WLP is the key pathway for acetogens to fix CO₂ and generate acetate, as illustrated in Figure 1 (Ljungdahl, 1986; Wood et al., 1986). WLP consists of two branches, the methyl branch and the carbonyl branch. The methyl branch is a linear metabolic branch that converts CO₂ to methyl groups through six sequential steps. When acetogens autotrophically grow on CO₂/H₂, the first reaction of methyl branch is the reduction of CO₂ to formate catalyzed by formate dehydrogenase (FDH) using two electrons. Formate then undergoes a reaction with tetrahydrofolate (THF) to form formyl-THF, catalyzed by formyl-THF synthetase (FTS) and consuming one ATP. Subsequently, formyl-THF is converted to methenyl-THF by formyl-THF cyclohydrolase (FTC), leading to the release of one H₂O. The reduction of methenyl-THF to methylene-THF, and eventually to methyl-THF, occurs through the successive actions of methylene-THF dehydrogenase (MDH) and methylene-THF reductase (MTHFR), with each step involving the consumption of two electrons. In the last step of methyl branch, the methyl group of methyl-THF is transferred to a corrinoid iron-sulfur-containing protein (CoFeSP) via the methyltransferase (MT). The reducing steps of methyl branch require a total of six electrons, while the specific coenzymes carrying the reducing equivalents are diverse and still partially unrevealed (Drake and Daniel, 2004; Schuchmann and Müller, 2016). The carbonyl-branch involves a single reaction facilitated by a multi-component enzyme complex known as carbon monoxide dehydrogenase/acetyl-CoA synthase (CODH/ACS), which catalyzes the reduction of CO₂ to CO using two electrons. In the final step of WLP, the bi-functional CODH/ACS enzyme complex catalyzes the reaction that converts CO, methyl-CoFeSP and CoA into acetyl-CoA. Certain acetogens like Moorella thermoacetica and Clostridium ljungdahlii exhibit the capability of thriving using CO as sole carbon and energy source. In this situation, CO₂ is generated from CO through the catalysis of CODH initially, and then enters the methyl branch (Ragsdale, 2004). Consequently, CO directly binds to the CODH/ACS. Ultimately, acetyl-CoA is phosphorylated to acetyl phosphate and further metabolized to acetate by phosphotransacetylase (PTA) and acetate kinase (ACK), releasing one molecule of ATP.

As described above, the fundamental chemistry of the methyl branch is the same in different acetogenic species. However, the enzymes that catalyze the reduction of CO₂ to formate and the reduction of methylene-THF to methyl-THF are different, involving different coenzymes. In the step of CO₂ reduction to formate, the standard redox potential (E₀ʹ) of the CO₂/formate couple is −432 mV (Thauer et al., 1977). Therefore, oxidation of NADH (E₀ʹ = −320 mV) is not sufficient to drive this reaction. Instead, reduced ferredoxin [(ferredoxin (Fd)/reduced ferredoxin (Fd²⁻)) E₀ʹ = −450 mV], NADPH [(NADP⁺/NADPH) E₀ʹ = −370 mV], and H₂ [(H⁺/H₂) E₀ʹ = −414 mV] all have physiological redox potentials that can potentially drive the reaction (Schuchmann and Müller, 2014). Different acetogens have evolved to use different electron donors (Figure 1A). In the model organism Acetobacterium woodii, a hydrogen-dependent CO₂ reductase (HDCR) directly employs H₂ as the electron donor for CO₂ reduction (Schuchmann and Müller, 2013). The HDCR complex comprises a hydrogenase module (HydA2) and a formate dehydrogenase module (FdhF1 or FdhF2), linked by two electron-transferring subunits (HycB2 and HycB3). In the case of M. thermoacetica, the reduction of CO₂ to formate is catalyzed by a two-subunit NADPH-dependent FDH, despite the unfavorable bioenergetics with NADPH (E₀ʹ = −370 mV) (Yamamoto et al., 1983; Rosenbaum and Müller, 2023). The genome of C. ljungdahlii harbors three gene clusters encoding potential FDHs (Schuchmann and Müller, 2014). The first gene cluster (Clju_c06990–07080), encodes an FDH and an electron-bifurcating hydrogenase, and this complex putatively use either H₂ or Fd^2- and NADPH as electron donors (Wang et al., 2013a; Nagarajan et al., 2013; Mock et al., 2015). The second gene cluster (Clju_c20030–20040) encodes an FDH and a small ferredoxin-like protein, and this complex presumably accepts electrons from Fd^2-. The third FDH is encoded by Clju_c08930, an isogene of Clju_c20040 (Schuchmann and Müller, 2014).

The MTHFR that catalyzes the reduction of methylene-THF to methyl-THF also exhibits diversity among various acetogens (Figure 1B). The MTHFR complex from A. woodii is composed of MetF, MetV, and RnfC2 in a stoichiometry of 1:1:1, with NADH serving as the electron donor (Ragsdale and Ljungdahl, 1984). In M. thermoacetica, the genes metVF encoding the MTHFR are co-transcribed together with the genes hdrCBA-mvhD. The complex MetFV-HdrABC-MvhD is hypothesized to be an electron-bifurcating enzyme, which uses NADH as reductant (Mock et al., 2014). However, the second electron acceptor is unknown yet. The MetF-MetV complex from C. ljungdahlii has been purified and characterized, indicating that the enzyme uses Fd^2- as the electron donor and is not capable of electron bifurcation (Yi et al., 2021). However, a proposed energy metabolic scheme suggests that this process cannot physiologically occur, as it would result in a negative ATP gain during autotrophic acetogenesis. It is postulated that the MTHFR may assemble into a large, flexible complex involving other electron-bifurcating enzymes (Öppinger et al., 2022).

Acetate kinase reaction during acetogenesis synthesizes one ATP via substrate-level phosphorylation (SLP). While, the activation of formate consumes one ATP, resulting in zero net ATP yield through SLP. This highlights a chemiosmotic energy conservation mechanism for ATP synthesis in acetogens, which relies on an Fd^2--driven ion pump complex (Figure 2). There are two different ion pump complexes in acetogens, Rnf (Rhodobacter nitrogen fixation) and Ech (Energy-converting hydrogenase) (Buckel and Thauer, 2013). The Rnf complex found in A. woodii, C. ljungdahlii, and Clostridium autoethanogenum, is a ferredoxin:NAD⁺ oxidoreductase, and it pumps Na⁺ or H⁺ outside the cell when transferring electrons from Fd^2- to NAD⁺ (Tremblay et al., 2012; Hess et al., 2013; Mock et al., 2015). In contrast, the Ech complex found in M. thermoacetica and Thermoanaerobacter kivui has ferredoxin:H⁺ oxidoreductase activity and pumps H⁺ (Katsyv and Müller, 2022; Rosenbaum and Müller, 2023). The transmembrane H⁺ or Na⁺ gradient created by Rnf or Ech drives the ATP synthase to generate ATP (Schoelmerich and Müller, 2019). In comparison to A. woodii, C. ljungdahlii, and C. autoethanogenum, M. thermoacetica exhibits a notable difference in the presence of cytochromes and quinones, which probably contribute to energy conservation (Gottwald et al., 1975; Rosenbaum and Müller, 2023). M. thermoacetica harbors a putative “headless” NADH dehydrogenase that may function as an H⁺ translocating enzyme, driving the ATP synthase by utilizing Fd^2- as an electron donor and quinone as an electron acceptor (Pierce et al., 2008). The largest thermodynamic barrier in the WLP is the reduction of CO₂ to CO, as the redox couple has a very low standard redox potential (E₀ʹ = −520 mV) (Thauer et al., 1977). Among the identified coenzymes in the WLP, only Fd^2- can provide electrons for the reaction (Schuchmann and Müller, 2014).

Every electron for autotrophic acetogenesis from H₂ and CO₂ is derived from the oxidation of hydrogen, facilitated by hydrogenases (Figure 2) (Schuchmann and Müller, 2014; Di Leonardo et al., 2022). A. woodii has a four-subunit soluble hydrogenase, HydABCD, which contains iron-sulphur centers and flavin (Schuchmann and Müller, 2012). The electron-bifurcating HydABCD catalyzes the oxidation of H₂, coupling the endergonic reduction of Fd and the exergonic reduction of NAD⁺. Fd and NAD⁺ are reduced simultaneously in a 1/1 stoichiometry (Schuchmann and Müller, 2012). The electron-bifurcating hydrogenases from C. ljungdahlii and C. autoethanogenum are composed of six subunits (HytABCDE1E2) and are NADP⁺ and Fd-dependent. The hydrogenase forms a tight complex with the formate dehydrogenase FdhA (Wang et al., 2013a; Nagarajan et al., 2013; Mock et al., 2015). The electron-bifurcating hydrogenase of M. thermoacetica, HydABC consists of three subunits, which catalyzes the oxidation of hydrogen with NAD⁺ and Fd as electron acceptors (Wang et al., 2013b). Part of the Fd²⁻ is used in the reductive reactions of acetogenesis and the remaining Fd²⁻ can be used by the Rnf or Ech complex to generate a chemiosmotic gradient. In addition, due to the involvement of different cofactors in the energy metabolism, some acetogens such as C. ljungdahlii, C. autoethanogenum, and M. thermoacetica have an electron-bifurcating ferredoxin-dependent transhydrogenase (Nfn complex) (Huang et al., 2012; Nagarajan et al., 2013; Mock et al., 2015). It is capable of producing NADPH from Fd²⁻ and NADH through electron bifurcation, regulating the redox pool.

In recent years, the understanding of acetogenic metabolism has gradually deepened, which supports the metabolic engineering of acetogens in principles. Meanwhile, complete genomic sequences have been published for the species including A. woodii, M. thermoacetica, C. ljungdahlii, C. autoethanogenum, and Clostridium aceticum, making genetic modification in those acetogens possible.

An efficient genetic system relies on a reliable DNA transfer method. DNA transfer in most acetogens is limited due to their Gram-positive cell wall (Bourgade et al., 2021). Currently, two usually used transformation methods for introducing foreign DNA into acetogens are electroporation and conjugation. The highly efficient electroporation of acetogens is usually described as treating early-log phase cells with cell-wall-weakening agents such as glycine or DL-threonine and osmotic stabilizers like sucrose; and then collecting and preparing the cells into competent cells; after that, electroporation is performed (Köpke et al., 2010; Kita et al., 2013; Leang et al., 2013; Straub et al., 2014; Shin et al., 2019; Zhao et al., 2019). The transformation efficiency varies depending on the species of the host cell and the type of plasmids, ranging from approximately 10¹–10⁵ CFU per μg DNA (Zhao et al., 2019; Moreira et al., 2023). The DNA transfer by conjugation involves a donor bacterial strain such as Escherichia coli to transfer single-stranded DNA to the acceptor cells through cell-to-cell contact. Conjugation is applied in C. autoethanogenum (Mock et al., 2015; Nagaraju et al., 2016) and Clostridium carboxidivorans (Cheng et al., 2019). It is worth noting that thermophilic acetogen T. kivui naturally took up plasmid DNA in exponential growth phase, with a transformation frequency of up to 3.9 × 10⁻⁶ (Basen et al., 2018).

The native restriction modification (RM) systems are widespread in acetogens, which is a barrier to acetogen transformation. RM systems consist of restriction endonucleases and corresponding methyltransferases. The endonucleases recognize and cut specific DNA sequences, while the methyltransferases methylate the corresponding sequences in the host genome to protect it from cleavage (Vasu and Nagaraja, 2013). Among the four types of RM systems identified in bacteria, types I, II, and III target and cleave unmethylated DNA, while type IV recognizes DNA with foreign methylation patterns. For hosts with types I, II, or III RM systems, to circumvent the protective systems, plasmids can be methylated prior to transformation. One feasible approach is to express the native methyltransferases of the host in E. coli to create a methylation strain (Yasui et al., 2009). An example is the transformation of M. thermoacetica, where three native methyltransferases were expressed in E. coli to methylate the plasmid in order to counter the type I and type II RM systems of M. thermoacetica (Kita et al., 2013). RM systems in acetogens are often species-specific. Updated information of identified restriction endonucleases and methyltransferases is available in REBASE database (Roberts et al., 2023). In addition, conjugation is believed to allow partial evasion of the host restriction-modification barriers (Purdy et al., 2002; Cheng et al., 2019). Some acetogens like C. autoethanogenum have type IV RM systems, and plasmids constructed from dam⁺/dcm⁺ E. coli strains like DH5α will be recognized and degraded (Woods et al., 2019). To prevent this, plasmids can be constructed in E. coli strains with different dam/dcm background. One successful example is the transformation of Clostridium sp. AWRP, which has never succeeded in electroporation with dam⁺/dcm⁺ DNA, whereas achieved transformation efficiency of 3.2 × 10² CFU/μg DNA with dam⁻/dcm⁻ DNA (Kwon et al., 2024). In addition to electroporation, the problem is also present in conjugative transfer. When C. autoethanogenum was the recipient, the commonly used conjugative donor strain E. coli CA434 (dam⁺/dcm⁺) exhibited a conjugation efficiency of 1.05 × 10⁻⁷ per donor cell. While the donor strain of E. coli sExpress which was obtained by introducing R-factor R702 into E. coli NEB Express strain (dam⁺/dcm⁻) had an efficiency of 7.94 × 10⁻⁵ per donor cell (Woods et al., 2019).

Genetic manipulation of acetogens inevitably involves the introduction of foreign DNA, which requires plasmid vectors. Acetogens are Gram-positive bacteria, whereas recombinant plasmids are typically constructed and amplified in E. coli and then “shuttled” into the target acetogens. Therefore, shuttle vectors must have both Gram-positive and Gram-negative replicons. The Gram-positive replicons are often derived from natural Gram-positive bacterial plasmids, such as pBP1 from Clostridium botulinum (van Treeck et al., 1981), pCB102 from Clostridium butyricum (Collins et al., 1985), pIM13 from Bacillus subtilis (Monod et al., 1986), pIP404 from Clostridium perfringens (Garnier and Cole, 1988), and pCD6 from Clostridium difficile (Minton et al., 2004). Recently, a low-copy native plasmid pCA from C. autoethanogenum was characterized and applied in heterologous expression, which may further facilitate physiological study and metabolic engineering of acetogens (Nogle et al., 2022). Table 1 shows a list of the Gram-positive replicons in shuttle plasmids that have been applied in acetogens. In certain cases, plasmids do not require replication and stable existence within the acetogen host. These plasmids, known as suicide vectors, have Gram-negative but not Gram-positive origins. Hence, they can be constructed and amplified in E. coli and transformed into acetogens for genomic editing (Tremblay et al., 2012). The Gram-negative replicons that are commonly used in E. coli-Acetogen shuttle vectors are ColE1, p15a, and pUC (Heap et al., 2009). In addition, antibiotic resistance genes including ermB, catP, and tetA are most extensively used selection markers for recombinant screening of acetogens. For thermophilic acetogens like M. thermoacetica, utilizing an auxotrophic mutant as a screening tool may be an optimal choice (Kita et al., 2013; Basen et al., 2018). In addition, the thermostable kanamycin resistance gene derived from Streptococcus faecalis (Trieucuot and Courvalin, 1983) has been used in thermophilic acetogens, including M. thermoacetica (Iwasaki et al., 2013) and T. kivui (Basen et al., 2018).

In efforts to improve the availability of shuttle plasmids tailored for clostridial hosts and specific applications, the pMTL80000 modular plasmid system was designed (Heap et al., 2009). This system facilitates the versatile construction of Clostridium-E. coli shuttle plasmids by integrating standardized modules comprising Gram-positive and Gram-negative replicons, selection markers, and other functional modules. The plasmid series has been extensively applied in acetogens, including Clostridium species such as C. autoethanogenum (Köpke et al., 2014), C. ljungdahlii (Ueki et al., 2014), and C. carboxidivorans (Cheng et al., 2019), as well as other acetogen species like A. woodii and Eubacterium limosum (Hoffmeister et al., 2016; Shin et al., 2019).

It is feasible to introduce genes into the hosts for overexpressing endogenous genes or heterologous metabolic pathways using replicable plasmids. However, maintaining plasmids for gene expressions in the host cells requires the use of antibiotics, which is unfavorable for industrial production. Moreover, there is a concern regarding the potential spread of resistance markers into the environment. In this context, several genome editing tools allowing for gene disruption and fragment insertion in acetogens have been developed (Figure 3; Table 2).

Acetone, an important commodity chemical and feedstock for poly (methyl methacrylate) production can be biosynthesized by ABE fermentation. The metabolic pathway involved has been well elucidated (Jones and Woods, 1986). Initially, two molecules of acetyl-CoA are converted to acetoacetyl-CoA by a thiolase (ThlA), and then CoA is transferred to acetate by a CoA transferase (CtfAB) to generate acetoacetate and another acetyl-CoA, and finally acetoacetate is decarboxylated by an acetoacetate decarboxylase (Adc) to generate acetone. However, ABE fermentation typically utilizes sugar as a substrate, leading to growing interest in C1 gas fermentation by acetogens. Acetone is not a natural product of any known acetogens. Proof-of-concept production of acetone has been achieved by plasmid-based heterologous expression of acetone synthesis pathways from Clostridium acetobutylicum in several acetogens, such as C. aceticum (Schiel-Bengelsdorf and Dürre, 2012), C. ljungdahlii (Banerjee et al., 2014), and A. woodii (Hoffmeister et al., 2016; Arslan et al., 2022). Studies have shown that acetone can be reduced to isopropanol by an endogenous primary-secondary alcohol dehydrogenase (sADH) in C. autoethanogenum (Köpke et al., 2014) and C. ljungdahlii (Bengelsdorf et al., 2016). The integration of acetone synthesis pathway into the genome of C. ljungdahlii resulted in the production of 0.6 mM acetone and 2.4 mM isopropanol by the engineered strain (Philipps et al., 2019). Instead, Jia et al. (2021) introduced an artificial isopropanol-producing pathway via an expression plasmid incorporating thiolase, CtfAB, Adc, and sADH from Clostridium species into C. ljungdahlii. The best-performing recombinant strain was able to produce 13.4 g/L of isopropanol in continuous gas fermentation. In addition, plasmid-based heterologous expression of the genes thlA-ctfAB-adc-sadh in A. woodii led to a maximum isopropanol production of 13.87 mM using CO₂ and H₂ as the carbon and energy source (Höfele et al., 2023). Recent advancements by LanzaTech have demonstrated commercial-ready levels of acetone and isopropanol production using genomically integrated strains, with continuous production rates reaching up to ∼3 g/L/h (Liew et al., 2022). Life cycle analysis confirmed a negative carbon footprint for their products. Additionally, Kato et al. successfully integrated a thermophilic acetone production pathway into the genome of M. thermoacetica, resulting in acetone yields of 0.04–0.09 g acetone/g-dry cell/h, making a notable contribution to the field of metabolic engineering of thermophilic acetogens (Kato et al., 2021).

3-Hydroxybutyrate (3-HB) is a chiral bioproduct of interest with diverse applications in the synthesis of fine chemicals, pharmaceuticals, biofuels, and bioplastics, particularly polyhydroxybutyrate (PHB) (Tokiwa and Ugwu, 2007). 3-HB can be synthesized from acetyl-CoA through a variety of pathways. There are two main pathways proposed to generate 3-HB in acetogens (Figure 4). Both pathways involve the action of a thiolase to combine two acetyl-CoA molecules, producing acetoacetyl-CoA. However, they differ in subsequent steps. In the Hbd pathway, (S)-3-hydroxybutyryl-CoA dehydrogenase (Hbd) reduces acetoacetyl-CoA to (S)-3-hydroxybutyryl-CoA, which is then converted to 3-HB by a thioesterase catalyzed CoA removal. In contrast, the CtfAB/3-HBdh pathway initially utilizes CtfAB to convert acetoacetyl-CoA to acetoacetate by CoA transfer to acetate. Acetoacetate is then reduced by the 3-HB dehydrogenase (3-HBdh), resulting in the formation of 3-HB. For the heterologous production of 3-HB, a novel CtfAB/3-HBdh pathway consisting of the genes thlA and ctfA/B from C. acetobutylicum, as well as bdhA encoding a 3-HBdh from Clostridioides difficile 630 was introduced into C. coskatii and C. ljungdahlii through an expression plasmid (Flüchter et al., 2019). The engineered strain of C. coskatii autotrophically grown on syngas produced 0.1 g/L of 3-HB. The recombinant C. ljungdahlii strain, however, did not produce any detectable amounts of 3-HB neither under heterotrophic nor under autotrophic growth conditions. Jones et al. (2016) proposed that acetoacetate can undergo native conversion to 3-HB in C. ljungdahlii, as their acetone-generating strain, containing all the necessary genes for the CtfAB/3HBdh pathway except for 3-HBdh itself, produced significant levels of 3-HB. Subsequently, an endogenous 3-HB dehydrogenase (3-HBdh) was identified in C. ljungdahlii, which could convert acetoacetate to 3-HB in an engineered strain for isopropanol production (Jia et al., 2021). Furthermore, the introduction of the bdhA gene from C. difficile 630 via an expression plasmid into the isopropanol-producing strain facilitated the efficient co-production of isopropanol, 3-HB, and ethanol, resulting in a 3-HB titer of 3 g/L in continuous gas fermentations (Jia et al., 2021). Through in vitro prototyping and rapid biosynthetic enzyme optimization from Clostridium, the plasmid-based expression of an Hbd pathway in C. autoethanogenum demonstrated the highest performance in 3-HB production (Karim et al., 2020). In optimized fermentations, the titer of 3-HB reached 14.6 g/L, with production rates exceeding 1.5 g L⁻¹ h⁻¹ in a continuous system. Recently, Lo et al. (2022) identified a native (S)-3-hydroxybutyryl-CoA dehydrogenase, Hbd2 in C. ljungdahlii. In conjunction with the integration of heterologous thiolase gene, atoB from E. coli and ctfAB from C. acetobutylicum into the pyrE locus, plasmid based overexpression of both the native hbd2 and thl2 from Clostridium kluyveri significantly improved yields of 3-HB on syngas, resulting in a titer of 9.2 g/L (Lo et al., 2022).

Poly-3-hydroxybutyrate (PHB) is a biodegradable plastic naturally produced by Cupriavidus necator and microbes from Alcaligenes, Azotobacter, Bacillus, Rhizobium, and Pseudomonas (Bhagowati et al., 2015; Jin et al., 2020). PHB is synthesized through the polymerization of 3-hydroxybutyryl-CoA by PHA synthase. For the production of PHB, the PHB synthesis pathway from C. necator was expressed in C. autoethanogenum under control of the native WLP promoter. Through bioprocess optimization, a PHB yield of 5.6% per CDW with a titer of 0.027 g/L was obtained on syngas in steady-state chemostat cultures (dilution rate of 1 day⁻¹) (De Souza Pinto Lemgruber et al., 2019). In addition, as a proof of concept, production of PHB was also achieved using recombinant C. coskatii and C. ljungdahlii strain expressing a Clostridial PHB synthesis pathway comprising the genes thlA, hbd, crt, phaJ, and phaEC (Flüchter et al., 2019) (Figure 4). The recombinant strains of both C. coskatii and C. ljungdahlii synthesized autotrophically 1.2% PHB per CDW.

Butyrate and butanol are also non-natural products that cannot be produced by wild-type C. ljungdahlii and C. autoethanogenum. In order to redirect carbon and electron flow to the production of butyrate, a fragment containing genes of thlA, hbd, crt, bcd, etfAB, ptb, and buk from C. acetobutylicum was integrated into the putative pta promoter region of C. ljungdahlii. Meanwhile, PTA-dependent acetate synthesis and a gene (Clju_c39430) encoding a homolog of CoA transferases, were disrupted. The final recombinant strain produced approximately 1.6 g/L of butyrate on CO/CO₂ (Ueki et al., 2014). In addition, to validate the efficiency of the phage serine integrase-mediated genome engineering, a butyrate production pathway from C. acetobutylicum was integrated into C. ljungdahlii, enabling the production of 1.01 g/L of butyrate by fermenting syngas (Huang et al., 2019).

The production of butanol by engineered acetogens presents challenges due to requirement of additional reducing equivalents for the conversion of butyryl-CoA into butanol. In a demonstration of principle, a plasmid containing most of the C. acetobutylicum butanol synthesis pathway, encompassing thlA, hbd, crt, bcd, adhE, and bdhA genes (Figure 4) under control of the phosphotransbutyrylase operon promoter P _ptb was introduced into C. ljungdahlii and a maximum amount of about 0.15 g/L butanol was detected in batch cultures of the engineered strain during mid-growth phase (Köpke et al., 2010). Another approach was carried out in C. autoethanogenum, which also expressed the C4 pathway of C. acetobutylicum including thlA, hbd, crt, bcd, etfAB genes (Köpke and Liew, 2011). The highest butanol production measured in the recombinant strain culture grown on CO was 1.54 g/L. In a recent study, an artificial biosynthesis pathway for butanol and hexanol was developed, incorporating 13 genes from C. kluyveri and C. acetobutylicum and integrated into the genome of C. ljungdahlii. In 2-L fermentations with CO₂ and H₂ continuous supply, the recombinant strain produced 451 mg/L of butanol, 122 mg/L of hexanol, as well as 199 mg/L of butyrate and 46 mg/L of caproate (Lauer et al., 2022). Vögeli et al. (2022) optimized and implemented the reverse β-oxidation pathway for the synthesis of C4-C6 acids and alcohols. Then engineered C. autoethanogenum strain was able to produce 1-hexanol from syngas, achieving a titer of 0.26 g/L in a 1.5 L continuous fermentation.

When utilizing 2,3-butanediol (2,3-BDO) as the substrate, the biosynthesis of 2-butanol involves two reactions: meso-2,3-BDO is converted to 2-butanone (MEK) by propanediol/glycerol dehydratase, and further reduced to 2-butanol by alcohol dehydrogenase (Adh) using NADPH. Acetogens such as C. ljungdahlii and C. autoethanogenum are capable of producing 2,3-BDO. However, the production of MEK and 2-butanol requires (R,S)-2,3-butanediol, not (R,R)-2,3-butanediol. Nevertheless, through the use of recombinant microorganisms adapted to overexpress crucial enzymes from Klebsiella species in the MEK and 2-butanol biosynthesis pathways, the production of MEK and 2-butanol was presumptively achieved (Mueller et al., 2018).

Isoprene is a hemiterpene (C5H8), used primarily in the production of cis-1,4-polyisoprene, the main component of natural rubber, and is also used in the creation of styrene-isoprene-styrene block copolymers, commonly employed in thermoplastic rubber and adhesives. Isopentenyl-pyrophosphate (IPP) and dimethylallyl pyrophosphate (DMAPP) are the initial precursors for all terpenoid biosynthesis, including isoprene (Schrader and Bohlmann, 2015). There are two distinct pathways existing for the generation of IPP and DMAPP: the 2-C-methyl-D-erythritol 4-phosphate (MEP) pathway, prevalent in most prokaryotes and starts with the condensation of pyruvate and glyceraldehyde-3-phosphate (GA3P), and the mevalonic acid (MVA) pathways, found in most eukaryotes, archaea, and some bacteria, which generally starts with the condensation of two acetyl-CoA molecules (Schrader and Bohlmann, 2015). In an effort to enable isoprene production from syngas, the MVA pathway was introduced into C. ljungdahlii, along with the overexpression of IPP isomerase and isoprene synthase (ipsS). However, the titer of isoprene during syngas fermentation only reached approximately 1.5 μg/L (Diner et al., 2017). Regrettably, the process of synthesizing isoprene from C1 gas is highly energy-intensive and requires additional ATP, which poses a significant challenge for acetogens with limited capacity for ATP generation. Therefore, efficient synthesis of isoprene is only possible by coupling it with other energy-generating pathways.

As mentioned above, efforts to genetically modify acetogens for the production of non-natural products have yielded only modest success, with low production efficiency that does not meet the demands of commercial scale production, except a few cases. This may be attributed to insufficient energy supply at autotrophic conditions, as acetogens thrive at the thermodynamic limit of life, and their ATP generation process is inherently impeded (Schuchmann and Müller, 2014). As a result, exogenous metabolic pathway modules introduced through genetic modification can only function sub-optimally under such energy constraints.

The synthesis of some non-natural products is thermodynamically favorable for acetogens, such as butyrate, acetone, and isopropanol. However, the production efficiency was still low in some earlier studies, which was caused by the limited metabolic flow.