This article is licensed under CC-BY 4.0 pubs.acs.org/est Article Lactic Acid from CO2: A Carbon Capture and Utilization Strategy Based on a Biocatalytic Approach Albert Carceller, Marina Guillén,* and Gregorio Á lvaro Cite This: Environ. Sci. Technol. 2023, 57, 21727−21735 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: The EU low-carbon economy aims to reduce the level of CO2 emission in the EU to 80% by 2050. High efforts are required to achieve this goal, where successful CCU (Carbon Capture and Utilization) technologies will have a high impact. Biocatalysts offer a greener alternative to chemical catalysts for the development of CCU strategies since biocatalysis conforms 10 of the 12 principles of green chemistry. In this study, a multienzymatic system, based on alcohol dehydrogenase (ADH), pyruvate decarboxylase (PDC), and lactate dehydrogenase (LDH), that converts CO2 and ethanol into lactic acid leading to a 100% atom economy was studied. The system allows cofactor regeneration, thus reducing the process cost. Through reaction media engineering and enzyme ratio study, the performance of the system was able to produce up to 250 μM of lactic acid under the best conditions using 100% CO2, corresponding to the highest concentration of lactic acid obtained up to date using this multienzymatic approach. For the first time, the feasibility of the system to be applied under a real industrial environment has been tested using synthetic gas mimicking real blast furnace off-gases composition from the iron and steel industry. Under these conditions, the system was also capable of producing lactic acid, reaching 62 μM. KEYWORDS: multienzymatic systems, carbon capture and utilization, biocatalysis, carbon dioxide 1. INTRODUCTION dissociation energy required to break the C�O bond in CO2 The Intergovernmental Panel on Climate Change (IPCC) molecules is high (749 kJ mol−1), which is a thermodynami- presented a special report on the impacts of global warming cally costly reaction.5 Thus, high conditions of temperature or considering that it has to be kept below 1.5 °C above pressure and/or highly efficient catalysts are required to carry preindustrial levels to efficiently response to the threat of out the bond breakage.6 Biocatalysts represent a greener alternative to CCUs based climate change.1 The EU low-carbon economy roadmap states on chemical catalysts, given that (i) they are biodegradable, that EU should cut emissions to 80% below 1990 levels by the safe, and nontoxic, (ii) they are produced from renewable middle of the century.2 Therefore, according to the EU resources, (iii) they work under mild conditions, thus leading guidelines, significant efforts should be made to reduce CO2 to less energy intensive processes, (iv) they show high emissions. substrate specificity and product selectivity, and (v) there is Carbon capture, utilization, and storage (CCUS) technol- no need of functional groups activation, protection, and ogies involve the capture of carbon dioxide from fuel deprotection steps leading to more step-economical processes combustion or industrial processes, the transport of this and less waste generation.7−10 Therefore, biocatalysis is playing carbon dioxide, and either its use as a resource to create an important role in the development of CCUs. Several valuable products or services or its permanent storage systems based on the use of enzymes to transform CO2 have underground in geological formations.3 According to the been described in the literature: formate dehydrogenase International Energy Agency, CCUS will need to form a key (FDH) to transform CO2 into formic acid, nitrogenase pillar of efforts to put the world on the path to net-zero emissions.4 CCU strategies are based on the use of CO2 as a carbon feedstock to produce several compounds such as fuels, Received: July 11, 2023 chemicals, and materials, obtaining a double benefit toward the Revised: October 30, 2023 climate change fight: a reduction in CO2 emissions and a Accepted: November 2, 2023 depletion of fossil fuels as feedstock. Published: December 11, 2023 CO2 is a highly stable molecule in which carbon is in the highest valence state (+4). Besides, as it is known, the © 2023 The Authors. Published by American Chemical Society https://doi.org/10.1021/acs.est.3c05455 21727 Environ. Sci. Technol. 2023, 57, 21727−21735 Environmental Science & Technology pubs.acs.org/est Article MoFe protein for conversion of CO2 into methane, and real blast furnace off-gases composition [24.5% CO2, 46.6% carbonic anhydrase to convert CO2 into bicarbonate.11−20 N2, 23.9% CO, 1.2% O2, and 3.8% H2] was obtained from However, most of the enzymatic systems are based on a single Carburos Metalicos (Barcelona, Spain). PDC from Zymobacter biocatalytic step, which limits the range of products that can be palmae (ZpPDC) was produced according to Alcover et al.16 enzymatically obtained from CO2. Multienzymatic systems can PDC from Saccharomyces cerevisiae (ScPDC), ADH from be also applied for the transformation of CO2, widening the Saccharomyces cerevisiae (ScADH) and Thermotoga maritima number of products of interest that can be produced compared LDH (TmLDH) were produced according to Benito et al. (see to the use of a single enzyme. Moreover, in most cases, the Supporting Information).17 enzymatic conversion of CO2 needs cofactors such as 2.2. Enzyme Activity Assays. PDC activity was NAD(P)H/NAD(P)+ which hampered the implementation determined by coupling the PDC with ADH and following of the enzymatic systems at an industrial scale due to the high NADH oxidation at 340 nm (εNADH = 6.22 mM−1 cm−1) and cost of these cofactors. The use of multienzymatic systems 25 °C with a Varian Cary 50 Bio UV−visible spectropho- allows in situ regeneration by coupling an enzymatic cofactor tometer (Agilent). The reaction mixture contained 33 mM regeneration reaction without the need to include chemical, sodium pyruvate, 0.11 mM NADH, 3.5 U mL−1 of commercial electrochemical, or photochemical systems.8,10,13−15 Therefore, ADH from Saccharomyces cerevisiae obtained from Sigma in order to widen the range of commodity chemicals that can Chemical Co. (St. Louis, USA), 0.1 mM TPP and 0.1 mM be obtained from carbon dioxide, CO2-fixing enzymes has to MgCl2 in citrate buffer 200 mM and pH 6. One unit of enzyme be used as a step of a multienzymatic cascade.10 activity corresponds to the amount of PDC that converts 1 In the present work, a multienzymatic system to produce μmol of pyruvate into acetaldehyde per minute. Analyses were lactic acid from CO2 and ethanol has been studied (Figure 1). carried out in duplicate. LDH activity was determined by following NADH oxidation at 340 nm (εNADH = 6.22 mM−1 cm−1) and 25 °C with a Varian Cary 50 Bio UV−visible spectrophotometer (Agilent). The reaction mixture contained 33 mM sodium pyruvate and 0.11 mM NADH, in 100 mM phosphate buffer, pH 6.7. One unit of enzyme activity corresponds to the amount of LDH that converts 1 μmol of pyruvate into lactate per minute. Figure 1. Multienzymatic system for the synthesis of lactic acid from Analyses were carried out in duplicate. CO2 and ethanol with an internal cofactor regeneration cycle. The ADH activity was determined by following NAD+ reduction system consists of three enzymes, ADH, PDC, and LDH. at 340 nm (εNADH = 6.22 mM−1 cm−1) and 25 °C with a Varian Cary 50 Bio UV−visible spectrophotometer (Agilent). The reaction mixture contained 567 mM ethanol and 1.7 mM The system is made up of three enzymes: a pyruvate NAD+, in phosphate buffer 100 mM at pH 8.8. One unit of decarboxylase (PDC) which converts CO2 and acetaldehyde enzyme activity corresponds to the amount of ADH that into pyruvic acid, a lactate dehydrogenase (LDH) which converts 1 μmol of ethanol into acetaldehyde per minute. transforms pyruvic acid into lactic acid consuming NADH, and Analyses were carried out in duplicate. an alcohol dehydrogenase (ADH) which produces acetalde- 2.3. Enzyme Stability and Activity over Different pHs. hyde from ethanol regenerating the NADH. This third reaction Enzyme stability of ADH, PDC, and LDH was measured by allows not only cofactor regeneration but also to produce incubating in 2 mL microtubes at a pH ranging from 5 to 10 acetaldehyde from ethanol, a cheaper, less toxic, and less for 24 h at 25 °C and 300 rpm using a Multi Therm hazardous substrate which can also be obtained by (Benchmark Scientific Inc.) Heat Block system. For pH 5, an fermentative processes from renewable resources. Following acetate buffer 100 mM was used, for pH 6, 7, and 8, a this multienzymatic system, all the reactant atoms will end up phosphate buffer 100 mM was used, and for pH 9 and 10, a in the desired product, representing an atom economy of bicarbonate buffer 100 mM was used. Samples were taken at 0, 100%. 2, and 24 h to assess enzyme activity using the corresponding In addition, the system has also been tested using synthetic activity test. gas mimicking real industrial off-gases from iron and steel Enzyme activity at different pHs was carried out by changing industry. the pH of the activity assay and performing the corresponding assay for each enzyme. Citrate buffer 200 mM (ZpPDC and 2. MATERIALS AND METHODS ScPDC), 100 mM (TmLDH), or 50 mM (ScADH) for pH 6, 2.1. Chemicals and Reagents. Sodium pyruvate, citric Tris−HCl buffer 200 mM (ZpPDC and ScPDC), phosphate acid, sodium citrate, thiamine pyrophosphate (TPP), and buffer 100 mM (TmLDH) or 50 mM (ScADH) for pH 7−8, magnesium chloride were obtained from Sigma Chemical Co. and bicarbonate buffer 200 mM (ZpPDC and ScPDC), 100 (St. Louis, USA) for the determination of enzyme activity and mM (TmLDH), or 50 mM (ScADH) for pH 9−10. reaction media. Ethanol, potassium phosphate, sodium acetate, 2.4. Enzymatic Reactions. 2.4.1. Single-Enzyme Reac- and sodium bicarbonate were obtained from Scharlab, S.L. tions. Each enzyme reaction was tested individually at 2 mL (Barcelona, Spain) for the determination of enzyme activity scale. For ADH, a bicarbonate buffer 250 mM at pH 7, 8, or 9 and reaction media. Acetoin and lactic acid standards were with ethanol 50 mM, NAD+ 10 mM was used. For PDC, a obtained from Sigma Chemical Co. (St. Louis, USA) and used bicarbonate buffer 250 mM at pH 7, 8 or 9 with acetaldehyde as analytical standards for gas chromatography and LC−MS 10 mM, TPP 1 mM, and MgCl2 1 mM was used. Regarding analysis, respectively. NADH and NAD+ were obtained from LDH, a bicarbonate buffer 250 mM at pH 7, 8, or 9 with Bontac Bioengineering Co. (Shenzhen, China) as reaction pyruvate 10 mM, NADH 20 mM was used. Each reaction was cofactors. Carbon dioxide gas as well as gas mixture mimicking performed at 25 °C with 500 rpm agitation using a Multi 21728 https://doi.org/10.1021/acs.est.3c05455 Environ. Sci. Technol. 2023, 57, 21727−21735 Environmental Science & Technology pubs.acs.org/est Article Therm (Benchmark Scientific Inc.) heating block system for In all analyses, compound standards of known concentration 24 h. Samples were taken after 1, 4, and 20 h (ADH reaction) were used for the calibration of the equipment. or 24 h (PDC and LDH reactions). 2.5.2. Spectrophotometric NADH Analysis. NADH was 2.4.2. Lactate Dehydrogenase-Coupled Pyruvate Decar- measured using a Varian Cary 50 Bio UV−visible spectropho- boxylase Reactions. Coupled-enzyme reactions were carried tometer (Agilent). Samples were measured in a 1.5 mL cuvette out by coupling LDH to PDC in a 50 mL reactor (Miniclave at 340 nm and 25 °C and diluted with distilled water when steel; Büchi) with a working volume of 25 mL. Reactions at 1 needed. The NADH concentration was calculated using the atm of CO2 pressure were performed in a phosphate buffer 250 Lambert−Beer equation with εNADH = 6.22 mM−1 cm−1. mM at pH 7 with acetaldehyde 5 mM, NADH 10 mM, TPP 1 mM, MgCl2 1 mM, 10 U mL−1 of purified PDC from Z. palmae 3. RESULTS AND DISCUSSION (ZpPDC) or S. cerevisiae (ScPDC) and LDH from T. maritima 3.1. Enzyme Characterization. All enzymes were (TmLDH) at 25 °C and 500 rpm magnetic stirring. CO2 was characterized in terms of activity and stability toward pH sparged into the reactor with bubbling into the reaction media. aiming to determine the most suitable pH range for the study The gas outlet was opened as soon as the reaction started. of the multienzymatic system. Aiming to work under favorable Then, the outlet was closed after 5 min, when dissolved CO2 conditions for CO2 solubility, the pH study was performed reached the equilibrium with the gas phase. under alkaline conditions. Regarding PDC, two enzymes were 2.4.3. Complete Multienzymatic System Reactions. tested: a bacterial PDC from Z. palmae (ZpPDC) and a fungal Complete multienzymatic system reactions were carried out PDC from S. cerevisiae (ScPDC). Both were characterized as in a 50 mL reactor (Miniclave steel; Büchi) with a working for the ScADH and TmLDH (Figures 2 and 3). volume of 20 mL. Reactions at 1 atm of CO2 pressure were performed in a phosphate, MOPS (3-(N-morpholino)- propanesulfonic acid) , Tris-HCl or citrate buffer 250 mM at pH 7 with ethanol 1 M, NAD+ 10 mM, TPP 1 mM, MgCl2 1 mM, using purified ADH from S. cerevisiae (ScADH), PDC from S. cerevisiae (ScPDC), and LDH from T. maritima (TmLDH) at 25 °C and 500 rpm magnetic stirring. CO2 was sparged into the reactor with bubbling into the reaction media. The gas outlet was opened as soon as the reaction started. Then, the outlet was closed after 5 min, when dissolved CO2 reached the equilibrium with the gas phase. Multienzymatic test using synthetic gas mimicking blast furnace off-gases composition were carried out under optimum conditions: MOPS buffer 250 mM at pH 7, ethanol 1 M, NAD+ 10 mM, TPP 1 mM, MgCl2 1 mM, 1 atm of synthetic gas mixture, ADH 75 U mL−1, PDC 150 U mL−1, and LDH 187.5 U/mL at 25 °C and 500 rpm agitation. The gas mixture composition was as follows according to the data provided by Arcelor Mittal in the frame of the BIOCON-CO2 project (24.5% CO2, 46.6% N2, 23.9% CO, 1.2% O2, and 3.8% H2). Figure 2. Enzyme stability at 2 and 24 h of incubation at pH 7, 8, 9, 2.5. Analytical Methods. 2.5.1. Acetaldehyde, Acetoin, and 10 with bicarbonate buffer 100 mM at 25 °C. Alcohol Ethanol, and Lactic Acid Quantification. The concentration dehydrogenase from S. cerevisiae (ScADH), PDC from Z. palmae of acetaldehyde, acetoin, and ethanol in the reaction mixture (ZpPDC), PDC from S. cerevisiae (ScPDC), and LDH from T. were measured with a Shimadzu GC 2010 system using a maritima (TmLDH) were used. Error bars correspond to standard Stabilwax-DA column (15 m × 0.33 × 1 μm). The conditions deviation (n = 2). were injection volume, 3 μL in split mode 20:1; injector temperature, 260 °C; carrier gas, He at a constant flow of 3 mL Regarding enzyme stability, all the biocatalysts showed high min−1. The initial oven temperature, 35 °C, was held for 2 min activities after 24 h at pH 7−8, reaching in all cases values and then programmed to increase at 15 °C min−1 to 120 °C, higher than 60% of the initial activity (Figure 2). At pH 10, all and finally programmed to increase at 40 °C min−1 to 240 °C, enzymes completely lost their activity after 2 h, except where it was held for 1 min. Before the analysis, reaction TmLDH which maintained 100% of its catalytic efficiency samples were inactivated by adding 20 μL of 36% (v v−1) HCl even after 24 h, probably due to its extremophile origin.21 At to 500 μL of sample. pH 9, TmLDH also showed high stability contrary to ScPDC In the case of lactic acid, the measurement of concentration which was completely inactive after 2 h. Even though ScADH was performed with a Shimadzu LCMS-2010A using an ICSep and ZpPDC maintained up to 80% of the initial activity after 2 87H USP L17 (Transgenomic) column. The conditions were h, ZpPDC was inactive after 24 h and ScADH showed less as follows: buffer solution 640 μL L−1 of acetic acid, injection than 20% of its initial activity. volume 5 μL, with a flow rate of 0.15 mL min−1; the nebulizing The pH activity profile of all enzymes is depicted in Figure gas was N2 with a flow of 1.5 L min−1; the CDL temperature, 3. ZpPDC and TmLDH showed their best performance at pH 200 °C; and the heat block temperature, 200 °C. The mass-to- 7, while ScPDC showed its highest activity at pH 6. All three charge ratio was set to 89 m/z to monitor the elution of lactic enzymes suffered an activity decrease as pH is increased up to a acid with a running time of 25 min. Before the analysis, complete deactivation at pH 10. On the contrary, ScADH reaction samples were inactivated by adding 20 μL of 98% showed activities higher than 60% in all tested pHs, reaching H2SO4 to 500 μL of sample. the highest biocatalytic performance at pH 9 and 10. 21729 https://doi.org/10.1021/acs.est.3c05455 Environ. Sci. Technol. 2023, 57, 21727−21735 Environmental Science & Technology pubs.acs.org/est Article completely inactivated after 20 h (Figure 4B). These results are in accordance with enzyme activity and stability profiles (Figures 2 and 3), showing that the pH has an opposite effect on the ADH activity and stability. Moreover, these results indicate a stronger effect of pH on stability than on activity (20-fold stability increase from pH 9 to 7 compared to 1.7-fold increase in activity from pH 7 to 9). The reaction catalyzed by LDH from T. maritima was also studied. TmLDH catalyzed the reduction of pyruvic acid to yield lactic acid. Results are depicted in Figure 5A, showing a lower dependence of TmLDH toward pH compared to the synthesis of acetaldehyde catalyzed by ScADH (Figure 5). The highest yield was obtained at pH 7 as shown in Figure 5A, with a 100% yield and conversion. Furthermore, the LDH of T. maritima shows a high stability at all tested pHs (Figure 5B), suggesting that a pH with a higher activity can be chosen while keeping the enzyme stable. Figure 3. pH activity profile of ADH from S. cerevisiae (ScADH), The high yield obtained as well as the high stability showed PDC from Z. palmae (ZpPDC), PDC from S. cerevisiae (ScPDC) and at all tested pHs by TmLDH, which catalyzes the last step of LDH from T. maritimae (TmLDH) at 25 °C. Enzyme activity is the system, represents a great advantage for the development expressed as a percentage of the maximum activity within the pH of the multienzymatic system by shifting the equilibriums range. Citrate buffer 200 mM (ZpPDC and ScPDC), 100 mM toward the lactic acid formation. (TmLDH), or 50 mM (ScADH) for pH 6, Tris−HCl buffer 200 mM When the synthesis of pyruvate was studied using PDC from (ZpPDC and ScPDC), phosphate buffer 100 mM (TmLDH) or 50 Z. palmae, the desired ketoacid was not produced at any of the mM (ScADH) for pH 7−8, and bicarbonate buffer 200 mM (ZpPDC and ScPDC), 100 mM (TmLDH), or 50 mM (ScADH and TmLDH) tested pHs. Otherwise, acetoin was detected in the sample for pH 9−10. Error bars correspond to standard deviation (n = 2). analysis at all pH values, showing that PDC catalyzes an undesired side reaction that converts two molecules of acetaldehyde into acetoin, instead of forming pyruvic acid According to these results, pH 10 was discarded due to from CO2 and acetaldehyde (Figures S1 and S2). This side inactivation of TmLDH and ZpPDC. Thus, the multi- reaction has already been reported in the literature for other enzymatic system was tested at pH 7, 8, and 9. PDCs from Zymomonas mobilis and Saccharomyces carlsbergen- 3.2. Single Enzymatic Reaction Study. Each reaction sis.22−24 According to previous research, the PDC enzyme within the multienzymatic system was first tested individually catalyzes the decarboxylation of pyruvate through the TPP at pH 7, 8, and 9, according to the results obtained in the coenzyme. After pyruvate binds to TPP, CO2 is liberated and enzyme characterization. All the tests were performed with the formed TPP-bound acetaldehyde is susceptible to the bicarbonate buffer as the CO2 source. addition of a second aldehyde, which results in the formation Figure 4A shows acetaldehyde formation from ethanol of acetoin. Moreover, as the results in this work also suggest, catalyzed by ADH from S. cerevisiae (ScADH). The highest this acetoin synthesis can occur when either pyruvate or yield was obtained at the most alkaline pH as shown in Figure acetaldehyde are the substrates. On the other hand, the 4A, reaching 2.45 mM acetaldehyde at pH 9. However, the obtained results showed that PDC from Z. palmae catalyzes highest enzyme stability was obtained at pH 7 with a half-life of this secondary reaction with a higher reaction rate than the 20 h, 20-fold higher than at pH 9 where the enzyme is desired pyruvate synthesis (Figure 6A) to the extent that no Figure 4. (A) Acetaldehyde formation from ethanol catalyzed by ScADH at different pHs after 20 h of reaction. (B) Stability of ADH from S. cerevisiae (ScADH), expressed as initial activity percentage, during the reaction at different bicarbonate buffer pHs. Reaction conditions: bicarbonate buffer 250 mM, ethanol 50 mM, NAD+ 10 mM, and ADH 7 U mL−1 at 25 °C and 500 rpm agitation. Error bars correspond to standard deviation (n = 2). 21730 https://doi.org/10.1021/acs.est.3c05455 Environ. Sci. Technol. 2023, 57, 21727−21735 Environmental Science & Technology pubs.acs.org/est Article Figure 5. (A) Lactic acid synthesis from pyruvic acid catalyzed by TmLDH at different pHs. Concentration at different pHs after 24 h. (B) Stability of LDH from T. maritima (TmLDH), expressed as initial activity percentage, during the reaction at different bicarbonate buffer pHs. Reaction conditions: bicarbonate buffer 250 mM, pyruvate 10 mM, NADH 20 mM, and LDH 20 U mL−1 at 25 °C and 500 rpm. Error bars correspond to standard deviation (n = 2). Figure 6. (A) Acetoin formation from acetaldehyde catalyzed by PDC at different pHs. Reaction conditions: bicarbonate buffer 250 mM, acetaldehyde 10 mM, TPP 1 mM, MgCl2 1 mM, PDC 5 U mL−1 at 25 °C and 500 rpm agitation. Reaction time was 24 h. PDC from Z. palmarum (ZpPDC) and PDC from S. cerevisiae (ScPDC) were used independently. (B) Stability of PDC from Z. palmae (ZpPDC) and S. cerevisiae (ScPDC) during the reaction at different bicarbonate buffer pHs. Reaction conditions: bicarbonate buffer 250 mM, acetaldehyde 10 mM, TPP 1 mM, MgCl2 1 mM, ZpPDC or ScPDC 5 U mL−1 at 25 °C, and 500 rpm agitation. Error bars correspond to standard deviation (n = 2). pyruvic acid could be detected. For this reason, ScPDC was modified, aiming to shift the equilibrium toward the formation also tested in the synthesis of pyruvic acid aiming to find an of the ketoacid. Thus, two strategies were considered: (i) enzyme with a higher specificity toward CO2. First, ScPDC was introducing gaseous CO2 to reach 1 atm of 100% CO2 and (ii) tested under the same reaction conditions. The results coupling the LDH reaction, which is highly shifted toward obtained show that PDC from S. cerevisiae produced lactic acid. Thus, lactic acid formation was analyzed in these approximately 20-fold less acetoin than PDC from Z. palmae experiments to evaluate the performance of both PDCs as (Figure 6A), indicating that using ScPDC can be more suitable enzymes for the multienzymatic system. Moreover, the favorable for the synthesis of pyruvic acid. However, pyruvic acetoin concentration was also analyzed as a parameter to acid was not detected under the tested conditions either. consider in the enzyme selection. The reaction was performed Regarding enzyme stability (Figure 6B), ScPDC showed an at pH 7 to favor the synthesis of lactic acid by TmLDH (Figure hyperactivation, increasing 5-fold the initial activity after 1 h of 7). It should be mentioned that since CO2 gas is added to reaction, while PDC from Z. palmae is more stable at all pHs reach a 100% CO2 composition in the gas phase, phosphate tested (Figure 6B). buffer was used in these experiments instead of bicarbonate 3.3. PDC Coupled to LDH Reaction Using Gaseous buffer to maintain the pH at 7. CO2. According to the results obtained in the PDC study, The results show that ScPDC produces significantly less pyruvate could not be detected neither with PDC from S. acetoin than ZpPDC under the same reaction conditions, cerevisiae nor with PDC from Z. palmae, meaning that a showing that ZpPDC is more efficient for acetoin synthesis selection of the most suitable enzyme could not be performed. than ScPDC. However, no lactic acid was detected in any of Therefore, the conditions studied for the reaction should be the coupled reactions, even when using reaction conditions 21731 https://doi.org/10.1021/acs.est.3c05455 Environ. Sci. Technol. 2023, 57, 21727−21735 Environmental Science & Technology pubs.acs.org/est Article Figure 7. Lactate and acetoin concentration obtained with PDC variants coupled with LDH. Reaction conditions: phosphate buffer 250 mM at pH 7, acetaldehyde 5 mM, NADH 10 mM, TPP 1 mM, MgCl2 1 mM, 1 atm of CO2, PDC 10 U mL−1, and LDH 10 U mL−1 at 25 °C and 500 rpm agitation. The reaction time was 24 h. PDC from Z. palmae (ZpPDC) and PDC from S. cerevisiae (ScPDC) were used independently. Error bars correspond to standard deviation (n = 2). that should favor product formation (substrate increase and ScADH and TmLDH activities are higher than ScPDC (1 coupling of a reaction highly shifted toward product), ScADH/ScPDC and 1 TmLDH/ScPDC). Similar results suggesting how challenging it is to reverse the decarboxylation (15.54 ± 0.83 μM) are reached when ScADH/ScPDC was reaction. Finally, ScPDC was selected for further studies for the 1.25 and TmLDH/ScPDC was 0.5. On the other hand, the multienzymatic system development considering the lower best results are obtained when ScADH activity is lower than amounts of byproduct obtained and the hyperactivation ScPDC (0.5 ScADH/ScPDC) and TmLDH activity is higher detected, which were promising features of this enzyme than ScPDC, reaching 29.29 ± 0.30 μM of lactic acid. compared to the ZpPDC. These results suggest that two events should be promoted at 3.4. Complete Multienzymatic Reaction. 3.4.1. Enzyme the same time to increase lactic acid synthesis: PDC reaction Ratio Optimization. Once the ScPDC was selected as the final rate should be favored over the formation of acetaldehyde, candidate, the complete multienzymatic reaction was tested at while a lactate synthesis rate should be high enough to shift the 1 atm of CO2. First, the enzyme ratio was studied since it has multienzymatic system toward product formation. been reported as a key parameter on the overall performance of 3.4.2. Reaction Media Optimization. Reaction media the multienzymatic systems.25 engineering was applied to increase the process metrics. In Since no lactic acid was detected when ScPDC was coupled that sense, other aqueous buffers were tested: Tris−HCl, to TmLDH, the initial activity of ScPDC was increased from citrate, and MOPS using the enzyme ratios and amounts that 10 to 150 U mL−1 aiming to increase the reaction rate of the gave the highest lactic acid concentration in phosphate buffer, limiting step: the conversion of acetaldehyde to pyruvate. Then, ScADH/ScPDC and TmLDH/ScPDC ratios of 0.5, 1, 0.5 ADH/PDC, 1.25 PDC/LDH, and 150 U mL−1 ScPDC. and 1.25 were tested maintaining constant the initial ScPDC The multienzymatic system performed with Tris−HCl led to activity. no lactic acid formation. Thus, this buffer was discarded for The obtained results are listed in Table 1. There is a strong further studies. Results using citrate buffer and MOPS are effect of enzyme ratios on the lactic acid concentration. The depicted in Figure 8A,B. Regarding the citrate buffer (Figure lowest value (10.87 ± 1.19 μM) is obtained when both 8A), the lactic acid concentration follows a sigmoidal shape, which may be explained by the increase of acetaldehyde from Table 1. Lactic Acid Formation from Ethanol and CO2 2.36 mM at 24 h to 3.95 mM at 96 h, where lactic acid has a Catalyzed by the Multi-Enzymatic System Formed by second increase due to a raise of substrate (acetaldehyde) ScADH, ScPDC, and TmLDH at Different Enzyme Ratios of available for ScPDC. At the end of the reaction, lactic acid ScADH/ScPDC and TmLDH/ScPDCa concentration reaches 49.24 μM, 1.7-fold higher compared to the result obtained with phosphate buffer. NADH has a peak ScADH/ScPDC TmLDH/ScPDC lactic acid (μM) concentration at 24 h (2.58 mM) and then decreases until the 1.25 0.50 15.54 ± 0.83 end of reaction, indicating that this cofactor is consumed at a 0.50 1.25 29.29 ± 0.30 higher rate than it is produced. However, the consumed 1.00 1.00 10.87 ± 1.19 NADH does not correspond to the lactic acid produced, a therefore this decrease may correspond to NADH instability Reaction conditions: phosphate buffer 250 mM pH 7, 1 atm CO2, ethanol 1 M, NAD+ 10 mM, MgCl2 1 mM, TPP 1 mM at 25 °C and under the reaction conditions. It should be mentioned that 500 rpm agitation, and 150 U mL−1 PDC. The reaction time was 24 ethanol decreases from 1050 to 900 mM at the very beginning h. ADH from S. cerevisiae, PDC from S. cerevisiae, and LDH from T. of the reaction probably due to evaporation during CO2 maritima. sparging. 21732 https://doi.org/10.1021/acs.est.3c05455 Environ. Sci. Technol. 2023, 57, 21727−21735 Environmental Science & Technology pubs.acs.org/est Article Figure 8. Concentration profile of substrate and products of the multienzymatic system over time. (A) Reaction performed using citrate buffer. Reaction conditions: citrate buffer 250 mM at pH 7, NAD+ 10 mM, ethanol 1 M, TPP 1 mM, MgCl2 1 mM, 1 atm of CO2, ADH 75 U mL−1, PDC 150 U mL−1, and LDH 187.5 U/mL at 25 °C and 500 rpm agitation. The reaction time was 96 h. (B) Reaction performed using MOPS buffer. Reaction conditions: MOPS buffer 250 mM at pH 7, ethanol 1 M, TPP 1 mM, MgCl2 1 mM, 1 atm of CO2, ADH 75 U mL−1, PDC 150 U mL−1, and LDH 187.5 U mL−1 at 25 °C and 500 rpm agitation. The reaction time was 144 h. ADH from S. cerevisiae, PDC from S. cerevisiae (ScPDC), and LDH from T. maritima were used. On the other hand, when the reaction was performed using Higher concentrations of lactic acid were obtained using MOPS buffer (Figure 8B), compared with citrate buffer, a citrate buffer and MOPS compared to the results using similar profile is obtained. However, a higher lactic acid phosphate buffer may be due to the ScPDC performance since concentration was detected, reaching 250 μM at the end of the it has been previously described that it is competitively reaction, surpassing by 5-fold and 8.5-fold the concentration inhibited by phosphates, with a relatively low Kip (inhibition using citrate and phosphate, respectively (Figures S3 and S4). constant) of 14.7 mM.27 Taking into account that MOPS has a pKa of 7.2,26 the In previous works where this multienzymatic system was first reaction can be buffered at pH 7, with a final reaction pH of described, a lactic acid concentration of 87 μM is reported as 6.6. As seen in previous experiments (Figures 2 and 3), ScPDC the maximum reached value.28 Therefore, the present work and TmLDH are most active and stable at pH 6−7. Since the represents a step-forward on the multienzymatic systems for reaction pH with MOPS is maintained in this range, ScPDC CO2 valorization into lactic acid thanks to the followed and TmLDH have a higher activity and stability, which could approach, leading to the highest lactic acid concentration ever explain the higher production of lactic acid. reported up to date. 21733 https://doi.org/10.1021/acs.est.3c05455 Environ. Sci. Technol. 2023, 57, 21727−21735 Environmental Science & Technology pubs.acs.org/est Article Figure 9. Concentration profile of the substrate and products of the multienzymatic system over time. Reaction conditions: MOPS buffer 250 mM at pH 7, ethanol 1 M, NAD+ 10 mM, TPP 1 mM, MgCl2 1 mM, 1 atm of synthetic gas mixture, ADH 75 U mL−1, PDC 150 U mL−1, and LDH 187.5 U/mL at 25 °C and 500 rpm agitation. Reaction time was 120 h. ADH from S. cerevisiae, PDC from S. cerevisiae (ScPDC), and LDH from T. maritima were used. 3.4.3. Reaction Using Synthetic Gas Mixture Mimicking affects the last enzyme of the multienzymatic system, that Real Iron and Steel Industry Off-Gases. In iron and steel requires NADH to reduce pyruvic acid into lactic acid. Despite production, about 248.4 Mtonnes of CO2 are being emitted these unfavorable conditions mimicking industrial gas each year.29,30 Steel works are optimized to achieve a tangible composition, the system is still able to produce lactic acid. environmental improvement and a commercial benefit through the following: (i) increasing the share of electric arc furnaces and (ii) integrating the utilization of the process gases for ■ ASSOCIATED CONTENT * Supporting Information sı energy generation. Though, it is an industrial activity that still The Supporting Information is available free of charge at represented approximately 6.7% of total world CO 2 https://pubs.acs.org/doi/10.1021/acs.est.3c05455. emissions.31 In order to test the feasibility of the multienzymatic system Materials and methods and analysis of standards and to be applied with real industrial off-gases, a synthetic gas reaction samples (PDF) mixture mimicking the blast furnace off-gas composition of iron and steel industry was tested (24.5% CO2, 46.6% N2, 23.9% CO, 1.2% O2, and 3.8% H2) (data provided by Arcelor Mittal in the frame of BIOCON-CO2 project). The conditions ■ AUTHOR INFORMATION Corresponding Author were those obtained previously after the reaction optimization. Marina Guillén − Department of Chemical, Biological and As shown in Figure 9, the reaction may be divided into Environmental Engineering, Universitat Autònoma de phases. In the first phase, ScADH enzyme consumes ethanol Barcelona, Bellaterra, Catalonia 08193, Spain; orcid.org/ and NAD+ to produce acetaldehyde and NADH, which 0000-0002-9740-9966; Phone: +34 93 581 4793; accumulates during the first hours of the reaction (up to 72 Email: marina.guillen@uab.cat; Fax: +34 93 581 2013 h). In the second phase, ScPDC and TmLDH begin to consume acetaldehyde to produce the final product, lactic acid, Authors which reaches a concentration of 62 μM. It can be observed Albert Carceller − Department of Chemical, Biological and that during this second phase, the byproduct produced by the Environmental Engineering, Universitat Autònoma de PDC begins to accumulate. This lower amount of lactic acid Barcelona, Bellaterra, Catalonia 08193, Spain and the production of acetoin can be caused by a lower Gregorio Á lvaro − Department of Chemical, Biological and concentration of CO2 in the media. After one molecule of Environmental Engineering, Universitat Autònoma de acetaldehyde binds to the TPP cofactor inside the active center Barcelona, Bellaterra, Catalonia 08193, Spain; of PDC, CO2 and a second molecule of acetaldehyde compete orcid.org/0000-0002-2924-8902 to produce pyruvic acid and acetoin, respectively. Therefore, Complete contact information is available at: this multienzymatic system shows that higher amounts of CO2 https://pubs.acs.org/10.1021/acs.est.3c05455 are required to favor the carboxylation reaction over the carboligation reaction. Moreover, NADH concentration Notes decreases before lactic acid is produced, which unfavorably The authors declare no competing financial interest. 21734 https://doi.org/10.1021/acs.est.3c05455 Environ. Sci. Technol. 2023, 57, 21727−21735 Environmental Science & Technology pubs.acs.org/est Article ■ ACKNOWLEDGMENTS This project has received funding from the European Union’s (15) Obert, R.; Dave, B. C. Enzymatic Conversion of Carbon Dioxide to Methanol: Enhanced Methanol Production in Silica Sol− Gel Matrices. J. Am. Chem. Soc. 1999, 121 (51), 12192−12193. Horizon 2020 Research and Innovation programme under (16) Alcover, N.; Carceller, A.; Á lvaro, G.; Guillén, M. Zymobacter grant agreement no. 761042 (BIOCON-CO2). This output palmae pyruvate decarboxylase production process development: reflects the views only of the author and the European Cloning in Escherichia coli, fed-batch culture and purification. Eng. Life Comission cannot be held responsible for any use which may Sci. 2019, 19, 502−512. be made of the information contained therein. The research (17) Benito, M.; Román, R.; Ortiz, G.; Casablancas, A.; Á lvaro, G.; group is recognized by Generalitat de Catalunya as 2017 SGR Caminal, G.; González, G.; Guillén, M. Cloning, expression, and one- 1462. Authors also thanks Agencia de Gestió d’Ajuts step purification/immobilization of two carbohydrate−binding module−tagged alcohol dehydrogenases. J. Biol. Eng. 2022, 16, 1−14. Universitaris I de Recerca (Generalitat de Catalunya) for the (18) Yang, Z. Y.; Moure, V. R.; Dean, D. R.; Seefeldt, L. C. Carbon PhD scholarship (FI: 2019FI_B 00139) of Albert Carceller. Dioxide Reduction to Methane and Coupling with Acetylene to Form Marina Guillén also thanks the “Programa Talent UAB-Banc Propylene Catalyzed by Remodeled Nitrogenase. Proc. Natl. Acad. Sci. de Santander”. U.S.A. 2012, 109 (48), 19644−19648. (19) Wen, H.; Zhang, L.; Du, Y.; Wang, Z.; Jiang, Y.; Bian, H.; Cui, ■ REFERENCES (1) Chen, Y.; Connors, S. Global Warming of 1.5 °C: Summary for J.; Jia, S. Bimetal Based Inorganic-Carbonic Anhydrase Hybrid Hydrogel Membrane for CO2 Capture. J. CO2 Util. 2020, 39, 101171. (20) Calzadiaz-Ramirez, L.; Meyer, A. S. Formate Dehydrogenases Policymakers, Technical Summary and Frequently Asked Questions; for CO2 Utilization. Curr. Opin. Biotechnol. 2022, 73, 95−100. Masson-Delmotte, V., Pörtner, H.-O., Skea, J., Zhai, P., Roberts, D., (21) Wrba, A.; Jaenicke, R.; Huber, R.; Stetter, K. O. Lactate Shukla, P. R., Pirani, A., Moufouma-Okia, W., Péan, C., Pidcock, R., Dehydrogenase from the Extreme Thermophile Thermotoga Connors, S., Matthews, J. B. R., Chen, Y., Zhou, X., Gomis, M. I., Maritima. Eur. J. Biochem. 1990, 188 (1), 195−201. Lonnoy, E., Maykock, T., Tignor, M., Waterfield, T., Eds.; (22) Sahm, H. Acetoin and Phenylacetylcarbinol Formation by the Intergovernmental Panel on Climate Change, 2019. Pyruvate Decarboxylases of Zymomonas Mobilis and Saccharomyces (2) European Comission. Energy Roadmap 2050; Global CCS Carlsbergensis Stephanie Bringer-Meyer. Biocatal. Biotransform. 1988, Institute: Luxembourg, 2012. 1 (4), 321−331. (3) Gowd, S. C.; Ganeshan, P.; Vigneswaran, V. S.; Hossain, M. S.; (23) Siegert, P.; McLeish, M. J.; Baumann, M.; Iding, H.; Kneen, M. Kumar, D.; Rajendran, K.; Ngo, H. H.; Pugazhendhi, A. Economic M.; Kenyon, G. L.; Pohl, M. Exchanging the Substrate Specificities of Perspectives and Policy Insights on Carbon Capture, Storage, and Pyruvate Decarboxylase from Zymomonas Mobilis and Benzoylfor- Utilization for Sustainable Development. Sci. Total Environ. 2023, mate Decarboxylase from Pseudomonas Putida. Protein Eng., Des. Sel. 883, 163656. 2005, 18 (7), 345−357. (4) International Energy Agency. Energy Technologies Perspective (24) Crout, D. H. G.; Dalton, H.; Hutchinson, D. W.; Miyagoshi, M. 2020: Special Report on Carbon Capture Utilisation and Storage; CCUS Studies on Pyruvate Decarboxylase: Acyloin Formation from in Clean Energy Transition: France, 2020. https://iea.blob.core. Aliphatic, Aromatic and Heterocyclic Aldehydes. J. Chem. Soc., Perkin windows.net/assets/181b48b4-323f-454d-96fb-0bb1889d96a9/ Trans. 1 1991, No. 5, 1329−1334. CCUS_in_clean_energy_transitions.pdf. (25) Gmelch, T. J.; Sperl, J. M.; Sieber, V. Optimization of a (5) Aurie Dean, J.; Adolph Lange, N. Lange’s Handbook of Chemistry, Reduced Enzymatic Reaction Cascade for the Production of L- 15th ed.; McGraw-Hill: Knoxville, US, 1999. Alanine. Sci. Rep. 2019, 9 (1), 11754. (6) Ampelli, C.; Perathoner, S.; Centi, G. Co2 Utilization: An (26) Häring, M.; Pérez-Madrigal, M.; Kühbeck, D.; Pettignano, A.; Enabling Element to Move to a Resource-and Energy-Efficient Quignard, F.; Díaz, D. DNA-Catalyzed Henry Reaction in Pure Water Chemical and Fuel Production. Philos. Trans. R. Soc., A 2015, 373 and the Striking Influence of Organic Buffer Systems. Molecules 2015, (2037), 20140177. 20 (3), 4136−4147. (7) Shi, J.; Jiang, Y.; Jiang, Z.; Wang, X.; Wang, X.; Zhang, S.; Han, (27) Stevenson, B. J.; Liu, J. W.; Ollis, D. L. Directed Evolution of P.; Yang, C. Enzymatic Conversion of Carbon Dioxide. Chem. Soc. Yeast Pyruvate Decarboxylase 1 for Attenuated Regulation and Rev. 2015, 44 (17), 5981−6000. Increased Stability. Biochemistry 2008, 47 (9), 3013−3025. (8) Schlager, S.; Dibenedetto, A.; Aresta, M.; Apaydin, D. H.; (28) Tong, X.; El-Zahab, B.; Zhao, X.; Liu, Y.; Wang, P. Enzymatic Dumitru, L. M.; Neugebauer, H.; Sariciftci, N. S. Biocatalytic and Synthesis of L-Lactic Acid from Carbon Dioxide and Ethanol with an Bioelectrocatalytic Approaches for the Reduction of Carbon Dioxide Inherent Cofactor Regeneration Cycle. Biotechnol. Bioeng. 2011, 108 Using Enzymes. Energy Technol. 2017, 5 (6), 812−821. (2), 465−469. (9) Sheldon, R. A.; Woodley, J. M. Role of Biocatalysis in (29) World Steel Association. www.worldsteel.org. Sustainable Chemistry. Chem. Rev. 2018, 118 (2), 801−838. (30) World Steel Association. Crude steel production for the 65 (10) Alissandratos, A.; Easton, C. J. Biocatalysis for the Application countries reporting to worldsteel. Press release; April 21, 2015. of CO2 as a Chemical Feedstock. Beilstein J. Org. Chem. 2015, 11, (31) International Energy Agency data. https://www.iea.org/data- 2370−2387. and-statistics. (11) Lu, Y.; Jiang, Z.; Xu, S. w.; Wu, H.; Wu, H. Efficient Conversion of CO2 to Formic Acid by Formate Dehydrogenase Immobilized in a Novel Alginate-Silica Hybrid Gel. Catal. Today 2006, 115 (1−4), 263−268. (12) Kim, S.; Kim, M. K.; Lee, S. H.; Yoon, S.; Jung, K. D. Conversion of CO2 to Formate in an Electro enzymatic Cell Using Candida Boidinii Formate Dehydrogenase. J. Mol. Catal. B: Enzym. 2014, 102, 9−15. (13) Long, N.; Lee, J.; Koo, K. K.; Luis, P.; Lee, M. Recent Progress and Novel Applications in Enzymatic Conversion of Carbon Dioxide. Energies 2017, 10 (4), 473. (14) Kuwabata, S.; Tsuda, R.; Yoneyama, H. Electrochemical Conversion of Carbon Dioxide to Methanol with the Assistance of Formate Dehydrogenase and Methanol Dehydrogenase as Biocata- lysts. J. Am. Chem. Soc. 1994, 116 (12), 5437−5443. 21735 https://doi.org/10.1021/acs.est.3c05455 Environ. Sci. Technol. 2023, 57, 21727−21735