Heterologous Production of Isopropanol Using Metabolically Engineered Acetobacterium woodii Strains

The depletion of fossil fuel resources and the CO2 emissions coupled with petroleum-based industrial processes present a relevant issue for the whole of society. An alternative to the fossil-based production of chemicals is microbial fermentation using acetogens. Acetogenic bacteria are able to metabolize CO or CO2 (+H2) via the Wood–Ljungdahl pathway. As isopropanol is widely used in a variety of industrial branches, it is advantageous to find a fossil-independent production process. In this study, Acetobacterium woodii was employed to produce isopropanol via plasmid-based expression of the enzymes thiolase A, CoA-transferase, acetoacetate decarboxylase and secondary alcohol dehydrogenase. An examination of the enzymes originating from different organisms led to a maximum isopropanol production of 5.64 ± 1.08 mM using CO2 + H2 as the carbon and energy source. To this end, the genes thlA (encoding thiolase A) and ctfA/ctfB (encoding CoA-transferase) of Clostridium scatologenes, adc (encoding acetoacetate decarboxylase) originating from C. acetobutylicum and sadH (encoding secondary alcohol dehydrogenase) of C. beijerinckii DSM 6423 were employed. Since bottlenecks in the isopropanol production pathway are known, optimization of the strain was investigated, resulting in a 2.5-fold increase in isopropanol concentration.


Introduction
Isopropanol, which is also known as isopropyl alcohol, is an important solvent and chemical intermediate for a variety of industrial branches.It is widely used as a solvent, cleaner or as a raw material for different derivatives, and it has gained interest since the coronavirus pandemic (COVID-19) due to its biocidal properties [1,2].As isopropanol is typically produced via the hydration of propylene, and propylene is synthesized via the steam cracking of petroleum feedstocks, the production process is fossil-based [1,3].Fossil fuel resources are shrinking, and petroleum-based processes are coupled with the excessive release of greenhouse gas emissions; therefore, alternative production processes need to be established [4].A promising alternative is found in fermentation using acetogenic bacteria.Acetogens, which are widespread throughout a huge variety of genera, employ the Wood-Ljungdahl pathway (WLP) for the fixation of CO or CO 2 (together with H 2 ) by producing acetate as the main product [5,6].Some acetogens are already used for the industrial production of solvents or alcohols, e.g., LanzaTech is using the organism Clostridium autoethanogenum for the commercial production of ethanol in anaerobic fermentation plants [7,8].The manufacture of high-value chemicals using acetogens has been favored during the past decades, as acetogens utilize greenhouse gas emissions instead of fossil fuels or renewable resources.Thereby, the food-vs.-fueldebate can be avoided.Furthermore, a carbon-neutral way of production can be established [2,9].The bacterial species Acetobacterium woodii comprises the genes of the WLP and produces acetate as the Bioengineering 2023, 10, 1381 2 of 19 sole product; hence, it is used as a model organism for understanding the WLP as well as for the metabolic engineering of acetogenic bacteria [10,11].The WLP consists of a methyl and a carbonyl branch.In the carbonyl branch, CO 2 is taken up via carbon monoxide dehydrogenase (CODH) and converted to CO by oxidation of a reduced ferredoxin (Fd 2− ).CO 2 is also taken up by the CODH in the methyl branch, followed by the conversion into formate via formate dehydrogenase.In an ATP-dependent step, formate is bound to tetrahydrofolate (THF) by formyl-THF synthetase.Employing the methenyl-THF cyclohydrolase, methylene-THF dehydrogenase and methenyl-THF reductase, formyl-THF is further processed to methyl-THF.After methyl transferase exchanges THF against an iron-sulfur cluster containing corrinoid protein (CoFeS-P), methyl-CoFeS-P is converted together with CO (from the carbonyl branch) to acetyl-CoA by acetyl-CoA synthase [12].In A. woodii, acetyl-coenzmye A (CoA) is converted to the sole product acetate along with the formation of ATP [10].
Over the past decades, A. woodii has been metabolically engineered, e.g., for the production of the bioplastic poly-3-hydroxybutyrate, lactate or the chemical intermediate acetone [13][14][15][16].The heterologous production of acetone includes the plasmid-based expression of the genes thlA, encoding a thiolase A, ctfA/ctfB, encoding a CoA-transferase, and adc, encoding an acetoacetate decarboxylase.Acetyl-CoA is produced via the WLP as a central intermediate and represents the starting point for acetone production.Two moieties of acetyl-CoA are converted to acetoacetyl-CoA via thiolase A. CoA-transferase transfers the CoA to one moiety of acetate by releasing acetoacetate.By cleaving off CO 2 via the acetoacetate decarboxylase, acetone is produced [15].During fermentation of the different recombinant A. woodii strains for acetone production, small amounts of isopropanol were also monitored.As it is known that the conversion of acetone to isopropanol is feasible via a secondary alcohol dehydrogenase (Sadh), the genome of A. woodii was screened for an appropriate enzyme for isopropanol production.Until now, no sadH gene was found in the genome of A. woodii [16].
In this study, the acetogenic bacterium A. woodii was employed for the autotrophic, recombinant production of isopropanol using CO 2 and H 2 as the carbon and energy resource.Therefore, the sadH genes of Clostridium beijerinckii DSM 6423 and C. ljungdahlii were chosen as the candidate enzymes for isopropanol production [17][18][19].Via BLAST analysis, another sadH gene was found in the genome of C. beijerinckii DSM 15410, which was also tested for recombinant isopropanol production.The production of isopropanol using different anaerobic Clostridium species already revealed bottlenecks of acetone and isopropanol production [2,20,21].Limitations regarding the conversion of acetone to isopropanol can be avoided by the insertion of additional copies of the sadH gene or expression of a gene named hydG, which can be found in the same operon as sadH.The gene hydG is annotated in both the C. beijerinckii and C. ljungdahlii genomes and it is predicted to function as an electron donor protein [20].Another bottleneck in the acetone production is the efficiency of the CoA-transferase, which can be bypassed by inserting an additional copy of the ctfA/ctfB gene [2,20].

Bacterial Strains and Media
The bacterial strains used in this study are listed in Table 1.Escherichia coli XL1-Blue MRF' was used for plasmid construction and amplification.Strains of E. coli XL1-Blue MRF' were cultivated in Lysogeny Broth (LB) medium (containing per liter: tryptone 10 g, NaCl 10 g, yeast extract 5 g) with the respective antibiotic aerobically at 37 • C with shaking at 180 rpm [22].

Construction of Plasmids
Standard molecular cloning techniques were carried out according to established protocols [22].For the construction of the acetone production plasmid pJIR750_ac2t2, the vector pJIR750_ac2t1 was digested with the restriction enzymes SalI and BamHI [16].The thiolase gene (thlA, Locus Tag: CKL_3698) of C. kluyveri and the promoter P thlA originating from C. acetobutylicum were ligated into the digested vector resulting in the plasmid pJIR750_ac2t2.For the construction of the plasmid pJIR750_ac3t3, pJIR750_ac2t2 was digested using the restriction enzymes SalI and KpnI.The acetoacetyl-CoA:acetate/butyrate CoA transferase genes (ctfA/ctfB, Locus tag: EG59DRAFT_00772-00773) and the thiolase A gene (thlA, Locus tag: EG59DRAFT_00774) of C. scatologenes were inserted into the vector.
For isopropanol production, the plasmids pJIR750_ac1t1, pJIR750_ac1t2, pJIR750_ac2t1, pJIR750_ac2t2 and pJIR750_ac3t3 were digested with the restriction enzyme EcoRI.The secondary alcohol dehydrogenase genes (sadH, Locus Tags: CLOBI_40010, K684DRAFT_02545 and CLJU_c24860, respectively) of C. beijerinckii DSM 6423, C. beijerinckii DSM 15410 and C. ljungdahlii were inserted into the different vectors, resulting in 15 different isopropanol production plasmids listed in Table 2.
For the construction of the two-plasmid system, the vector pMTL83251 was digested with the restriction enzymes BamHI and KpnI [24].The gene hydG (Locus Tag: CLOBI_40020) and the sadH-hydG gene cluster (Locus tag: CLOBI_40010-40020) with an intergenic region were inserted under control of the P thlA promoter resulting in the plasmids pMTL83251_PthlA_h1 and pMTL83251_PthlA_sh1, respectively.In a second step, pMTL83251_PthlA_sh1 was digested with the restriction enzymes NcoI and AatII, and the genes ctfA/ctfB (Locus tag: EG59DRAFT_00772-00773) of C. scatologenes were inserted, resulting in the plasmid pMTL83251_PthlA_sh1c3.

Isolation of Genomic and Plasmid DNA
The genomic DNA of Gram-positive organisms was isolated using the MasterPure TM Gram-Positive DNA Purification Kit (Epicentre, Madison, WI, USA) according to the manufacturer's instructions.The isolation of plasmid DNA from E. coli strains was performed using the Zyppy TM Plasmid Miniprep Kit (ZYMO Research Europe GmbH, Freiburg, Germany) according to the manufacturer's instructions.

Transformation of A. woodii
For the preparation of electrocompetent A. woodii cells and transformation, the recently published protocol of Hoffmeister et al. was used [15].Transformation was verified by the isolation of genomic DNA and 16S rDNA amplification via PCR.Subsequently, the amplified DNA fragment was sent for sequencing.Using the retransformation of E. coli XL1-Blue MRF', plasmid integrity was verified.

Growth Conditions of Batch Experiments
Growth experiments with A. woodii were performed anaerobically at 30 • C in a modified DSM 135 medium without and with shaking at 130 rpm for heterotrophic and autotrophic experiments, respectively.For heterotrophic cultivation, 50 mL of medium were filled into 125 mL serum bottles (SGD Pharma, Paris, France), and 60 mM fructose were added as the carbon and energy source.Autotrophic growth was carried out in 50 mL of medium filled into 500 mL serum bottles (SGD Pharma, Paris, France), and the gas phase was exchanged to 1 bar CO 2 + H 2 (33% + 67%, MTI Industriegase AG, Elchingen, Germany).At a minimal pressure of 0.3 bar, the gas phase was replenished during growth experiments.
For adaption to the respective growth conditions, the A. woodii strains were transferred once into fresh medium prior to the growth experiments.Experiments were performed as biological triplicates.Integrity of the plasmid and strain verification was carried out before and after the growth experiments as described above.
OD 600 , pH, substrate consumption and end-product concentrations were measured during growth.

Analytics 2.6.1. Optical Density and pH Measurements
The growth of A. woodii was monitored by means of the optical density (OD 600 : absorption of the culture at 600 nm) and pH, measured as recently described [13].

High-Performance Liquid Chromatography
Fructose consumption and acetate production during the heterotrophic and autotrophic growth of recombinant A. woodii strains were measured via HPLC analysis.Therefore, 500 µL of cell suspension was centrifuged (30 min at 4 • C and 13,000× g), the supernatant was transferred to a glass vial (CS-Chromatographie Service GmbH, Langerwehe, Germany) and sealed with a crimp cap (BGB Analytik Vertrieb, Rheinfelden, Germany).For the detection of fructose and acetate, the Agilent 1260 Infinity Series HPLC System (Agilent Technologies, Santa Clara, CA, USA) equipped with a diode array detector, a refractive index detector and the CS-Chromatographie organic acid column with a length of 150 mm (CS-Chromatographie Service, Langerwehe, Germany) was used.The applied method was recently described by Höfele and Dürre 2023 [13].

Gas Chromatography
The production of acetone and isopropanol was monitored via GC analysis.For this purpose, 2 mL of culture broth samples were taken during the growth of A. woodii and centrifuged (30 min at 4 • C and 13,000× g).The supernatant was transferred to glass vials (CS-Chromatographie Service GmbH, Langerwehe, Germany), acidified with 2 M HCl and sealed with screwcaps (CS-Chromatographie Service GmbH, Langerwehe, Germany).The acetone and isopropanol concentrations were analyzed with the gas chromatograph Perkin Elmer Clarus 680 GC system (Perkin Elmer, Waltham, MA, USA) equipped with an Elite-FFAP capillary column (Perkin Elmer, Waltham, MA, USA) with a length of 30 m, inner diameter of 0.32 mm, and a density film of 0.25 µM.An amount of 1 µL of the supernatant was injected with a split ratio of 20.Amounts of 45 mL min −1 H 2 and 450 mL min −1 synthetic air served as the carrier gas.The separation of acetone and isopropanol was performed by using a temperature program (temperature held for 4 min at 50 • C, temperature ramp with 40 • C per minute to a maximum of 240 • C and temperature held at 240 • C for 3 min).Solvents were detected by a flame ionization detector (FID).External quantification standards were used for calibration.

Isopropanol Production with Recombinant A. woodii Strains
For acetone production, the plasmids pJIR750_ac2t2 and pJIR750_ac3t3 were constructed.After transformation into competent A. woodii cells, heterotrophic and autotrophic growth experiments were performed.During the heterotrophic growth of A. woodii [pJIR750_ac2t2] and A. woodii [pJIR750_ac3t3] with 60 mM fructose as the substrate, 1.61 mM and 16.89 mM acetone together with 1.31 mM and 1.19 mM isopropanol were produced, respectively (data not shown).The autotrophic growth of A. woodii [pJIR750_ac2t2] and A. woodii [pJIR750_ac3t3] with CO 2 + H 2 as the substrate resulted in the production of 0.38 mM and 1.6 mM acetone.During the metabolism of CO 2 + H 2 , no isopropanol was produced by the recombinant strains (Supplementary Figure S1).
Heterologous isopropanol production was established by insertion of the sadH genes of C. beijerinckii DSM 6423, C. beijerinckii DSM 15410 and C. ljungdahlii into the acetone production plasmids pJIR750_ac1t1, pJIR750_ac1t2, pJIR750_ac2t1, pJIR750_ac2t2 and pJIR750_ac3t3 (Table 3) [16].The 15 different plasmids for isopropanol production were transformed into electrocompetent A. woodii cells, and heterotrophic and autotrophic growth was performed and monitored.As fructose was consumed completely by all recombinant strains, a maximum OD 600 of 3.3-3.5 was reached, while the pH dropped to 6.2-6.6.The natural product acetate was accumulated by the recombinant A. woodii strains in concentrations of 118-181 mM (Supplementary Figure S2).The production of isopropanol during autotrophic growth reached a maximum of 5.64 ± 1.08 mM by A. woodii [pJIR750_ac3t3s1], while producing 10.79 ± 2.25 mM acetone (Figure 1D-F).A. woodii [pJIR750_ac1t1s1] and A. woodii [pJIR750_ac1t2s1] produced 5.95 ± 1.02 mM and 3.38 ± 0.22 mM acetone and 0.76 ± 0.25 mM and 0.38 ± 0.09 mM isopropanol, respectively.The strains A. woodii [pJIR750_ac2t1s1] and A. woodii [pJIR750_ac2t2s1] reached a maximum acetone concentration of 7.72 ± 0.51 mM and 7.40 ± 1.23 mM and an isopropanol concentration of 0.49 ± 0.14 mM and 1.57 ± 0.77 mM, respectively.A maximum OD 600 in the range of 1.45 to 1.68 was reached during consumption of 1.25 to 2.18 bar of CO 2 + H 2 .During the production of 168-181 mM acetate, the pH of all strains dropped similarly (Supplementary Figure S3). A. woodii wild-type and A. woodii [pJIR750] served as the control strains and showed no production of acetone and isopropanol.

Isopropanol Production with the sadH Gene of C. beijerinckii DSM 15410
As the BLAST analysis revealed a 97% similarity of the Locus tag K684DRAFT_02545 in the genome of C. beijerinckii DSM 15410 to the sadH gene of C. beijerinckii DSM 6423, the sadH gene of C. beijerinckii DSM 15410 was inserted into the acetone production plasmids.The isopropanol and acetone production of the newly constructed strains A. woodii [pJIR750_ac1t1s2], A. woodii [pJIR750_ac1t2s2], A. woodii [pJIR750_ac2t1s2], A. woodii [pJIR750_ac2t2s2] and A. woodii [pJIR750_ac3t3s2] were examined by a heterotrophic growth experiment using 60 mM fructose as the carbon and energy source (Figure 2A-C).The highest isopropanol production was measured after the cultivation of A. woodii [pJIR750_ac3t3s2] with a concentration of 20.44 ± 0.16 mM in combination with a production of 9.22 ± 0.01 mM acetone. A. woodii [pJIR750_ac1t1s2] and A. woodii [pJIR750_ac1t2s2] reached concentrations of 5.89 ± 0.41 and 11.25 ± 0.59 mM acetone and 1.78 ± 0.31 and 15.27 ± 0.48 mM isopropanol, respectively.The strains A. woodii [pJIR750_ac2t1s2] and A. woodii [pJIR750_ac2t2s2] produced 3.45 ± 0.92 mM and 6.16 ± 1.28 mM acetone as well as 6.37 ± 0.43 mM and 10.90 ± 2.16 mM isopropanol, respectively.During growth, an OD 600 of 3.3-3.9was reached by the recombinant strains, and the production of acetate in concentrations of 84 mM to 145 mM was measured (Supplementary Figure S4).Strains that produced less isopropanol and acetone showed a pH drop to 6.3 as stronger producers showed a pH of 6.9 to 7.1.
The autotrophic production of isopropanol was carried out using CO 2 + H 2 as the substrate, which resulted in a maximum isopropanol and acetone production of A. woodii [pJIR750_ac3t3s2] with 4.20 ± 0.69 mM and 10.30 ± 3.06 mM, respectively (Figure 2D-F).In total, 4.37 ± 1.41 mM and 2.80 ± 0.55 mM acetone together with 0.17 ± 0.14 mM and 0.83 ± 0.22 mM isopropanol were produced by A. woodii [pJIR750_ac1t1s2] and A. woodii [pJIR750_ac1t2s2], respectively.The strains A. woodii [pJIR750_ac2t1s2] and A. woodii [pJIR750_ac2t2s2] showed the autotrophic production of 6.45 ± 1.47 mM and 6.59 ± 0.97 mM acetone and 0.54 ± 0.02 and 1.53 ± 0.49 mM isopropanol.Due to an uptake of 1.7 to 2.2 bar of CO 2 + H 2 , maximum OD 600 values of 1.5 to 1.6 were reached by the recombinant strains.Acetate was produced in a concentration range of 112 mM to 174 mM and the pH dropped similarly for all strains (Supplementary Figure S5).
[pJIR750_ac1t2s2], respectively.The strains A. woodii [pJIR750_ac2t1s2] and A. woodii [pJIR750_ac2t2s2] showed the autotrophic production of 6.45 ± 1.47 mM and 6.59 ± 0.97 mM acetone and 0.54 ± 0.02 and 1.53 ± 0.49 mM isopropanol.Due to an uptake of 1.7 to 2.2 bar of CO2 + H2, maximum OD600 values of 1.5 to 1.6 were reached by the recombinant strains.Acetate was produced in a concentration range of 112 mM to 174 mM and the pH dropped similarly for all strains (Supplementary Figure S5).

Isopropanol Production with the sadH Gene of C. ljungdahlii
The sadH gene of C. ljungdahlii demonstrated no significant similarity to the ones of C. beijerinckii DSM 6423 and DSM 15410, which was why it was chosen as a third candidate for isopropanol production in recombinant A. woodii strains.To examine isopropanol production, the strains A. woodii [pJIR750_ac1t1s3], A. woodii [pJIR750_ac1t2s3], A. woodii [pJIR750_ac2t1s3], A. woodii [pJIR750_ac2t2s3] and A. woodii [pJIR750_ac3t3s3] were cultivated heterotrophically with 60 mM fructose as the substrate (Figure 3A-C; Supplementary Figure S6).The highest production of isopropanol was measured for A. woodii [pJIR750_ac3t3s3] with 18.06 ± 0.65 mM together with 11.17 ± 0.62 mM acetone. A. woodii

Promotion of Isopropanol Production with the Gene hydG of C. beijerinckii DSM 6423
During fermentation of the recombinant A. woodii strains (described above), the heterologous production of isopropanol could be established.Despite expression of the sadH During autotrophic growth with CO 2 + H 2 as the carbon and energy source, a maximum of 5.33 ± 0.37 mM isopropanol and 9.54 ± 0.41 mM acetone was produced by A. woodii [pJIR750_ac3t3s3] (Figure 3D-F; Supplementary Figure S7).Isopropanol was also produced by A. woodii [pJIR750_ac1t1s3] and A. woodii [pJIR750_ac1t2s3] with a concentration of 0.26 ± 0.07 mM and 0.57 ± 0.16 mM together with 3.54 ± 0.12 mM and 1.84 ± 0.20 mM acetone.The strains A. woodii [pJIR750_ac2t1s3] and A. woodii [pJIR750_ac2t2s3] showed autotrophic production of 0.72 ± 0.10 mM and 4.27 ± 0.50 mM isopropanol in combination with 4.89 ± 0.29 mM and 11.58 ± 1.09 mM acetone.An uptake of 1.4-2.3bar of CO 2 + H 2 resulted in maximum OD 600 values of 1.5-1.9.The pH dropped similarly for all recombinant strains.During fermentation of the recombinant A. woodii strains (described above), the heterologous production of isopropanol could be established.Despite expression of the sadH gene, encoding a secondary alcohol dehydrogenase, acetone was not completely converted to isopropanol.To promote the conversion of acetone to isopropanol, the hydG gene, which is annotated as a gene for a potential electron donor protein [20], should be included in the best-performing isopropanol plasmids.After several unsuccessful trials to insert hydG into the plasmid pJIR750_ac3t3s1, it was decided to construct a two-plasmid system.Therefore, the gene hydG of C. beijerinckii DSM 6423 was

Overcoming the Bottlenecks of Recombinant Isopropanol Production
As the additional insertion of hydG originating from C. beijerinckii DSM 6423 did not lead to the enhanced conversion of acetone to isopropanol, but to a decrease in total solvent production, the location of hydG in the genome and its function was further analyzed.Thereby, it was found that hydG harbors a σ 54 interaction domain, indicating that it serves as an σ 54 activator protein [26].Further analysis of the intergenic region between sadH and hydG revealed the typical −24/−12 consensus recognition sequence for σ 54 binding [27].With this background of Wang et al. [18], showing enhanced isopropanol production by mM and 13.87 ± 1.97 mM, together with acetone concentrations of 0.65 ± 0.20 mM and 1.35 ± 0.14 mM, respectively.In comparison, the control strain A. woodii [pJIR750_ac3t3s1] produced 3.26 ± 0.30 mM of isopropanol and 3.19 ± 0.59 mM of acetone.

Discussion
The discovery of how to expend fossil fuels, such as natural gas, petroleum or coal, has played a severe role in the changes to human society since the late 18th century.Due to the use of fossil fuels, the standards of living, wealth and human health could be improved and led us to the way of today's living [28].Besides the undeniable positive effect of industrialization and use of fossil fuels on society, there are also crucial disadvantages and threats coming up with the utilization of fossil resources.As the burning of fossil fuels comes together with the emission of CO2, it drives the greenhouse gas effect and accelerates global warming [29].The estimated global warming of 1-5 °C during the next century will generate environmental changes, e.g., melting of glaciers, rising sea levels and extreme weather events, with serious impacts on flora, fauna and human life [30,31].Additionally, fossil resources are limited and will be depleted in a few years if today's consumption will continue.To counteract this phenomenon and retard global warming, alternatives for fossil-based processes should be developed and investigated.A promising

Discussion
The discovery of how to expend fossil fuels, such as natural gas, petroleum or coal, has played a severe role in the changes to human society since the late 18th century.Due to the use of fossil fuels, the standards of living, wealth and human health could be improved and led us to the way of today's living [28].Besides the undeniable positive effect of industrialization and use of fossil fuels on society, there are also crucial disadvantages and threats coming up with the utilization of fossil resources.As the burning of fossil fuels comes together with the emission of CO 2 , it drives the greenhouse gas effect and accelerates global warming [29].The estimated global warming of 1-5 • C during the next century will generate environmental changes, e.g., melting of glaciers, rising sea levels and extreme weather events, with serious impacts on flora, fauna and human life [30,31].Additionally, fossil resources are limited and will be depleted in a few years if today's consumption will continue.To counteract this phenomenon and retard global warming, alternatives for fossilbased processes should be developed and investigated.A promising alternative, which has gained interest throughout the last decades, is the production of high-value chemicals by fermentation using acetogens [32,33].The exchange of fossil-based procedures against acetogenic fermentation comes up with two major advantages: Fossil resources are saved since acetogens are capable of metabolization of single carbon gases (C 1 -gases).Further, the emission of C 1 -gases that accelerate global warming can be diminished by the production of high-value chemicals using acetogens and result in the best case in a carbon-neutral way of manufacturing [2].Since acetogenic bacteria are defined as a class of microorganisms harboring the enzymes of the WLP, members of acetogens are widely spread throughout many genera [6].Thereby, the natural product spectrum of acetogens is versatile, comprising, e.g., ethanol, 2,3-butanediol or hexanol [34,35].Further, the development of toolboxes for metabolic engineering expands the economic potential of some acetogens, which are genetically accessible.One model organism with established protocols for metabolic engineering and recently also genomic editing via its endogenous CRISPR/Cas system is A. woodii [15,36,37].The natural product spectrum of A. woodii comprises solely acetate and traces of ethanol under certain circumstances [38,39].By successful metabolic engineering, A. woodii was recently employed to produce lactate, poly-3-hydroxybutyrate or acetone together with the unexpected accumulation of isopropanol [13][14][15][16].
The reduction of acetone to isopropanol is also known in some natural isopropanolproducing Clostridia species, e.g., C. beijerinckii and C. aurantibutyricum; the latter was recently reclassified as C. felsineum [40,41].Clostridium species capable of isopropanol production harbor a primary/secondary alcohol dehydrogenase (SadH) specific for acetone, which is strictly NADPH-dependent [17].The production of isopropanol occurs in clostridial species in combination with butanol and ethanol and is a modification of the acetone-butanol-ethanol (ABE) fermentation.A prominent candidate for ABE fermentation is C. acetobutylicum, which was formerly employed for acetone production during World War I [42].For C. beijerinckii species, butanol is the favored product, whereby the production of isopropanol at an industrial scale reveals that it is unprofitable.Therefore, recombinant isopropanol production was examined using E. coli, resulting in a maximum production of up to 40 g/L [43].Due to enormous high solvent titers and the biocidal properties of isopropanol, cost-intensive gas stripping for the removal of produced isopropanol was necessary.Furthermore, fermentation using glucose as a substrate proved to be expensive, which is why researchers focused on the production of high-value chemicals using cheap waste gases as feedstock.Efficient and economically competitive processes have been recently developed by LanzaTech, which are using genomically modified C. autoethanogenum strains for the production of ethanol based on synthesis gas (syngas; a mixture of H 2 , CO 2 and CO) [8,9].LanzaTech is also working on the economically profitable production of isopropanol with recombinant C. autoethanogenum using syngas as the substrate.By the screening of a huge industrial genome databank and use of kinetic models, Liew et al. constructed a strain capable of producing approximately 3 g/L/h of isopropanol with syngas as the feedstock.Therefore, the native sadH gene of C. autoethanogenum was employed [2].
During the recombinant production of acetone with A. woodii, traces of isopropanol could be detected.Since the analysis of the genome of A. woodii regarding a sadH gene, responsible for the conversion of acetone to isopropanol, was not successful until now, this study focused on the heterologous production of isopropanol [16].Therefore, the plasmidbased expression of the required genes thlA (encoding a thiolase A), ctfA/ctfB (encoding a CoA-transferase), adc (encoding an acetoacetate decarboxylase) and sadH (encoding a secondary alcohol dehydrogenase) was established (Figure 6).To find the best recombinant isopropanol producer, a library with genes originating from different Clostridium species was established.As acetone production in A. woodii was first examined using thlA, ctfA/ctfB and adc of C. acetobutylicum, those genes were also chosen for isopropanol production [15].Due to a high K m value of ctfA/ctfB originating from C. acetobutylicum, the genes of C. aceticum, C. kluyveri and C. scatologenes were also tested [16,44].Variation of the gene encoding the thiolase A proved to be promising for acetone production, which is why it was also examined regarding isopropanol accumulation in this study.The gene library was completed by the sadH genes of C. beijerinckii and C. ljungdahlii [17][18][19].Since isopropanol was an unexpected byproduct of recombinant acetone production in C. autethanogenum, the responsible gene of C. autoethanogenum was also taken into consideration [2].However, BLAST analysis revealed a 100% similarity of the respective gene to the sadH of C. ljungdahlii.The same result was found for comparison with the genome of C. coskatii.Genome analysis of C. beijerinckii DSM 15410, which was formerly known as C. diolis, revealed a candidate gene coding for a secondary alcohol dehydrogenase with 97% identity to the sadH of C. beijerinckii DSM 6423 [45].
gene encoding the thiolase A proved to be promising for acetone production, which is why it was also examined regarding isopropanol accumulation in this study.The gene library was completed by the sadH genes of C. beijerinckii and C. ljungdahlii [17][18][19].Since isopropanol was an unexpected byproduct of recombinant acetone production in C. autethanogenum, the responsible gene of C. autoethanogenum was also taken into consideration [2].However, BLAST analysis revealed a 100% similarity of the respective gene to the sadH of C. ljungdahlii.The same result was found for comparison with the genome of C. coskatii.Genome analysis of C. beijerinckii DSM 15410, which was formerly known as C. diolis, revealed a candidate gene coding for a secondary alcohol dehydrogenase with 97% identity to the sadH of C. beijerinckii DSM 6423 [45].The combination of the acetone production genes of C. acetobutylicum with the candidate genes for sadH in recombinant A. woodii confirmed that autotrophic production of isopropanol is possible.As productions levels ranged between 0.17 and 0.76 mM, different gene combinations were examined.Thereby, it was found that any plasmid containing the thlA or ctfA/ctfB genes originating from C. acetobutylicum resulted in similar isopropanol concentrations during autotrophic growth (Table 4).Higher isopropanol accumulation was reached by exchanging thlA and ctfA/ctfB against genes originating from C. kluyveri and C. aceticum, respectively.The combination of those genes with the sadH of C. beijerinckii DSM 6423 and DSM 15410 resulted in two-fold increased isopropanol concentrations of 1.53-1.57mM.By the expression of the SadH originating from C. ljungdahlii, the isopropanol production was increased to 4.27 mM.The most promising recombinant isopropanol producers were found using the thlA and ctfA/ctfB of C. scatologenes.Combining those genes with the sadH of C. ljungdahlii and C. beijerinckii DSM 6423, the isopropanol concentration could be raised up to 5.33-5.64mM.The combination of the acetone production genes of C. acetobutylicum with the candidate genes for sadH in recombinant A. woodii confirmed that autotrophic production of isopropanol is possible.As productions levels ranged between 0.17 and 0.76 mM, different gene combinations were examined.Thereby, it was found that any plasmid containing the thlA or ctfA/ctfB genes originating from C. acetobutylicum resulted in similar isopropanol concentrations during autotrophic growth (Table 4).Higher isopropanol accumulation was reached by exchanging thlA and ctfA/ctfB against genes originating from C. kluyveri and C. aceticum, respectively.The combination of those genes with the sadH of C. beijerinckii DSM 6423 and DSM 15410 resulted in two-fold increased isopropanol concentrations of 1.53-1.57mM.By the expression of the SadH originating from C. ljungdahlii, the isopropanol production was increased to 4.27 mM.The most promising recombinant isopropanol producers were found using the thlA and ctfA/ctfB of C. scatologenes.Combining those genes with the sadH of C. ljungdahlii and C. beijerinckii DSM 6423, the isopropanol concentration could be raised up to 5.33-5.64mM.Since isopropanol production was already established in C. acetobutylicum and C. saccharoperbutylacetonicum, as well as in the acetogens C. ljungdahlii and C. autoethanogenum, bottlenecks of the production pathway were already discovered [2,[19][20][21].One limitation was found in the reduction efficiency of acetone to isopropanol by the SadH, indicated by leftover acetone concentrations.In case of fermentation using recombinant C. saccharoperbutylacetonicum, the conversion efficiency was increased from 50% to approximately 95% by additionally expressing the gene hydG, which is located in the same operon as sadH.Wang et al. [20] hypothesized that hydG encodes an electron transfer protein, delivering electrons to raise the NADPH supply necessary for the reduction of acetone.A positive effect by the co-expression of hydG and sadH was only detected if the sadH-hydG operon was integrated into the genome of C. saccharoperbutylacetonicum.Plasmid-based expression of the operon did not lead to enhanced acetone reduction [20].The co-expression of hydG and sadH was also tested in this study to improve isopropanol concentrations and reduce acetone accumulation.Since the construction of a plasmid harboring all enzymes required for isopropanol production and hydG was not successful, a two-plasmid system was established.Therefore, a plasmid harboring the gene hydG under the control of the P thlA promoter was constructed.The effect of the newly constructed plasmid was investigated using the best performing isopropanol producers of previous experiments.Thereby, recombinant strains harboring two plasmids showed both a significantly inferior isopropanol and also total solvent production.Further analysis of HydG exposed several σ 54 interaction domains, indicating that the protein acts as a σ 54 -dependent transcription activator.The sigma factor σ 54 is involved in the regulation of gene transcription via binding to the RNA polymerase.The σ 54 holoenzyme is not sufficient to bind to the RNA polymerase, but it requires an activator protein, which is bound by the hydrolysis of ATP.The transcription regulating complex binds to the conserved −24 (GG)/−12 (GC) promoter sequence, which was found in the intergenic region between hydG and sadH in the genome of C. beijerinckii DSM 6423 [26,27].For that reason, a new plasmid was constructed harboring the sadH-hydG operon including the σ 54 -binding promoter sequence in the intergenic region.The effect of the newly constructed plasmid was examined in combination with the acetone production plasmid pJIR750_ac3t3 to maintain the copy number of sadH.In comparison to isopropanol production with A. woodii [pJIR750_ac3t3s1], the co-expression of both plasmids resulted in similar isopropanol accumulation.Based on those results, it remains unclear whether active HydG is expressed.It is also not known if σ 54 is expressed in A. woodii, although the gene encoding the sigma factor is present in the genome.Transcriptomic analysis or fusion of hydG to the fluorescence-activating and absorption-shifting tag (FAST) for expression control would be clarifying [46].As acetogens are living at the thermodynamic limit of life, ATP restriction could also hinder the ATP-dependent binding of hydG to σ 54 .A second solution to enhance acetone reduction is the multiplication of sadH gene copy numbers.With this approach, isopropanol production could be successfully increased to 7.02 ± 0.44 mM.
A second bottleneck discovered in the isopropanol production pathway is the efficiency of the CoA-transferase.Via examination of the ctfA/ctfB and thlA genes originating from different bacterial species, isopropanol production could be increased from an initial 0.76 ± 0.25 mM to a maximum 5.64 ± 1.08 mM, which means a 7.5-fold improvement.For the recombinant production of isopropanol using metabolically engineered C. autoethanogenum, insertion of a second gene copy of ctfA/ctfB led to significantly raised isopropanol levels [2].The same approach was considered for improving isopropanol production using A. woodii.The best performing A. woodii strain, harboring two gene copies of both sadH and ctfA/ctfB produced 13.87 ± 1.97 mM (0.83 g/L) of isopropanol, indicating an increase by more than 18-fold compared to the initial isopropanol accumulation.This is still only a third of the production ability of C. autoethanogenum, but provides so far the most successful example of isopropanol formation from CO 2 + H 2 under anaerobic conditions.It can be summarized that the heterologous production of isopropanol using recombinant A. woodii strains is possible.Variation in gene origin as well as the addition of multiple gene copies can help to improve the isopropanol yield.In this case, gene variation and the

Figure 1 .
Figure 1.Production of isopropanol using recombinant A. woodii strains, harboring the secondary alcohol dehydrogenase (sadH) gene of C. beijerinckii DSM 6423, cultivated with 60 mM fructose (A-C) or CO2 + H2 (D-F) as substrate.Growth was performed in biological triplicates with A. woodii wild-type and A. woodii [pJIR750] as control strains (depicted as black line with filled circles or filled black column and dashed black line with empty circles or dashed black column, respectively).During growth, (A,D) the optical density at 600 nm (OD600), (B,E) changes in pH and (C,F) product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac1t1s1] and A. woodii [pJIR750_ac1t2s1] are depicted as dark blue lines with filled rectangles or filled dark blue column and dashed dark blue lines with empty rectangles or dashed dark blue column, respectively.A. woodii [pJIR750_ac2t1s1] and A. woodii [pJIR750_ac2t2s1] are shown as light blue lines with filled rhombus or light blue column and dashed light blue lines with empty rhombus or dashed light blue column, respectively.The pastel blue lines with filled triangle or pastel blue column depict A. woodii [pJIR750_ac3t3s1].

Figure 1 .
Figure 1.Production of isopropanol using recombinant A. woodii strains, harboring the secondary alcohol dehydrogenase (sadH) gene of C. beijerinckii DSM 6423, cultivated with 60 mM fructose (A-C) or CO 2 + H 2 (D-F) as substrate.Growth was performed in biological triplicates with A. woodii wild-type and A. woodii [pJIR750] as control strains (depicted as black line with filled circles or filled black column and dashed black line with empty circles or dashed black column, respectively).During growth, (A,D) the optical density at 600 nm (OD 600 ), (B,E) changes in pH and (C,F) product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac1t1s1] and A. woodii [pJIR750_ac1t2s1] are depicted as dark blue lines with filled rectangles

Figure 2 .
Figure 2. Production of isopropanol using recombinant A. woodii strains, harboring the sadH gene of C. beijerinckii DSM 15410, cultivated with 60 mM fructose (A-C) or CO2 + H2 (D-F) as substrate.Growth was performed in biological triplicates with A. woodii wild-type and A. woodii [pJIR750] as control strains (depicted as black line with filled circles or filled black column and dashed black line with empty circles or dashed black column, respectively).During growth, (A,D) OD600, (B,E) changes in pH and (C,F) product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac1t1s2] and A. woodii [pJIR750_ac1t2s2] are depicted as green lines with filled rectangles or filled green column and dashed green lines with empty rectangles or dashed green column, respectively.A. woodii [pJIR750_ac2t1s2] and A. woodii [pJIR750_ac2t2s2] are shown as light green lines with filled rhombus or filled light green column and dashed light green lines with empty rhombus or dashed light green column, respectively.The brown-green lines with filled triangle or brown-green column depict A. woodii [pJIR750_ac3t3s2].

Figure 2 .
Figure 2. Production of isopropanol using recombinant A. woodii strains, harboring the sadH gene of C. beijerinckii DSM 15410, cultivated with 60 mM fructose (A-C) or CO 2 + H 2 (D-F) as substrate.Growth was performed in biological triplicates with A. woodii wild-type and A. woodii [pJIR750] as control strains (depicted as black line with filled circles or filled black column and dashed black line with empty circles or dashed black column, respectively).During growth, (A,D) OD 600 , (B,E) changes in pH and (C,F) product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac1t1s2] and A. woodii [pJIR750_ac1t2s2] are depicted as green lines with filled rectangles or filled green column and dashed green lines with empty rectangles or dashed green column, respectively.A. woodii [pJIR750_ac2t1s2] and A. woodii [pJIR750_ac2t2s2] are shown as light green lines with filled rhombus or filled light green column and dashed light green lines with empty rhombus or dashed light green column, respectively.The brown-green lines with filled triangle or brown-green column depict A. woodii [pJIR750_ac3t3s2].

Figure 3 .
Figure 3. Production of isopropanol using recombinant A. woodii strains, harboring the sadH gene of C. ljungdahlii, cultivated with 60 mM fructose (A-C) or CO2 + H2 (D-F) as substrate.Growth was performed in biological triplicates with A. woodii wild-type and A. woodii [pJIR750] as control strains (depicted as black line with filled circles or filled black column and dashed black line with empty circles or dashed black column, respectively).During growth, (A,D) OD600, (B,E) changes in pH and (C,F) product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac1t1s3] and A. woodii [pJIR750_ac1t2s3] are depicted as pink lines with filled rectangles or filled pink column and dashed pink lines with empty rectangles or dashed pink column, respectively.A. woodii [pJIR750_ac2t1s3] and A. woodii [pJIR750_ac2t2s3] are shown as purple lines with filled rhombus or filled purple column and dashed purple lines with empty rhombus or dashed purple column, respectively.The rosa lines with filled triangle or rosa column depict A. woodii [pJIR750_ac3t3s3].

Figure 3 .
Figure 3. Production of isopropanol using recombinant A. woodii strains, harboring the sadH gene of C. ljungdahlii, cultivated with 60 mM fructose (A-C) or CO 2 + H 2 (D-F) as substrate.Growth was performed in biological triplicates with A. woodii wild-type and A. woodii [pJIR750] as control strains (depicted as black line with filled circles or filled black column and dashed black line with empty circles or dashed black column, respectively).During growth, (A,D) OD 600 , (B,E) changes in pH and (C,F) product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac1t1s3] and A. woodii [pJIR750_ac1t2s3] are depicted as pink lines with filled rectangles or filled pink column and dashed pink lines with empty rectangles or dashed pink column, respectively.A. woodii [pJIR750_ac2t1s3] and A. woodii [pJIR750_ac2t2s3] are shown as purple lines with filled rhombus or filled purple column and dashed purple lines with empty rhombus or dashed purple column, respectively.The rosa lines with filled triangle or rosa column depict A. woodii [pJIR750_ac3t3s3].

Figure 4 .
Figure 4. Production of isopropanol using recombinant A. woodii strains, harboring the hydG gene of C. beijerinckii DSM 6423, cultivated with 60 mM fructose as substrate.Growth was performed in biological triplicates with A. woodii wild-type, A. woodii [pJIR750] and A. woodii [pJIR750] [pMTL83251] as control strains (depicted as black line with filled circles or filled black column, dashed black line or column and do ed black line or column with empty circles, respectively).During growth, (A) OD600, (B) changes in pH and (C) maximum product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac3t3s1] and A. woodii [pJIR750_ac3t3s1] [pMTL83251_PthlA_h1] are depicted as pastel blue lines with filled rectangles or filled pastel blue columns and dashed pastel blue lines with empty rectangles or dashed pastel blue columns, respectively.A. woodii [pJIR750_ac3t3s2] and A. woodii [pJIR750_ac3t3s2] [pMTL83251_PthlA_h1] are shown as brown-green lines with filled rhombus or filled brown-green columns and dashed brown-green lines with empty rhombus or dashed brown-green column, respectively.The rosa lines with filled triangle or filled rosa columns and dashed rosa lines with empty rectangles or dashed rosa columns depict A. woodii [pJIR750_ac3t3s3] and A. woodii [pJIR750_ac3t3s3] [pMTL83251_PthlA_h1], respectively.

Figure 4 .
Figure 4. Production of isopropanol using recombinant A. woodii strains, harboring the hydG gene of C. beijerinckii DSM 6423, cultivated with 60 mM fructose as substrate.Growth was performed in biological triplicates with A. woodii wild-type, A. woodii [pJIR750] and A. woodii [pJIR750] [pMTL83251] as control strains (depicted as black line with filled circles or filled black column, dashed black line or column and dotted black line or column with empty circles, respectively).During growth, (A) OD 600 , (B) changes in pH and (C) maximum product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac3t3s1] and A. woodii [pJIR750_ac3t3s1] [pMTL83251_PthlA_h1] are depicted as pastel blue lines with filled rectangles or filled pastel blue columns and dashed pastel blue lines with empty rectangles or dashed pastel blue columns, respectively.A. woodii [pJIR750_ac3t3s2] and A. woodii [pJIR750_ac3t3s2] [pMTL83251_PthlA_h1] are shown as brown-green lines with filled rhombus or filled brown-green columns and dashed browngreen lines with empty rhombus or dashed brown-green column, respectively.The rosa lines with filled triangle or filled rosa columns and dashed rosa lines with empty rectangles or dashed rosa columns depict A. woodii [pJIR750_ac3t3s3] and A. woodii [pJIR750_ac3t3s3] [pMTL83251_PthlA_h1], respectively.

Figure 5 .
Figure 5. Production of isopropanol using recombinant A. woodii strains, harboring the sadH-hydG gene cluster of C. beijerinckii DSM 6423 and an additional copy of ctfA/ctfB originating from C. scatologenes, cultivated with 60 mM fructose (A-C) or CO2 + H2 (D-F) as substrate.Growth was performed in biological triplicates with A. woodii wild-type, A. woodii [pJIR750] and A. woodii [pJIR750] [pMTL83251] as control strains (depicted as black line with filled circles or filled black column, dashed black line or column and do ed black line or column with empty circles, respectively).During growth, (A,D) OD600, (B,E) changes in pH and (C,F) maximum product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac3t3], A. woodii [pJIR750_ac3t3] [pMTL83251_PthlA_sh1] and A. woodii [pJIR750_ac3t3] [pMTL83251_PthlA_sh1c3] are depicted as dark blue lines or column with filled rhombus, dashed dark blue lines or column with empty rhombus and do ed dark blue lines or column with empty rhombus, respectively.A. woodii [pJIR750_ac3t3s1], A. woodii [pJIR750_ac3t3s1] [pMTL83251_PthlA_sh1] and A. woodii [pJIR750_ac3t3s1] [pMTL83251_PthlA_sh1c3] are shown as pastel blue lines or column with filled triangles, dashed pastel blue lines or column with empty triangles and do ed pastel blue lines or column with empty triangles, respectively.

Figure 5 .
Figure 5. Production of isopropanol using recombinant A. woodii strains, harboring the sadH-hydG gene cluster of C. beijerinckii DSM 6423 and an additional copy of ctfA/ctfB originating from C. scatologenes, cultivated with 60 mM fructose (A-C) or CO 2 + H 2 (D-F) as substrate.Growth was performed in biological triplicates with A. woodii wild-type, A. woodii [pJIR750] and A. woodii [pJIR750] [pMTL83251] as control strains (depicted as black line with filled circles or filled black column, dashed black line or column and dotted black line or column with empty circles, respectively).During growth, (A,D) OD 600 , (B,E) changes in pH and (C,F) maximum product concentrations of acetone and isopropanol were monitored.The recombinant strains A. woodii [pJIR750_ac3t3], A. woodii [pJIR750_ac3t3] [pMTL83251_PthlA_sh1] and A. woodii [pJIR750_ac3t3] [pMTL83251_PthlA_sh1c3] are depicted as dark blue lines or column with filled rhombus, dashed dark blue lines or column with empty rhombus and dotted dark blue lines or column with empty rhombus, respectively.A. woodii [pJIR750_ac3t3s1], A. woodii [pJIR750_ac3t3s1] [pMTL83251_PthlA_sh1] and A. woodii [pJIR750_ac3t3s1] [pMTL83251_PthlA_sh1c3] are shown as pastel blue lines or column with filled triangles, dashed pastel blue lines or column with empty triangles and dotted pastel blue lines or column with empty triangles, respectively.

Table 2 .
Plasmids used in this work.

Table 3 .
Designation scheme for the genes of the isopropanol production plasmids according to their origin.
* a is used as abbreviation for adc originating from C. acetobutylicum.