The WL pathway is mainly composed of two linear metabolic branches: a methyl- and a carbonyl-branch (Figure 1). The methyl-branch consists of six reactions, starting with the reduction of CO2 to formate using two reducing equivalents, and formate is transferred to tetrahydrofolate (THF) to generate formyl-THF, consuming one molecule of ATP. Water splits from formyl-THF to generate methenyl-THF, which is further reduced via methylene-THF to methyl-THF. Finally, the methyl group is transferred to acetyl-CoA synthase (ACS) via the corrinoid iron–sulfur protein (CoFeSP). Most genes involved in the methyl-branch are strongly conserved among phylogenetically diverse acetogens. However, some genes have diverse characteristics depending on the species, such as the formation of various protein complexes or the use of different cofactors (Shin et al., 2016). Unlike the methyl-branch, the carbonyl-branch undergoes a one-step reaction by a multi-component enzyme called carbon monoxide dehydrogenase/acetyl-CoA synthase (CODH/ACS), which synthesizes acetyl-CoA by attaching a carbonyl-group from the reduction of CO₂, or CO directly, to methyl-CoFeSP generated from the methyl-branch. Acetyl-CoA can be used as a building block for various chemicals. Consequently, C1 gas fixation using the WL pathway requires only one molecule of ATP, which is the lowest energy requirement among all existing biological CO₂ fixing metabolic pathways (Fast and Papoutsakis, 2012). Thus, acetogens have received attention as biocatalysts that can efficiently fix C1 gases. In this section, we compare and summarize the processes of C1 assimilation by several major acetogens.

As one molecule of ATP is consumed for fixation of C1 gases through the WL pathway, additional cellular energy is required for acetogens to grow C1 gas conditions. Although the oxidation of CO generates Fd_red, acetogens usually require an additional energy source, such as H₂, to obtain reducing equivalents and to fix the residual CO₂ produced from the oxidation of CO. Hydrogenase has diverse protein characteristics in different microbial species but the common role is to generate Fd_red or NAD(P)H from H₂. Electron bifurcation found in acetogens is a mechanism of biological energy conservation that couples the exergonic oxidation and endergonic reduction reactions. (Peters et al., 2016). Acetogens have electron-bifurcating hydrogenases that couple the exergonic reduction of NAD⁺ to endergonic reduction of ferredoxin with exergonic oxidation of H₂ by making the overall reaction exergonic (Figure 1; Schuchmann and Müller, 2014). During C1 gas fixation, reducing equivalents are supplied via electron carriers to the redox reactions involved in the WL or other metabolic pathways for cell growth. Interestingly, Fd_red produced by CODH or hydrogenase generates an ion gradient between cellular membranes via the membrane-bound respiratory enzyme complex of acetogens and yields additional ATP using membrane-bound ATP synthase. Generation of ATP via these respiratory enzyme complexes involved in energy conservation systems is the mode used by acetogens to sustain their lives.

There are two different complexes used by acetogens, the Rnf and Ech complexes (Figure 1). The Rnf complex is composed of six subunits, as found in A. woodii, C. ljungdahlii, and C. autoethanogenum, and has a ferredoxin:NAD⁺ oxidoreductase activity (Biegel and Müller, 2010). It receives electrons from Fd_red and transfers them to NAD⁺ to generate NADH. It pumps cations such as H⁺ or Na⁺ from inside the cells. In contrast, the Ech complex, which consists of eight or nine subunits and is found in M. thermoacetica and T. kivui, has ferredoxin:H⁺ oxidoreductase activity that uses protons instead of NAD⁺ as the final electron acceptor and pumps out protons_, during the generation of hydrogen (Hedderich and Forzi, 2005). The generated electrochemical ion gradient results in ATP synthesis by the membrane-bound ATP synthase. The Rnf-containing, Na⁺-dependent acetogen A. woodii is known to produce one molecule of ATP per 3.3 Na⁺, as experimentally determined by solving the crystal structure of A. woodii ATP synthase (Matthies et al., 2014). In contrast, a one molecule yield of ATP per 4 H⁺ is assumed for other species such as C. ljungdahlii and M. thermoacetica (Schuchmann and Müller, 2014).

In addition, some acetogens such as C. autoethanogenum and S. ovata have been reported to have NADH-dependent reduced ferredoxin:NADP⁺ oxidoreductase (Nfn) and Sporomusa-type Nfn (Stn), respectively (Figure 1; Wang et al., 2013; Kremp et al., 2020; Mahamkali et al., 2020). The Nfn complex is composed of two subunits of NfnAB, each of which is known to have several Fe-S reaction centers. It is known for its redox balancing potential by producing NADPH from Fd_red and NADH through electron bifurcation (Mahamkali et al., 2020). Therefore, both energy conservation and redox balancing systems are highly important for acetogens to generate cellular energy during autotrophic growth under C1 gas conditions.

To increase the C1 gas fixation efficiency via the WL pathway, the most straightforward approach is to overexpress genes encoding enzymes of the methyl-branch (Figure 2A). When four THF-dependent enzymes (FHS, FCH, MDH, and MTHFR) of C. ljungdahlii were overexpressed in A. woodii, its growth rate increased approximately 1.1-fold and acetate production increased approximately 1.2-fold compared to the empty vector control under C1 autotrophic batch cultivation. In addition, the engineered strain showed approximately a 1.6-fold increase in the specific activity of ACK, whereas strains overexpressing PTA or ACK showed approximately a 1.2-fold increase in specific activity of ACK (Straub et al., 2014). This result indicated that CO₂ fixation efficiency can be increased by upregulating the genes involved in the methyl-branch.

Another way to increase CO₂ fixation efficiency is to introduce a new metabolic pathway that can fix CO₂ in connection with the WL pathway (Figure 2B). An example is the glycine synthase-reductase pathway (GSRP). Clostridium drakei has a unique CO₂-fixing metabolic pathway, in which the WL pathway and GSRP are connected. Methylene-THF, an intermediate of the methyl-branch of the WL pathway, is converted to glycine or serine through GSRP or the reductive glycine pathway (RGP), and these amino acids are converted to acetate through acetyl-phosphate. In this process, one ATP molecule can be synthesized through SLP. Heterologous expression of GSRP in Eubacterium limosum showed that the CO₂ consumption rate and acetate production rate increased 1.5-fold (Song et al., 2020). These results suggest the possibility of increasing the C1 gas-fixing efficiency by introducing an additional C1 gas-fixing metabolic pathway that can be connected to the WL pathway.

As mentioned above, the only way to utilize CO as carbon and energy sources in acetogens is using the carbonyl-branch of the WL pathway, and CODH/ACS complex is involved in this process. C. autoethanogenum and T. kivui can grow well under CO autotrophic condition (Weghoff et al., 2016; Liew et al., 2017), whereas some acetogens such as A. woodii and E. limosum show low CO oxidation rate and growth retardation at a high concentration of CO (Bertsch and Müller, 2015b; Kang et al., 2020). Hence, engineering the CODH/ACS complex is necessary for efficient utilization of CO and improving growth on CO in these acetogens. To increase the CO oxidation rate, the most straightforward approach is to overexpress all genes encoding proteins of the CODH/ACS complex (Figure 2C). The overexpression of acsA-encoding CODH, acsB encoding ACS, and cooC2 encoding maturation protein in Eubacterium callanderi KIST612 using plasmids led to an increase in the CO oxidation and acetate production rates by 3.1-fold and 1.4-fold, respectively, compared to the control strain, whereas there was no difference when individual genes were overexpressed (Kang et al., 2021).

In one case, the CO oxidation rate was increased by changing the protein sequence of CODH/ACS but not by overexpression of the CODH/ACS complex (Figure 2D). In this study, adaptive laboratory evolution was performed on the E. limosum ATCC8486 strain under 44% CO conditions to enhance tolerance to CO. In the evolved strain, a C290A single-nucleotide variation (SNV) was found in the acsA-encoding CODH catalytic subunit. This SNV caused an A97E amino acid change, and the mutant strain showed a 1.4-fold increase in both the growth rate and the CO consumption rate under autotrophic conditions of 44% CO syngas compared to the wild-type strain (Kang et al., 2020). These results show that C1 gas utilization efficiency can be increased by altering the kinetics of the CODH/ACS protein. Overexpression of such mutant CODH proteins may facilitate higher C1 gas utilization than wild-type CODH/ACS.

The electrochemical or photochemical reduction of CO₂ generates formate and methanol, which are liquid forms of C1 feedstock with the benefits of mass transfer and energy efficiency (Cotton et al., 2020). Unlike C1 gases, they are easily soluble in water and are transportable, storable, and safe. The most promising aspect is their energy efficiency, as the conversion of methanol or formate is higher than that achieved with H₂/CO₂ or CO (Claassens et al., 2019).

Methanol and formate are directly assimilated into the WL pathway (Figure 3A). The utilization of these substrates in acetogens has been shown to improve cell growth and product yields of reduced chemicals such as butyrate. For example, E. limosum cultivated on methanol has not only shown higher growth rates than cells cultivated with C1 gases (Genthner and Bryant, 1987; Loubière et al., 1992), but also high product yields with the production of 12 mM acetate and 3.7 mM butyrate from 20 mM methanol (Litty and Müller, 2021). The yield of butyrate under methanol conditions was significantly higher than that obtained under H₂/CO₂ or CO conditions, where acetate was the major product while butyrate was produced in trace amounts. As methanol retains more reducing equivalents than gaseous substrates, a surplus of reducing equivalents can be used to drive biosynthesis and improve the conversion yield of butyrate. The increased butyrate-to-acetate ratio in the presence of methanol was assumed to be due to the role of butyrate production in the NAD(P)H-NAD(P)⁺ balance during methanol assimilation, as NAD(P)H generation from methanol is catalyzed by 3-hydroxybutyryl-CoA dehydrogenase to regenerate NAD(P)⁺ (Flaiz et al., 2021).

Similarly, A. woodii has been regarded as an interesting organism for anaerobic formate-based bio-production because of the high-energy efficiency achieved through acetate production from formate compared to that from H₂/CO₂ (Cotton et al., 2020). A recent study showed that the cultivation of A. woodii with formate as the sole carbon and energy source resulted in conversion of formate to acetate at a higher rate and efficiency than cultivation on gaseous substrates (Moon et al., 2021; Neuendorf et al., 2021). The study also compared the energetic efficiency of different acetogens and other common microbial hosts (e.g., formatotrophs or engineered strains capable of utilizing formate or methanol) during growth and product formation on C1 or sugar substrates, revealing that acetogens show superior energy efficiency on all substrates analyzed, with the highest values for C1 substrates. Formate showed an even higher energy potential than gaseous substrates. Among acetogens, A. woodii and E. limosum are promising biocatalysts for the conversion of formate into acetate, as they form acetate as a major product during growth on formate (Litty and Müller, 2021).

It should be noted that utilization of formate and methanol has several disadvantages. During formate consumption, pH of the medium can increase slightly (Litty and Müller, 2021; Moon et al., 2021), and the produced alcohols are toxic to some microorganisms at high concentrations, presumably because they damage the cell membrane and cause end-product inhibition of glycolytic enzymes (Dürre et al., 1988; Caldwell, 1989). The increase in pH during formate consumption can be controlled through addition of buffers during fermentation. Regarding the toxicity of methanol, adaptive evolution of acetogens in high concentrations of methanol can increase their tolerance to methanol, as shown by a previous study wherein an adapted strain of S. ovata achieved a 5-fold increase in the growth rate of methanol with a higher tolerance to the solvent (Tremblay et al., 2015). More importantly, the reduction of formate to acetate produces two molecules of CO₂ in the absence of excess reducing equivalents to refix the released CO₂, resulting in a carbon efficiency of only 50%. To reduce the carbon loss, addition of H₂ can be a strategy to enable complete fixation of CO₂, improve the carbon efficiency, and even facilitate net CO₂ uptake (Molitor et al., 2016; Novak et al., 2021).

H₂ and CO are poor electron donors compared to high-energy substrates, such as methanol and glucose, resulting in insufficient electron supply in acetogens (Emerson and Stephanopoulos, 2019). Instead, the direct supply of such electron sources could be a potential strategy to overcome cellular energy limitations in acetogens. Replacing H₂ or CO with renewable reducing powers for gas fermentation has been demonstrated in acetogens with added advantages of the enhanced efficiency of C1 gas conversion and the metabolic shift toward reduced products (Nevin et al., 2011; Kracke et al., 2016; Im et al., 2018; Cheng et al., 2022).

Microbial electrosynthesis (MES) is a process in which microbes use electrons derived from an electrode to reduce CO₂ to multi-carbon products (Nevin et al., 2010). Several acetogens have been reported to perform MES via extracellular electron transfer. For example, direct electron uptake from the electrode has been proposed for S. ovata, Clostridium aceticum, C. ljungdahlii, and M. thermoacetica, showing reduction of CO₂ into organic acids; however, clear evidence of direct electron transfer is still lacking (Nevin et al., 2010, 2011). In contrast, C. autoethanogenum requires an electron mediator such as methyl viologen (MV) or neutral red to transfer electrons from the electrode to cells (Kracke et al., 2016). As some acetogens grow on the electrode by forming a biofilm (Nevin et al., 2010), electron transfer can be spatially limited to cells in contact with the electrode and affected by the biofilm density (Pirbadian et al., 2020), which can result in complex spatial patterns of activity in the bioelectrochemical system (BES).

Extracellular electron supply in the BES induced a significant metabolic shift in C. autoethanogenum, resulting in a significant reduction in acetate production, with increased production of lactate and 2,3-BDO by 35-fold and 3-fold, respectively, compared to that under conditions without the electron mediator (Kracke et al., 2016). Recently, Clostridium carboxidivorans was found to be capable of MES with MV or biochar as an electron mediator, which facilitated electricity-driven autotrophic CO₂ fixation (Cheng et al., 2022). The standard redox potential of MV is −446 mV, which is sufficiently low to reduce ferredoxin (−450 to −500 mV) and NADH (−320 mV). Hence, it can regulate the redox balance and redistribute the redox flux in acetogens (Michaelis and Hill, 1933; Schuchmann and Müller, 2014). However, increasing the amount of MV supplementation has been reported to have toxic effects on cells (Cheng et al., 2022). Alternatively, neutral red is another electron shuttle with low toxicity and a standard redox potential similar to NADH (Harrington et al., 2015). Addition of neutral red was found to improve the production of volatile fatty acids from CO by increasing the reducing power (Im et al., 2018).

MES powered by solar energy enables artificial photosynthesis in acetogens, with the same net overall reaction as plant-based photosynthesis, in which CO₂ and water are converted to organic compounds and oxygen (Nevin et al., 2010). Light-capturing catalysts, such as cadmium sulfide (CdS) or gold (Au) nanoparticles (NPs), are attached to cells and deliver electrons obtained from light to the cells. For example, M. thermoacetica displaying biologically synthesized CdS-NPs or Au-NPs were found to convert CO₂ to acetate (Sakimoto et al., 2016; Zhang et al., 2018). However, the proposed system was limited by the biosynthesis of photo-responsive CdS, which decreases chemical production due to a defense mechanism against heavy metals in bacteria. To circumvent this issue, the chemically synthesized CdS-NP was attached to C. autoethanogenum, where the successful conversion of CO₂ into acetate was demonstrated using only light-induced electrons from CdS-NPs (Jin et al., 2021). However, the mechanism of extracellular electron transfer from nanoparticles remains unclear. Transcriptional analysis of C. autoethanogenum attached to CdS-NPs suggested that electrons from CdS-NPs are transferred to cells indirectly through mediators such as iron or flavin mononucleotide (FMN). Another study conducted in M. thermoacetica-CdS suggested two types of electron transfer pathways: The H₂ generation pathway mediated by membrane-bound hydrogenase and an H₂-independent pathway (Kornienko et al., 2016). Interestingly, a recent study found that neither CdS-NPs nor light was involved in CO₂ fixation, but cysteine present in the medium was used as the carbon source for acetate production in M. thermoacetica (Göbbels et al., 2021). These findings indicate that the underlying mechanism of this system is not fully understood; hence, more research is needed to exploit the potential of artificial photosynthesis systems with acetogens.

Acetogenic metabolism operates at a thermodynamic limit, which results in an insufficient energy supply during autotrophy (Schuchmann and Müller, 2014). This problem becomes more critical during the cultivation of engineered acetogens, where antibiotics are added to maintain the plasmid, or when the engineered strains express heterologous biosynthesis pathways for energetically high-cost metabolites. In a study, limited autotrophic growth was observed in cells harboring an empty plasmid (Shin et al., 2019), and the non-native chemical mevalonate was only produced under fructose conditions because of the limited availability of reducing equivalents and ATP during C1 gas fermentation (Diner et al., 2018). To overcome the thermodynamic problem and improve autotrophic growth under such conditions, sugar co-feeding is a promising strategy. With a broad range of substrate utilization, acetogens can improve bioenergetics by using sugars along with C1 feedstocks in a process called mixotrophy, where carbon is fixed through both the glycolytic and WL pathways (Fast and Papoutsakis, 2012; Tracy et al., 2012; Fast et al., 2015). Sugar oxidation via glycolysis generates eight reducing equivalents and two moles of CO₂, all of which are re-assimilated via the WL pathway (Figure 3B). Theoretically, it produces a total of three molecules of acetyl-CoA and hence, increases the yield of acetyl-CoA by 50% compared to standard Embden-Meyerhof-Parnas (EMP) glycolysis (Fast et al., 2015). However, in practice, some of the reducing equivalents are required for biomass generation and maintenance, which indicates that CO₂ evolved in glycolysis cannot be fully converted to acetyl-CoA, resulting in carbon loss as CO₂ and a decrease in the overall carbon yield. Hence, the choice of substrates co-fed with sugars is crucial for achieving complete carbon fixation and high product yields. Supplying exogenous reducing power such as H₂ or syngas during sugar fermentation of acetogens has been demonstrated, which showed high productivity that exceeded the sum of individual substrate productivities and carbon efficiency over 90%, and promoted the generation of more reduced products (Jones et al., 2016; Maru et al., 2018; Cheng et al., 2019).

Another advantage of mixotrophy is the improvement in cellular growth due to the increased availability of acetyl-CoA obtained from additional substrates and improved cellular energy in acetogens. Glucose co-feeding has been shown to improve the methanol uptake rate in E. limosum with an increased growth rate and biomass yield due to the improved availability of both carbon and energy for anabolic reactions (Loubière et al., 1992). Recently, A. woodii showed mixotrophic growth on formate and fructose, where co-feeding of formate increased the acetate production rate by 50% compared to fructose utilization alone (Neuendorf et al., 2021). Formate utilization improved cell-specific acetate productivity, whereas fructose increased the overall bioenergetics, increasing the amount of ATP wasted. However, co-utilization of formate and fructose resulted in an equal production of CO₂, which was released for growth solely on formate, with a carbon efficiency of only 50%. Although co-feeding of formate and fructose is not preferable to formate utilization alone in terms of carbon efficiency, addition of fructose could be used to improve the bioenergetics of formate-utilizing acetogens and facilitate the redirection of carbon flux from acetate toward desired products, such as ethanol and lactate, in metabolically engineered strains (Neuendorf et al., 2021). Further research is needed to improve carbon yield when formate is co-fed for mixotrophic cultivation.

Although mixotrophy is a promising strategy for increasing the bioenergetics of acetogens, a key concern in its implementation is the possibility of carbon catabolite repression (CCR) of the WL pathway in the presence of a preferred sugar substrate. CCR has been found to be dependent on species and culture conditions (Liu et al., 2015; Maru et al., 2018). For example, CCR in C. aceticum (Braun et al., 1981), M. thermoacetica (Huang et al., 2012), and Blautia coccoides GA-1 (Liu et al., 2015) inhibits autotrophic metabolism and causes poor H₂/CO₂ consumption when glucose or fructose is present at high concentrations. In contrast, A. woodii, Butyribacterium methylotrophicum, C. autoethanogenum, C. carboxidivorans, C. ljungdahlii, and E. limosum concurrently utilize sugars (glucose or fructose) and C1 gases without CCR (Jones et al., 2016; Maru et al., 2018). However, E. limosum was found to be affected by CCR when co-fed with methanol and glucose (Loubière et al., 1992).

Despite the promising aspects of mixotrophy, few studies have elucidated the CCR mechanism in acetogens or the interaction between the WL pathway and glycolysis. Therefore, a thorough understanding of mixotrophy is crucial to fully exploit its potential without CCR. Nevertheless, researchers have recently discovered solutions to overcome CCR in mixotrophy. Controlling the glucose feeding rate with kinetically limiting concentrations during the continuous fermentation of M. thermoacetica successfully prevented CCR (Park et al., 2019). This shifted the carbon substrate preferences toward CO₂ and helped achieve mixotrophic growth without CCR. Another recent study similarly employed xylose-limited conditions, in which the simultaneous uptake of xylose and CO for acetate production was observed in C. autoethanogenum (Mann et al., 2020).

The development of genetic tools and synthetic biology approaches has opened up possibilities for the metabolic engineering of several acetogens (Köpke et al., 2010; Shin et al., 2019; Zhao et al., 2019; Jin et al., 2020). Using such tools, the heterologous expression of biosynthetic pathways for producing desired chemicals has expanded the product spectrum of acetogens, including isoprene, isopropanol, acetone, and 3-hydroxybutyrate (3-HB; Jones et al., 2016; Diner et al., 2018; Woolston et al., 2018; Karim et al., 2020; Jia et al., 2021). Although they were proof-of-concept production, as their titers were considerably low due to concurrent production of native metabolites as byproducts, significant improvements can be achieved with strain engineering that redirects carbon fluxes toward desired chemicals.

Blocking competing pathways is an effective method for redirecting the carbon flux. Deletion of three genes encoding PTA, AdhE1, and CoA transferase in C. ljungdahlii increased butyrate synthesis and reduced acetate and ethanol byproducts (Ueki et al., 2014). CRISPR-mediated downregulation of adhE1 also redirected the carbon from ethanol to butyrate in engineered butyrate-producing C. ljungdahlii (Zhao et al., 2019).

Considering that acetate synthesis is coupled with ATP formation and is essential for energy supply in acetogens living in energy-limited autotrophic conditions, blocking acetate-producing pathways can result in poor cell growth (Huang et al., 2016). A feasible strategy to overcome this issue is to reassimilate the acetate to form the target product. The AOR present in acetogens converts acetate into ethanol with acetaldehyde as an intermediate. Its role in acetate reassimilation has been found in a study in which knockout of AOR enzymes significantly impaired the conversion of acetate into alcohols in C. autoethanogenum (Liew et al., 2017). The positive effect of this strategy was confirmed by the overexpression of native AOR in C. carboxidivorans, which resulted in higher ethanol production (Cheng et al., 2019). Recently, a novel acetate reassimilation pathway containing two acyl-CoA synthetases (ACS from C. ljungdahlii and FadKM1/2 from E. coli) that catalyze acetate conversion to form acetyl-CoA was tested in C. ljungdahlii along with overexpression of native AOR enzymes (Jia et al., 2021). Introduction of the pathway into an engineered strain to produce isopropanol successfully reinforced acetate assimilation, achieving significantly reduced acetate formation and increased production of ethanol and isopropanol through syngas fermentation.