR], and C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] under autotrophic conditions. Growth experiments were performed as biological triplicates in 500-ml glass flasks containing 50 ml Tanner medium with 3. Results 87

6.0

] 1 5.5

5.0 600nm OD pH [ 4.5 0.1

Growth Growth 4.0

3.5 35 100

30 80 25

20 60

15 40 Acetate [mM] Acetate Fructose [mM] Fructose 10 20 5

0 0 25 3.0

2.5 20

2.0 15 1.5 10 1.0 2,3-BD [mM] 2,3-BD Ethanol [mM] Ethanol 5 0.5

0 0.0 0 50 100 150 200 250 0 50 100 150 200 250 Time [h] Time [h]

Figure 33 : Growth, pH, fructose consumption, and production pattern of C. ljung dahlii WT ( ), mutated C. ljungdahlii [pMTL82151] ( ), C. ljungdahlii [pMTL82151_23BD_PFOR] ( ), and C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] ( ) growing in 50 ml Tanner medium containing 30 mM fructose. syngas (1 bar overpressure) as energy and carbon source. All recombinant C. ljungdahlii strains reached a lower max. OD 600nm (1.2-1.3) compared to C. ljungdahlii WT (1.7). The strain

C. ljungdahlii [PMTL82151_23BD_oBDH_PFOR] exhibited the lowest OD600nm of 0.8. After 2,207 h, pH decreased to 4 for all C. ljungdahlii strains. Regarding acetate production, all strains produced very similar concentrations after 2,207 h from 103.8 mM to 110.2 mM. In contrast, ethanol and 2,3-BD production showed a completely different picture. It was striking 88 3. Results

6.0 ] 1 5.5 600nm 5.0 pH 0.1 4.5 Growth [OD 4.0 0.01 120 60

100 50

80 40

60 30

40 Ethanol [mM] 20 Acetate [mM]

20 10

0 0 6 0 500 1000 1500 2000 Time [h] 5

4

3

2 2,3-BD[mM]

1

0 0 500 1000 1500 2000 Time [h]

Figure 34 : Growth, pH, and produc tion pattern of C. ljung dahlii WT ( ), mutated C. ljungdahlii [pMTL82151] ( ), C. ljungdahlii [pMTL82151_23BD_PFOR] ( ), and C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] ( ) growing in 50 ml Tanner medium with syngas (1 bar overpressure). that C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] produced much less ethanol (14.1 mM) compared to the other strains (43.5 mM-50.9 mM) and most importantly, comparing 2,3-BD production C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] produced the highest 2,3-BD concentration (4.7 mM) besides mutated C. ljungdahlii [pMTL82151] (5.6 mM), which is more than twice as much compared to C. ljungdahlii WT (2.2 mM). However, recombinant strain 3. Results 89

C. ljungdahlii [pMTL82151_23BD_PFOR] produced similar amounts of 2,3-BD (2.5 mM) compared to C. ljungdahlii WT.

In summary, only C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] led to an improvement of 2,3-BD production, but it was still lower compared to mutated C. ljungdahlii [pMTL82151] with syngas as energy and carbon source (Figure 34). Table 9 presents the summarized growth and production pattern of the investigated C. ljungdahlii strains under heterotrophic and autotrophic growth conditions.

Table 9: Summary of max. OD 600nm , max. acetate production, max. ethanol production, and max. 2,3-BD production of recombinant C. ljungdahlii strains in comparison to C. ljungdahlii WT under heterotrophic and autotrophic growth conditions.

Strain Max. Max. acetate Max. ethanol Max. 2,3 -BD

OD 600nm production [mM] production production [mM] [mM]/[mg/l] Heterotrophic growth conditions C. ljungdahlii WT 2.4 77.7 6.6 1.0 /090.1 Mutated C. ljungdahlii 1.8 55.7 23.1 2.4 /216.3 [pMTL82151] C. ljungdahlii 1.9 67.0 5.8 0.8 /072.1 [pMTL82151_23BD_PFOR] C. ljungdahlii 1.9 71.4 7.1 2.2 /198.3 [pMTL82151_23BD_oBDH_PFOR] Autotrophic growth conditions C. ljungdahlii WT 2.0 108.5 50.9 2.2 /198.3 Mutated C. ljungdahlii 1.3 110.2 44.2 5.6 /504.7 [pMTL82151] C. ljungdahlii 1.5 103.8 43.5 2.5 /225.3 [pMTL82151_23BD_PFOR] C. ljungdahlii 0.9 109.0 14.1 4.7 /423.6 [pMTL82151_23BD_oBDH_PFOR]

90 3. Results

3.2 Heterologous butanol production Another aim of this study was the heterologous production of butanol in C. ljungdahlii . In a previous study, the heterologous production of 2 mM butanol in C. ljungdahlii with the genes originating from C. acetobutylicum was reported (Köpke et al ., 2010). The genes for butanol production from C. acetobutylicum , which are thlA (thiolase), hbd (3-hydroxybutyryl-CoA dehydrogenase), crt (crotonase), bcd (butyryl-CoA dehydrogenase), bdhA (butanol dehydrogenase), and adhE (bifunctional aldehyde/alcohol dehydrogenase) were located on the plasmid pSOB ptb under control of the ptb promoter. The plasmid pSOB ptb does not contain the genes etfA and etfB , which were shown to be essential for reduction of crotonyl-CoA to butyryl-CoA in C. kluyveri (Li et al ., 2008) as well as C. acetobutylicum (Lütke-Eversloh and Bahl, 2011). Therefore, plasmid pSB3C5-UUMKS 3 was synthesized by the company ATG:biosynthetics GmbH (Freiburg, Germany) in a further study, containing the genes for butanol production ( thlA , hbd , crt , bcd , and adhE2 ) as well as etfA and etfB from C. acetobutylicum (Schuster, 2011). Thus, pSB3C5-UUMKS 3 harbors all genes encoding enzymes necessary for heterologous butanol production in C. ljungdahlii (Figure 35). However, it only contains an origin of replication for Gram-negative bacteria ( ori p15A) and butanol production was only shown in E. coli (Schuster, 2011). For enabling the butanol plasmid to

Figure 35: Butanol production in C. ljungdahlii with enzymes originating from C. acetobutylicum starting from acetyl-CoA. Thl (thiolase), Hbd (3-hydroxybutyryl-CoA dehydrogenase), Crt (crotonase), Bcd (butyryl-CoA dehydro genase), EtfA/B (electron transfer flavoproteins), AdhE2 (bifunctional aldehyde/alcohol dehydrogenase). 3. Results 91 replicate in C. ljungdahlii , the origins of replication from C. perfringen s (ori pIP404) and C. butyricum (ori pCB102) were cloned into plasmid pSB3C5-UUMKS 3. For this purpose, ori pIP404 and ori pCB102 were amplified (pIP404PstI_fwd, pIP404SalI_rev, pCB102SalI_fwd pCB102PstI_rev) using vector pJIR750 and pMTL83151 as template, respectively. Afterwards, plasmid pSB3C5-UUMKS 3 and amplified DNA fragments were digested using restriction enzymes Pst I and Sal I. Subsequently, ligation of the digested and purified ori pIP404 and ori pCB102 into the linearized plasmid pSB3C5-UUMKS 3 was performed. The resulting plasmids were named pCE3 and pJR3 (Figure 36). Restriction enzyme digestion using Pst I and Sal I of pCE3 (Figure 37A; expected DNA fragments 12,495 bp, 2,191 bp) and pJR3 (Figure 37A; expected DNA fragments 12,495 bp, 1,704 bp) as well as sequencing (pIP404Seq1, pIP404Seq2, pIP404Seq3, pJR3Seq1, pJR3Seq2, pJR3Seq3, pJR3Seq4) by GATC Biotech AG (Constance, Germany) confirmed successful cloning. Prior to transformation of the two butanol plasmids pCE3 and pJR3 into C. ljungdahlii , in vivo methylation with E. coli ER2275 [pACYC184_MCljI] was performed. However, for this purpose, the origin of replication for Gram-negative bacteria p15A of the respective plasmid had to be exchanged, since pCE3 as well as pJR3 harbor the identical ori. The pSC101-replicon also shows a very low copy number just like the p15A-replicon. Therefore, p15A-replicon of in vivo methylation plasmid

Figure 36 : Cloning strategy for construction of plasmids pCE3 and pJR3. Plasmid pSB3C5 -UUMKS 3 as well as repH (pCB102) and rep (pIP404) were digested using restriction enzymes Pst I and Sal I. Rep (pIP404) and repH (pCB102) were ligated into linearized pSB3C5-UUMKS 3, resulting in plasmids pCE3 and pJR3, respectively. P ptb , promoter region upstream of ptb-buk operon from C. acetobutylicum ; thlA , thiolase from C. acetobutylicum ; hbd , 3-hydroxybutyryl-CoA dehydrogenase from C. acetobutylicum ; crt , crotonase from C. acetobutylicum ; bcd , butyryl-CoA dehydrogenase from C. acetobutylicum ; etfA , electron transfer flavoprotein from C. acetobutylicum ; etfB , electron transfer flavoprotein from C. acetobutylicum ; adhE2 , bifunctional aldehyde/alcohol dehydrogenase from C. acetobutylicum ; rep (p15A) , p15A-replicon for Gram-negative bacteria; rep (pIP404) , pIP404- replicon for Gram-positive bacteria from C. perfringens ; repH (pCB102) , pCB102-replicon for Gram- positive bacteria from C. butyricum ; catP , chloramphenicol resistance gene. 92 3. Results

Figure 37: Analysis of DNA fragments resulting from analytical restriction enzyme diges tion of plasmids pCE3 and pJR3 (A), analytical restriction enzyme digestion of plasmid pACYC184_MCljI_pSC101 (B), and analytical restriction enzyme digestion of plasmids pCE3_traJ and pJR3_traJ (C) using 0.8 % agarose gels. GeneRuler DNA Ladder Mix (M). (A ) Analytical restriction enzyme digestion using Pst I and Sal I of pCE3 with expected DNA fragments 12,495 bp and 2,191 bp (1) as well as pJR3 with expected DNA fragments 12,495 bp and 1,704 bp (2). (B) Analytical restriction enzyme digestion using Bsu 15I and Sac II of pACYC184_MCljI_pSC101 with expected DNA fragments 6,730 bp and 1,597 bp (1). (C) Analytical restriction enzyme digestion using Pst I of pCE3_traJ with expected DNA fragments 14,686 bp and 766 bp (2) as well as pJR3_traJ with expected DNA fragmen ts 14,199 bp and 7,66 bp (1). pACYC184_MCljI was exchanged with pSC101-replicon. For this purpose, rep101/ori (pSC101) was amplified (pSC101SacII_fwd, pSC101ClaI_rev) using vector pSC101 as template. Subsequently, plasmid pACYC184_MCljI and amplified DNA fragment were digested using restriction enzymes Bsu 15I and Sac II. Afterwards, ligation of the digested and purified ori pSC101 into the linearized plasmid pACYC184_MCljI was performed and the resulting plasmid was named pACYC184_MCljI_pSC101 (Figure 38). Restriction enzyme digestion using Bsu 15I and Sac II of pACYC184_MCljI_pSC101 (Figure 37B; expected DNA fragments 6,730 bp, 1,597 bp) as well as sequencing (pSC101SacII_fwd) by GATC Biotech AG (Constance, Germany)

Figure 38 : Cloning strategy for construction of plasmid pACYC184_MCljI_pSC101. Plasmid pACYC184_MCljI and rep101 /ori (pSC101) were digested using restriction enzymes Bsu 15I and Sac II. Rep101 /ori (pSC101) was ligated into linearized pACYC184_MCljI, resulting in plasmid pACYC184_MCljI_pSC101. P ptb , promoter region upstream of ptb-buk operon from C. acetobutylicum ; CLJU_c03310, locus tag encoding methylase subunit; CLJU_c033 20, locus tag encoding specifity subunit; rep (p15A) , p15A-replicon for Gram-negative bacteria; tetR , tetracycline resistance gene; rep101 /ori (pSC101) , pSC101-replicon for Gram-negative bacteria. 3. Results 93 confirmed successful cloning. Thus, in vivo methylation of pCE3 and pJR3 was achieved using E. coli ER2275 [pACYC184_MCljI_pSC101] and for plasmid isolation “QIAGEN Plasmid Midi Kit” (QIAGEN-tip 100; Qiagen GmbH, Hilden, Germany) was applied in order to get high yields of plasmid concentration. Nevertheless, no successful transformation was achieved by using electroporation method with competent C. ljungdahlii cells. As a further attempt, conjugation experiments were performed to transform the butanol plasmids into C. ljungdahlii . Since pCE3 and pJR3 do not contain any genes for conjugal transfer, the oriT /traJ region of vector pMTL82151 was amplified (traJInfusionPstI_fwd, traJInfusionPstI_rev) and cloned into the respective plasmids. Plasmids pCE3 and pJR3 were linearized by restriction enzyme digestion using Pst I and oriT /traJ was inserted by “In-Fusion HD cloning”. On the one hand, successful cloning of pCE3_traJ (Figure 37C, expected DNA fragments 14,199 bp, 766 bp) and pJR3_traJ (Figure 37C, expected DNA fragments 14,686 bp, 766 bp) was confirmed by analytical restriction digestion with Pst I, and on the other hand by sequencing (pJR3-traJ Seq, traJseq1, traJSeq_rev) of plasmids by GATC Biotech AG (Constance, Germany). The resulting plasmids

Figure 39 : Cloning strategy for construction of plasmids pCE3_t raJ and pJR3_traJ. Plasmids pCE3 and pJR3 and DNA fragments were digested with restriction enzyme Pst I. OriT/traJ was cloned into linearized pCE3 and pJR3 by “In-Fusion HD Cloning”, resulting in plasmids pCE3_traJ and pJR3_traJ, respectively. P ptb , promoter region upstream of ptb-buk operon from C. acetobutylicum ; thlA , thiolase; hbd , 3-hydroxybutyryl-CoA dehydrogenase; crt , crotonase; bcd , butyryl-CoA dehydrogenase; etfA , electron transfer flavoprotein; etfB , (lectron transfer flavoprotein; adhE2 , bifunctional aldehyde/alcohol dehydrogenase; rep (p15A) , p15A-replicon for Gram-negative bacteria; rep (pIP404) , pIP404-replicon for Gram-positive bacteria from C. perfringens ; rep H (pCB101) , pCB102-replicon for Gram-positive bacteria from C. butyricum ; catP , chl oramphenicol resistance gene; traJ , gene for conjugal transfer function; oriT , origin of conjugal transfer function. 94 3. Results were named pCE3_traJ and pJR3_traJ (Figure 39). However, conjugation experiments using pCE3_traJ and pJR3_traJ for transformation of C. ljungdahlii failed to produce recombinant strains.

4. Discussion 95

4. Discussion

4.1 Natural microbial 2,3-BD production Microbial 2,3-BD production is often accompanied by mixed-acid fermentation, a metabolism found in most members of the family Enterobacteriaceae and some other microorganisms including both aerobic and anaerobic types. Acidic products of mixed acid fermentation are acetate, lactate, formate, and succinate, which lower the intracellular pH. Thus, production of 2,3-BD is assumed to prevent excessive acidification by switching the metabolism towards this neutral compound (Vivijs et al ., 2014a; b). Moreover, external supplementation of acids induces 2,3-BD biosynthesis (Nakashimada et al ., 2000; Lee et al ., 2016). The reversible reaction in-between acetoin and 2,3-BD with concomitant NAD(P)H/ NAD(P) + formation might also play a role in regulation of the NAD(P)H/NAD(P) + ratio (Celińska and Grajek, 2009). Bacteria can reuse 2,3-BD very often in case other carbon and energy sources are scarce, so 2,3-BD production is also considered to be a carbon and energy-storing strategy (Xiao and Xu, 2007). In the past years, several microorganisms including yeasts and bacteria from different genera were reported to produce 2,3-BD (Ji et al ., 2011a). Table 10 shows an overview of microorganisms capable of 2,3-BD production, utilized substrate, culture method, and titer since 2009. Since the best natural 2,3-BD producers belong to the risk group 2, research is also focussing in metabolic engineering of risk group 1 organisms to avoid high safety requirements, which are unfavorable for industrial-scale processes (Celińska and Grajek, 2009; Ji et al ., 2011a), but in most cases 2,3-BD concentrations do not reach yields of risk group 2 organisms. So far, Serratia marcescens and Klebsiella pneumoniae are the most efficient 2,3- BD production strains with 152 g/l and 150 g/l, respectively (Table 10). However recently, some bacteria generally regarded as safe (GRAS) such as Paenibacillus polymyxa , Bacillus amyloliquefaciens , and Bacillus licheniformis were shown to produce comparable amounts of 2,3-BD as risk group 2 organisms (Table 10). Many 2,3-BD production strains use sugars like hexoses and pentoses as substrate, which is the reason why more sustainable substrates such as glycerol, hydrolysates of cellulose, corn stover hydrolysates, and starch have been recently used for 2,3-BD production (Table 10). In 2011, the publication reporting that acetogens are capable of 2,3-BD production using steel mill waste gas attracted great attention, although only producing low amounts of 1.4 mM-2 mM (Köpke et al ., 2011b). Furthermore, it was

96 4. Discussion

Table 10: Microbial 2,3-BD production (titer) of different bacterial species and applied substrate and culture condition (since 2009)

Strain Substrate Culture method Titer Reference (g/l) Engineered natural 2,3 -BD production strains Bacillus amyloliquefaciens B10 - Glucose Fed -batch 092.3 Yang et al ., 2011 127 B. amyloliquefac iens B10 -127 Glucose Fed -batch 061.4 Yang et al ., 2012 B. amyloliquefaciens pBG Glucose Fed -batch 132.9 Yang et al ., 2013 B. amyloliquefaciens GAR Crude glycerol Fed -batch 102.3 Yang et al ., 2015b B. amyloliquefaciens TUL -308 Sugarcane Fed -batch 060.0 Sikora et al ., molasses 2016 B. atrophaeus NRS -213 Glucose Fed -batch 029.9 Kallbach et al ., 2017 B. licheni formis BL8 Xylose Batch 013.8 Wang et al ., 2012b B. licheniformis 10 -1-A Glucose Fed -batch 115.7 Li et al ., 2 013 B. licheniformis DSM8785 Glucose Fed -batch 144.7 Jurchescu et al ., 2013 B. licheniformis X-10 Corn stover Fed -batch 074.0 Li et al ., 2014 b hydrolysate B. licheniformis AT CC Inulin SSF* batch 103.0 Li et al ., 2014 c

14580 hydrolysate B. licheniformis WX -02 ΔbudC Glucose Shake flask 030.8 Vivijs et al ., 2014b B. licheniformis NCIMB* * Apple pomace Fed -batch 113.0 Biał kowska et al ., 8059 hydrolysates 2015 B. mojavensis B-14698 Glucose Batch 037.8 Kallbach et al ., 2017 B. subtilis ATCC ** * 23857 Glucose Batch 00 6.1 Biswas et al ., 2012 *simoultaneous saccharification and fermentation **National Collections of Industrial, Marine and Food Bacteria ***American Type Culture Collection 4. Discussion 97

Table 10: Microbial 2,3-BD production (titer) of different bacterial species and applied substrate and culture condition (continued)

Strain Substrate Culture method Titer Reference (g/l) Engineered natural 2,3 -BD production strains B. subtilis BSF20 Glucose Shake flask 049.3 Fu et al ., 2014 B. subti lis AFYL Glucose Shake flask 044.2 Yang et al ., 2015 c B. subtilis WN1394 Glucose Batch, IPTG 00 2.4 De Oliveira and Nicholson, 2016 B. subtilis TUL 322 Sugarcane Fed -batch 075.0 Bia łkowska et al ., molasses 2016 B. vallismortis B-14891 Glucose Batch 060.4 Kallbach et al ., 2017 Clostridium autoethanogenum Steel mill gas Batch 00 0.4 Kö pke and Chen, DSM23693 2013 C. autoethanogenum LZ1561 Synthetic steel Batch 00 1.1 Kö pke and mill gas Mueller, 2016 C. autoethanogenum LZ1561 Synthetic steel Continous 00 9.0 Kö pke and mill gas fermentation Mueller, 2016 Enterobacter cloacea subsp. Cassava SSF * batch 093.9 Wang et al ., dissolvens SDM powder 2012a E. cloacea SDM 09 Glucose Fed -batch 119.4 Li et al ., 2015 a E. cloacae CGMCC* * 605 Sugarcane Fed -batch 099.5 Dai et al ., 2015 molasses E. aerogenes EMY 01 Glucose Fed -batch 118.1 Jung et al ., 2012 E. aerogenes EMY -68 Sugarcane Fed -batch 098.7 Jung et al ., 2015 molasses E. aerogenes EMY -70S Sugarcane Fed -batch 140.0 Jung et al ., 2015 molasses Klebsiella pneumoniae SDM Glucose Fed -batch 150.0 Ma et al ., 2009 *simoultaneous saccharification and fermentation **China General Microbiological Culture Collection Center

98 4. Discussion

Table 10: Microbial 2,3-BD production (titer) of different bacterial species and applied substrate and culture condition (since 2009) (continued)

Strain Substrate Culture Tit er Reference method (g/l)

Engineered natural 2,3 -BD production strains K. pneumoniae G31 Glycerol Fed -batch 049.2 Petrov and Petrova, 2009 K. pneumoniae CICC* 10011 Jerusalem ** SSF batch 084.0 Sun et al ., 2009 a artichoke tuber K. pneumoniae CICC * 10 011 Jerusalem ** SSF batch 067.4 Li et al ., 2010 a artichoke stalk and tuber K. pneumoniae G31 Glycerol Fed -batch 070.0 Petro v and Petrova, 2010 K. pneumoniae SDM Corncob Fed -batch 078.9 Wang et al ., 2010 molasses K. pneumoniae SGJSB04 Glucose Fed -batch 101.5 Kim et al ., 2012 K. pneumoniae KG -rs Glucose Shake flask 024.5 Guo et al ., 2014 b K. pneumoniae ΔadhE ΔldhA Glucose Fed -batch 115.0 Guo et al ., 2014 a K. pneumo niae KMK -05 Glucose Shake flask 031.1 Jung et al ., 2014 K. pneumoniae SGSB105 Glucose Fed -batch 090.0 Kim et al ., 2014a K. pneumoniae G31 -A Starch ** SSF batch 053.8 Tsvetanova et al ., 2014 K. pneumoniae M3 Crude glycerol Fed -batch 131.5 Cho et al ., 2015a K. pneumoniae SGSB112 Glucose Fed -batch 061.0 Lee et al ., 2015 K. oxytoca ME -UD -3 Glucose Batch 095.5 Ji et al ., 2009 K. oxytoca ACCC 10370 Cornc ob acid Fed -batch 035.7 Cheng et al ., 2010 hydrolysate K. oxytoca ME -XJ -8 Glucose Fed -batch 130.0 Ji et al ., 2010 K. oxytoca ME -CRPin Glucose, Xylose Shake flask 023.9 Ji et al ., 2011b K. oxytoca ME -UD -3 Glucose Fed -batch 127.9 Nie et al ., 2011 K. oxytoca GSC12206 ( ΔldhA) Glucose Fed -batch 115.0 Kim et al ., 2013 a *China Center of Industrial Culture Collection **simoultaneous saccharification and fermentation 4. Discussion 99

Table 10: Microbial 2,3-BD production (titer) of different bacterial species and applied substrate and culture condition (since 2009) (continued)

Strain Substrate Culture method Titer Reference (g/l) Engineered natural 2,3 -BD production strains K. oxytoca ΔldhAΔpflB Glucose Fed -batch 113.0 Park et al ., 2013 K. oxytoca M1 Glucose Fed -batch 142.5 Cho et al ., 2015b K. oxytoca KMS005 -73T Glucose Fed -batch 117.4 Jantama et al ., 2015 K. oxytoca Glucose Fed -batch 106.7 Park et al ., 2015 ΔldhAΔpflBΔbudC::PBDH Paenibacillus polymyxa ZJ -9 Jerusalem Batch 036.9 Gao et al ., 2010 artichoke tuber P. polymyxa DSM365 Sucrose Fed -batch 111.0 Häßler et al ., 2012 Serra tia marcescens H30 Sucrose Fed -batch 139.9 Zhang et al ., 2010a S. marcescens H30 Sucrose Fed -batch 152.0 Zhang et al ., 2010b S. marcescens ΔslaC -bdhA Sucrose Fed -batch 089.8 Bai et al ., 2015 Na tural 2,3 -BD production strains Raoultella planticola CECT * 843 Pure glycerol Batch 030.5 Ripoll et al ., 2016

R. terrigena CECT * 4519 Raw glycerol Batch 033.7 Ripoll et al ., 2016 Construction of pathway for 2,3 -BD production Clostridium acetobutylicum Glucose Batch 00 2.0 Siemerink et al ., pWUR460 2011 Clostridium sp. MBD136 Syngas Continous 00 9.2 Tyurin and fermentation Kiriukhin, 2013 Corynebact erium glutamicum Glucose Two -stage 00 6.3 Radoš et al ., 2015 ATCC**13032 fermentation *The Spanish type culture collection of microorganisms ** American Type Culture Collection

100 4. Discussion

Table 10: Microbial 2,3-BD production (titer) of different bacterial species and applied substrate and culture condition (since 2009) (continued)

Strain Substrate Culture Titer Reference method (g/l)

Construction of pathway for 2,3 -BD production C. glutamicum SGSC102 Glucose/ Fed -batch 018.9/ Yang et al ., 2015 a cassava powder 012.0 Escherichia coli JCL260 Glucose Shake flask 00 6.1 Yan et al ., 2009 E. coli JM109LPAP Glucose Batch 014.5 Li et al ., 2010 b E. coli YYC202 ΔldhAΔilvC Glucose Shake flask 00 1.1 Nielsen et al ., 2010 E. coli SGSB03 Gluc ose Batch 015.7 Lee et al ., 2012 E. coli SGSB03 Crude glycerol Batch 00 6.9 Lee et al ., 2012 E. coli BW25113/ ΔackA Glycerol Shake flask 00 9.6 Shen et al ., 2012 E. coli Cellodextrin SSF * batch 00 5.5 Shin et al ., 2012 E. coli DSM02 Algal Fed -batch 014.1 Mazumdar et al ., hydrolysate 2013 E. coli Diacetyl Fed -batch 026.8 Wang et al ., 2013b E. coli (pETDuet -bdhfdh) Diacetyl Fed -batch 031.7 Wang et al ., 2013b E. coli mlc -XT7 -LAFC -YSD XL Glucose + xylose Shake flask 054.0 Nakashima et al ., 2014 E. coli BL21/pET -RABC Glucose Fed -batch 073.8 Xu et al ., 2014 E. coli budBbudC Glucose Shake flask 00 2.2 Chu et al ., 2015 E. coli MQ1 Glucose Fed -batch 115.0 Ji et al ., 2015 E. coli S30 Glucose Lysate with 082.0 Kay and Jewett, NAD +, ATP 2015 E. coli YJ2 Glucose Shake flask 030.5 Tong et al ., 2016 Sacharomyces cerevisiae B2C - Glucose Shake flask 00 2.3 Ng et al ., 2012 a1a3a5 *simoultaneous saccharification and fermentation

4. Discussion 101

Table 10: Microbial 2,3-BD production (titer) of different bacterial species and applied substrate and culture condition (since 2009) (continued)

Strain Substrate Culture method Titer Reference (g/l)

Construction of pathway for 2,3 -BD production S. cerevisiae BD4 Glucose Fed -batch 096.2 Kim et al ., 2013 b S. cerevisiae BD4X Xylose Fed -batch 043.6 Kim et al ., 2014 b S. cerevisiae JL0432 Glucose + Fed -batch 100.0 Lian et al ., 2014 galactose S. cerevisiae BD5_t2nox Glucose Shake flask ~31.0 Kim et al ., 2015

Synechococcus elongatus CO 2 shake flask 002.4 Oliver et al ., 2013

Synechocystis sp. PCC6830 CO 2 shake flask 00 0.4 Savakis et al ., 2013 reported that LanzaTech uses already 2,3-BD produced by C. autoethanogenum from steel mill waste gas as a precursor for the production of 1,3-butadiene in pilot/demo scale (Bengelsdorf et al ., 2013; Köpke and Havill, 2014). 2,3-BD formation from CO and water is coupled to CO 2 formation (1).

11 CO + 5 H 2O → C4O2H10 + 7 CO 2 (1)

Acetogens are unique organisms since they can fix CO 2 and/or CO via the non-circular Wood- Ljungdahl pathway (Figure 2), which is considered to be one of the oldest metabolic pathways in life (Russell and Martin, 2004). Moreover, it is assumed to be the most efficient carbon fixation pathway since it acts in a linear manner (Fast and Papoutsakis, 2012). There are over 100 different acetogens, including at least 25 different genera (Daniell et al ., 2016) while the best characterised species either belong to the group of Gram-positive bacteria with a low GC content (e.g. C. ljungdahlii , C. autoethanogenum , C. ragsdalei , C. coskatii , A. woodii , Clostridium aceticum , and M. thermoacetica ) (Schiel-Bengelsdorf and Dürre, 2012) or to anaerobic spirochaetes, which live as symbionts in the gut of termites (Fuchs, 2007).

Acetogens can use a variety of different substrates including C 1-compounds such as formate,

CO, CO 2, and methanol (Küsel and Drake, 2005), sugars (hexoses and pentoses), glycerol and cellulose (Olson et al., 2012) , carboxylic acids (pyruvate, lactate), aldehydes (benzaldehyde, glyoxylate), and many more (Drake et al ., 2008). Moreover, acetogenic clostridia were shown 102 4. Discussion

to utilize electricity as energy source in order to fix CO 2 (Nevin et al ., 2010; Lovley, 2011; Nevin et al ., 2011; Lovley and Nevin, 2013). Besides the main product acetate of the Wood-Ljungdahl pathway, acetogens can also produce ethanol, lactate, butyrate, butanol, hexanol, hexanoate and 2,3-BD (Liou et al., 2005; Drake et al ., 2008; Köpke et al ., 2011b; Phillips et al ., 2015; Li et al ., 2015b). The acetogens A. woodii , M. thermoacetica , and C. aceticum mainly are considered to produce acetate as sole end product, while B. methylotrophicum and C. carboxidivorans are capable of producing butanol as by-product, and C. ljungdahlii , C. autoethanogenum , C. ragsdalei , and C. coskatii are mainly considered for ethanol and 2,3-BD production besides acetate (Daniell et al ., 2016). In this study, C. ljungdahlii , C. ragsdalei , and C. autoethanogenum were first investigated for natural 2,3-BD production with synthetic syngas as energy and carbon source and furthermore C. aceticum and C. carboxidivorans were also tested for natural 2,3-BD production (Figure 5). Since these organisms produce 94 % of 2R ,3R -BD, while only 6 % of meso -2,3-BD is formed (Köpke et al ., 2011b), only the D(-)-form was investigated in analytics.

The growth experiment was conducted to find the best natural 2,3-BD production strain in order to work on genetic engineering leading to enhanced 2,3-BD production. However, 2,3- BD production was only detected in the organsims C. ljungdahlii , C. autoethanogenum , and C. ragsdalei , while C. ljungdahlii produced the highest 2,3-BD concentration of 4.8 mM (Figure 5B). Furthermore, Köpke et al. (2011) reported that the genes, which are involved in 2,3-BD formation in C. autoethanogenum are upregulated massively during stationary growth phase (15-fold increased to normalized mRNA level). This is also the growth phase, when the majority of 2,3-BD is produced. Furthermore, only gene bdh was shown to be expressed constitutively at high normalized mRNA levels over the whole period of growth (Köpke et al ., 2011b). In this study, formation of 2,3-BD in the stationary growth phase was also shown for the organisms C. ljungdahlii , C. autoethanogenum , and C. ragsdalei . Thus, it might be concluded that formation of the neutral compound 2,3-BD in acetogens also represents a protection mechanism of both excessive acidification and excessive reducing equivalents, like speculated for other 2,3-BD producing bacteria (Ji et al ., 2011a). Therefore, 2,3-BD production starts at the transition to the stationary growth phase, when pH decreases due to acetate formation and a potential excess of reducing equivalents is present. 4. Discussion 103

4.2 Enhancement of 2,3-BD production using acetogens

4.2.1 Enhancement of 2,3-BD production via overexpression of alsS , budA , and bdh Since testing of natural 2,3-BD production in some acetogens resulted in C. ljungdahlii showing the highest 2,3-BD concentration (4.8 mM), this organism was chosen for further enhancement of 2,3-BD production. C. ljungdahlii (Figure 3A) was isolated from chicken yard waste in 1988 (Barik et al., 1988). It is a Gram-positive, rod-shaped, spore-forming, motile, mesophilic, and obligate anaerobic bacterium with a GC content of 31 mol % (Tanner et al .,

1993). Moreover, it can grow autotrophically using H 2 + CO 2 and CO, but it also uses substrates such as formate, ethanol, pyruvate, fumarate as well as sugars such as fructose and xylose for heterotrophic growth (Tanner et al ., 1993). A few years later, the description of a very close relative (genome similarity of 99.3 %; Bengelsdorf et al ., 2016) named C. autoethanogenum was published, which was isolated from rabbit feces (Abrini et al ., 1994). Furthermore, the organism C. coskatii , which was was isolated from estuary sediment collected from Coskata- Coatue Wildlife Refuge (Zahn and Saxena, 2011), was shown to be a very close relative of C. ljungdahlii , C. autoethanogenum , and C. ragsdalei (Bengelsdorf et al ., 2016). Therefore, C. coskatii was considered to be a natural 2,3-BD producer. In 2011, C. ljungdahlii as well as C. autoethanogenum were reported to produce 2,3-BD from steel mill waste gas (Köpke et al ., 2011b). Furthermore, toxicity tests using external addition of 2,3-BD with C. autoethanogenum showed that growth and acetate production ceased with addition of 40 to 50 g/l 2,3-BD (Köpke et al ., 2011b). This shows that C. ljungdahlii might also be considered as a potential commercial 2,3-BD production strain displaying a high 2,3-BD tolerance. Another advantage of choosing this organism for enhancement of 2,3-BD production is the efficient and reproducible transformation protocol, enabling genetic engineering of C. ljungdahlii (Leang et al ., 2013). Furthermore, the acetogen C. coskatii , which was not tested for natural 2,3-BD production by this time, harbors the same genes alsS (CCOS_39110), budA (CCOS_38400), and bdh (CCOS_06940) for 2,3-BD formation as C. ljungdahlii (CLJU_c38920, CLJU_c08380, CLJU_c23220) and C. autoethanogenum (CAETHG_1740, CAETHG_2932, CAETHG_0385). This organism produces mainly acetate but also ethanol is formed as side product from syngas (Zahn and Saxena, 2011). Overexpression of the genes alsS , budA , and bdh encoding the acetolactate synthase (AlsS), acetolactate decarboxylase (BudA), and 2,3-BD dehydrogenase (Bdh) might enhance natural 2,3-BD 104 4. Discussion production of C. ljungdahlii and C. coskatii using glycolysis and/or Wood-Ljungdahl pathway. The enzyme AlsS converts pyruvate to S-acetolactate, which is further decarboxylated to R- acetoin by BudA (Figure 6), while acetoin is a molecule, which is extremely prone to spontaneous racemization via an enolate intermediate (Köpke et al ., 2014). Thus, it is possible that small amounts of R-acetoin are in vivo spontaneously racemized by enzyme BudA. Finally, the enzyme Bdh reduces R-acetoin to 2R,3R -BD and S-actoin to meso -2,3-BD, respectively (Figure 6). The three genes alsS , budA , and bdh were synthesized in form of an operon under control of the constitutive promoter P pta-ack and cloned into two E. coli /Clostridium shuttle- vectors namely pMTL82151 and pJIR750. These vectors harbor different replicons for Gram- positive bacteria (pMTL82151, pBP1 from C. botulinum ; pJIR750, pIP404 from C. perfringens ), which were shown to yield good transformation efficiencies for C. ljungdahlii (Leang et al ., 2013). Since C. coskatii is a very close relative of C. ljungdahlii (Bengelsdorf et al ., 2016), the same transformation protocol was successfully applied for transformation of this organism. Furthermore, the pBP1-replicon and pIP404-replicon might influence the plasmid copy number differently in C. ljungdahlii as well as C. coskatii and thus, lead to changes in 2,3-BD production. After transformation of the vectors pMTL82151 and pJIR750 as well as the 2,3-BD production plasmids pMTL82151_23BD and pJIR750_23BD into C. ljungdahlii and C. coskatii , growth experiments with fructose as well as syngas as energy and carbon source were conducted to provide insights into growth and production pattern in comparison to the respective wildtype strains.

Growth experiments using fructose as carbon source showed only slight differences in terms of growth and production pattern of the recombinant C. ljungdahlii strains compared to C. ljungdahlii WT (Figure 9). The strain C. ljungdahlii [pJIR750_23BD] produced the highest 2,3-BD concentration of 1.2 mM, which is 1.5-fold higher compared to C. ljungdahlii WT (0.8 mM). Moreover, C. ljungdahlii [pMTL82151_23BD] showed a slight increase in 2,3-BD production (1.1 mM) compared to C. ljungdahlii WT. The growth experiment using syngas as energy and carbon source (Figure 10) confirmed the results of the heterotrophic growth experiment (Figure 9) in C. ljungdahlii [pJIR750_23BD] being the best 2,3-BD production strain. This strain produced the highest concentration of 2,3-BD (6.4 mM), which is a 1.5-fold increase compared to C. ljungdahlii WT (4.3 mM). In contrast, C. ljungdahlii [pMTL82151_23BD] did not produce higher amounts of 2,3-BD (3.8 mM) compared to C. ljungdahlii WT growing with 4. Discussion 105 syngas. This is contrary to the results of another autotrophic growth experiment with C. ljungdahlii [pMTL82151_23BD] (Figure 24, Table 7), showing a 1.8-fold increase in 2,3-BD production (3.1 mM) compared to C. ljungdahlii WT (1.7 mM). Furthermore, it was striking that all C. ljungdahlii strains except C. ljungdahlii [pMTL82151_23BD] showed higher product formation and thus higher 2,3-BD concentrations in the autotrophic growth experiment depicted in Figure 10, compared to the other autotrophic growth experiment (Figure 24). This might be explained due to higher overpressure of syngas (1.8 bar) conducted in this experiment (Figure 10) compared to 1.0 bar (Figure 24). It was shown in various fermentation experiments that increasing pressure enhances gas solubility and improves mass transfer and therefore production (Ungerman and Heindel, 2007). The gas-to-liquid mass transfer of substrate into the culture medium represents the rate-limiting step in the process of gas fermentation due to the low solubility of CO and H 2 in water (Daniell et al ., 2016). This is in accordance to one of the gas laws formulated by William Henry in 1803, which states that the solubility of a gas in a liquid is directly proportional to the pressure of the gas above the liquid. Furthermore, LanzaTech considered that providing sufficient or elevated levels of CO during fermentation process, leads to production of higher energy products, such as 2,3-BD (Simpson et al ., 2011). Moreover, Table 10 shows that the same C. autoethanogenum LZ1561 strain produced 1.1 g/l 2,3-BD during batch fermentation, while 9.0 g/l 2,3-BD (eight-fold increase) were formed during continuous fermentation. This shows that the 2,3-BD production of C. ljungdahlii [pJIR750_23BD] might be further enhanced via continuous fermentation experiments. Another important observation in the growth experiments using C. ljungdahlii is the fact that although overexpression of the genes for 2,3-BD formation was realized using a constitutive promoter from C. ljungdahlii , 2,3-BD production did not start earlier during growth compared to C. ljungdahlii WT. 2,3-BD production was still formed during late stationary growth phase, which would support the assumption that 2,3-BD production is heavily dependent on either acetate formation or excess of reducing equivalents.

Regarding heterotrophic growth experiments of C. coskatii strains using fructose as carbon source, it was striking that none of the recombinant C. coskatii strains produced higher 2,3-BD concentrations compared to C. coskatii WT (Figure 12). Furthermore, C. coskatii WT (1.1 mM) and C. coskatii [pJIR750_23BD] (1.0 mM) were the only strains with 2,3-BD concentrations above the detection limit. Moreover, all C. coskatii strains showed lower 106 4. Discussion fructose consumption compared to C. ljungdahlii strains. The recombinant C. coskatii strains did only consume 20.7-26.4 mM fructose compared to recombinant C. ljungdahlii strains (40-30 mM) growing in Tanner medium. Additionally, C. coskatii required higher yeast extract concentrations (2 g instead of 0.5 g) to reach a comparable OD 600nm in Tanner medium as C. ljungdahlii . These heterotrophic growth experiments already indicated that C. coskatii is not an efficient 2,3-BD production strain, nevertheless, C. coskatii WT and the recombinant C. coskatii strains were also investigated under autotrophic growth conditions using syngas as energy and carbon source. All C. coskatii strains showed decreased OD 600nm values of 0.5-0.8 (Figure 13) compared to C. ljungdahlii strains (0.7-1.8) (Figure 10) but this did not lead to a decreased acetate production. In contrast, ethanol production was significantly decreased in all C. coskatii strains compared to C. ljungdahlii strains growing with syngas. In C. autoethanogenum and other acetogens it was shown that the enzyme aldehyde:ferredoxin oxidoreductase (AOR) plays an important role in ethanol formation (Fast and Papoutsakis, 2012; Bertsch and Müller, 2015; Mock et al ., 2015; Richter et al ., 2016; Liew et al ., 2017). The enzyme AOR catalyzes the reversible reduction of an acid to its corresponding aldehyde (White et al ., 1989). In several acetogens, undissociated acetic acid is reduced to acetaldehyde (Napora-Wijata et al ., 2014), which is further metabolised to ethanol via a monofunctional alcohol dehydrogenase (Adh) or a bifunctional aldehyde/alcohol dehydrogenase (AdhE) (Figure 41). This pathway was postulated recently (Bertsch and Müller, 2015) alongside the classic pathway of ethanol formation via the bifunctional AdhE (Figure 41). One great advantage of the ethanol pathway via AOR is that one mol ATP is generated by the enzyme acetate kinase (Ack) via substrate-level phosphorylation during acetate formation. Moreover, proteome studies of C. ljungdahlii revealed that both AORs (aor1 CLJU_c20110, aor2 CLJU_c20210) and Adh (adh2 , CLJU_c39950) were highly abundant during syngas fermentation under acidogenesis as well as solventogenesis (Richter et al ., 2016), which is in accordance to the assumption that ethanol formation via AOR is more advantageous due to ATP generation. Thus, ethanol production can immediately take place, as soon as overflow reactants such as undissociated acetic acid and reducing equivalents reach a critical concentration (Richter et al ., 2016). This is the case when an excess of acetyl-CoA as well as reducing equivalents cannot be further utilized in biomass formation (Richter et al ., 2016). Furthermore, Richter et al. (2016) postulated that ethanol production is exclusively achieved via reduction of undissociated acetic acid during fermentation of syngas. The genomes of 4. Discussion 107

C. autoethanogenum (aor1 CAETHG_0092, aor2 CAETHG_0102) as well as C. ljungdahlii (aor1 CLJU_c20110, aor2 CLJU_c20210) contain two aor genes and it turned out that these genes are missing in C. coskatii (Bengelsdorf et al ., 2016). This would explain why all C. coskatii strains produced significantly less ethanol under autotrophic growth conditions. Furthermore, the recombinant C. coskatii [pMTL82151_23BD] strain produced higher ethanol concentrations (15.7 mM/7.6 mM) compared to C. coskatii WT (6.1 mM/2.3 mM). In contrast, no higher 2,3-BD amounts were detected growing with fructose as carbon source as well as with syngas, respectively. It might be assumed that flux towards 2,3-BD originating from pyruvate via enzymes AlsS, BudA, and Bdh is not sufficient and in this case Bdh might rather be converting acetaldehyde to ethanol than acetoin to 2,3-BD. The enzyme Bdh from C. ljungdahlii was characterized recently but acetaldehyde was not tested as possible substrate of this enzyme (Tan et al ., 2015). Moreover, the primary-secondary alcohol dehydrogenase (sAdh) from C. autoethanogenum (CAETHG_0553), which is also present in C. ljungdahlii (CLJU_c24860), is not only capable of reducing acetoin to 2,3-BD, but it can also utilize acetaldehyde as substrate (Köpke et al ., 2014). Thus, it cannot be excluded that Bdh from C. ljungdahlii is also capable of reducing acetaldehyde to ethanol. In this case, Bdh would play the same role as ethanol formation via AOR in acetogens: getting rid of excess of reducing equivalents when growth is limited during stationary growth phase and biomass can no longer be produced. However, C. coskatii [pJIR750_23BD] harboring the same synthetic 2,3-BD operon, but a different replicon (pIP404 from C. perfringens ) for Gram-postive bacteria did not show the effect of higher ethanol production. This might indicate that the pBP1-replicon from C. botulinum (pMTL82151) leads to higher plasmid copy numbers in C. coskatii compared to pIP404-replicon (pJIR750). Nevertheless, this assumption needs to be proven by experimental data such as investigation of transcript alsS -budA -bdh from C. coskatii by Northern blot analysis or determination of plasmid copy numbers of pMTL82151_23BD and pJIR750_23BD in C. coskatii via qRT-PCR. In summary, heterotrophic as well as autotrophic growth experiments with C. coskatii strains showed that this acetogen is not a suitable 2,3-BD production strain.

4.2.2 Modifications of 2,3-BD production plasmid to optimize 2,3-BD production For further enhancement of 2,3-BD production using C. ljungdahlii , several modifications of the 2,3-BD production plasmid pJIR750_23BD were conducted. Firstly, the sequence of budA from the synthetic 2,3-BD operon alsS -budA -bdh was shortened at the 3'-end of budA and 108 4. Discussion cloned into the plasmid pJIR750_23BD to avoid a potential rho-independent terminator structure (Figure 14). Although the sequence of the terminator structure does not show a poly-A tail and the Gibbs free energy is rather low (-9.3 kcal/mol; Figure 14) compared to most rho-independent terminator structures, for example of the Gram-positive bacterium B. subtilis (~15 kcal/mol; De Hoon et al ., 2005), abortion of transcription due to this hairpin loop cannot be excluded. As a second modification of pJIR750_23BD, the gene bdh of the alsS -budA -bdh operon was exchanged with adh (CLJU_c24860), encoding the primary-secondary alcohol dehydrogenase sAdh from C. ljungdahlii and the identical gene is also present in C. autoethanogenum . This enzyme was shown to be capable of reducing S-acetoin to 2R,3R - BD as well as acetone to isopropanol in C. autoethanogenum (Köpke et al ., 2014). If enzyme sAdh shows a better substrate specifity or a higher enzymatic activity compared to Bdh this would also lead to a higher 2,3-BD production in C. ljungdahlii . As a third option for modification of pJIR750_23BD, the budABC operon from Raoultella terrigena was cloned into the vector pJIR750 under control of P pta-ack from C. ljungdahlii and T sol-adc from C. acetobutylicum. This organism was first described as K. terrigena in 1981 and it was reported as a Gram-negative bacterium, which can be found in water and soil (Izard et al ., 1981). In 2001, it was renamed R. terrigena together with Raoultella planticola and Raoultella ornithinolytica due to 16S rDNA and rpoB gene sequencing studies (Drancourt et al ., 2001). Via overexpression of the 2,3-BD operon budABC from R. terrigena with a constitutive promoter from C. ljungdahlii and a terminator from C. acetobutylicum , 2,3-BD overproduction using heterologous genes might be achieved in C. ljungdahlii (Figure 40). Investigation of the acetoin reductase (Acr) encoded by budC from R. terrigena via BLAST analysis on protein level revealed high similarity (95 %) to Acr from K. pneumoniae . This enzyme installs a S stereocenter during reduction of R-acetoin, which leads to production of meso -2,3-BD (Yan et al ., 2009; Zhang et al ., 2012a). Thus, it is very likely that Acr from R. terrigena also installs a S-stereocenter, which would lead to formation of meso -2,3-BD in C. ljungdahlii , if correct transcription and translation of the heterologous genes takes place. Transformation of the modified pJIR750_23BD plasmids into C. ljungdahlii via electroporation was only successful after in vivo methylation using E. coli ER2275 [pACYC184_MCljI]. This plasmid contains the modification (methylase) subunit (CLJU_c03310) and the specifity subunit (CLJU_c03320) from the C. ljungdahlii restriction-modification (R-M) system (Matthias Beck, unpublished). R-M systems degrade exogenous DNA and thus previous in vivo methylation of plasmids might 4. Discussion 109

Figure 40: 2,3-BD production in C. ljungdahlii with genes originating from R. terrigena starting from pyruvate with budB (acetolactate synthase), budA (acetolactate decarboxylase), and budC (acetoin reductase). be beneficial to overcome restriction barriers of the organism of choice by stabilizing transformed DNA and therefore boosting transformation efficiency (Chen et al ., 2008; Suzuki, 2012). Although the modified plasmids are smaller in size (10,867 bp-10,342 bp) compared to pJIR750_23BD (10,960 bp), it might be possible that the batch of electrocompetent C. ljundahlii cells was not as efficient as the one used in transformation of pJIR750_23BD and therefore in vivo methylation enhanced transformation of the modified pJIR750_23BD plasmids into C. ljungdahlii . After successful transformation, the recombinant C. ljungdahlii strains harboring the modified 2,3-BD production plasmid were investigated in terms of growth behavior and production pattern under heterotrophic and autotrophic growth conditions.

In growth experiments using fructose as carbon source C. ljungdahlii [pJIR750_23BD] showed decreased growth with an OD600nm of 0.9 compared to C. ljungdahlii WT (OD 600nm 2.7), but also a higher ethanol concentration (29.5 mM) compared to C. ljungdahlii WT (8.5 mM) (Figure 23). This might be explained by analytical problems either in acetate detection via HPLC or ethanol analysis via GC, since calculation of carbon recovery shows that carbon concentration of product formation (266.6 mM) exceeds carbon concentration of substrate consumption (237.6 mM). The strain C. ljungdahlii [pJIR750_23BD] did not produce higher ethanol concentrations in the other growth experiment with fructose (Figure 9), which suggests that problems in ethanol analytics occurred. Regarding the recombinant strains harboring the 110 4. Discussion modified 2,3-BD production plasmids, only C. ljundahlii [pJIR750_23BD_budAshort_adh] as well as C. ljungdahlii [pJIR750_budABCoperon] achieved slightly increased 2,3-BD concentrations of 2.1 mM and 1.9 mM compared to C. ljungdahlii WT (1.8 mM). The growth experiment using syngas as energy and carbon source (Figure 24) showed a slightly different picture, especially regarding 2,3-BD production. Under these conditions, C. ljungdahlii [pJIR750_23BD_budAshort_adh] showed the highest 2,3-BD production of 3.3 mM, which is 1.9-fold higher compared to C. ljungdahlii WT (1.7 mM). Recently, it was shown that sAdh of C. autoethanogenum displays a lower enzyme activity (0.4 U/mg) compared to Bdh (0.8 U/mg) (Köpke et al., 2016). However, overexpression of adh in C. ljungdahlii [pJIR750_23BD_budAshort_adh] led to similar 2,3-BD concentrations (3.3 mM) compared to C. ljungdahlii [pJIR750_23BD] (2.9 mM) as well as C. ljungdahlii [pMTL82151_23BD] (3.1 mM), in which bdh is overexpressed. This shows that lower enzyme activity of sAdh is either not present in C. ljungdahlii or it does not have a massive effect on 2,3-BD production. The strain C. ljungdahlii [pJIR750_23BD_budAshort] did not produce higher 2,3-BD amounts compared to C. ljungdahlii [pJIR750_23BD] neither using fructose (1.5 mM vs. 2.7 mM) nor syngas (2.8 mM vs. 2.9 mM) as energy and carbon source. This shows that the potential rho- independent terminator structure downstream of budA does not seem to affect expression of the synthetic 2,3-BD operon. Moreover, the recombinant strain C. ljungdahlii [pJIR750_budABCoperon], harboring the heterologous genes for 2,3-BD production from R. terrigena produced similar 2,3-BD concentrations (1.8 mM) compared to C. ljungdahlii WT (1.7 mM). Thus, neither using fructose as carbon source nor syngas as energy and carbon source led to higher meso -2,3-BD production in C. ljungdahlii [pJIR750_budABCoperon] compared to the natural 2R,3R -BD production of C. ljungdahlii WT. The budABC operon from R. terrigena was cloned directly into the vector pJIR750, but comparison of the GC content of R. terrigena (56.7 %) with C. ljungdahlii (31 %) shows that there is a major difference. Also, the codon usage table of R. terrigena compared to C. ljungdahlii (Table 11) displays differences in the abundance of triplets coding for the different amino acids. The complete codon usage table is presented in Table 13 (chapter 7. Supplement). For example, Table 11 shows that C. ljungdahlii prefers GCA triplet to encode alanine, whereas R. terrigena uses GCC triplet. Further differences are present for amino acids glycine, isoleucine, leucine, asparagine, proline, glutamine, arginine, serine, threonine, and valine. Thus, it is possible that codon usage 4. Discussion 111 optimization of the budABC operon from R. terrigena would lead to higher 2,3-BD concentrations in C. ljungdahlii , due to faster translation rates or higher translation efficiency.

Table 11: Comparison of codon usage in R. terrigena and C. ljungdahlii

Amino acid (abbr.) Triplet Abundance [%] R. terrigena * Abundance [%] C. ljungdahlii ** Alanine (A) GCA 18 41 GCC 30 15 GCG 29 6 GCT 23 38 Cysteine (C) UGC 62 44 UGU 38 56 Glycine (G) GGA 12 42 GGC 39 16 GGG 18 15 GGU 30 27 Isoleucine (I) AUA 12 43 AUC 35 17 AUU 53 40 Leucine (L) CUA 10 14 CUC 8 7 CUG 36 10 CUU 14 20 UUA 17 32 UUG 14 17 Asparagine (N) AAC 50 27 AAT 50 73 Proline (P) CCA 21 38 CCC 15 15 CCG 37 8 CCU 27 39

*according to Codon usage database (http://www.kazusa.or.jp/codon/cgi-bin/showcodon.cgi?species=577&aa=1&style=N, 14 th of June, 2017.) **according to Anja Poehlein, Göttingen Genomics Laboratory (Göttingen, Germany) 112 4. Discussion

Table 11: Comparison of codon usage in R. terrigena and C. ljungdahlii (continued)

Amino acid (abbr.) Triplet Abundance [%] R. terrigena * Abundance [%] C. ljungdahlii ** Glutamine (Q) CAA 37 61 CAG 63 39 Arginine (R) AGA 13 48 AGG 8 30 CGA 11 6 CGC 30 4 CGG 10 5 CGU 29 7 Serine (S) AGC 24 14 AGU 15 21 UCA 14 21 UCC 14 16 UCG 17 3 UCU 16 25 Threonine (T) ACA 17 36 ACC 38 19 ACG 29 7 ACU 15 38 Valine (V) GUA 14 38 GUC 22 12 GUG 32 17 GUU 32 33 *according to Codon usage database (http://www.kazusa.or.jp/codon/cgi-bin/showcodon.cgi?species=577&aa=1&style=N, 14 th of June, 2017.) **according to Anja Poehlein, Göttingen Genomics Laboratory (Göttingen, Germany) 4.2.3 PFOR enzyme as potential bottle-neck in 2,3-BD production? In order to further enhance natural 2,3-BD production in C. ljungdahlii , the gene nifJ encoding the pyruvate:ferredoxin oxidoreductase (PFOR) from C. ljungdahlii was overexpressed. This organism harbors two nifJ genes (CLJU_c29340 and CLJU_c09340), which are also present in C. autoethanogenum . Köpke et al . (2011) reported that in C. autoethanogenum only one nifJ gene (CAETHG_0928) is constituvely expressed at a high level, while the other (CAETHG_3029) 4. Discussion 113 is only upregulated at the end of the growth in a batch. This was the reason for choosing the homologous gene from C. ljungdahlii (CLJU_c09340) for overexpression in this organism. Moreover, it was shown that PFOR exhibits a lower enzyme activity (0.11 U/mg) in C. autoethanogenum compared to Bdh and sAdh (1.2 U/mg) and therefore conversion of acetyl-CoA to pyruvate seems to be the rate limiting step in 2,3-BD formation (Köpke et al ., 2016). They concluded that overexpression of nifJ together with alsS and budA is sufficient to remove the bottleneck in 2,3-BD production (Köpke et al ., 2016). Hence, overexpressing nifJ can be useful in order to increase the amount of PFOR and thus enhance flux towards 2,3-BD in C. ljungdahlii . This was tested by cloning either nifJ additionally into pMTL82151_23BD as well as pJIR750_23BD or exchange of bdh with nifJ in both plasmids. Transformation of these plasmids via electroporation did not lead to any success and via conjugation only pMTL82151_23BD_PFOR and pMTL82151_23BD_oBDH_PFOR were transformed successfully into C. ljungdahlii . The newly constructed plasmids harboring nifJ were the largest plasmids for 2,3-BD production in this study with sizes of 11,898 bp to 15,274 bp. A correlation between plasmid size and transformation effiency was reported in many bacteria, e. g. E. coli (Hanahan, 1983; Sheng et al ., 1995; Chan et al ., 2002) and B. subtilis (Ohse et al ., 1995) only to mention a few of them. A more detailed description why conjugation was used for transformation of large plasmids is given in chapter 4.6.

Growth experiments using fructose (Figure 33) as well as syngas as energy and carbon source (Figure 34) were conducted to give insights into growth behavior and production pattern of C. ljungdahlii overexpressing nifJ . Under both growth conditions only C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] showed increased 2,3-BD concentrations of 2.2 mM and 4.7 mM, respectively, compared to C. ljungdahlii WT (1 mM and 2.2 mM). This correlates to a 2.1 and 2.2-fold, respectively, higher 2,3-BD production of C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] compared to C. ljungdahlii WT. These results demonstrate that overexpression of nifJ together with alsS and budA is leading to an increased flux towards 2,3-BD. Nevertheless, the production is 2.6-fold lower compared to results of the company Lanzatech (1.1 g/L = 12.2 mM), which were achieved with C. autoethanogenum overexpressing alsS and budA in batch culture using synthetic syngas es energy and carbon source (Table 10; Köpke et al ., 2016). There are several possible reasons why the recombinant C. ljungdahlii strain in this study produced lower 2,3-BD concentrations. Firstly, LanzaTech 114 4. Discussion used the plasmid pMTL83195 harboring the pCB102-replicon from C. butyricum compared to the pBP1-replicon from C. botulinum used in this study. It is possible that pMTL83195 exhibits a higher plasmid copy number in C. autoethanogenum compared to pMTL82151 in

C. ljungdahlii . Secondly, Lanzatech uses the ferredoxin gene promoter (P fdx ) from

C. perfringens instead of P pta-ack , which is reported as one of the strongest promoters in the genus Clostridum (Takamizawa et al ., 2004). Thirdly, Lanzatech uses higher overpressure (2.0 bar) in batch fermentations combined with utilization of a shaking incubator, which both lead to a better gas-to-liquid mass transfer of CO. These are all factors, which might be starting points to further increase 2,3-BD production in C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR]. The best 2,3-BD production of the recombinant C. ljungdahlii strains constructed in this study (0.4-0.6 g/l), are much below 2,3-BD concentrations of recombinant S. marcescens (152 g/l) and K. pneumoniae (150 g/l) strains (Table 10). However, these strains use glucose as carbon source and are unattractive for industrial fermentation processes, due to belonging to the risk group 2. Moreover, Lanzatech already showed that 2,3-BD production from syngas has great potential for industry by opening a pilot plant at BlueScope Steel (Glenbrook, New Zealand) producing CO-based 2,3- BD with 15,000 gal/year (Regalbuto et al ., 2017).

The autotrophic growth experiment (Figure 34) revealed that C. ljungdahlii

[pMTL82151_23BD_oBDH_PFOR] showed a decreased max. OD 600nm (0.9) compared to the other recombinant C. ljungdahlii strains (1.3 and 1.5) as well as compared to C. ljungdahlii WT (2.0). Moreover, this strain produced less ethanol (14.1 mM) compared to the other C. ljungdahlii strains (43.5 mM-50.9 mM). A possible explanation might be presented, if consumption and generation of reducing equivalents are taken into account (Figure 41). This figure was obtained taking the data of a recently published study on C. autoethanogenum into account (Richter et al ., 2016). Overexpression in C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] leads to higher conversion of acetyl-CoA to pyruvate and thereby reduced ferredoxin (Fd 2-) is consumed. This explains higher 2,3-BD production compared to C. ljungdahlii WT. Moreover, if more reduced Fd 2- is consumed via the PFOR reaction, less Fd 2- is available for ethanol production via the AOR pathway. It was already 4. Discussion 115

Figure 41 : Enzymes of the Wood -Ljungdahl pathway with corresponding reducing equivalents in C. ljungdahlii . Fdh and electron bifurcating hydrogenase HytCBDE 1AE 2 can form a complex to directly reduce CO 2 to formate with H 2. Rnf, Rhodobacter nitrogen fixation; ATPase, adenosine triphosphatase; Nfn, NADH-dependent reduced ferredoxin:NADP + oxidoreductase; CODH, carbon monoxide dehydrogenase; FdhA, formate dehydrogenase; Acs/CODH, complex of acetyl-CoA synthase with carbon monoxide dehydrogenase; THF, tetrahydrofolate; [CO], enzyme-bound CO; Fdh, formate dehydrogenase; Fhs, formyl-THF synthetase; Mtc, methenyl-THF cyclohydrolase; Mtd, methylene-THF deyhdrogenae; Mtr, methylene-THF reductase; Met, methyl transferase; Pfor, pyruvate:ferredoxin oxidoreductase; AlsS, acetolactate synthase; BudA, acetolactate decarboxylase; Bdh, 2,3-BD dehydrogenase; sAdh, primary-secondary alcohol dehydrogenas e; Pta, phosphotransacetylase; Ack, acetate kinase; AdhE, bifunctional aldehyde/alcohol dehydrogenase; Adh, monofunctional alcohol dehydrogenase; Aor, aldehyde:ferredoxin oxidoreductase; Co-FeS-P, corrinoid iron-sulfur protein; HS-CoA, coenzyme A. Based on the model of C. ljungdahlii published by Richter et al. (2016). mentioned above that ethanol production is expected to be exclusively achieved via reduction of undissociated acetic acid in C. ljungdahlii during fermentation on syngas (Richter et al ., 2016), which would explain lower ethanol production in C. ljungdahlii 116 4. Discussion

[pMTL82151_23BD_oBDH_PFOR]. Furthermore, reduction of acetyl-CoA to ethanol via acetate by AOR pathway yields one ATP due to substrate level phosphorylation, which is lacking if AOR reaction cannot take place, due to lower amounts of Fd 2-. Additionally, the decreased growth of C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] might be explained by the minimized Fd 2- pool for the Rnf (Rhodobacter nitrogen fixation; ferredoxin:NAD + oxidoreductase) complex. This complex needs Fd 2- to reduce oxidized nicotinamide adenine dinucleotide (NAD +) with concomitant translocation of cations (H + in C. ljungdahlii , Na + in A. woodii ) over the cell membrane (Imkamp et al ., 2007; Müller et al ., 2008; Biegel and Müller, 2010; Tremblay et al ., 2012; Hess et al ., 2016). Thereby, a proton gradient or sodium gradient is generated, which is further used by an ATPase to provide additional ATP, since the Wood- Ljungdahl pathway alone leads to no net ATP gain (Reidlinger and Müller, 1994). The ATP, which is generated during acetate production via substrate level phosphorylation is needed for activation of formate to formyl-THF. A detailed discussion of the phenotype of mutated C. ljungdahlii [pMTL82151] is given in section 4.4.

The recombinant strain C. ljungdahlii [pMTL82151_23BD_PFOR], which harbors the plasmid for overexpression of alsS , budA , bdh , and nifJ did not produce higher 2,3-BD concentrations (0.8 mM and 2.5 mM) compared to C. ljungdahlii WT (1.0 mM and 2.2 mM) neither using fructose nor syngas as energy and carbon source. The first three genes of the plasmid pMTL82151_23BD_PFOR are identical to pMTL82151_23BD_oBDH_PFOR, so C. ljungdahlii [pMTL82151_23BD_PFOR] was expected to produce similar or even higher amounts of 2,3-BD, due to additionally harboring the bdh gene. The plasmid pMTL82151_23BD_PFOR was verified successfully in C. ljungdahlii [pMTL82151_23BD_PFOR] after performing the growth experiments, but it cannot be excluded that mutations were present in this plasmid which might affect 2,3-BD production of this strain. Thus, the sequence of pMTL82151_23BD_PFOR from recombinant strain C. ljungdahlii [pMTL82151_23BD_PFOR] needs to be further elucidated.

4.3 Detection of transcripts from synthetic 2,3-BD operons in C. ljungdahlii Northern blot experiments were conducted to investigate, if the transcripts of the synthetic 2,3-BD operons alsS -budA -bdh , alsS -budAshort -bdh , and alsS -budAshort -adh (Figure 25) are completely expressed in C. ljungdahlii . Moreover, it was investigated if utilization of one promoter is sufficient for transcription of three genes. RNA was isolated in the early stationary 4. Discussion 117 growth phase (after 65 h), when 2,3-BD formation starts and transcripts for 2,3-BD production were expected to be present. This was also shown in C. autoethanogenum via quantitative PCR, where a slight up-regulation of genes involved in 2,3-BD formation was shown after 50 h in the late exponential growth phase (Köpke et al ., 2011b). Detection of transcripts alsS, budA , and bdh from C. ljungdahlii chromosome was successful in C. ljungdahlii WT with sizes of 1,670 bases, 1,240 bases, and 1,074 bases, respectively (Figure 26). The transcript budA was expected to have a size of 720 bases, but Figure 25B shows that budA is located downstream of a coding sequence (CLJU_c08370), which is annotated as a hypothetical protein. Thus, it seems budA and CLJU_c08370 form an operon, leading to a polycistronic mRNA with a size of 1,240 bases compared to 720 bases, in case of budA being a monocistronic mRNA. Since genes for natural 2,3-BD production are located on the chromosome of C. ljungdahlii , they are expected to be present in all investigated C. ljungdahlii strains. However, transcript alsS was only detected in C. ljungdahlii WT, transcript budA in C. ljungdahlii WT, C. ljungdahlii [pMTL82151], as well as C. ljungdahlii [pJIR750_23BD_budAshort_adh], and transcript bdh in C. ljungdahlii WT as well as C. ljungdahlii [pMTL82151] (Figure 26). A possible explanation might be that the respective mRNAs were not formed yet after 65 h, when RNA was isolated from the different C. ljungdahlii strains. Furthermore, alsS -budA -bdh trancript of the synthetic 2,3-BD operon is only present in C. ljungdahlii [pJIR750_23BD], which is in accordance with results of growth experiments using fructose as well as syngas as energy and carbon source (Figure 9 and Figure 10) showing C. ljungdahlii [pJIR750_23BD] as best 2,3-BD producer compared to C. ljungdahlii WT. Moreover, the transcript alsS -budA -bdh was not detected in C. ljungdahlii [pMTL82151_23BD], although it produced a higher 2,3-BD concentration compared to C. ljungdahlii WT in a growth experiment using syngas as carbon source (Figure 24). There might be much less transcript present in C. ljungdahlii [pMTL82151_23BD] cells compared to C. ljungdahlii [pJIR750_23BD], which would show that plasmid replication of vector pMTL82151 is lower compared to pJIR750 in C. ljungdahlii . Furthermore, the transcripts of the other 2,3-BD operons alsS -budAshort -bdh and alsS -budAshort -adh could not be detected via Northern blot analysis. This was unexpected, since all synthetic 2,3-BD operons share the same constitutive promoter from C. ljungdahlii (P pta -ack ) with the only difference in exchange of one gene of the operon. Either the transcripts were not formed yet after 65 h, which is otherwise unlikely since a constitutive promoter was used for expression, or the transcript was absent, due to a so far unknown reason. Moreover, Northern blot analysis 118 4. Discussion revealed several shorter signals at sizes of around 500 bases, 750 bases, 1,000 bases, and 1,500 bases. In contrast, these shortened signals were never present in C. ljungdahlii WT, so they might be the result of unspecific binding of the probes either to the vector or to the plasmid. However, alignment of the probes with the respective plasmids via in silico analysis did not reveal any high homologies, so there might be another unknown reason for these truncated signals.

4.4 Investigation of mutated C. ljungdahlii [pMTL82151] showing increased flux towards ethanol and 2,3-BD The mutated C. ljungdahlii [pMTL82151] strain produced slightly higher 2,3-BD concentrations of 2.4 mM with fructose as carbon source (Figure 33) and 5.6 mM with syngas as energy and carbon source (Figure 34) compared to C. ljungdahlii [pMTL82151_23BD_oBDH_PFOR] (2.2 mM and 4.7 mM). This correlates to a 2.4-fold and 2.5-fold higher 2,3-BD production compared to C. ljungdahlii WT, which produced 1.0 mM and 2.2 mM 2,3-BD, respectively. Furthermore, this strain did not produce higher 2,3-BD concentrations (1.5 mM) in another growth experiment using syngas as energy and carbon source (Figure 24) compared to C. ljungdahlii WT (41.9 mM and 1.7 mM, respectively). This did not reflect the results from the growth experiment using fructose as carbon source (Figure 23) and the other growth experiment using syngas as energy and carbon source (Figure 34). It might be assumed that mutated C. ljungdahlii [pMTL82151] did not show the phenotype of higher ethanol production as well as higher 2,3-BD production in this growth experiment (Figure 24), due to worse CO supply. As mentioned above, sufficient CO supply is an important feature for 2,3-BD production (Simpson et al ., 2011). Since C. ljungdahlii [pMTL82151] did still contain the vector pMTL82151, which was confirmed by isolation of genomic DNA with concomitant retransformation into E. coli XL-1 Blue MRF' as well as sequencing by GATC Biotech AG (Constance, Germany), it was assumed that one or more mutations located on the chromosome were responsible for the phenotype of higher ethanol as well as 2,3-BD production. Recently, a spontaneous C. ljungdahlii mutant showing higher ethanol production was reported (Whitham et al ., 2017). The strain designated as C. ljungdahlii OTA-1 was obtained after repeated subculturing of C. ljungdahlii WT in a rich mixotrophic medium and storage for 2 weeks at room temperature (Whitham et al ., 2017). Thus, it seems like C. ljungdahlii has a potential to develop mutations during cultivation over a long period. The 4. Discussion 119 assumption that mutation of C. ljungdahlii also appeared in this study was confirmed by sequencing genomes of C. ljungdahlii WT and C. ljungdahlii [pMTL82151] with subsequent SNP analysis. Several SNPs leading to a change in the translated AA were found in mutated C. ljungdahlii [pMTL82151] (Table 8). The spontaneous mutant strain showed this phenotype after lyophilisation and reactivation. Since most of the SNPS were present in genes, which were not assumed to be involved in ethanol or 2,3-BD production, only SNPs in gene adhE1 were examined more closely via comparison of protein domains (Figure 43) and protein structure modelling (Figure 42). Two of the not investigated locus tags are CLJU_C07590 as well as CLJU_c07600, which are annotated as putative secretion protein HlyD and putative transporter protein, respectively (Table 8). Ethanol and 2,3-BD are small uncharged polar molecules, which can pass the cell membrane via passive diffusion (Lodish et al ., 2004). Hence, it is unlikely that transporter or secrection proteins are involved in higher ethanol or higher 2,3-BD production. Moreover, locus tags CLJU_c24850 and CLJU_c24870 are designated as genes stc1 and stc2 , which encode signal-transduction and transcriptional regulator proteins. They are also present in the stc cluster of the fungus Aspergillus nidulans , encoding enzymes needed for synthesis of a toxic and carcinogenic secondary metabolite called sterigmatocystin (precurser of the fungal toxin aflatoxin) (Hicks et al ., 1997). The acetogen C. ljungdahlii does not contain such a stc cluster, thus it might be concluded that genes stc1 and stc2 were obtained during evolution via horizontal gene transfer, but do not fulfill a specific function in this organism. The SNP introducing a stop codon in adhE1 (CLJU_c16510) (Table 8) seems to

Figure 42 : Structure of enzyme models for AdhE1 from C. ljungdahlii WT (A) and mutated C. ljungdahlii [pMTL82151] (B). Modelling of protein sequences was performed using I-TASSER: Protein Structure & Function Predictions (http://zhanglab.ccmb.med.umich.edu/I-TASSER/). 120 4. Discussion

Figure 43: Protein domains of AdhE1 from C. ljungdahlii WT (A) and truncated AdhE1 from mutated C. ljungdahlii [pMTL82151] (B ). Analysis of protein sequences were performed using InterPro: protein sequence analysis & classification (http://www.ebi.ac.uk/interpro/search/sequence- search). be the most important SNP of this study, since this gene encodes a bifunctional aldehyde/alcohol dehydrogenase, which catalyzes the reduction of acetyl-CoA to acetaldehyde as well as the subsequent reduction of acetaldehyde to ethanol. This enzyme is typically composed of a N-terminal acetylating aldehyde dehydrogenase (Ald) domain and a C-terminal Fe-dependent Adh domain (Membrillo-Hernandez et al ., 2000; Extance et al ., 2013). Figure 42 and Figure 43 show the impact of the stop codon on the enzyme structure of AdhE1 and its protein domains in comparison to C. ljungdahlii WT. The stop codon leads to a truncated version of AdhE1 with 202 AA in mutated C. ljungdahlii [pMTL82151] (Figure 42B and Figure 43B) compared to AdhE1 of C. ljungdahlii WT with 870 AA (Figure 42A and Figure 43A). There is another gene ( adhE2 ) encoding a bifunctional aldehyde/alcohol dehydrogenase, which is located directly downstream of adhE1 on the chromosome of C. ljungdahlii (Köpke et al ., 2010; Leang et al ., 2013). The two genes are also present in C. autoethanogenum and supposed to be a potential result of gene duplication (Humphreys et al ., 2015). Moreover, it was reported that deletion of adhE2 showed no effect on ethanol 4. Discussion 121 production in C. ljungdahlii using fructose as carbon source (Leang et al ., 2013) and overexpression of adhE2 in an adhE1 /adhE2 -deficient C. ljungdahlii strain did not result in restoring ethanol production (Banerjee et al ., 2014). These findings mean either that adhE2 does not encode a functional bifunctional aldehyde/alcohol dehydrogenase or posttranscriptional factors limit expression and/or enzyme activity of AdhE2 (Banerjee et al ., 2014). In contrast, Liew et al . (2017) reported that adhE2 inactivation resulted in 63 % lower ethanol production in C. autoethanogenum compared to C. autoethanogenum WT. Hence, although being very similar at the genetic level (Bengelsdorf et al ., 2016), both organisms show differences in the phenotype concerning ethanol production (Cotter et al ., 2009; Köpke et al ., 2010; Brown et al ., 2014; Liew et al ., 2016; Martin et al ., 2016; Marcellin et al ., 2016). The truncated AdhE1 of mutated C. ljungdahlii [pMTL82151] in this study was assumed to be equal to adhE1 knouckout due to the missing alcohol dehydrogenase domain and the truncated aldehyde dehydrogenase N-terminal domain (Figure 43B). The inactivated adhE1 phenotype of this study, showing higher ethanol and 2,3-BD production with fructose as well as syngas as energy and carbon source, reflects more the results observed in C. autoethanogenum (Liew et al ., 2017) than in C. ljungdahlii (Banerjee et al ., 2014). The deletion of adhE1 in C. ljungdahlii was reported to result in a strain, which produced 6-fold less ethanol compared to C. ljungdahlii WT (Banerjee et al ., 2014). This is contrary to the results presented in this study. However, adhE1 inactivation in C. autoethanogenum was shown to yield higher ethanol production (171-183 %) with CO as carbon source but decreased 2,3-BD production compared to C. autoethanogenum WT (Liew et al ., 2017). Furthermore, the increased ethanol production only occured, when strains were growing with CO but not during heterotrophic growth with fructose (Liew et al ., 2017). This shows that mutated C. ljungdahlii [pMTL82151] represents a new phenotype compared to adhE1 inactivation strains described in literature. The increase in ethanol production and 2,3-BD production of mutated C. ljungdahlii [pMTL82151] might be explained considering the pool of reducing equivalents (Figure 41). Higher ethanol concentrations are in agreement with the assumption that ethanol formation via AOR pathway is more favourable for heterotrophic as well as autotrophic ethanol biosynthesis, due to generation of ATP (Fast and Papoutsakis, 2012; Bertsch and Müller, 2015; Mock et al ., 2015; Richter et al ., 2016; Valgepea et al ., 2017a). Furthermore, Valgepea et al . (2017a) demonstrated that ethanol production is strongly correlated with biomass concentrations during steady-state culturing of C. autoethanogenum . The organism channels more carbon 122 4. Discussion into reduced by-products such as ethanol and 2,3-BD at high biomass concentrations. Additionally, they showed that several transcripts were down-regulated with higher ethanol production in C. autoethanogenum , for example the transcript of bifunctional aldehyde/alcohol dehydrogenase AdhE (CAETHG_3747) (Valgepea et al ., 2017a). This observation fits to the results being observed during this study that a truncated AdhE1 enzyme leads to higher ethanol production in mutated C. ljungdahlii [pMTL82151]. The higher 2,3-BD production might also result from inactivation of adhE1 leading to an excess of reducing equivalents, which can be consumed during 2,3-BD formation. This shows that mutated C. ljungdahlii [pMTL82151] might be a target for further enhancement of 2,3-BD production in C. ljungdahlii . Several passages of the strain without addition of antibiotic would result in loss of the vector pMTL82151, due to missing selection pressure. Subsequently, allelic exchange (Al-Hinai et al ., 2012) or ClosTron mutagenesis (Heap et al ., 2007) might be applied to inactivate genes encoding AOR. These methods of genetic manipulation to inactivate genes in clostridia were successfully applied for example in C. autoethanogenum recently (Mock et al ., 2015; Minton et al ., 2016; Liew et al ., 2017). Inactivation of aor would result in knock down of the AOR pathway. Afterwards, pMTL82151_23BD_oBDH_PFOR plasmid could be transformed into this strain to create flux towards 2,3-BD and not towards ethanol, since excess of reducing equivalents would be consumed solely during 2,3-BD formation. A further method to enhance 2,3-BD production in mutated C. ljungdahlii [pMTL82151] might be represented by the arginine deiminase pathway, which was recently reported to provide C. autoethanogenum with additional ATP, if arginine is used as nitrogen resource (Valgepea et al ., 2017b). This pathway would also lead to higher biomass concentrations and as mentioned above thereby producing higher 2,3-BD concentrations. However, RNA sequencing showed that the Wood-Ljungdahl pathway was down-regulated in strains growing with arginine (Valgepea et al ., 2017b), which might be unfavourable for 2,3-BD production using syngas as energy and carbon source.

4.5 A. woodii as suitable host for 2,3-BD production from H 2 + CO 2? The plasmids for 2,3-BD production were transformed into A. woodii to investigate, if

A. woodii can form 2,3-BD heterologously while growing on H 2 + CO 2. In order to test if 2,3-BD formation from H 2 + CO 2 as sole energy and carbon source is energetically profitable in A. woodii , stoichiometric energy balancing was performed. The calculation was done with 4. Discussion 123 incorporation of relevant energy mechanisms and reducing equivalents consuming-reactions depicted in Figure 44 (Poehlein et al ., 2012). Furthermore, the postulation that Mtr is not reducing oxidized ferredoxin and therefore is not electron-bifurcating was incorporated into the following equations (Bertsch et al ., 2015). The formation of 2,3-BD as sole end-product would be energetically unfavorable in A. woodii resulting in a negative ATP gain of -1.7 ATP (Bertsch et al ., 2015) (2).

4 CO 2 + 11 H2 → C4O2H10 + 4 H 2O + -1.7 ATP (2)

Figure 44 : Stoichiometrical heterologous 2,3 -BD production in A. wo odii with H 2 + CO 2 as sole energy and carbon source as well as relevant enzymes. Rnf, Rhodobacter nitrogen fixation; ATPase, adenosine triphosphatase; Acs/CODH, complex of acetyl-CoA synthase with carbon monoxide dehydrogenase; THF, tetrahydrofolate; [CO], enzyme-bound CO; Fdh, formate dehydrogenase; Fhs, formyl-THF synthetase; Mtc, methenyl-THF cyclohydrolase; Mtd, methylene-THF deyhdrogenase; Mtr, methylene-THF reductase; Met, methyl transferase; PFOR, pyruvate:ferredoxin oxidoreductase; AlsS, acetolactate synthase; BudA, acetolactate decarboxylase; Bdh, 2,3-BD dehydrogenase; sAdh, primary-secondary alcohol dehydrogenase; Co-FeS-P, corrinoid iron-sulfur protein; HS-CoA, coenzyme A.

124 4. Discussion

In conclusion, production of 6 mol acetate/mol 2,3-BD is needed to get a positive ATP balance of 0.1 ATP in A. woodii using H 2 + CO 2 as sole energy and carbon source (3).

- 16 CO 2 + 35 H2 → C4O2H10 + 6 C2H3OO + 16 H2O + 0.1 ATP (3)

Nevertheless, no recombinant A. woodii strain was able to produce 2,3-BD heterologously using H 2 + CO 2 as sole energy and carbon source (Figure 28). By this time, it was unknown that A. woodii can use divalent alcohols such as 2,3-BD as well as 1,2-propanediol via oxidation as carbon source (Hess et al ., 2015; Schuchmann and Müller, 2016). In this reaction 2,3-BD is oxidized to acetate with CO 2 as electron acceptor, which is reduced to an additional acetate (Hess et al ., 2015) (4).

- C4O2H10 + 1.5 CO 2 + 1.3 ADP + 1.3 P i→ 2.75 C2H3OO + 1.3 ATP (4)

This finding explains why 2,3-BD was not detected in the conducted growth experiment. However, A. woodii [pJIR750_23BD], A. woodii [pJIR750_23BD_budAshort], and A. woodii [pJIR750_23BD_budAshort_adh] displayed higher acetate concentrations of 163.9 mM, 166.0 mM, and 156.3 mM, respectively, compared to A. woodii WT (148.5 mM). These results would support the assumption that 2,3-BD was produced in A. woodii , but subsequently used as carbon source. Nevertheless, it was shown in a previous study that vector pJIR750 leads to higher acetate concentrations in A. woodii (Straub, 2012). Hence, it cannot be excluded that increased acetate concentrations were plasmid-based. This might have been elucidated, if samples for HPLC analysis would have been taken more often to see if 2,3-BD was formed and subsequently consumed in case of A. woodii producing 2,3-BD concentrations above the detection limit of HPLC analysis.

4.6 Heterologous butanol production in C. ljungdahlii - the conflict of large plasmid transformation In the past, heterologous butanol production (2 mM) from CO using C. ljungdahlii with the genes for butanol formation from C. acetobutylicum was reported (Köpke et al ., 2010). Butanol production from CO as sole carbon and energy source is coupled to the production of

CO 2 (5).

12 CO + 5 H 2O→ C 4OH 10 + 8 CO 2 (5) 4. Discussion 125

However, glycerol cultures of C. ljungdahlii [pSOB ptb ] could not be reactivated successfully. Thus, a new recombinant C. ljungdahlii strain producing butanol heterologously had to be constructed. For this purpose the plasmid pSB3C5-UUMKS3 (Schuster, 2011) was used containing two important differences compared to pSOB ptb . Firstly, this plasmid contains etfA and eftB , which were shown to be essential for reduction of crotonyl-CoA to butyryl-CoA in C. kluyveri (Li et al ., 2008) and are assumed to be also needed in C. acetobutylicum for this reaction (Lütke-Eversloh and Bahl, 2011). Furthemore, pSB3C5-UUMKS 3 harbors the gene adhE2 instead of bdhA and adhE1 , since recently published data showed that AdhE2 is the key enzyme responsible for butanol formation under acidogenesis and alcoholgenesis in C. acetobutylicum (Yoo et al ., 2016). Prior to transformation of the butanol plasmids pCE3 and pJR3 into C. ljungdahlii in vivo methylattion using E. coli ER2275 [pACYC184_MCljI] was performed. Therefore, the p15A-replicon of pACYC184_MCljI was exchanged, since pCE3 as well as pJR3 harbor the identical p15A-replicon . It is known from literature that two identical origins of replication cannot coexist in one organism (Novick, 1987). The pSC101-replicon was chosen to maintain the low-copy number of pACYC184_MCljI and furthermore p15A-replicon was described to coexist in one cell with pSC101-replicon (Peterson and Phillips, 2008). However, transformation of the large butanol plasmids (14,199 bp-15,425 bp; Figure 36 and Figure 39) was not successful via electroporation. The effect of decreasing transformation efficiency with increasing plasmid size was also shown for the acetogen A. woodii (Jonathan Baker, University of Nottingham, unpublished). Although electrotransformation is a method generally simple and efficient, conjugation is still a commonly used and frequently more efficient method, most notably for transformation of large plasmids (Schweizer, 2008). Bacterial conjugation requires direct contact between donor and recipient cell and was first described between different E. coli strains (Lederberg and Tatum, 1946). Conjugation shows at least two important advantages over electroporation. As mentioned above, electroporation efficiencies decrease with increasing plasmid sizes (Szostková and Horáková, 1998), while conjugation is lacking this limitation since even entire genomes have been successfully transformed into E. coli using this method (Isaacs et al ., 2011). The second advantage is that methylation of plasmid DNA prior to transformation is obsolete, since during conjugation the plasmid is transferred from the donor cell to recipient cell in form of a single strand, which is afterwards methylated in second strand synthesis (Dominguez and O’Sullivan, 2013). Thus, overcoming R-M system barriers of the organism can be achieved by using conjugation. 126 4. Discussion

However, also conjugation experiments with the butanol plasmids into C. ljungdahlii were not successful in this study. It might be concluded that the limit of transformable plasmid size was reached. A patent published by the company LanzaTech reported recombinant C. autoethanogenum DSM23693 and C. ljungdahlii DSM13528 strains harboring the butanol production plasmid pMTL85245-thlA-crt-hbd (Köpke and Liew, 2012). This plasmid contains the pIM13 replicon (Bacillus subtilis ) and the genes thlA (thiolase), hbd (3-hydroxybutyryl-CoA dehydrogenase), crt (crotonase), bcd (butyryl-CoA dehydrogenase), etfA (electrontransfer flavoprotein), and etfB (electrontransfer flavoprotein) under control of the promoter P pta-ack for butanol synthesis. Furthermore, genes encoding phosphotransbutyrylase, butyrate kinase, aldehyde:ferredoxin oxidoreductase, and butanol dehydrogenase were expressed. These genes allow C. autoethanogenum and C. ljungdahlii to produce two molecules of acetyl-CoA, which are afterwards combined to acetoacetyl-CoA and eventually converted via 3- hydroxybutyryl-CoA and crotonyl-CoA to butyryl-CoA. Thereafter, butyryl-CoA is transformed to butyryl phosphate catalyzed by phosphotransbutyrylase and this compound is further converted to butyrate via the enzyme butyrate kinase (Dürre, 2016b). Recombinant C. ljungdahlii strains produced 6 mM butanol under static conditions (without shaking), while C. autoethanogenum produced higher 1-butanol concentrations of 25.7 mM and butyrate production occurred (3.7 mM) (Köpke and Liew, 2012). Hence, this approach is beneficial due to improved energetics of the pathway, as the butyrate kinase reaction yields an additional ATP (Dürre, 2016b) and should be considered for further investigations on butanol production in acetogens.

5. Summary 127

5. Summary

1) Natural 2,3-BD production was detected in the acetogens C. ljungdahlii , C. autoethanogenum , and C. ragsdalei using syngas as energy and carbon source. Acetate was the main product and ethanol was a side product of these organisms. 2,3-BD was only a minor side product, while C. ljungdahlii produced the highest 2,3-BD concentration of 4.8 mM. 2) 2,3-BD production plasmids were constructed using the genes encoding acetolactate synthase ( alsS ), acetolactate decarboxlyase ( budA ), and 2,3-BD dehydrogenase ( bdh ) from

C. ljungdahlii under control of the constitutive P pta -ack promoter from the same organism. The synthetic 2,3-BD operon alsS -budA -bdh was cloned successfully into the E. coli - Clostridium shuttle vectors pMTL82151 and pJIR750 containing different replicons for Gram-positive bacteria. 3) Overexpression of the synthetic 2,3-BD operon alsS -budA -bdh using pJIR750_23BD resulted in a 1.5-fold higher 2,3-BD production (1.2 and 6.4 mM 2,3-BD, respectively) compared to C. ljungdahlii WT (0.8 and 4.3 mM 2,3-BD, respectively) with fructose as well as syngas as energy and carbon source. Overexpression of the synthetic 2,3-BD operon alsS -budA -bdh using pMTL82151_23BD resulted in a 1.8-fold increase of 2,3-BD production (3.1 mM) compared to C. ljungdahlii WT (17 mM) using syngas as energy and carbon source. 4) Overexpression of the synthetic 2,3-BD operon alsS -budA -bdh did not lead to an improvement of 2,3-BD production in C. coskatii . 5) Three modified 2,3-BD production plasmids were constructed using a shortened sequence of budA , exchange of bdh with primary-secondary alcohol dehydrogenase ( adh ) from

C. ljungdahlii , and the budABC operon from R. terrigena under control of P pta -ack promoter

from C. ljungdahlii and T sol -adc terminator from C. acetobutylicum . 6) Heterotrophic growth experiments with modified 2,3-BD production plasmids showed nearly no improvement of 2,3-BD production in C. ljungdahlii , whereas autotrophically grown C. ljungdahlii strains overexpressing adh produced 2.9-fold higher 2,3-BD concentrations (3.3 mM) compared to C. ljungdahlii WT (1.7 mM). Prevention of the potential rho-independent terminator structure as well as heterologous 2,3-BD 128 5. Summary

production using budABC operon from R. terrigena did not enhance 2,3-BD production compared to overexpression of the alsS -budA -bdh operon. 7) Recombinant A. woodii strains harboring 2,3-BD production plasmids did not produce 2,3-

BD heterologously using H 2 + CO 2 as energy and carbon source. 8) Genes encoding pyruvate:ferredoxin oxidoreductase (nifJ ) and conjugal transfer region (traJ /oriT ) were cloned successfully into pMTL82151_23BD and pJIR750_23BD. Conjugation into C. ljungdahlii was only successful for pMTL82151-based plasmids. 9) Overexpression of nifJ via alsS -budA -nifJ operon resulted in 2.2-fold (2.2 mM 2,3-BD) and 2.1-fold (4.7 mM 2,3-BD) higher 2,3-BD concentrations compared to C. ljungdahlii WT (1.0 mM and 2.2 mM 2,3BD) using fructose and syngas as energy and carbon source, respectively. Overexpression of nifJ via alsS -budA -nifJ -bdh operon did not improve 2,3-BD production compared to C. ljungdahlii WT. 10) Recombinant strain C. ljungdahlii [pMTL82151] showed the phenotype of decreased acetate production with concomitant increased ethanol and 2,3-BD production compared to C. ljungdahlii WT. 2,3-BD production was 2.4-fold increased (2.4 mM 2,3-BD) using fructose and 2.5-fold increased (5.6 mM 2,3-BD) using syngas as energy and carbon source compared to C. ljungdahlii WT (1.0 mM and 2.2 mM 2,3-BD, respectively). Genome sequencing with subsequent SNP analysis revealed several SNPs located in the chromosome of the spontaneously mutated C. ljungdahlii [pMTL82151] strain. The respective strain was designated mutated C. ljungdahlii [pMTL82151]. Most important, one SNP leading to a stop codon and therefore to a truncated AdhE1 enzyme seems to be the reason for the observed phenotype. 11) Transcripts alsS , budA , and bdh were successfully detected in C. ljungdahlii WT via Northern blot analysis. Transcript of synthetic 2,3-BD operon alsS -budA -bdh was verified exclusively in C. ljungdahlii [pJIR750_23BD]. 12) Replicons pIP404 ( C. perfringens ) and pCB102 ( C. butyricum ) as well as traJ /oriT region were cloned successfully into butanol production plasmid pSB3C5-UUMKS 3. Replicon p15A was exchanged successfully with replicon pSC101 in methylation plasmid pACYC184_MCljI. Transformation of butanol production plasmids pCE3 and pJR3 into C. ljungdahlii was not successful neither using electroporation nor conjugation. 6. Zusammenfassung 129

6. Zusammenfassung

1) Die acetogenen Stämme C. ljungdahlii , C. autoethanogenum und C. ragsdalei zeigten eine natürliche 2,3-Butandiol (2,3-BD)-Produktion mit Syngas als Kohlenstoff- und Energiequelle. Dabei war Acetat das Hauptprodukt und Ethanol bzw. 2,3-BD traten als Nebenprodukte auf, wobei 2,3-BD nur ein geringfügiges Nebenprodukt war. Dabei produzierte C. ljungdahlii am meisten 2,3-BD mit einer Konzentration von 4,8 mM. 2) Es wurden 2,3-BD-Produktionsplasmide konstruiert, welche die Gene für die Acetlactat- Synthase ( alsS ), Acetlactat-Decarboxylase ( budA ) und 2,3-Butandiol-Dehydrogenase ( bdh )

unter Kontrolle des konstitutiven Promoters Ppta -ack aus C. ljungdahlii tragen. Das synthetische 2,3-BD-Operon alsS -budA -bdh wurde erfolgreich in die E. coli -Clostridium Schaukel-Vektoren pMTL82151 und pJIR750 kloniert, die unterschiedliche Replikons für Gram-positive Bakterien enthalten. 3) Die Überexpression des synthetischen 2,3-BD-Operons alsS -budA -bdh mit Hilfe des Plasmids pJIR750_23BD führte zu einer 1,5-fachen 2,3-Butandiol-Steigerung (1,2 mM unter heterotrophen Wachstumsbedingungen bzw. 6,4 mM 2,3-BD unter autotrophen Wachstumsbedingungen) im Vergleich zu C. ljungdahlii WT (0,8 mM mit Fructose bzw. 4,3 mM mit Syngas als Energie- und Kohlenstoffquelle). Unter autotrophen Wachstumsbedingungen führte die Überexpression des synthetischen 2,3-BD-Operons alsS -budA -bdh mit Hilfe des Plasmids pMTL82151_23BD zu einer 1,8-fachen Steigerung der 2,3-BD-Produktion (3,1 mM) im Vergleich zu C. ljungdahlii WT (1,7 mM). 4) Die Überexpression des synthetischen 2,3-BD-Operons alsS -budA -bdh führte zu keiner Steigerung der 2,3-BD-Produktion in C. coskatii . 5) Es wurden drei modifizierte 2,3-BD Produktionsplasmide konstruiert. Das erste Plasmid weist eine kürzere Sequenz des budA -Gens auf, das zweite Plasmid entstand durch den Austausch von bdh mit adh (primäre-sekundäre Alkohol-Dehydrogenase) und das dritte

Plasmid enthält das budABC -Operon aus R. terrigena unter Kontrolle des Ppta -ack -Promoters

aus C. ljungdahlii und des Tsol -adc -Terminators aus C. acetobutylicum . 6) Heterotrophe Wachstumsversuche mit den modifizierten 2,3-BD-Produktionsplasmiden hatten keine Verbesserung der 2,3-BD-Produktion zur Folge. Jedoch führte die Überexpression von adh in C. ljungdahlii zu einer 2,9-fachen Steigerung der 2,3-BD- Konzentration (3,3 mM) im Vergleich zu C. ljungdahlii WT (1,7 mM). Sowohl das Entfernen 130 6. Zusammenfassung

der Rho-unabhängigen Terminatorstruktur, als auch die heterologe Überexpression des budABC -Operons aus R. terrigena konnten eine Steigerung der 2,3-BD Produktion im Vergleich zur Überexpression des alsS -budA -bdh-Operons herbeiführen.

7) Unter autotrophen Wachstumsbedingen (H 2 + CO 2) waren rekombinante A. woodii - Stämme nicht in der Lage, mit Hilfe der 2,3-BD-Produktionsplasmide heterolog 2,3-BD zu produzieren. 8) Das Gen für die Pyruvat:Ferredoxin-Oxidoreduktase und die Genregion zum Transfer über Konjugation ( traJ /oriT ) wurden erfolgreich in die Plasmide pMTL82151_23BD und pJIR750_23BD kloniert. Die Konjugation mit C. ljungdahlii war nur für die pMTL82151- basierten Plasmide erfolgreich. 9) Sowohl unter heterotrophen, als auch autotrophen Wachstumsbedingungen konnte durch die Überexpression von nifJ über das alsS -budA -nifJ -Operon eine 2,2-fache (2,2 mM) bzw. 2,1-fache (4.7 mM) Steigerung der 2,3-BD Konzentration im Vergleich zu C. ljungdahlii WT (1,0 mM bzw. 2,2 mM) erreicht werden. Die Überexpression von nifJ über das alsS -budA -nifJ -bdh -Operon führte zu keiner Verbesserung der 2,3-BD-Produktion im Vergleich zu C. ljungdahlii WT. 10) Der rekombinante Stamm C. ljungdahlii [pMTL82151] zeigte im Verlauf dieser Arbeit einen veränderten Phänotyp mit einer verringerten Acetat-Produktion zusammen mit einer gesteigerten Ethanol- und 2,3-BD-Produktion im Vergleich zu C. ljungdahlii WT. Mit Fructose als Substrat war die Produktion von 2,3-BD 2,4-fach (2,4 mM) gesteigert und mit Syngas 2,5-fach (5,6 mM) erhöht im Vergleich zu C. ljungdahlii WT (1,0 mM bzw. 2,2 mM 2,3-BD). Durch Sequenzierung des Genoms mit anschließender SNP-Analyse konnten mehrere SNPs im Chromosom des spontan mutierten Stamms C. ljungdahlii [pMTL82151] detektiert werden. Der entsprechende Stamm wurde als mutierter C. ljungdahlii [pMTL82151] beichnet. Der bedeutsamste SNP führte zu einem Stoppcodon und damit zu einem verkürzten AdhE1-Enzym, was der Grund für den beobachteten Phänotyp zu sein scheint. 11) Die Transkripte alsS , budA und bdh konnten in C. ljungdahlii WT über die Northern Blot- Analyse erfolgreich nachgewiesen werden. Das Transkript des synthetischen 2,3-BD- Operons konnte nur in C. ljungdahlii [pJIR750_23BD] erfolgreich detektiert werden. 12) Die Replikons pIP404 (C. perfringens ) und pCB102 ( C. butyricum ), als auch die traJ /oriT - Region konnten erfolgreich in das Butanol-Produktionsplasmid kloniert werden. In dem 6. Zusammenfassung 131

Plasmid pACYC184_MCljI wurde das Replikon p15A erfolgreich gegen das Replikon pSC101 ausgetauscht. Die Butanol-Produktionsplasmide pCE2 und pJR3 konnten weder über Elektroporation noch über Konjugation in C. ljungdahlii transformiert werden.

132 7. References

7. References

Abrini J, Naveau H, Nyns EJ . 1994. Clostridium autoethanogenum , sp. nov., an anaerobic bacterium that produces ethanol from carbon monoxide. Arch. Microbiol. 161 :345–351.

Aeschelmann F, Carus M . 2015. Biobased building blocks and polymers in the world: capacities, production, and applications–status quo and trends towards 2020. Ind. Biotechnol. 11 :154–159.

Al-Hinai MA, Fast AG, Papoutsakis ET. 2012. Novel system for efficient isolation of Clostridium double-crossover allelic exchange mutants enabling markerless chromosomal gene deletions and DNA integration. Appl. Environ. Microbiol. 78 :8112–8121.

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ . 1990. Basic local alignment search tool. J. Mol. Biol. 215 :403–410.

Alwine JC, Kemp DJ, Stark GR . 1977. Method for detection of specific RNAs in agarose gels by transfer to diazobenzyloxymethyl-paper and hybridization with DNA probes. Proc. Natl. Acad. Sci. USA 74 :5350–5354.

Ash C, Priest FG, Collins MD . 1993. Molecular identification of rRNA group 3 bacilli (Ash, Farrow, Wallbanks and Collins) using a PCR probe test. Proposal for the creation of a new genus Paenibacillus . Ant. Leeuwenhoek 64 :253–260.

Bai F, Dai L, Fan J, Truong N, Rao B, Zhang L, Shen Y . 2015. Engineered Serratia marcescens for efficient (3 R)-acetoin and (2 R,3 R)-2,3-butanediol production. J Ind. Microbiol. Biotechnol. 42 :779–786.

Banerjee A, Leang C, Ueki T, Nevin KP, Lovley DR 2014. Lactose-inducible system for metabolic engineering of Clostridium ljungdahlii . Appl. Environ. Microbiol. 80 :2410–2416.

Bannam TL, Rood JI . 1993. Clostridium perfringens -Escherichia coli shuttle vectors that carry single antibiotic resistance determinants. Plasmid 29 :233–235.

Barik S, Prieto S, Harrison SB, Clausen EC, Gaddy JL . 1988. Biological production of alcohols from coal through indirect liquefaction. Appl. Biochem. Biotechnol. 18 :363–378. 7. References 133

Bartowsky EJ, Henschke PA . 2004. The “buttery” attribute of wine-diacetyl-desirability, spoilage and beyond. Int. J. Food Microbiol. 96 :235–252.

Bengelsdorf FR, Straub M, Dürre P . 2013. Bacterial synthesis gas (syngas) fermentation. Environ. Technol. 34 :1639–1651.

Bengelsdorf FR, Poehlein A, Linder S, Erz C, Hummel T, Hoffmeister S, Daniel R, Dürre P . 2016. Industrial acetogenic biocatalysts: a comparative metabolic and genomic analysis. Front. Microbiol. 7:1036.

Berlyn MKB . 1998. Linkage map of Escherichia coli K-12: the traditional map. Microbiol. Mol. Biol. Rev. 62 :814–984.

Bertani G . 1951. Studies on lysogenesis I. The mode of phage liberation by lysogenic Escherichia coli . J. Bacteriol. 62 :293-300.

Bertsch J, Müller V . 2015. Bioenergetic constraints for conversion of syngas to biofuels in acetogenic bacteria. Biotechnol. Biofuels 8:210

Bertsch J, Öppinger C, Hess V, Langer JD, Müller V . 2015. Heterotrimeric NADH-oxidizing methylenetetrahydrofolate reductase from the acetogenic bacterium Acetobacterium woodii . J. Bacteriol. 197 :1681–1689.

Białkowska AM, Gromek E, Krysiak J, Sikora B, Kalinowska H, Jędrzejczak-Krzepkowska M, Kubik C, Lang S, Schütt F, Turkiewicz M . 2015. Application of enzymatic apple pomace hydrolysate to production of 2,3-butanediol by alkaliphilic Bacillus licheniformis NCIMB 8059. J. Ind. Microbiol. Biotechnol. 42 :1609–1621.

Białkowska AM, Jędrzejczak-Krzepkowska M, Gromek E, Krysiak J, Sikora B, Kalinowska H, Kubik C, Schütt F, Turkiewicz M . 2016. Effects of genetic modifications and fermentation conditions on 2,3-butanediol production by alkaliphilic Bacillus subtilis . Appl. Microbiol. Biotechnol. 100 :2663–2676.

Biegel E, Müller V . 2010. Bacterial Na +-translocating ferredoxin:NAD + oxidoreductase. Proc. Natl. Acad. Sci. USA 107 :18138–18142. 134 7. References

Biswas R, Yamaoka M, Nakayama H, Kondo T, Yoshida K, Bisaria VS, Kondo A . 2012. Enhanced production of 2,3-butanediol by engineered Bacillus subtilis . Appl. Microbiol. Biotechnol. 94 :651–658.

Blomqvist K, Nikkola M, Lehtovaara P, Suihko ML, Airaksinen U, Str aby KB, Knowles JK, Penttilä ME . 1993. Characterization of the genes of the 2,3-butanediol operons from Klebsiella terrigena and Enterobacter aerogenes . J. Bacteriol. 175 :1392–1404.

Bolger AM, Lohse M, Usadel B . 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30 :2114–2120.

Brown SD, Nagaraju S, Utturkar S, De Tissera S, Segovia S, Mitchell W, Land ML, Dassanayake A, Köpke Michael . 2014. Comparison of single-molecule sequencing and hybrid approaches for finishing the genome of Clostridium autoethanogenum and analysis of CRISPR systems in industrial relevant clostridia. Biotechnol. Biofuels 7:40.

Bruant G, Levesque MJ, Peter C, Guiot SR, Masson L . 2010. Genomic analysis of carbon monoxide utilization and butanol production by Clostridium carboxidivorans strain P7. PLoS ONE 5:e13033.

Buckel W, Thauer RK. 2013. Energy conservation via electron bifurcating ferredoxin reduction and proton/Na + translocating ferredoxin oxidation. Biochim. Biophys. Acta 1827 :94-113.

Celińska E, Grajek W . 2009. Biotechnological production of 2,3-butanediol - current state and prospects. Biotechnol. Adv. 27 :715–725.

Chakraborty S, Piszel PE, Hayes CE, Baker RT, Jones WD . 2015. Highly selective formation of n-butanol from ethanol through the Guerbet process: a tandem catalytic approach. J. Am. Chem. Soc. 137 :14264–14267.

Chan V, Dreolini LF, Flintoff KA, Lloyd SJ, Mattenley AA . 2002. The effect of increasing plasmid size on transformation efficiency in Escherichia coli . J. Exp. Microbiol. Immunol. 2:207–223.

Chen Q, Fischer JR, Benoit VM, Dufour NP, Youderian P, Leong JM . 2008. In vitro CpG methylation increases the transformation efficiency of Borrelia burgdorferi strains harboring the endogenous linear plasmid lp56. J. Bacteriol. 190 :7885–7891. 7. References 135

Cheng KK, Liu Q, Zhang JA, Li JP, Xu JM, Wang GH . 2010. Improved 2,3-butanediol production from corncob acid hydrolysate by fed-batch fermentation using Klebsiella oxytoca . Process Biochem. 45 :613–616.

Cho S, Kim T, Woo HM, Kim Yunje, Lee Jinwon, Um Y . 2015a. High production of 2,3- butanediol from biodiesel-derived crude glycerol by metabolically engineered Klebsiella oxytoca M1. Biotechnol. Biofuels 8:146.

Cho S, Kim T, Woo HM, Lee J, Kim Y, Um Y . 2015b. Enhanced 2,3-butanediol production by optimizing fermentation conditions and engineering Klebsiella oxytoca M1 through overexpression of acetoin reductase. PLoS ONE 10 :e0138109.

Chu H, Xin B, Liu P, Wang Y, Li L, Liu X, Zhang X, Ma C, Xu P, Gao C . 2015. Metabolic engineering of Escherichia coli for production of (2 S,3 S)-butane-2,3-diol from glucose. Biotechnol. Biofuels 8:143.

Cohen SN, Chang AC . 1977. Revised interpretation of the origin of the pSC101 plasmid. J. Bacteriol. 132 :734-737.

Cotter JL, Chinn MS, Grunden AM . 2009. Ethanol and acetate production by Clostridium ljungdahlii and Clostridium autoethanogenum using resting cells. Bioprocess Biosyst. Eng. 32 :369–380.

Dai JY, Zhao P, Cheng XL, Xiu ZL . 2015. Enhanced production of 2,3-butanediol from sugarcane molasses. Appl. Biochem. Biotechnol. 175:3014–3024.

Daniel SL, Hsu T, Dean SI, Drake HL . 1990. Characterization of the H 2- and CO-dependent chemolithotrophic potentials of the acetogens Clostridium thermoaceticum and Acetogenium kivui . J. Bacteriol. 172 :4464–4471.

Daniell J, Nagaraju S, Burton F, Köpke M, Simpson SD . 2016. Low-carbon fuel and chemical production by anaerobic gas fermentation. Adv. Biochem. Eng. Biotechnol. 156 :293-321.

De Hoon MJL, Makita Y, Nakai K, Miyano S . 2005. Prediction of transcriptional terminators in Bacillus subtilis and related species. PLoS Comput. Biol. 1:e25. 136 7. References

De Oliveira RR, Nicholson WL . 2016. Synthetic operon for ( R,R )-2,3-butanediol production in Bacillus subtilis and Escherichia coli . Appl. Microbiol. Biotechnol. 100 :719–728.

Debuissy T, Pollet E, Avérous L . 2016. Synthesis of potentially biobased copolyesters based on adipic acid and butanediols: kinetic study between 1,4- and 2,3-butanediol and their influence on crystallization and thermal properties. Polymer 99 :204–213.

Dominguez W, O’Sullivan DJ . 2013. Developing an efficient and reproducible conjugation- based gene transfer system for bifidobacteria. Microbiology 159 :328–338.

Dower WJ, Miller JF, Ragsdale CW . 1988. High efficiency transformation of E. coli by high voltage electroporation. Nucl. Acids Res. 16 :6127–6145.

Drake HL, Gößner AS, Daniel SL. 2008. Old acetogens, new light. Ann. N. Y. Acad. Sci. 1125 :100–128.

Drancourt M, Bollet C, Carta A, Rousselier P . 2001. Phylogenetic analyses of Klebsiella species delineate Klebsiella and Raoultella gen. nov., with description of Raoultella ornithinolytica comb. nov., Raoultella terrigena comb. nov. and Raoultella planticola comb. nov. Int. J. Syst. Evol. Microbiol. 51 :925–932.

Dürre P . 2005. Formation of solvents in clostridia. In: Dürre P (ed.), Handbook on clostridia, CRC Press, Boca Raton, FL, USA:671–693.

Dürre P . 2008. Fermentative butanol production. Ann. N. Y. Acad. Sci. 1125 :353–362.

Dürre P . 2016a. Gas fermentation - a biotechnological solution for today’s challenges. Microb. Biotechnol. 10 :14–16.

Dürre P. 2016b. Butanol formation from gaseous substrates. FEMS Microbiol. Lett. 363 :6.

Dürre P, Eikmanns BJ . 2015. C 1-carbon sources for chemical and fuel production by microbial gas fermentation. Curr. Opin. Biotechnol. 35 :63–72.

Extance J, Crennell SJ, Eley K, Cripps R, Hough DW, Danson MJ . 2013. Structure of a bifunctional alcohol dehydrogenase involved in bioethanol generation in Geobacillus thermoglucosidasius . Acta Crystallogr. D Biol. Crystallogr. 69 :2104–2115. 7. References 137

Faba L, Díaz E, Ordóñez S . 2016. Base-catalyzed reactions in biomass conversion: reaction mechanisms and catalyst deactivation. In: Schlaf M and Zhang ZC (eds.), Reaction pathways and mechanisms in thermocatalytic biomass conversion I: cellulose structure, depolymerization and conversion by heterogeneous catalysts, Springer Singapore, Singapore, Singapore:87–122.

Fast AG, Papoutsakis ET . 2012. Stoichiometric and energetic analyses of non-photosynthetic

CO 2-fixation pathways to support synthetic biology strategies for production of fuels and chemicals. Curr. Opin. Chem. Eng. 1:380–395.

Faveri DD, Torre P, Molinari F, Perego P, Converti A . 2003. Carbon material balances and bioenergetics of 2,3-butanediol bio-oxidation by Acetobacter hansenii . Enzyme Microb. Technol. 33 :708–719.

Fontaine FE, Peterson WH, McCoy E, Johnson MJ, Ritter GJ . 1942. A new type of glucose fermentation by Clostridium thermoaceticum . J. Bacteriol. 43 :701–715.

Formanek J, Mackie R, Blaschek HP . 1997. Enhanced butanol production by Clostridium beijerinckii BA101 grown in semidefined P2 medium containing 6 percent maltodextrin or glucose. Appl. Environ. Microbiol. 63 :2306–2310.

Fu J, Wang Z, Chen T, Liu W, Shi T, Wang G, Tang Y, Zhao X . 2014. NADH plays the vital role for chiral pure D-(−)-2,3-butanediol production in Bacillus subtilis under limited oxygen conditions. Biotechnol. Bioeng. 111 :2126–2131.

Fuchs G. 2007. Acetogenese: CO 2 als Elektronenakzeptor. In: Fuchs G (ed.), Allgemeine Mikrobiologie, 8 th edition, Georg Thieme Verlag, Stuttgart, Germany:400-402.

Fuchs G . 2011. Alternative pathways of carbon dioxide fixation: insights into the early evolution of life? Annu. Rev. Microbiol. 65 :631–658.

Fulmer EI, Christensen LM, Kendali AR . 1933. Production of 2,3-butylene glycol by fermentation. Ind. Eng. Chem. 25 :798–800. 138 7. References

Gao J, Xu H, Li Q, Feng X, Li S . 2010. Optimization of medium for one-step fermentation of inulin extract from Jerusalem artichoke tubers using Paenibacillus polymyxa ZJ-9 to produce R,R-2,3-butanediol. Bioresour. Technol. 101 :7087–7093.

Garg SK, Jain A . 1995. Fermentative production of 2,3-butanediol: a review. Bioresour. Technol. 51 :103–109.

Gibson DG, Young L, Chuang R-Y, Venter JC, Hutchison CA, Smith HO . 2009. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods 6:343–345.

Gräfje H, Körnig W, Weitz HM, Reiß W, Steffan G, Diehl H, Bosche H, Schneider K, Kieczka H . 2000. Butanediols, butenediol, and butynediol. In: Bailey JE, Brinker CJ, Cornils B (eds.), Ullmann’s encyclopedia of industrial chemistry, 6 th edition Wiley-VCH, Weinheim, Germany:407-414.

Green MR, Sambrook J . 2012. Molecular cloning: a laboratory manual. 4 th edition, Cold Spring Harbor Laboratory Press, New York, NY, USA.

Grethlein AJ, Worden RM, Jain MK, Datta R . 1991. Evidence for production of n-butanol from carbon monoxide by Butyribacterium methylotrophicum . J. Ferment. Bioeng. 72 :58–60.

Gubbels E, Jasinska-Walc L, Koning CE . 2013. Synthesis and characterization of novel renewable polyesters based on 2,5-furandicarboxylic acid and 2,3-butanediol. J. Polymer Sci. A Polymer Chem. 51 :890–898.

Guo X, Cao C, Wang Y, Li C, Wu M, Chen Y, Zhang C, Pei H, Xiao D . 2014a. Effect of the inactivation of lactate dehydrogenase, ethanol dehydrogenase, and phosphotransacetylase on 2,3-butanediol production in Klebsiella pneumoniae strain. Biotechnol. Biofuels 7:1-11.

Guo XW, Zhang YH, Cao CH, Shen T, Wu MY, Chen YF, Zhang CY, Xiao DG. 2014b. Enhanced production of 2,3-butanediol by overexpressing acetolactate synthase and acetoin reductase in Klebsiella pneumoniae . Biotechnol. Appl. Biochem. 61 :707–715.

Hahn HD, Dämbkes G, Rupprich N, Bahl H, Frey GD . 2000. Butanols. In: Bailey JE, Brinker CJ, Cornils B (eds.), Ullmann’s encyclopedia of industrial chemistry, 6 th edition, Wiley-VCH, Weinheim, Germany:417-428. 7. References 139

Hanahan D . 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166 :557–580.

Harden A, Walpole GS . 1906. Chemical action of Bacillus lactis aerogenes (Escherich ) on glucose and mannitol: production of 2:3-butyleneglycol and acetylmethylcarbinol. Proc. R. Soc. Lond. B Biol. Sci. 77 :399-405.

Häßler T, Schieder D, Pfaller R, Faulstich M, Sieber V . 2012. Enhanced fed-batch fermentation of 2,3-butanediol by Paenibacillus polymyxa DSM 365. Bioresour. Technol. 124 :237–244.

Haveren J van, Scott EL, Sanders J . 2008. Bulk chemicals from biomass. Biofuels, Bioprod. Bioref. 2:41–57.

Heap JT, Pennington OJ, Cartman ST, Carter GP, Minton NP . 2007. The ClosTron: a universal gene knock-out system for the genus Clostridium . J. Microbiol. Methods 70 :452–464.

Heap JT, Pennington OJ, Cartman ST, Minton NP . 2009. A modular system for Clostridium shuttle plasmids. J. Microbiol. Methods 78 :79–85.

Hess V, Oyrik O, Trifunović D, Müller V . 2015. 2,3-Butanediol metabolism in the acetogen Acetobacterium woodii . Appl. Environ. Microbiol. 81 :4711–4719.

Hess V, Gallegos R, Jones JA, Barquera B, Malamy MH, Müller V . 2016. Occurrence of ferredoxin:NAD + oxidoreductase activity and its ion specificity in several Gram-positive and Gram-negative bacteria. PeerJ. 4:e1515.

Hicks JK, Yu JH, Keller NP, Adams TH . 1997. Aspergillus sporulation and mycotoxin production both require inactivation of the FadA Gα protein-dependent signaling pathway. EMBO J. 16 :4916–4923.

Hoffmeister, S. 2017. Production of platform chemicals with acetogenic bacteria. Ph. D. Thesis. Ulm University, Ulm, Germany.

Hoffmeister S, Gerdom M, Bengelsdorf FR, Linder S, Flüchter S, Öztürk H, Blümke W, May A, Fischer RJ, Bahl H, Dürre P . 2016. Acetone production with metabolically engineered strains of Acetobacterium woodii . Metab. Eng. 36 :37–47. 140 7. References

Hu X, Shen X, Huang M, Liu C, Geng Y, Wang R, Xu R, Qiao H, Zhang Liqun . 2016. Biodegradable unsaturated polyesters containing 2,3-butanediol for engineering applications: synthesis, characterization and performances. Polymer 84 :343–354.

Humphreys CM, McLean S, Schatschneider S, Millat T, Henstra AM, Annan FJ, Breitkopf R, Pander B, Piatek P, Rowe P, Wichlacz AT, Woods C, Norman R, Blom J, Goesman A, Hodgman C, Barrett D, Thomas NR, Winzer K, Minton NP . 2015. Whole genome sequence and manual annotation of Clostridium autoethanogenum , an industrially relevant bacterium. BMC Genomics 16 :1085.

Imkamp F, Biegel E, Jayamani E, Buckel W, Müller V . 2007. Dissection of the caffeate respiratory chain in the acetogen Acetobacterium woodii : identification of an Rnf-type NADH dehydrogenase as a potential coupling site. J. Bacteriol. 189 :8145–8153.

Inoue H, Nojima H, Okayama H . 1990. High efficiency transformation of Escherichia coli with plasmids. Gene 96 :23–28.

Isaacs FJ, Carr PA, Wang HH, Lajoie MJ, Sterling B, Kraal L, Tolonen AC, Gianoulis TA, Goodman DB, Reppas NB, Emig CJ, Bang D, Hwang SJ, Jewett MC, Jacobson JM, Church GM . 2011. Precise manipulation of chromosomes in vivo enables genome-wide codon replacement. Science 333 :348-353.

Izard D, Ferragut C, Gavini F, Kersters K, De Ley J, Leclerc H . 1981. Klebsiella terrigena , a new species from soil and water. Int. J. Syst. Evol. Microbiol. 31 :116–127.

Jantama K, Polyiam P, Khunnonkwao P, Chan S, Sangproo M, Khor K, Jantama SS, Kanchanatawee S . 2015. Efficient reduction of the formation of by-products and improvement of production yield of 2,3-butanediol by a combined deletion of alcohol dehydrogenase, acetate kinase-phosphotransacetylase, and lactate dehydrogenase genes in metabolically engineered Klebsiella oxytoca in mineral salts medium. Metab. Eng. 30 :16–26.

Ji XJ, Huang H, Du J, Zhu JG, Ren LJ, Hu N, Li S . 2009. Enhanced 2,3-butanediol production by Klebsiella oxytoca using a two-stage agitation speed control strategy. Bioresour. Technol. 100 :3410–3414. 7. References 141

Ji XJ, Huang H, Zhu JG, Ren LJ, Nie ZK, Du J, Li S . 2010. Engineering Klebsiella oxytoca for efficient 2,3-butanediol production through insertional inactivation of acetaldehyde dehydrogenase gene. Appl. Microbiol. Biotechnol. 85 :1751–1758.

Ji XJ, Huang H, Ouyang PK . 2011a. Microbial 2,3-butanediol production: a state-of-the-art review. Biotechnol. Adv. 29 :351–364.

Ji XJ, Nie ZK, Huang H, Ren LJ, Peng C, Ouyang PK . 2011b. Elimination of carbon catabolite repression in Klebsiella oxytoca for efficient 2,3-butanediol production from glucose-xylose mixtures. Appl. Microbiol. Biotechnol. 89 :1119–1125.

Ji XJ, Huang H, Nie ZK, Qu L, Xu Q, Tsao GT . 2012. Fuels and chemicals from hemicellulose sugars. Adv. Biochem. Eng. Biotechnol. 128 :199–224.

Ji XJ, Xia ZF, Fu NH, Nie ZK, Shen MQ, Tian QQ, Huang H . 2013. Cofactor engineering through heterologous expression of an NADH oxidase and its impact on metabolic flux redistribution in Klebsiella pneumoniae . Biotechnol. Biofuels 6:7.

Ji XJ, Liu LG, Shen MQ, Nie ZK, Tong YJ, Huang H . 2015. Constructing a synthetic metabolic pathway in Escherichia coli to produce the enantiomerically pure (R,R )-2,3-butanediol. Biotechnol. Bioeng. 112 :1056–1059.

Jones DT, Woods DR . 1986. Acetone-butanol fermentation revisited. Microbiol. Rev. 50 :484– 524.

Jung MY, Jung HM, Lee J, Oh MK . 2015. Alleviation of carbon catabolite repression in Enterobacter aerogenes for efficient utilization of sugarcane molasses for 2,3-butanediol production. Biotechnol. Biofuels 8:106.

Jung MY, Mazumdar S, Shin SH, Yang KS, Lee J, Oh MK. 2014. Improvement of 2,3-butanediol yield in Klebsiella pneumoniae by deletion of the pyruvate formate-lyase gene. Appl. Environ. Microbiol. 80 :6195–6203.

Jung MY, Ng CY, Song H, Lee J, Oh MK . 2012. Deletion of lactate dehydrogenase in Enterobacter aerogenes to enhance 2,3-butanediol production. Appl. Microbiol. Biotechnol. 95 :461–469. 142 7. References

Jurchescu IM, Hamann J, Zhou X, Ortmann T, Kuenz A, Prusse U, Lang S . 2013. Enhanced 2,3- butanediol production in fed-batch cultures of free and immobilized Bacillus licheniformis DSM8785. Appl. Microbiol. Biotechnol. 97 :6715–6723.

Kallbach M, Horn S, Kuenz A, Prüße U . 2017. Screening of novel bacteria for the 2,3- butanediol production. Appl. Microbiol. Biotechnol. 101 :1025–1033.

Kay JE, Jewett MC . 2015. Lysate of engineered Escherichia coli supports high-level conversion of glucose to 2,3-butanediol. Metab. Eng. 32 :133–142.

Kim B, Lee S, Jeong D, Yang J, Oh MK, Lee J . 2014a. Redistribution of carbon flux toward 2,3- butanediol production in Klebsiella pneumoniae by metabolic engineering. PLoS ONE 9:e105322.

Kim B, Lee S, Park J, Lu M, Oh M, Kim Y, Lee J . 2012. Enhanced 2,3-butanediol production in recombinant Klebsiella pneumoniae via overexpression of synthesis-related genes. J. Microbiol. Biotechnol. 22 :1258–1263.

Kim DK, Rathnasingh C, Song H, Lee HJ, Seung D, Chang YK . 2013a. Metabolic engineering of a novel Klebsiella oxytoca strain for enhanced 2,3-butanediol production. J. Biosci. Bioeng. 116 :186–192.

Kim JW, Seo SO, Zhang GC, Jin YS, Seo JH . 2015. Expression of Lactococcus lactis NADH oxidase increases 2,3-butanediol production in Pdc-deficient Saccharomyces cerevisiae . Bioresour. Technol. 191 :512–519.

Kim SJ, Seo SO, Jin YS, Seo JH . 2013b. Production of 2,3-butanediol by engineered Saccharomyces cerevisiae . Bioresour. Technol. 146 :274–281.

Kim SJ, Seo SO, Park YC, Jin YS, Seo JH . 2014b. Production of 2,3-butanediol from xylose by engineered Saccharomyces cerevisiae . J. Biotechnol. 192 :376–382.

Koboldt DC, Zhang Q, Larson DE, Shen D, McLellan MD, Lin L, Miller CA, Mardis ER, Ding L, Wilson RK . 2012. VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res. 22 :568–576. 7. References 143

Koda K, Matsu-ura T, Obora Y, Ishii Y . 2009. Guerbet reaction of ethanol to n-butanol catalyzed by iridium complexes. Chem. Lett. 38 :838–839.

Köpke M, Chen WY. 2013. Recombinant microorganisms and uses therefor. US 2013/0323806 A1. Washington, D.C.: U.S. Patent and Trademark Office.

Köpke M, Gerth ML, Maddock DJ, Mueller AP, Liew F, Simpson SD, Patrick WM. 2014. Reconstruction of an acetogenic 2,3-butanediol pathway involving a novel NADPH-dependent primary-secondary alcohol dehydrogenase. Appl. Environ. Microbiol. 80 :3394–3403.

Köpke M, Havill A. 2014. LanzaTech’s route to bio-butadiene. Catal. Rev. 27 :7–12.

Köpke M, Held C, Hujer S, Liesegang H, Wiezer A, Wollherr A, Ehrenreich A, Liebl W, Gottschalk G, Dürre P. 2010. Clostridium ljungdahlii represents a microbial production platform based on syngas. Proc. Natl. Acad. Sci. USA 107 :13087–13092.

Köpke M, Liew F. 2012. Butanol production from carbon monoxide by a recombinant microorganism. WO2012053905 A1. World patent.

Köpke M, Mihalcea C, Bromley JC, Simpson SD . 2011a. Fermentative production of ethanol from carbon monoxide. Curr. Opin. Biotechnol. 22 :320–325.

Köpke M, Mihalcea C, Liew F, Tizard JH, Ali MS, Conolly JJ, Al-Sinawi B, Simpson SD. 2011b. 2,3-Butanediol production by acetogenic bacteria, an alternative route to chemical synthesis, using industrial waste gas. Appl. Environ. Microbiol. 77 :5467–5475.

Köpke M, Mueller AP. 2016. Recombinant microorganisms exhibiting increased flux through a fermentation pathway. US 2016/0160223 A1. Washington, D.C.: U.S. Patent and Trademark Office.

Kozlowski JT, Davis RJ . 2013. Heterogeneous catalysts for the Guerbet coupling of alcohols. ACS Catal. 3:1588–1600.

Kuhz H, Kuenz A, Prüße U, Willke T, Vorlop K-D. 2017. Products components: alcohols. Adv. Biochem. Biotechnol. doi:10.1007/10_2016_74. 144 7. References

Küsel K, Drake H . 2005. Acetogenic clostridia. In: Dürre P (ed.), Handbook on clostridia, CRC Press, Boca Raton, FL, USA:719–746.

Langmead B, Salzberg SL . 2012. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9:357–359.

Leang C, Ueki T, Nevin KP, Lovley DR. 2013. A genetic system for Clostridium ljungdahlii : a chassis for autotrophic production of biocommodities and a model homoacetogen. Appl. Environ. Microbiol. 79 :1102–1109.

Lechner M, Findeiß S, Steiner L, Marz M, Stadler PF, Prohaska SJ . 2011. Proteinortho: detection of (Co-)orthologs in large-scale analysis. BMC Bioinformatics 12 :124.

Lederberg J, Tatum EL . 1946. Gene recombination in Escherichia coli . Nature 158 :558.

Lee S, Kim B, Park K, Um Y, Lee J . 2012. Synthesis of pure meso -2,3-butanediol from crude glycerol using an engineered metabolic pathway in Escherichia coli . Appl. Biochem. Biotechnol. 166 :1801–1813.

Lee S, Kim B, Yang J, Jeong D, Park S, Lee J . 2015. A non-pathogenic and optically high concentrated ( R,R )-2,3-butanediol biosynthesizing Klebsiella strain . J. Biotechnol. 209 :7–13.

Lee SJ, Thapa LP, Lee JH, Choi HS, Kim SB, Park C, Kim SW . 2016. Stimulation of 2,3-butanediol production by upregulation of alsR gene transcription level with acetate addition in Enterobacter aerogenes ATCC 29007. Process Biochem. 51 :1904–1910.

Lehrach H, Diamond D, Wozney JM, Boedtker H . 1977. RNA molecular weight determinations by gel electrophoresis under denaturing conditions, a critical reexamination. Biochemistry 16 :4743–4751.

Li D, Dai JY, Xiu ZL . 2010a. A novel strategy for integrated utilization of Jerusalem artichoke stalk and tuber for production of 2,3-butanediol by Klebsiella pneumoniae . Bioresour. Technol. 101 :8342–8347. 7. References 145

Li F, Hinderberger J, Seedorf H, Zhang J, Buckel W, Thauer RK. 2008. Coupled ferredoxin and crotonyl coenzyme A (CoA) reduction with NADH catalyzed by the butyryl-CoA dehydrogenase/Etf complex from Clostridium kluyveri . J. Bacteriol. 190 :843–850.

Li H . 2011. A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data. Bioinformatics 27 :2987–2993.

Li J, Baral NR, Jha AK . 2014a. Acetone–butanol–ethanol fermentation of corn stover by Clostridium species: present status and future perspectives. World J. Microbiol. Biotechnol. 30 :1145–1157.

Li L, Chen C, Li K, Wang Y, Gao C, Ma C, Xu P . 2014c. Efficient simultaneous saccharification and fermentation of inulin to 2,3-butanediol by thermophilic Bacillus licheniformis ATCC 14580. Appl. Environ. Microbiol. 80 :6458–6464.

Li L, Li K, Wang K, Chen C, Gao C, Ma C, Xu P . 2014b. Efficient production of 2,3-butanediol from corn stover hydrolysate by using a thermophilic Bacillus licheniformis strain. Bioresour. Technol. 170 :256–261.

Li L, Li K, Wang Y, Chen C, Xu Y, Zhang L, Han B, Gao C, Tao F, Ma C, Xu P . 2015a. Metabolic engineering of Enterobacter cloacae for high-yield production of enantiopure (2 R,3 R)-2,3- butanediol from lignocellulose-derived sugars. Metabolic Engineering 28 :19–27.

Li L, Zhang L, Li K, Wang Y, Gao C, Han B, Ma C, Xu P . 2013. A newly isolated Bacillus licheniformis strain thermophilically produces 2,3-butanediol, a platform and fuel bio- chemical. Biotechnol. Biofuels 6:123.

Li N, Yang J, Chai C, Yang S, Jiang W, Gu Y . 2015b. Complete genome sequence of Clostridium carboxidivorans P7T, a syngas-fermenting bacterium capable of producing long-chain alcohols. J. Biotechnol. 211 :44–45.

Li ZJ, Jian J, Wei XX, Shen XW, Chen GQ . 2010b. Microbial production of meso -2,3-butanediol by metabolically engineered Escherichia coli under low oxygen condition. Appl. Microbiol. Biotechnol. 87 :2001–2009. 146 7. References

Lian J, Chao R, Zhao H . 2014. Metabolic engineering of a Saccharomyces cerevisiae strain capable of simultaneously utilizing glucose and galactose to produce enantiopure ( 2R,3R )- butanediol. Metab. Eng. 23 :92–99.

Liew F, Henstra AM, Kӧpke M, Winzer K, Simpson SD, Minton NP . 2017. Metabolic engineering of Clostridium autoethanogenum for selective alcohol production. Metab. Eng. 40 :104–114.

Liew F, Martin ME, Tappel RC, Heijstra BD, Mihalcea C, Köpke M . 2016. Gas fermentation – a flexible platform for commercial scale production of low-carbon-fuels and chemicals from waste and renewable feedstocks. Front. Microbiol. 7: 694.

Liou JSC, Balkwill DL, Drake GR, Tanner RS . 2005. Clostridium carboxidivorans sp. nov., a solvent-producing clostridium isolated from an agricultural settling lagoon, and reclassification of the acetogen Clostridium scatologenes strain SL1 as Clostridium drakei sp. nov. Int. J. of Syst. Evol. Microbiol. 55 :2085–2091.

Liu Z, Qin J, Gao C, Hua D, Ma C, Li L, Wang Y, Xu P . 2011. Production of (2 S,3 S)-2,3-butanediol and (3 S)-acetoin from glucose using resting cells of Klebsiella pneumonia and Bacillus subtilis . Bioresour. Technol. 102 :10741–10744.

Ljungdahl LG . 1986. The autotrophic pathway of acetate synthesis in acetogenic bacteria. Annu. Rev. Microbiol. 40 :415–450.

Lodish H, Berk A, Baltimore D, Matsudaira P, Zipursky SL, Darnell J . 2004. Overview of membrane transport. In: Lodish H, Berk A, Matsudaira P, Kaiser C, Krieger M, Zipursky SL, Darnell J(eds.), Molecular cell biology, 5 th edition, W.H. Freeman and Co., New York, NY, USA:245-251.

Lovley DR, Nevin KP . 2013. Electrobiocommodities: powering microbial production of fuels and commodity chemicals from carbon dioxide with electricity. Curr. Opin. Biotechnol. 24 :385–390.

Lovley DR. 2011. Powering microbes with electricity: direct electron transfer from electrodes to microbes. Environ. Microbiol. Rep. 3:27–35. 7. References 147

Lütke-Eversloh T, Bahl H . 2011. Metabolic engineering of Clostridium acetobutylicum : recent advances to improve butanol production. Curr. Opin. Biotechnol. 22 :634–647.

Ma C, Wang A, Qin J, Li L, Ai X, Jiang T, Tang H, Xu P . 2009. Enhanced 2,3-butanediol production by Klebsiella pneumoniae SDM. Appl. Microbiol. Biotechnol. 82 :49–57.

Magee RJ, Kosaric N . 1987. The microbial production of 2,3-butanediol. Adv. Appl. Microbiol. 32 :89–161.

Marcellin E, Behrendorff JB, Nagaraju S, DeTissera S, Segovia S, Palfreyman RW, Daniell J, Licona-Cassani C, Quek L, Speight R, Hodson MP, Simpson SD, Mitchell WP, Köpke M, Nielsen LK. 2016. Low carbon fuels and commodity chemicals from waste gases – systematic approach to understand energy metabolism in a model acetogen. Green Chem. 18 :3020–3028.

Martin ME, Richter H, Saha S, Angenent LT. 2016. Traits of selected Clostridium strains for syngas fermentation to ethanol. Biotechnol. Bioeng. 113 :531–539.

Mazumdar S, Lee J, Oh MK . 2013. Microbial production of 2,3 butanediol from seaweed hydrolysate using metabolically engineered Escherichia coli . Bioresour. Technol. 136 :329–336.

Membrillo-Hernandez J, Echave P, Cabiscol E, Tamarit J, Ros J, Lin EC . 2000. Evolution of the adhE gene product of Escherichia coli from a functional reductase to a dehydrogenase. Genetic and biochemical studies of the mutant proteins. J. Biol. Chem. 275 :33869–33875.

Mermelstein LD, Papoutsakis ET. 1993. In vivo methylation in Escherichia coli by the Bacillus subtilis phage ɸ3T I methyltransferase to protect plasmids from restriction upon transformation of Clostridium acetobutylicum ATCC 824. Appl. Environ. Microbiol. 59 :1077– 1081.

Minton NP, Ehsaan M, Humphreys CM, Little GT, Baker J, Henstra AM, Liew F, Kelly ML, Sheng L, Schwarz K, Zhang Y . 2016. A roadmap for gene system development in Clostridium . Anaerobe 41 :104–112.

Mock J, Zheng Y, Mueller AP, Ly S, Tran L, Segovia S, Nagaraju S, Köpke M, Dürre P, Thauer

RK. 2015. Energy conservation associated with ethanol formation from H 2 and CO 2 in Clostridium autoethanogenum involving electron bifurcation. J. Bacteriol. 197 :2965–2980. 148 7. References

Müller V, Imkamp F, Biegel E, Schmidt S, Dilling S . 2008. Discovery of a ferredoxin:NAD +- oxidoreductase (Rnf) in Acetobacterium woodii : a novel potential coupling site in acetogens. Ann. N. Y. Acad. Sci. 1125 :137–146.

Mullis K, Faloona F, Scharf S, Saiki R, Horn G, Erlich H . 1986. Specific enzymatic amplification of DNA in vitro : the polymerase chain reaction. Cold Spring Harb. Symp. Quant. Biol. 51 :263– 273.

Myszkowski J, Zielinski AZ. 1965. Synthesis of glycerol monochlorohydrins from allyl alcohol. Przem. Chem. 44 :249-252.

Nakashima N, Akita H, Hoshino T . 2014. Establishment of a novel gene expression method, BICES (biomass-inducible chromosome-based expression system), and its application to the production of 2,3-butanediol and acetoin. Metab. Eng. 25 :204–214.

Nakashimada Y, Marwoto B, Kashiwamura T, Kakizono T, Nishio N . 2000. Enhanced 2,3- butanediol production by addition of acetic acid in Paenibacillus polymyxa . J. Biosci. Bioeng. 90 :661–664.

Napora-Wijata K, Strohmeier GA, Winkler M . 2014. Biocatalytic reduction of carboxylic acids. Biotechnol. J. 9:822–843.

Ndou A, Plint N, Coville N. 2003. Dimerisation of ethanol to butanol over solid-base catalysts. Appl. Catal. A 251 :337–345.

Nevin KP, Woodard TL, Franks AE, Summers ZM, Lovley DR. 2010. Microbial electrosynthesis: feeding microbes electricity to convert carbon dioxide and water to multicarbon extracellular organic compounds. mBio 1:e00103–10–e00103–10.

Nevin KP, Hensley SA, Franks AE, Summers ZM, Ou J, Woodard TL, Snoeyenbos-West OL, Lovley DR . 2011. Electrosynthesis of organic compounds from carbon dioxide is catalyzed by a diversity of acetogenic microorganisms. Appl. Environ. Microbiol. 77 :2882–2886.

Ng CY, Jung MY, Lee J, Oh MK . 2012. Production of 2,3-butanediol in Saccharomyces cerevisiae by in silico aided metabolic engineering. Microb. Cell Fact. 11 :68. 7. References 149

Nie ZK, Ji XJ, Huang H, Du J, Li ZY, Qu L, Zhang Q, Ouyang PK . 2011. An effective and simplified fed-batch strategy for improved 2,3-butanediol production by Klebsiella oxytoca . Appl. Biochem. Biotechnol. 163 :946–953.

Nielsen DR, Yoon SH, Yuan CJ, Prather KLJ . 2010. Metabolic engineering of acetoin and meso - 2,3-butanediol biosynthesis in E. coli . Biotechnol. J. 5:274–284.

Novick RP . 1987. Plasmid incompatibility. Microbiol. Rev. 51 :381-395.

O’Lenick AJ . 2001. Guerbet chemistry. J. Surfactants Deterg. 4:311–315.

Ohse M, Takahashi K, Kadowaki Y, Kusaoke H . 1995. Effects of plasmid DNA sizes and several other factors on transformation of Bacillus subtilis ISW1214 with plasmid DNA by electroporation. Biosci. Biotechnol. Biochem. 59 :1433–1437.

Oliver JWK, Machado IMP, Yoneda H, Atsumi S. 2013. Cyanobacterial conversion of carbon dioxide to 2,3-butanediol. Proc. Natl. Acad. Sci. USA 110 :1249–1254.

Olson DG, McBride JE, Joe Shaw A, Lynd LR . 2012. Recent progress in consolidated bioprocessing. Curr. Opin. Biotechnol. 23 :396–405.

Park JM, Rathnasingh C, Song H . 2015. Enhanced production of ( R,R )‑2,3 ‑butanediol by metabolically engineered Klebsiella oxytoca . J. Ind. Microbiol. Biotechnol. 42 :1419–1425.

Park JM, Song H, Lee HJ, Seung D . 2013. Genome-scale reconstruction and in silico analysis of Klebsiella oxytoca for 2,3-butanediol production. Microb. Cell Fact. 12 :20.

Pasteur L. 1862. Quelques résultats nouveaux relatifs aux fermentations acétique et butyrique. Bull. Soc. Chim. Paris, May 1862 :52–53.

Peterson J, Phillips GJ . 2008. New pSC101-derivative cloning vectors with elevated copy numbers. Plasmid 59:193–201.

Petrini P, De Ponti S, Fare S, Tanzi MC . 1999. Polyurethane-maleamides for cardiovascular applications: synthesis and properties. J. Mater. Sci. Mater. Med. 10 :711–714. 150 7. References

Petrov K, Petrova P . 2009. High production of 2,3-butanediol from glycerol by Klebsiella pneumoniae G31. Appl. Microbiol. Biotechnol. 84 :659–656.

Petrov K, Petrova P . 2010. Enhanced production of 2,3-butanediol from glycerol by forced pH fluctuations. Appl. Microbiol. Biotechnol. 87 :943–949.

Phillips JR, Atiyeh HK, Tanner RS, Torres JR, Saxena J, Wilkins MR, Huhnke RL . 2015. Butanol and hexanol production in Clostridium carboxidivorans syngas fermentation: medium development and culture techniques. Bioresour. Technol. 190 :114–121.

Playne MJ . 1985. Determination of ethanol, volatile fatty acids, lactic and succinic acids in fermentation liquids by gas chromatography. J. Sci. Food Agric. 36 :638–644.

Poehlein A, Schmidt S, Kaster A-K, Goenrich M, Vollmers J, Thürmer A, Bertsch J, Schuchmann K, Voigt B, Hecker M, Daniel R, Thauer RK, Gottschalk G, Müller V . 2012. An ancient pathway combining carbon dioxide fixation with the generation and utilization of a sodium ion gradient for ATP synthesis. PLoS ONE 7:e33439.

Purdy D, O’keeffe TA, Elmore M, Herbert M, McLeod A, Bokori-Brown M, Ostrowski A, Minton NP . 2002. Conjugative transfer of clostridial shuttle vectors from Escherichia coli to Clostridium difficile through circumvention of the restriction barrier. Mol. Microbiol. 46 :439– 452.

Quan J, Tian J . 2011. Circular polymerase extension cloning for high-throughput cloning of complex and combinatorial DNA libraries. Nat. Protoc. 6:242–251.

Radoš D, Carvalho AL, Wieschalka S, Neves AR, Blombach B, Eikmanns BJ, Santos H. 2016. Engineering Corynebacterium glutamicum for the production of 2,3-butanediol. Microb. Cell Fact. 14 :171.

Ragsdale SW . 2007. Nickel and the carbon cycle. J. Inorg. Biochem. 101 :1657–1666.

Ragsdale SW, Pierce E . 2008. Acetogenesis and the Wood-Ljungdahl pathway of CO 2 fixation. Biochim. Biophys. Acta. 1784 :1873–1898. 7. References 151

Rajagopalan S, Datar RP, Lewis RS . 2002. Formation of ethanol from carbon monoxide via a new microbial catalyst. Biomass Bioenergy 23 :487–493.

Ramió-Pujol S, Ganigue R, Baneras L, Colprim J . 2015. Incubation at 25 °C prevents acid crash and enhances alcohol production in Clostridium carboxidivorans P7. Bioresour. Technol. 192 :296–303.

Regalbuto JR, Almalki F, Liu Q, Banerjee R, Wong A, Keels J . 2017. Hydrocarbon fuels from lignocellulose . In: Murad S, Baydoun E, Daghir N (eds.), Water, energy & food sustainability in the Middle East, Springer International Publishing AG, Cham, Switzerland:127–159.

Reidlinger J, Müller V. 1994. Purification of ATP synthase from Acetobacterium woodii and

+ identification as a Na -translocating F 1F0-type enzyme. Eur. J. Biochem. 223 :275–283.

Richter H, Molitor B, Wei H, Chen W, Aristilde L, Angenent LT. 2016. Ethanol production in syngas-fermenting Clostridium ljungdahlii is controlled by thermodynamics rather than by enzyme expression. Energy Environ. Sci. 9:2392–2399.

Ripoll V, De Vicente G, Morán B, Rojas A, Segarra S, Montesinos A, Tortajada M, Ramón D, Ladero M, Santos VE . 2016. Novel biocatalysts for glycerol conversion into 2,3-butanediol. Process Biochem. 51 :740–748.

Russell MJ, Martin W . 2004. The rocky roots of the acetyl-CoA pathway. Trends Biochem. Sci. 29 :358–363.

Savakis PE, Angermayr SA, Hellingwerf KJ . 2013. Synthesis of 2,3-butanediol by Synechocystis sp. PCC6803 via heterologous expression of a catabolic pathway from lactic acid- and enterobacteria. Metab. Eng. 20 :121–130.

Schiel-Bengelsdorf B, Dürre P . 2012. Pathway engineering and synthetic biology using acetogens. FEBS Letters 586 :2191–2198.

Schochetman G, Ou CY, Jones WK . 1988. Polymerase chain reaction. J. Infec. Dis. 158 :1154– 1157. 152 7. References

Schuchmann K, Müller V . 2012. A bacterial electron-bifurcating hydrogenase. J. Biol. Chem. 287 :31165-31171.

Schuchmann K, Müller V. 2014. Autotrophy at the thermodynamic limit of life: a model for energy conservation in acetogenic bacteria. Nat. Rev. Microbiol. 12 :809-821.

Schuchmann K, Müller V . 2016. Energetics and application of heterotrophy in acetogenic bacteria. Appl. Environ. Microbiol. 82 :4056–4069.

Schuster S . 2011. Expression von identifizierten clostridiellen Butanol-Synthese-Genen im Zwischenwirt Escherichia coli . Ph. D. Thesis. Ulm University, Germany.

Schweizer H . 2008. Bacterial genetics: past achievements, present state of the field, and future challenges. Biotechniques 44 :633–641.

Shen X, Lin Y, Jain R, Yuan Q, Yan Y . 2012. Inhibition of acetate accumulation leads to enhanced production of ( R,R )-2,3-butanediol from glycerol in Escherichia coli . J. Ind. Microbiol. Biotechnol. 39 :1725–1729.

Sheng Y, Mancino V, Birren B . 1995. Transformation of Escherichia coli with large DNA molecules by electroporation. Nucl. Acids Res. 23 :1990–1996.

Shin HD, Yoon SH, Wu J, Rutter C, Kim SW, Chen RR . 2012. High-yield production of meso - 2,3-butanediol from cellodextrin by engineered E. coli biocatalysts. Bioresour. Technol. 118 :367–373.

Siemerink MAJ, Kuit W, Lopez Contreras AM, Eggink G, Van der Oost J, Kengen SWM . 2011. D-2,3-Butanediol production due to heterologous expression of an acetoin reductase in Clostridium acetobutylicum . Appl. Environ. Microbiol. 77 :2582–2588.

Sikora B, Kubik C, Kalinowska H, Gromek E, Białkowska A, Jędrzejczak-Krzepkowska M, Schütt F, Turkiewicz M . 2016. Application of byproducts from food processing for production of 2,3-butanediol using Bacillus amyloliquefaciens TUL 308. Prep. Biochem. Biotechnol. 46 :610–619. 7. References 153

Simpson SD, Tran PL, Mihalcea CD, Fung JMY, Liew F. 2011. Production of butanediol by anaerobic microbial fermentation. EP2307556 A1. European Patent Office, Munich, Germany.

Straub, M. 2012. Fermentative Acetatproduktion durch Homoacetat-Gärung bzw. Acetacetatbildung. Ph. D. Thesis. Ulm University, Germany.

Sun J, Wang Y . 2014. Recent advances in catalytic conversion of ethanol to chemicals. ACS Catal. 4:1078–1090.

Sun LH, Wang XD, Dai JY, Xiu ZL . 2009a. Microbial production of 2,3-butanediol from Jerusalem artichoke tubers by Klebsiella pneumoniae . Appl. Microbiol. Biotechnol. 82 :847– 852.

Sun QY, Ding LW, He LL, Sun YB, Shao JL, Luo M, Xu ZF . 2009b. Culture of Escherichia coli in SOC medium improves the cloning efficiency of toxic protein genes. Anal. Biochem. 394 :144– 146.

Suzuki H . 2012. Host-mimicking strategies in DNA methylation for improved bacterial transformation. In: Dricu A (ed.), Methylation - from DNA, RNA and histones to diseases and treatment, InTech Rijeka, Croatia:219-236.

Syu MJ . 2001. Biological production of 2,3-butanediol. Appl. Microbiol. Biotechnol. 55 :10–18.

Szostková M, Horáková D . 1998. The effect of plasmid DNA sizes and other factors on electrotransformation of Escherichia coli JM109. Bioelectrochem. Bioenerg. 47 :319–323.

Takamizawa A, Miyata S, Matsushita O, Kaji M, Taniguchi Y, Tamai E, Shimamoto S, Okabe A. 2004. High-level expression of clostridial sialidase using a ferredoxin gene promoter-based plasmid. Protein Expr. Purif. 36 :70–75.

Tan Y, Liu ZY, Liu Z, Li FL . 2015. Characterization of an acetoin reductase/2,3-butanediol dehydrogenase from Clostridium ljungdahlii DSM 13528. Enzyme Microb. Technol. 79-80 :1–7.

Tanner RS . 2007. Cultivation of bacteria and fungi. In: Hurst C, Crawford R, Garland J, Lipson D, Mills A, Stetzenbach L (eds.) Manual of environmental microbiology, 3 rd edition, American Society of Microbiology, Washington, D.C., WA, USA:69-78. 154 7. References

Tanner RS, Miller LM, Yang D . 1993. Clostridium ljungdahlii sp. nov., an acetogenic species in clostridial rRNA homology group I. Int. J. Syst. Evol. Microbiol. 43 :232–236.

Tong YJ, Ji XJ, Shen MQ, Liu LG, Nie ZK, Huang H . 2016. Constructing a synthetic constitutive metabolic pathway in Escherichia coli for ( R,R )-2,3-butanediol production. Appl. Microbiol. Biotechnol. 100 :637–647.

Tran AV, Chambers RP . 1987. The dehydration of fermentative 2,3-butanediol into methyl ethyl ketone. Biotechnol. Bioeng. 29 :343–351.

Tremblay PL, Zhang T, Dar SA, Leang C, Lovley DR. 2012. The Rnf complex of Clostridium ljungdahlii is a proton-translocating ferredoxin:NAD + oxidoreductase essential for autotrophic growth. mBio 4:e00406–00412.

Tschech A, Pfennig N . 1984. Growth yield increase linked to caffeate reduction in Acetobacterium woodii . Arch. Microbiol. 137 :163–167.

Tsvetanova F, Petrova P, Petrov K . 2014. 2,3-butanediol production from starch by engineered Klebsiella pneumoniae G31-A. Appl. Microbiol. Biotechnol. 98 :2441–2451.

Tyurin M, Kiriukhin M . 2013. Synthetic 2,3-butanediol pathway integrated using Tn 7-tool and powered via elimination of sporulation and acetate production in acetogen biocatalyst. Appl. Biochem. Biotechnol. 170 :1503–1524.

Ungerman AJ, Heindel TJ . 2007. Carbon monoxide mass transfer for syngas fermentation in a stirred tank reactor with dual impeller configurations. Biotechnol. Prog. 23 :613–620.

Valgepea K, De Souza Pinto Lemgruber R, Meaghan K, Palfreyman RW Abdalla T, Heijstra BD, Behrendorff JB, Tappel R, Köpke M, Simpson SD, Nielsen LK, Marcellin E . 2017a. Maintenance of ATP homeostasis triggers metabolic shifts in gas-fermenting acetogens. Cell Syst. 4:505–515.

Valgepea K, Loi KQ, Behrendorff JB, Lemgruber RSP, Plan M, Hodson MP, Köpke M, Nielsen LK, Marcellin E . 2017b. Arginine deiminase pathway provides ATP and boosts growth of the gas-fermenting acetogen Clostridium autoethanogenum . Metab. Eng. 41 :202–211. 7. References 155

Veibel S, Nielsen JI . 1967. On the mechanism of the Guerbet reaction. Tetrahedron 23 :1723– 1733.

Visioli LJ, Enzweiler H, Kuhn RC, Schwaab M, Mazutti MA . 2014. Recent advances on biobutanol production. Sustain. Chem. Process. 2:15.

Vivijs B, Moons P, Aertsen A, Michiels CW . 2014a. Acetoin synthesis acquisition favors Escherichia coli growth at low pH. Appl. Environ. Microbiol. 80 :6054–6061.

Vivijs B, Moons P, Geeraerd AH, Aertsen A, Michiels CW . 2014b. 2,3-Butanediol fermentation promotes growth of Serratia plymuthica at low pH but not survival of extreme acid challenge. Int. J. Food Microbiol. 175 :36–44.

Wang A, Wang Y, Jiang T, Li L, Ma C, Xu P . 2010. Production of 2,3-butanediol from corncob molasses, a waste by-product in xylitol production. Appl. Microbiol. Biotechnol. 87 :965–970.

Wang A, Xu Y, Ma C, Gao C, Li L, Wang Y, Tao F, Xu P . 2012a. Efficient 2,3-butanediol production from cassava powder by a crop-biomass-utilizer, Enterobacter cloacae subsp. dissolvens SDM. PLoS ONE 7:e40442.

Wang Q, Chen T, Zhao X, Chamu J . 2012b. Metabolic engineering of thermophilic Bacillus licheniformis for chiral pure D-2,3-butanediol production. Biotechnol. Bioeng. 109 :1610–1621.

Wang S, Huang H, Kahnt J, Mueller AP, Köpke M, Thauer RK. 2013a. NADP-specific electron- bifurcating [FeFe]-hydrogenase in a functional complex with formate dehydrogenase in Clostridium autoethanogenum grown on CO. J. Bacteriol. 195 :4373-4386.

Wang Y, Li L, Ma C, Gao C, Tao F, Xu P . 2013b. Engineering of cofactor regeneration enhances (2 S,3 S)-2,3-butanediol production from diacetyl. Sci. Rep. 3:2643.

Weisburg WG, Barns SM, Pelletier DA, Lane DJ . 1991. 16S ribosomal DNA amplification for phylogenetic study. J. Bacteriol. 173 :697–703.

Weiss B, Jacquemin-Sablon A, Live TR, Fareed GC, Richardson CC . 1968. Enzymatic breakage and joining of deoxyribonucleic acid VI. Further purification and properties of polynucleotide ligase from Escherichia coli infected with bacteriophage T4. J. Biol. Chem. 243 :4543–4555. 156 7. References

White H, Strobl G, Feicht R, Simon H . 1989. Carboxylic acid reductase: a new tungsten enzyme catalyses the reduction of non-activated carboxylic acids to aldehydes. Eur. J. Biochem. 184 :89–96.

Whitham JM, Schulte MJ, Bobay BG, Bruno-Barcena JM, Chinn MS, Flickinger MC, Pawlak JJ, Grunden AM . 2017. Characterization of Clostridium ljungdahlii OTA1: a non-autotrophic hyper ethanol-producing strain. Appl. Microbiol. Biotechnol. 101 :1615–1630.

Wolin E, Wolin M, Wolfe R . 1963. Formation of methane by bacterial extracts. J. Biol. Chem. 238 :2882–2886.

Wood HG . 1991. Life with CO or CO 2 and H2 as a source of carbon and energy. FASEB J. 5:156– 163.

Xiao Z, Xu P . 2007. Acetoin metabolism in bacteria. Crit. Rev. Microbiol. 33 :127–140.

Xiao Z, Lv C, Gao C, Qin J, Ma C, Liu Z, Liu P, Li L, Xu P . 2010. A novel whole-cell biocatalyst with NAD + regeneration for production of chiral chemicals. PLoS ONE 5:e8860.

Xu Y, Chu H, Gao C, Tao F, Zhou Z, Li K, Li L, Ma C, Xu P . 2014. Systematic metabolic engineering of Escherichia coli for high-yield production of fuel bio-chemical 2,3-butanediol. Metab. Eng. 23:22–33.

Yan Y, Lee CC, Liao JC . 2009. Enantioselective synthesis of pure ( R,R )-2,3-butanediol in Escherichia coli with stereospecific secondary alcohol dehydrogenases. Org. Biomol. Chem. 7:3914–3917.

Yang C, Meng ZY . 1993. Bimolecular condensation of ethanol to 1-butanol catalyzed by alkali cation zeolites. J. Catal. 142 :37–44.

Yang J, Kim B, Kim H, Kweon Y, Lee S, Lee J . 2015a. Industrial production of 2,3-butanediol from the engineered Corynebacterium glutamicum . Appl. Biochem. Biotechnol. 176 :2303– 2313. 7. References 157

Yang T, Rao Z, Hu G, Zhang X, Liu M, Dai Y, Xu M, Xu Z, Yang ST . 2015c. Metabolic engineering of Bacillus subtilis for redistributing the carbon flux to 2,3-butanediol by manipulating NADH levels. Biotechnol. Biofuels 8:129.

Yang T, Rao Z, Zhang X, Lin Q, Xia H, Xu Z, Yang S . 2011. Production of 2,3-butanediol from glucose by GRAS microorganism Bacillus amyloliquefaciens . J. Basic Microbiol. 51 :650–658.

Yang T, Rao Z, Zhang X, Xu M, Xu Z, Yang ST . 2015b. Enhanced 2,3-butanediol production from biodiesel-derived glycerol by engineering of cofactor regeneration and manipulating carbon flux in Bacillus amyloliquefaciens . Microb. Cell Fact. 14 :122.

Yang T, Zhang X, Rao Z, Gu S, Xia H, Xu Z . 2012. Optimization and scale-up of 2,3-butanediol production by Bacillus amyloliquefaciens B10-127. World J. Microbiol. Biotechnol. 28 :1563– 1574.

Yang TW, Rao ZM, Zhang X, Xu MJ, Xu ZH, Yang ST . 2013. Fermentation of biodiesel-derived glycerol by Bacillus amyloliquefaciens : effects of co-substrates on 2,3-butanediol production. Appl. Microbiol. Biotechnol. 97 :7651–7658.

Yoo M, Croux C, Meynial-Salles I, Soucaille P . 2016. Elucidation of the roles of adhE1 and adhE2 in the primary metabolism of Clostridium acetobutylicum by combining in-frame gene deletion and a quantitative system-scale approach. Biotechnol. Biofuels 9:92.

Zahn JA, Saxena J. 2011. Novel ethanologenic species Clostridium coskatii . US 2011/0229947 A1. Washington, D.C.: U.S. Patent and Trademark Office.

Zeng AP, Sabra W . 2011. Microbial production of diols as platform chemicals: recent progresses. Curr. Opin. Biotechnol. 22 :749–757.

Zhang G, Bao P, Zhang Y, Deng A, Chen N, Wen T . 2011. Enhancing electro-transformation competency of recalcitrant Bacillus amyloliquefaciens by combining cell-wall weakening and cell-membrane fluidity disturbing. Anal. Biochem. 409 :130–137.

Zhang GL, Wang CW, Li C . 2012a. Cloning, expression and characterization of meso -2,3- butanediol dehydrogenase from Klebsiella pneumoniae . Biotechnol. Lett. 34 :1519–1523. 158 7. References

Zhang L, Sun J, Hao Y, Zhu J, Chu J, Wei D, Shen Y . 2010b. Microbial production of 2,3- butanediol by a surfactant (serrawettin)-deficient mutant of Serratia marcescens H30. J. Ind. Microbiol. Biotechnol. 37 :857–862.

Zhang L, Yang Y, Sun J, Shen Y, Wei D, Zhu J, Chu J. 2010a. Microbial production of 2,3- butanediol by a mutagenized strain of Serratia marcescens H30. Bioresour. Technol. 101 :1961–1967.

Zhang W, Yu D, Ji X, Huang H . 2012b. Efficient dehydration of bio-based 2,3-butanediol to butanone over boric acid modified HZSM-5 zeolites. Green Chem. 14 :3441–3450.

8. Supplement 159

8. Supplement

Changed AA type type AA Changed Stop codon Aliphatic aromatic, Acidic, basic, basic, sulfur-containing, acidic Aliphatic, aliphatic, basic, aliphatic, cyclic, aliphatic, basic, aromatic, basic Basic, aliphatic, aliphatic, stop aliphatic, aliphatic codon, acidic, Aliphatic, sulfur-containing, sulfur-containing, acidic, acidic Acidic

changed amino acid type acid amino changed

type AA Original Hydroxyl Aliphatic hydroxyl, Aliphatic, basic, basic, aliphatic aliphatic, aliphatic, Acidic, sulfur- aliphatic, containing, acidic, aliphatic, cyclic, aromatic, aliphatic Sulfur-containing, aliphatic, aliphatic, sulfur- aliphatic, cyclic containing, acidic, Acidic, sulfur-containing, acidic, aliphatic, acidic Aliphatic WT WT with locus tag, of annotation respective

Changed AA AA Changed No AA G T, R, D,R, F, D K, V, G, V, V, R P, F, H, no R, V, A, A, L AA, D, E,A, S, S, D D C. C. ljungdahlii to to

of of april, 2017.)

th Original AA AA Original Y A, V, Y, H, H, G A, M, V, D, G, P, V, Q,I F, A, M, A, A, S, P D,S, I, E, E, E V 25

,

cgi)

original and changed amino acid, and original and and original acid, and amino changed and original [pMTL82151] compared [pMTL82151] bin/mer/main.

-

changed AA position 1283 305 298, 1144, 1126, 1123, 1171 1148, 1146, 2806, 2797, 2794, 2836, 2830, 2821, 2854 2848, 2844, 599, 596, 568, 544, 634 607, 532, 530, 523, 520, 1769 1768, 19 C. C. ljungdahlii

Annotation Annotation Phage terminase subunit large corrinoid Putative protein methyltransferase secretion putative protein HlyD transporter putative protein bifunctional aldehyde/alcohol dehydrogenase bifunctional aldehyde/alcohol dehydrogenase protein hypothetical

Locus tag* Locus tag* CLJU_c03530 CLJU_c07520 CLJU_c07590 CLJU_c07600 CLJU_c16510 CLJU_c16520 CLJU_c18630 Table Table 12: Results of SNP analysis of with (AA) acid amino changed of tag, locusposition *according to /MERIMG (https://img.jgi.doe.gov/cgi 160 8. Supplement

ag, ag,

) Changed AA Changed type Acidic sulfur- Aliphatic, containing Aliphatic, aliphatic Acidic Aliphatic Aromatic Sulfur- containing

Original AA type type AA Original Aliphatic basic Acidic, Sulfur-containing, sulfur-containing Basic Acidic Aliphatic Aliphatic

(continued acid type no

with locus tag,of t WT locus respective annotation Changed AA AA Changed D V, T I, I Q V F T

C. C. ljungdahlii Original AA AA Original A E, R M M, H E L I

of of april, 2017.)

th , 25

changed AA position 1759 1342 802, 1818 1817, 509 91 227 91 nd changed amino acid, and original and changed ami changed and original and acid, amino changed nd [pMTL82151] compared compared to [pMTL82151] bin/mer/main.cgi) -

C. ljungdahlii C. ljungdahlii

Annotation Annotation surface/cell-adhesion putative protein and signal-transduction protein transcriptional-control and signal-transduction protein transcriptional-control permease putative NADH related RnfC putative dehydrogenase transporter ABC cobalt putative permease protein hypothetical

Locus tag* Locus tag* CLJU_c22360 CLJU_c24850 CLJU_c24870 CLJU_c27200 CLJU_c39770 CLJU_c40730 CLJU_c41870

Table 12: SNP of Results analysis of a original with (AA) acid amino changed of position *according to /MERIMG (https://img.jgi.doe.gov/cgi 8. Supplement 161

Table 13: Comparison of codon usage in R. terrigena and C. ljungdahlii

Amino acid (abbr.) Triplet Abundance [%] R. terrigena * Abundance [%] C. ljungdahlii ** Alanine (A) GCA 18 41 GCC 30 15 GCG 29 6 GCT 23 38 Cysteine (C) UGC 62 44 UGU 38 56 Aspartic acid (D) GAC 36 24 GAU 64 76 Glutamic acid (E) GAA 51 72 GAG 49 28 Phenylalanine (F) UUC 39 29 UUU 61 71 Glycine (G) GGA 12 42 GGC 39 16 GGG 18 15 GGU 30 27 Histidine (H) CAC 37 28 CAU 63 72 Isoleucine (I) AUA 12 43 AUC 35 17 AUU 53 40 Lysine (K) AAA 54 69 AAG 46 31 Leucine (L) CUA 10 14 CUC 8 7 CUG 36 10 CUU 14 20

*according to Codon usage database (http://www.kazusa.or.jp/codon/cgi- bin/showcodon.cgi?species=577&aa=1&style=N, 14 th of June, 2017.) **according to Anja Poehlein, Göttingen Genomics Laboratory (Göttingen, Germany)

162 8. Supplement

Table 13: Comparison of codon usage in R. terrigena and C. ljungdahlii

Amino acid (abbr.) Triplet Abundance [%] R. terrigena * Abundance [%] C. ljungdahlii ** Leucine (L) UUA 17 32 UUG 14 17 Methionine (M) AUG 100 100 Asparagine (N) AAC 50 27 AAT 50 73 Proline (P) CCA 21 38 CCC 15 15 CCG 37 8 CCU 27 39 Glutamine (Q) CAA 37 61 CAG 63 39 Arginine (R) AGA 13 48 AGG 8 30 Arginine (R) CGA 11 6 CGC 30 4 CGG 10 5 CGU 29 7 Serine (S) AGC 24 14 AGU 15 21 UCA 14 21 UCC 14 16 UCG 17 3 UCU 16 25 Threonine (T) ACA 17 36 ACC 38 19 ACG 29 7 ACU 15 38 Valine (V) GUA 14 38 GUC 22 12 *according to Codon usage database (http://www.kazusa.or.jp/codon/cgi- bin/showcodon.cgi?species=577&aa=1&style=N, 14 th of June, 2017.) **according to Anja Poehlein, Göttingen Genomics Laboratory (Göttingen, Germany) 8. Supplement 163

Table 13: Comparison of codon usage in R. terrigena and C. ljungdahlii

Amino acid (abbr .) Triplet Abundance [%] R. terrigena * Abundance [%] C. ljungdahlii ** Valine (V) GUG 32 17 GUU 32 33 Tryptophan (W) UGG 100 100 Tyrosine (Y) UAC 44 29 UAU 56 71 Stop codon UAA 48 53 UAG 35 23 UGA 17 24 *according to Codon usage database (http://www.kazusa.or.jp/codon/cgi- bin/showcodon.cgi?species=577&aa=1&style=N, 14 th of June, 2017.) **according to Anja Poehlein, Göttingen Genomics Laboratory (Göttingen, Germany)

Contribution to this work 165

Contribution to this work

The research leading to these results has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no. 311815 (SYNPOL project).

Danksagung 167

Danksagung

Aus Datenschutzgründen wurde die Danksagung auf den Seiten 167-169 entfernt. stark, denn ich fürchte vor der Verteidigung wird es noch schlimmer… You’re simply the best!

168 Danksagung

Danksagung 169

Curriculum vitae 171

Curriculum vitae

Aus Datenschutzgründen wurde der Curriculum vitae auf den Seiten 171 und 172 entfernt.

172 Curriculum vitae

Statement 173

Statement

I hereby declare that the present thesis is the result of my own work and that all sources of information and support used for its achievement have been mentioned in the text and in the references. All citations are explicitly identified, all figures contain only the original data and no figure was edited in a way that substantially modified its content. All existing copies of the present thesis are identical regarding their form and content.

Ulm, July 2017

______Catarina Erz