EP3788157A1 - Method for converting co2 by means of biological reduction - Google Patents

Abstract

The invention relates to a method for converting CO2 by means of biological reduction, comprising a step of contacting a liquid phase containing the bacterium Stenotrophomonas maltophilia with a gaseous phase containing CO2 under conditions that allow the production of formate and/or methane from said CO2. The method according to the invention can be carried out, in particular, in a closed reactor, a semi-closed reactor or a continuous reactor, optionally electrochemically assisted.

Description

 Process for recovering CO2 by biological reduction

The invention relates to a microbiological process for reducing C0 ₂ to formate and / or methane, catalyzed by Stenotrophomonas maltophilia. According to the invention, this CO2 reduction process can be intensified by the addition of electrons and exogenous protons. This contribution can be achieved by the implementation of the bacterium in an electrolytic device (bio-electrolyser). The process according to the invention can advantageously be used to treat industrial or agricultural fumes rich in CO2 or biogas containing mainly a mixture of methane and CO2, in order to reduce the CO2 content of said fumes or biogas.

Technological background

Carbon dioxide (CO2) accounts for more than 75% of global greenhouse gas (GHG) emissions, with annual anthropogenic emissions of around 25 to 35 Gt. These emissions come mainly from heavy industries and industrial production. such as cement plants, aluminum and steel production sites, coal or oil-fired power plants, etc., as well as road transport.

The use of CO2 as raw material at the industrial level is currently considered as one of the possible solutions to reduce the emissions of this gas to the atmosphere.

In this context, the direct conversion of anthropogenic CO2 into alternative fuels (such as methane, C ₁ -C ₅ alcohols or formic acid) has become a major research topic for fifty years. However, the CO2 molecule has a high chemical stability and the activation of CO2 therefore requires relatively high temperatures and pressures as well as efficient and selective catalysts.

Homogeneous or heterogeneous catalysts have been developed which catalyze hydrogenation, artificial photosynthesis or electrochemical reduction reactions to obtain fuels from CO2. Nevertheless, these chemical conversion routes still have low yields, durability and / or selectivity.

Biological processes of CO2 activation have also been developed. These biological processes are particularly interesting because they consume little energy and generally avoid reagents and / or by-products that could have a negative impact on the environment. To date, however, only processes implementing Photosynthetic microorganisms such as microalgae or cyanobacteria have reached an industrial scale. The main products of the recovery of C0 ₂ by these processes are lipids or sugars which can then be transformed into biodiesel or biofuel. Open basin technology has allowed scale-up in the valuation of C0 ₂ by micro-algae. However, the development of this process is still limited because of (i) the need for sunshine which implies low depths of photo-bioreactors (and therefore large areas), (ii) mass transfer problems in nutrients, ( iii) a significant need for water, (iv) the need to maintain a basic pH (8-10) and (v) the need to harvest microorganisms to extract the compounds of interest. All these constraints translate into low productivity and relatively high costs.

At the laboratory scale, cyanobacteria have been genetically modified to directly produce alcohols from C0 ₂ . Very recently, the carbon assimilation pathway of the cyanobacterium Synechococcus elongatus PCC 7942 has been directed to the production of 2,3-butanediol from CO ₂ and glucose in the dark. Such an advance overcomes the limitations of the need for sunshine. Nevertheless, the need to add glucose makes this bioprocess unattractive from an economic point of view. In addition, the biological stability of modified cyanobacteria is still insufficient to allow its use on an industrial scale.

An enzymatic synthesis route of formate (basic formic acid form) from carbon dioxide (C0 ₂ ) was also studied. Specifically, formate dehydrogenase (FDH) enzymes extracted from microorganisms have been purified and modified to improve their catalysis performance, and used to reduce CO ₂ to formate (Alissandratos et al., 2013, Appl Environ Microbiol., 79 (2), 741-4, Spinner et al., 2012, Catal, Sci., Technol., 2, 19-28). Recently, nitrogenase enzymes have also been reported as capable of catalyzing the reduction of C0 ₂ to formate, carbon monoxide, acetate or methane, in addition to their N ₂ nitrogen reduction activity (Khadka et al. 2016, Inorg Chem., 55, 8321-8330, Zheng et al., 2018, Nature Microbiology, 3, 281-286). Nevertheless, the addition of an expensive cofactor is necessary to perform the bioconversion reaction, making these alternatives incompatible with industrial application.

Electrochemically assisted reactors (also called bio-electrolyzers) which are inoculated with electro-active litho-autotrophic bacteria constitute another biological pathway for the conversion of C0 ₂ . Various products such as formate, acetate, C3-C5 alcohols, polyhydroxybutyrate (PHB) or biomass have been obtained. In these systems, two configurations have been implemented: (1) the electrons necessary for the reduction of C0 ₂ are directly supplied by the polarized cathode and the protons come from the oxidation of the water at the anode or (2) the electrons of the cathode are transported to the bacteria by electrochemical mediators (such as dihydrogen or formate) that can be added to the medium or generated in situ at the polarized cathode. However, until now, the feasibility of the configuration (1) has been demonstrated only in the case where an organic electron donor is also present in the medium, which decreases the economic attractiveness of this system. Furthermore, the use or in situ production of formate or a clean and expensive energy carrier such as dihydrogen (¾) remains problematic in the configuration (2).

Summary of the invention

By working on the valorization of C0 _2, the inventors have discovered that the bacterium Stenotrophomonas maltophilia, known for its ability to fix CO2 by its carboxylase enzymes, is also able to reduce CO2 into formate and / or methane. Methane is particularly interesting because it is a fuel whose combustion produces few atmospheric pollutants (of the NO _x or SO _{x type} ) and releases less CO2 per unit of energy (nearly 30% less) than the combustion of energy. fossil such as oil or coal. Methane is also a precursor to synthesis gases (H2 / CO / CO2 mixtures) which are currently used for the industrial production of methanol, ammonia and hydrocarbons (Société Chimique de Lrance, 2013). Formic acid, or methanoic acid, which is the acid form of formate, is a naturally occurring aliphatic carboxylic acid, which is in the form of a colorless liquid with a persistent odor. Formic acid has many applications in the leather and textile industry, in the composition of dyes and treatment products, insecticides, solvents, fumigants, anti-mold products, etc., but also in perfumery and the food industry as an aromatic molecule. Recently, it has been shown that formic acid can be used as fuel in fuel cells (Yu et al., 2008, Journal of Power Sources, 182 (1), 124-132) or to store hydrogen which can then be released by dehydrogenation of formic acid (Lellay et al., 2008, Angewandte Chemie, 120 (21), 4030-4032). The possibility of using formic acid as an energy carrier suggests that the need for formic acid will increase in the coming years. In addition, the formate can be used in secondary bioprocesses to obtain energy sources such as methane (Conrad, Contribution of hydrogen to methane production and control of hydrogen concentrations in methanogenic soils and sediments, FEMS Microbiology Ecology, 28 (3 ) (1999) 193-202) or alcohols (Li et al., Integrated Electromicrobial Conversion of CO2 to higher alcohols, Science, 335 (2012) 1596; Pen et al., An innovative membrane bioreactor for methane bio hydroxylation. Bioresource Technology, 174 (2014) 42-52).

This new pathway of reduction by S. maltophilia has many advantages over existing bioprocesses. Indeed, this biocatalyst is a native cell. Compared to enzymes, it is therefore not necessary to carry out protein extractions or to add cofactors under stoichiometric conditions, since the cells already have their own stock. This cell does not require genetic modifications either, which are poorly supported on an industrial scale. Moreover, no contribution of dihydrogen (H ₂ ) or photons (light) is necessary for the reduction reaction of C0 ₂ because the cell has an intracellular stock of proton and electron donor, which it accumulates during the pre-culture phase (for example on acetate derived from methane oxidation or on a mixture of peptone and yeast extract).

The subject of the invention is therefore a process for the recovery of C0 ₂ by biological reduction comprising a step of contacting a liquid phase containing the bacterium Stenotrophomonas maltophilia with a gas phase containing C0 ₂ under conditions permitting the reduction of said C0 _2, especially in formate and / or methane.

In particular, the contacting step can be carried out at 30 ° C., +/- 10 ° C., in particular at 30 ° C. +/- 5 ° C., under atmospheric pressure or slight overpressure (up to 0.5 bar). ).

The liquid phase advantageously comprises mineral salts useful for the maintenance or survival of the bacteria during the CO ₂ reduction reaction. Those skilled in the art will be able to adapt the composition to mineral salts. For example, the liquid phase contains water and phosphates and / or MgCl ₂ , etc.

Advantageously, the step of bringing the aqueous liquid phase into contact with the CO ₂ is carried out in a closed reactor or in a semi-closed reactor.

The subject of the invention is also the intensification of said process for the biological reduction of C0 ₂ . The intracellular store of proton and electron donor being limited, the inventors have demonstrated that it is possible to provide exogenous or extracellular protons and electrons, that is to say other than intracellular protons and electrons. Stenotrophomonas maltophilia so that the C0 ₂ reduction reaction persists over time. For example, it is possible to introduce into the reactor a donor of electrons and protons, such as a biopolymer and in particular polyhydroxybutyrate (PHB) and / or a electrochemical assistance type bioelectrolysis. Indeed, the inventors have shown that Stenotrophomonas maltophilia is capable to use the electrons of a weakly polarized cathode and the protons resulting from the oxidation of the water at the anode. In one embodiment, the energy provided to the bio-electrolysis is of green origin, solar in particular, which makes this process autonomous from the energy point of view.

Thus, the subject of the invention is also a process for intensifying the reduction of CO ₂ by the bacterium Stenotrophomonas maltophilia comprising a step according to which exogenous PHB or an electrochemical assistance (bioelectrolysis) is added to the reactor.

According to the invention, identified products of the CO ₂ reduction are formate and / or methane. It is also possible to couple the use of Stenotrophomonas maltophilia to another microorganism, in particular a microorganism able to use the formate contained in the liquid to produce other molecules of interest, such as methane, lactate or alcohols, especially C1-C5 (such as methanol, isobutanol or methyl-1-butanol).

The invention also relates to the use of a bacterium Stenotrophomonas maltophilia for the production of formate and / or methane and / or more generally any product whose carbon oxidation number is lower than that of C0 _2, by reduction. of C0 ₂ .

The subject of the invention is also a process for the treatment of a biogas or of industrial fumes rich in C0 ₂ or of agricultural fumes rich in C0 _2, comprising a step of reduction of C0 ₂ by biological reduction, comprising a step according to which contacting said biogas or industrial fumes or agricultural fumes with a liquid phase containing the bacterium Stenotrophomonas maltophilia under conditions allowing the reduction of C0 ₂ .

The invention also relates to a process for producing formate from CO ₂ , comprising a step according to which a CO ₂ -containing gas phase is brought into contact with a liquid phase containing the bacterium Stenotrophomonas maltophilia under conditions allowing the reduction C0 ₂ in formate.

The subject of the invention is also a process for producing methane from CO ₂ , comprising a stage in which a CO ₂ -containing gas phase is brought into contact with a liquid phase containing the bacterium Stenotrophomonas maltophilia under conditions allowing the reduction C0 ₂ in methane. Brief description of the figures

Figure 1 NMR spectra of a Stenotrophomonas maltophilia suspension implementation for reduction tests of ¹³ C0 ₂ in a closed reactor and under baseline conditions: (a) at t = 0 days, (b) at t = 8 days and (c) at t = 35 days.

Figure 2. Production kinetics in ¹³ C-formate, obtained during ¹³ CO ₂ reduction tests carried out in a closed reactor and under reference conditions with the bacterium Stenotrophomonas maltophilia. Each color represents an independent test.

Figure 3. Reduction Tests of ¹³ C0 ₂ in a closed reactor with a suspension of S. maltophilia prepared in a reaction medium enriched with ammonium (ABSA); the other reaction parameters being unchanged compared to the reference conditions. A ¹³ CO ₂ reduction test is also conducted simultaneously under reference conditions. (A) Evolution of the concentration of ammonium ions (NH ₄ ⁺ ): in the MREA (O) and in the reference reaction medium (#). (B) Evolution of ¹³ C-formate concentration (H ¹³ COO): in MREA (D) and in the reference reaction medium (A).

Figure 4. NMR spectrum of a suspension of S. maltophilia initially enriched with 1.5 mM ¹³ C-formate (H ¹³ COO).

Figure 5. NMR spectra of the reaction medium containing the consortium S. maltophilia / M. trichosporium OB3b obtained in the presence of ¹³ C0 ₂ (a) t = 0 days, ¹³ where the C0 ₂ and its basic form H ¹³ C0 ₃ ^_ are present and (b) t = 8 days showing the appearance of a ¹³ C-formate peak.

Figure 6. NMR spectra of the reaction medium containing the S. maltophilia / M. oxidans consortium obtained in the presence of ¹³ CO ₂ at (a) t = 5 days; (b) t = 8 days and (c) t = 12 days.

Figure 7. Scheme and principle of an embodiment of a semi-closed reactor, assisted by electrolysis (or bio-electrolyser) for the CO ₂ reduction reaction according to the method of the invention.

Figure 8. Evolution of the cathodic current density measured over time in a bioelectrolyser inoculated with Stenotrophomonas maltophilia, according to one embodiment of the method of the invention. The bias voltage is set at -0.7 V vs. Ag / AgCl. Six phases are distinguished, corresponding to the various operating conditions used: (a) bubbling of CO ₂ (25 mL.min ¹ ); (b) Arg argon bubbling (10 mL.min ¹ ); (c) stopping argon bubbling; (d) bubbling CO ₂ (5 mL.min ¹ ); (e) argon bubbling (100 mL.min ¹ ) and (f) bubbling of CO2 (25 mL.min ¹ ). The symbol (#) indicates the polarization interrupts for the acquisition of cyclic voltammetries carried out at 10 mV-s ¹ .

Detailed description of the invention

The inventors have discovered that Stenotrophomonas maltophilia is capable of producing formate and methane (CH ₄ ) by direct reduction of CO2 under mild operating conditions. In a particularly interesting way, the inventors have demonstrated that Stenotrophomonas maltophilia is able to produce formate and methane (CH ₄ ) using CO2 as the only carbon source, without adding cofactors, organic molecules or dihydrogen. (¾) or expensive growth factors.

In the context of the invention, CO2 reduction is understood to mean any reaction that makes it possible to pass the carbon oxidation number of CO 2 to a lower degree than in CO 2.

The bacterium Stenotrophomonas maltophilia is a Gram-negative aerobic bacterium of the family Pseudomonadaceae. It has so far never been used for industrial purposes. However, the inventors have discovered that this bacterium is particularly interesting for the production of molecules of interest from CO2. Indeed, the inventors have discovered that this bacterium has the capacity to produce formate and / or methane from CO2 as the only carbon source, after having been advantageously put beforehand in culture under autotrophic conditions, that is to say in aerobic pure culture on a usual Lysogeny Broth Miller type organic medium, or in co-culture with a methanotrophic bacterium, such as Methylosinus trichosporium OB3b, on a mineral medium placed in contact with a methane / air mixture (1: lv / v). The co-cultivation route is particularly advantageous because it ensures an optimal carbon balance; indeed, methane, of renewable origin, is the only source of carbon required for cultivation.

Advantageously, the liquid phase comprises at least 3 g of dry cells / L of the bacterium Stenotrophomonas maltophilia, at least 10 g of dry cells / L, at least 20 g of dry cells / L,

Dry cells / L, 40 g dry cells / L, 50 g dry cells / L, 60 g dry cells / L, 70 g dry cells / L,

80 g of dry cells / L, 90 g of dry cells / L, 100 g of dry cells / L.

Advantageously, and whatever the mode of culture, the cultures are carried out at 30 ° C., +/- 10 ° C. and at atmospheric pressure or slight overpressure (up to 0.5 bar).

According to the invention, the gas phase containing CO ₂ brought into contact with the liquid reaction medium containing the bacterium Stenotrophomonas maltophilia may be atmospheric air. The gaseous phase may otherwise be pure C0 ₂ or a gaseous mixture, such as a CCh-air mixture, CO2-N2, CO2-O2, CO2-CH4, CO2-H2, CO2-N2-H2, CO2-CH4- H2, C0 ₂ -CH ₄ -air, CO2-CH4-N2 or CO2-CH4-O2. In one embodiment, the gas phase contains or consists of biogas. In another embodiment, the gas phase contains or consists of industrial or agricultural fumes rich in CO2. In a particular embodiment, the gaseous phase comprises at least 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% of CO2. Advantageously, the gaseous phase comprises at least 30% of CO2. It is also possible to use a gaseous phase comprising 100% CO2. In one embodiment, the gas phase comprises between 30 and 100% of CO2.

In a reaction mode, the reaction medium contains 20 mM phosphate buffer at pH 7.0 ± 0.1 and 5.0 mM MgCl ₂ . In another embodiment, the reaction medium contains 1.3 mM KCl, 28.0 mM NH ₄ Cl, 29.8 mM NaHCO 3 and 5.0 mM NaH ₂ PO ₄ . Those skilled in the art are able to choose the appropriate reaction medium for the reduction of CO2 to formate by Stenotrophomonas maltophilia.

Formate and / or methane and / or any other product whose carbon oxidation number is lower than that of CO2 being obtained directly by reduction of CO2, the yield of formate and / or methane and / or any other product of which the number of oxidation of carbon is lower than that of the CO2 of the process according to the invention depends directly on the contribution of CO2. Those skilled in the art will be able to adapt the concentrations of CO2 to the desired yields.

According to the invention, the CO2 contribution can be achieved by any known means and in particular by membrane contactors (Pen et al., 2014, Bioresource Technology, 174, 42-52), a porous gas distribution sinter (Kim et al. 2010, Biotechnology and Bioprocess Engineering, 15 (3), 476-480, Duan et al., Bioresource Technology, 102 (15), 7349-7353) or a simple gas-liquid contact in a closed reactor (Pen et al. 2014, Bioresource Technology, 174, 42-52).

In one embodiment, the CO2 recovery process is implemented in a closed reactor ("batch"). Alternatively, the process can be implemented in a semi-closed reactor ("fed-batch").

According to the invention, the gaseous phase containing the CO2 can be injected into the headspace of the closed reactor. In one embodiment, it is possible to remove the air present in the headspace of the closed reactor prior to injection of the gas phase, so as to place the reaction under an atmosphere consisting solely of the gas phase. Alternatively or additionally, the gaseous phase can be injected at the center of the closed or semi-closed reactor, at the heart of the liquid phase containing the bacterium Stenotrophomonas maltophilia, by means in particular of a gas distributor.

In the case of a semi-closed reactor ("fed-batch"), the gaseous phase containing CO2 is provided continuously, preferably by bubbling, inside the reactor.

In an exemplary embodiment, the pH in the reaction medium, that is to say the liquid phase, is maintained between 5 and 8, preferably at pH 6.5, +/- 0.5. Advantageously, the CO2 dissolved in the water forms a buffer sufficient to allow the maintenance of the pH, without external regulation.

Similarly, the temperature in the reaction medium is preferably maintained between 20 ° C and 50 ° C, preferably between 25 ° C and 35 ° C, and more preferably at 30 ° C, +/- 1.

According to the invention, it is possible, prior to the reaction step, that is to say prior to bringing the bacterium into contact with the CO2 to produce the formate and / or the methane, to place the bacterium in a culture medium under conditions of temperature and pH favorable to its growth.

By "culture medium" is meant the medium in which the bacteria are grown, to produce bacterial biomass. The culture medium conventionally comprises the chemical elements strictly necessary for bacterial growth in a form that can be used by the bacteria, namely a source of carbon, mineral salts and water. Standard media are available commercially, described in the scientific literature or in the catalogs of suppliers of bacterial strains. Those skilled in the art know what are the minimum required components in the absence of which Stenotrophomonas maltophilia can not grow, and can easily grow such a bacterium. The general conditions of culture (temperature, composition of the medium, agitation, etc.) allowing the growth or maintenance of Stenotrophomonas maltophilia, are easily defined by those skilled in the art.

This prior culture step may, for example, enable the bacterium to accumulate proton and electron donors required for the reaction to reduce CO2 to formate and / or methane according to equations (1) and (2) and / or (1) and (3):

CO2 + H ⁺ + 2e-> HCOO (Equation 1)

C0 ₂ + 8H ⁺ + 8e-> CH ₄ + 2H ₂ 0 (equation 2) HCOO + 7H ⁺ + 6e- ® CH ₄ + 2 H ₂ 0 (Equation 3)

Alternatively or additionally, it is possible to add into the reactor an electron donor and exogenous protons, such as a bio-polymer, so as to provide the bacteria with the electrons and protons necessary for the reaction, even in the absence or after exhaustion of intracellular donors. In particular, it is possible to add polyhydroxybutyrate (PHB) to the reactor. In one embodiment, the PHB is previously produced by methanotrophic bacteria from methane, especially renewable (Pieja et al., Distribution and selection of poly-3-hydroxybutyrate production capacity in methanotrophic proteobacteria, Microbial Ecology, 62 (3 ) (2011) 564-573). Such an embodiment is particularly advantageous from an economic point of view.

In one embodiment, exogenous PHB is added to the reactor. Preferably, between 30 mg / L and 3 g / L, exogenous PHB in the liquid phase is added, in particular approximately 0.3 g / L.

Alternatively or additionally, and particularly in the case of a semi-closed reactor, it may be particularly advantageous to use an electrochemical assistance device. Such a device, housed inside the reactor and in contact with the bacterium, provides electrons and protons to the bacteria. The electrochemical assistance may especially be of the bio-electrolysis type (FIG. 7). Advantageously, the energy used to operate the bioelectrolyser is solar energy.

Different materials can be used for the anode and the cathode, as long as they are electron conductors. For example, the cathode may be stainless steel, graphite or felt. The anode can be platinum, stainless steel, graphite or felt.

Advantageously, the cathode / reactor volume ratio is between 1 and 100 m ² / m ³ , preferably about 10 m ² / m ³ . By "active surface" is meant the geometrical surface exposed to the anode. The projected geometric projected anode / projected geometric cathode surface ratio is preferably between 1 and 10, more preferably approximately equal to 2. The range of polarization potential can for example go from -1.5 Y to 0 Y vs Ag / AgCl, especially -1 Y at -0.5 V, for example -0.7 V or -0.8 V.

In order to optimize the CO ₂ reduction reaction _, it is possible to add ammonium and / or ammonia in the liquid phase and / or ammonia in the gas phase, in particular between 5 mmol / L and 100 mmol / L, preferably between 20 and 30 mmol / L. Alternatively or additionally, the liquid phase comprises, in addition to the bacteria Stenotrophomonas maltophilia, another micro-organism able to use the formate resulting from the reduction of CO ₂ to form more complex metabolites, such as methane, lactate or alcohols. The inventors have shown that such a consortium can lead to a synergy capable of intensifying the flow of CO2 reduced by the bacteria Stenotrophomonas maltophilia.

All or part of the process according to the invention can be implemented on a laboratory scale or on an industrial scale, that is to say on fermenters or reactors of average capacity (approximately 0.1 L at 100 L) or large capacity (100 L to several hundred m ³ ).

The invention also relates to the intensification of the CO2 reduction process by the bacterium Stenotrophomonas maltophilia by means of addition of exogenous PHB or electrochemical assistance (bioelectrolysis) as described above.

In a particular embodiment, the CO2 reduction reaction by the bacterium Stenotrophomonas maltophilia can be promoted by continuously removing the product formate in the liquid phase of the reactor in the form of formate and / or formic acid. Those skilled in the art are aware of the possible techniques for continuously removing the product formate.

In another embodiment of the process according to the invention, the formate-enriched liquid and inorganic reaction medium may be used as recovered in a secondary process or bioprocess to enhance the product formate into higher value molecules. added. The skilled person knows the possible techniques to value the formate product. In a particular embodiment, the process for producing formate according to the invention comprises the steps of:

Introduce Stenotrophomonas maltophilia into a reactor containing a reaction medium; Supply the reaction medium, or liquid phase, in a gaseous phase containing CO2; and optionally recovering the formate produced in the reactor by reduction of CO2 in the form of formate and / or formic acid.

In another embodiment of the process according to the invention, the formate-enriched mineral reaction medium may be used as recovered in a secondary process for upgrading the product formate to higher value-added molecules. In another embodiment of the process according to the invention, the product formate is recovered in the form of formic acid from the reaction medium.

In a particular embodiment, the step of recovering formic acid from the reaction medium is carried out by electrodialysis or by liquid-liquid extraction with a co-solvent. These recovery devices make it possible to continuously recover the formic acid and thus avoid the possible inhibition of the bacteria by an excess of product, while obtaining a formic acid free of impurities (such as the salts of the reaction medium).

In the case of continuous recovery of the product, a continuous supply of reaction medium can be carried out in the bioreactor. In the case of an electrodialysis recovery process, a bipolar electrodialysis module is advantageously coupled with a microfiltration module for continuously separating and recovering the formate produced in its acid form (formic acid). In particular, bipolar electrodialysis can be performed on a reaction medium without biocatalysts, obtained continuously after separating the bacteria from the medium by a tangential microfiltration module. Bipolar electrodialysis involves combining a bipolar membrane to acidify formate into formic acid and monopolar membranes to extract salts from the reaction medium.

The formate and / or formic acid obtained from such a microbiological process can be advantageously used in any industry that may need it, and in particular in the leather and textile industry, in perfumery , in the food industry, etc.

Preferably, the formate and / or formic acid obtained from such a microbiological process is used in a secondary biological process (in situ or ex situ) to obtain organic compounds of higher added value, and in particular methane, lactate or C ₁ -C ₅ alcohols. Alternatively, it is possible to directly use the reaction medium containing formate and / or formic acid obtained (s) for the implementation of the secondary biological method. If the compounds obtained are alcohols, a stripping process carried out by a bubbling of gas containing CO ₂ can be implemented to continuously recover the alcohols produced.

In a particular embodiment, it is possible, prior to the reaction step, to co-cultivate Stenotrophomonas maltophilia with one or more other methanotrophic bacterial species. It is thus possible to produce organic nutrients from methane, said nutrients being then used by Stenotrophomonas maltophilia as a source of carbon and energy. For example, the inventors have demonstrated that the co-culture under a methane / air mixture and on a mineral medium of Stenotrophomonas maltophilia with a bacterium methanotroph, such as Methylosinus trichosporium OB3b, allows the production of acetate usable by Stenotrophomonas maltophilia for its growth. The inventors have also shown that the presence of such a methanotrophic bacterium in the liquid phase during the subsequent reaction step does not interfere with the reduction of CO ₂ or the production of formate and / or methane by Stenotrophomonas maltophilia.

Alternatively or additionally, it is possible during the reaction step (that is to say putting Stenotrophomonas maltophilia into contact with the CO ₂ -containing gas phase) to add to the liquid phase one or more other microorganisms, and especially bacteria, archaea and / or yeasts, able to use formate to produce molecules with higher added value. It is also possible to co-cultivate Stenotrophomonas maltophilia with one or more of these microorganisms prior to the reaction step, and then to add all the microorganisms obtained in the liquid reaction medium for the reaction step. In one embodiment, the bacterium Microbacterium oxydans is added to Stenotrophomonas maltophilia. In one embodiment, hydrogenotrophic methanogenic archaea are added to obtain methane. In one embodiment, the bacterium Ralstonia eutropha H16 (LH74D) is added to Stenotrophomonas maltophilia, in particular to obtain C4-C5 alcohols. In one embodiment, a methanotrophic bacterium and an inhibitor of methanol dehydrogenase are added to Stenotrophomonas maltophilia in order to obtain methanol.

Alternatively, the reaction step for reducing CO2 to formate by Stenotrophomonas maltophilia is carried out in a first reactor, and the product formate is recovered to feed a second reactor in which the microorganism (s) capable of using the formate for produce molecules with higher added value. The second reactor can be fed directly with the reaction medium of the first reactor, containing the product formate, or on the contrary contain a medium to which is added the formate extracted from the reaction medium of the first reactor.

The invention also relates to the use of Stenotrophomonas maltophilia, for the production of formate and / or methane and / or any other product whose carbon oxidation number is lower than that of C0 ₂ by CO2 reduction as exposed above. The use of Stenotrophomonas maltophilia makes it possible to consume significant amounts of CO2 and thus to answer the environmental problems related to the accumulation of CO2 in the atmosphere while valuing the CO2 consumed. Other aspects and advantages of the present invention will appear in the following experimental part, which must be regarded as an illustration, in no way limiting the extent of the protection sought.

Examples

Al Materials and Methods

 1. Culture of the bacterium Stenotrophomonas maltophilia

Two culture routes of S. maltophilia have been implemented:

(i) Aerobic co-culture with methanotrophic bacterium Methylosinus trichosporium OB3b

In this case, the source of carbon and energy used is methane, which is converted by the methanotrophic bacterium into organic nutrients assimilated by S. maltophilia bacteria for its growth. In particular, acetate, at a concentration of the order of 35 mg.L ¹ , is measured in the culture reactor at the end of growth. Such a synergy between these two types of bacteria has already been reported in nature and for a consortium involving them, the production of acetate by methanotrophs has already been demonstrated under limiting oxygen conditions (Costa et al. , Denitrification with methane as electron donor in oxygen-limited bioreactors, Microbiol Appl., Biotechnol., 53 (6) (2000) 754-762).

The culture medium is a mineral medium enriched in copper and iron. The base medium is composed of: 1.06 g / L KH ₂ PO ₄ , 4.34 g / L Na ₂ HPO ₄ 12H ₂ O, 1.7 g / L NaNO 3, 0.34 g / LK ₂ SO ₄ and 0.074 g / L MgSO ₄ -7H ₂ 0. Once prepared, the pH of this basal medium is adjusted to 7.0 ± 0.1 with 0.1 M NaOH, 0.1 M or HCl, and then autoclaved. The mineral, copper and iron solutions are prepared independently and sterilized by filtration on 0.2 μm pore size cellulose acetate filters before being added to the base medium. The final mineral concentrations are: 0.57 mg / L ZnSO ₄ 7H ₂ 0, 0.446 mg / L MnSO ₄ H ₂ O, 0.124 mg / L H3BO3, 0.096 mg / L Na ₂ MoO ₄ 2H ₂ 0, 0.096 mg / L LKI and 7.00 mg / L CaCl ₂ 2H ₂ 0. The final concentrations of copper and iron are: 0.798 mg / L CuSO ₄ and 11.20 mg / L LeSO ₄ 7H ₂ 0.

The cultures are conducted in sealed closed reactors, incubated at 30 ° C. and under constant rotation (160 rpm). The headspace of the reactors is filled with a mixture of air and methane (1: / v / v); the volume of gas is three times greater than that of the liquid. Inoculation percentages of S. maltophilia can range from 10 to 50% v / v. The cultures are stopped when the optical density at 600 nm (DCboo _nm ) is constant. Two successive cultures are conducted before implementing the consortium for the C0 ₂ reduction reaction.

(il) Aerobic pure culture of bio catalyst

The bacterium S. maltophilia was first isolated from a consortium formed with the bacterium M. trichosporium OB3b (obtained according to the protocol described above). Isolation was achieved by successive transplantation of S. maltophilia colonies onto Lysogeny Broth (LB) Miller medium agars (Sigma Aldrich, France). The purity and identification of the isolated bacterium is confirmed by biochemical analyzes (see point 2 below). The strain is preserved on the one hand on LB agars stored at 4 ° C., and on the other hand, in the freezer at -20 ° C. in its native liquid culture medium to which glycerol (20% v / v) is added. The culture reactor is a closed reactor with a permeable plug which is dedicated to sterile cultures. A LB liquid culture medium is inoculated with the bacterium S. maltophilia, 5% v / v from a frozen aliquot or with all the colonies of an agar, then incubated for 24 hours incubated at 30 ° C. and under constant rotation (160 rpm). The cultures are stopped when the stationary phase starts (DCLoo _nm is constant).

Two successive cultures are conducted before implementing the biocatalyst for the C0 ₂ reduction reaction.

A correlation between the DCLoo _nm (-) and the mass concentration [X] in S. maltophilia bacteria (g dry cells.L ¹ ) is established to determine the productivities in formate; the correlation is obtained on the basis of five independently replicated points:

[X] (g dry cells.L ^-1 ) = 0.3035 X ODeOnmn (R ² = 0.995)

2. Biochemical analyzes

The Stenotrophomonas maltophilia isolate from co-culture was first characterized with Gram stain and mass spectrometry. Gram staining has shown that the isolate is a Gram-negative bacillus. The mass spectrometry analysis confirmed that the isolate is pure and identify the bacterium S. maltophilia which is indeed a Gram-negative bacillus. In mass spectrometry, it should be noted that a score above 1.90 indicates reliable identification and that the score obtained for S. maltophilia is 2.18. Before each culture, a qualitative analysis of the colonies of S. maltophilia spread on agar plates is carried out in order to detect a possible contamination. In addition, RNA sequencing 16S of the isolate confirmed that the bacterium identified is S tenotrophomonas maltophilia.

3. Preparation of the bacterial suspension implemented in the CO2 reduction tests

At the end of the culture, the cells are collected by centrifugation at 4 ° C and 4000 g for 20 min. They are then resuspended in a 20 mM phosphate buffer at pH 7.0 and then centrifuged again under the same conditions as above and the bacterial pellet is recovered. A reaction medium containing 20 mM phosphate buffer pH 7.0 and 5 mM MgCk was used to resuspend the bacterial pellet and obtain a DCLoo _nm of (i) 10.3 ± 0.5 (ie 3.1 ± 0 , 2 g of dry cell / L) for closed reactor tests and (ii) 6.6 ± 0.3 (ie 2.0 ± 0.1 g of dry cell / L) for electrolysis assisted assays. A sample of the prepared suspension is always stored at -20 ° C with glycerol (20% v / v) for subsequent biochemical analyzes. The rest of the suspension is used immediately for CO2 reduction tests. The reaction medium does not contain organic nutrients for the growth of the biocatalyst, so it is used as a resting cell.

4. ¹³ CO ₂ reduction tests in closed reactor

The bacterial suspension obtained in the previous step is distributed in sealed 60 mL closed reactors by adding 6 mL of suspension to each reactor. To assess the ability of the biocatalyst to reduce CO2, the headspace is filled with a gaseous mixture containing ¹³ CC> 2 which is sterilized by filtration on Teflon filter pore diameter of 0.2 mih. Different gas mixtures were tested: ¹³ C0 ₂ : atmospheric air (3: 7 v / v), ¹³ CC> 2: N2 (3: 7 v / v) and ¹³ CC> 2 pure. For each experiment, a set of several reactors is prepared under identical conditions and incubated at 30 ° C. with constant stirring (160 rpm). Each kinetic is followed for more than 20 days and up to 35 days; for this, at each sampling time, a reactor is sacrificed for the analyzes. Different measurements are made: optical density at 600 nm (DCLoonm), pH, assay of the formate by Gas Chromatography coupled to mass spectrometry (GC-MS), characterization of the gaseous composition of the headspace by GC-MS, determination of intracellular poly-3-hydroxybutyrate (PHB) by GC-MS and determination of ammonium ions (NH ₄ ⁺ ) by ion chromatography (IC). The products from ¹³ CC> 2 are highlighted in the reaction medium and in the headspace by GC-MS and NMR. Counts of viable bacteria are also performed at the beginning and at the end of the kinetics; for this, the samples are grown under aerobic autotrophic conditions. Finally, whites made with the reaction medium alone (without bacteria) in a mixture of ¹³ C0 ₂ : air atmospheric (3: 7 v / v) showed no contamination over the duration of the tests, confirming the sterility of closed reactor tests.

5. CO2 reduction tests in semi-closed reactor, assisted by electrolysis

A glass reactor is used. Its useful liquid volume is 60 mL and its head space represents about 40% of its total volume. A lid is screwed on the reactor and the sealing of the junction lid-reactor is guaranteed by a seal. This cover has an inlet port for the gas supply ( ¹² C0 ₂ 100%) which is carried out at the heart of the solution by a gas distributor, a port for the gas outlet and 3 ports for the positioning of the electrodes . The reactor is thermostated at 30 ° C. with constant stirring (300 rpm).

The electrochemical system is a conventional 3-electrode device (working electrode, reference electrode and counter-electrode). The working electrode (or cathode) is a 2.5 cm x 2.5 cm and 0.5 cm thick (Goodfellow) surface graphite coupon, electrically connected to a titanium rod (Goodfellow). Before each experiment, the graphite coupon is washed for 1 h in a solution of 1 N HCl to dissolve any adsorbed species and then in a solution of 1 N NaOH to neutralize the acidity and finally rinsed with ultrapure sterile water before being left overnight in 1 L of sterile ultrapure water to flush out any soda residues included in the pores of the graphite. The titanium rod is cleaned with acetone and then autoclaved. In this device, the ratio of the active surface of the cathode to the volume of the reactor is therefore 10.4 m ² / m ³ ; where the active surface is defined as the geometric surface being exposed to the counter electrode.

The counter-electrode (or anode) is a platinum grid, previously cleaned and disinfected by flame. In this device, the ratio of the projected geometric surface of the anode to that of the cathode is about 1.5 to not limit the cathode phenomena.

The potentials are monitored and expressed with respect to an Ag / AgCl reference electrode (potential of 0.240 V / ESH, Radiometer analytical).

The bias potential of the working electrode is applied with a single-channel potentiostat (Ametek VersaSTAT3) and the current is recorded every 900 s. The chronoamperometry (CA) is periodically stopped to acquire cyclic voltammetry (CVs) between -1.2 and 1.0 V vs. Ag / AgCl at a scanning rate of between 1 and 10 mV.s ¹ . As a reminder, a cyclic voltammetry (or cyclic voltammetry) consists of carrying out a potential sweep at the working electrode and measuring the current flowing in the electrochemical system. This technique allows: (i) to check the coherence of the CA with the CV, (ii) to acquire kinetic information about the system and (iii) highlighting redox compounds in the suspension.

In a first step, the reactor is filled with a bacterial suspension of S. maltophilia in reaction medium (prepared according to 2.c), in which a continuous bubbling of 100% CO ₂ or a CO _{2 2} : CH gas mixture is carried out. ₄ (1: 1 or 1: 2 v / v) is produced at a flow rate of between 2 and 25 ml.min ¹ . Then a polarization ranging from -0.7 V to -1.0 V vs Ag / AgCl is applied. Polarization is started at the same time as chronoamperometry. In parallel, a control reactor without electrodes (and therefore without polarization) is implemented in the same manner as the electrochemical reactor. Samples are taken over time, under microbiological hood, to measure the optical density at 600 nm (DCLoo _nm ) and the pH of the liquid medium; the pH was found to be constant and equal to 6.4 ± 0.2 (corresponding to pKa of C0 _2, H ₂ 0 / HC0 ₃ ^_). The gas composition is analyzed at the reactor inlet and outlet by Gas Chromatography coupled to a katharometer (GC-TCD) to determine the experimental flow of reduced CO ₂ .

6. Physico-chemical analyzes

 • Treatment of liquid samples

The complete reaction medium (with bacteria) and the supernatant (without bacteria) are analyzed. The complete reaction medium is either directly analyzed or frozen at -20 ° C as soon as collected for further analysis. The supernatant is obtained by centrifugation of the complete medium just taken; the centrifugation is carried out for 10 min at 10,000 g and 10 ° C and the supernatant is collected at the end of the centrifugation for immediate analysis or freezing at -20 ° C. The pellet is stored at -20 ° C for later analysis.

• Nuclear Magnetic Resonance (NMR) analyzes:

For these analyzes, 500 pL sample liquid (reaction carried out with ¹³ C02, 2.d) are introduced into a NMR tube of 5 mm in diameter with 50 pL of D ₂ 0 and then analyzed by NMR Bruker Avance III - 500MHz - CryoProbe Helium for 4 hours. Different standards (labeled and unlabelled sodium formate, sodium acetate, methanol, ethanol, formaldehyde, acetaldehyde, sodium lactate, glycerol, isobutanol, sodium succinate, sodium fumarate, sodium pyruvate, sodium oxaloacetate) prepared in The reaction medium was previously analyzed by NMR to obtain the fingerprints of these molecules. On this apparatus, the detection threshold of molecules labeled with ¹³ C is 6 mg.L ¹ . However, NMR analyzes can only detect ¹³ C isotopes which have a relative abundance of 1.1% compared to their ¹² C isotope in nature [W. Mook and P. Grootes, International Journal of Mass Spectrometry and Physics, 1973, 12 (3): p. 273-298] Therefore, a concentration of unlabeled agent at ¹³ C of about 600 mg L ¹ is required to highlight these compounds.

• Analyzes carried out by Gas Chromatography coupled to Mass Spectrometry (GC-MSi:

 Equipment

Gas chromatography (Clarus 580, Perkin Elmer) is conducted with a capillary column (30m x 0.25mm ID, 8μm film thickness, Rt-Q-Bond Plot, Restek), coupled to mass spectrometry (Clarus SQ -8-MS, Perkin Elmer) equipped with a selective quadrupole mass detector on electronic impact (El) operated at 70 eV. A sample changer (Turbomatrix Headspace 16S, Perkin Elmer) is used for the injection of head spaces obtained after heating at 100 ° C. of the liquid samples. For analysis of the closed reactor head space composition maintained at 30 ° C, the samples are taken directly into the reactor with a gas-tight syringe and manually injected into the GC-MS. Helium is used as a carrier gas at a flow rate of 1.5 ml / min.

Determination of ¹² C-formate and ¹³ C-formate in liquid samples

An esterification method based on the protocol described by Wallage et al (2008) (Formic acid and methanol concentrations in death investigations, Journal of Analytical Toxicology, 32 (2008) 241-248) was used for quantification of the formate. The esterification reaction is carried out between the organic acid present in the liquid sample and the methanol added under acidic conditions. In a GC-MS analysis vial (22 mL), sample 600 pL, 100 pL of acetonitrile (157.2 mg.L ¹⁾ as an internal standard, 100 pL of concentrated sulfuric acid and 100 pL of methanol as a branching agent are introduced in this order. The assay vial is then placed in the autosampler and heated at 100 ° C for 15 minutes to complete the esterification reaction. In the case of the analysis of the complete medium (with bacteria), a spread of the mixture on agar made it possible to verify that these conditions allow the lysis of the bacteria because no re-growth has been demonstrated. The temperature of the injector is set at 200 ° C. and the split at 1: 16. In the oven, a temperature gradient is carried out at 40 to 150 ° C. at a rate of 10 ° C./min. The ions detected, in single-ion recording (SIR) mode, correspond to one of the mass / charge ratios m! z 60 for ¹² C methyl formate and 41 for acetonitrile. To determine whether the carbon is ¹² C or ¹³ C, [M + 1] has also been evaluated for the ^C-13 methyl formate (ni / z 61). The presence of marked species is considered only for levels significantly higher than 1%, which therefore exceed the percentage related to natural isotopic abundance. Calibration curves were previously established with formate and acetate standards: Na (HCOO) and Na (H ¹³ COO) obtained from Sigma-Aldrich.

Analysis of the gases in the headspace of the closed reactor

The headspace of the closed reactor is analyzed to determine the gaseous species, marked ¹³ C or not, present over time. For each assay 250 μl of headspace were taken from the reactor incubated at 30 ° C using a gas-tight syringe and 50 μl was injected into the GC-MS. The temperature of the injector is set at 200 ° C and the split at 1:16. In the oven, a temperature gradient is achieved from 100 to 150 ° C at a rate of 10 ° C / min. In SIR mode, the ions detected are m! z 28 (N ₂ ), m / z 44 (C0 ₂ ), m / z 45 ( ¹³ C0 ₂ ), m / z 16

• Determination of ammonium ions (NH ₄₎ by ion chromatography (IC)

The NH ₄ ⁺ ions are assayed in the supernatants. The analysis is carried out by injecting 25 μl of sample into a Dionex ICS-1000 chromatographic device (Thermo Scientifc) equipped with an IonPAc AS19 capillary column (0.4 × 250 mm). The elution program is as follows: 10 mM KOH (from 0 to 10 min), then 10-45 mM KOH (from 10 to 30 min).

• Analyzes performed by gas chromatography coupled to a catharo meter TCD) Material

Gas chromatography (Clarus 580, Perkin Elmer) is conducted with a PE-Q column (Perkin Elmer, 30m) in series with a PE-MOLESIEVE molecular sieve column (Perkin Elmer, 30 m) coupled to a thermal conductivity detector, again called katharometer (Perkin Elmer). A 10-way loop valve is used to charge and inject gaseous samples.

Analysis of the composition of gases, whether or not marked

For each analysis, a minimum of 20 mL of gas to be analyzed is sent into the loop valve and 20 μL of sample is injected into the columns. Helium is the carrier gas at a flow rate of 10 mL.min ¹ . In the oven, a temperature gradient is formed from 40 ° C to l20 ° C at a rate of l0 C.min ^{o _1.} A calibration is performed to determine the C0 ₂ . Determination of the% by mass of intracellular PHB

The method based on the digestion of PHB and the analysis of its monomer (3-hydroxybutyrate) by GC-MS (AJ Pieja et al., Applied and Environmental Microbiology, 201 1: AEM.00509-11; G. Braunegg et al., European Journal of Applied Microbiology and Biotechnology, 1978, 6 (1): pp. 29-37) was carried out, with the difference that the bacterial pellets were frozen instead of being freeze-dried before step of digestion of the PHB.

The bacterial pellets stored at -20 ° C. are thawed at room temperature and then digested for 3 hours at 100 ° C. in 2 ml of methanol containing 3% of concentrated sulfuric acid and 0.1% of benzoic acid supplemented with 2 ml of chloroform. Regrowth tests on LB medium showed that the bacteria contained in the pellets and treated by this acid digestion were totally deactivated. Once the sample is cooled, 1 mL of demineralized water is added to induce phase separation. The aqueous phase is then removed and 2 μl of organic phase are injected into the GC-MS. The analyzes are conducted with the same GC-MS equipment as that used for formate analyzes. A DB-1 (Agilent J & W) column is used for separation. The vector gas is helium at a flow rate of 32 mL.min ¹ . The oven temperature is programmed at 80 ° C for 1 min, up to 120 ° C at a rate of 10 ° C / min, then the temperature is increased to 270 ° C at a rate of 45 ° C / min. . The temperature of the injector is set at 200 ° C and split at 30 mL / min.

 The use of a PHB standard (Sigma Aldrich) made it possible to establish a calibration curve giving the ratio of the areas of the 3-hydroxybutyrate and benzoate peaks as a function of the mass of PHB introduced.

Ratio of peak areas (3-hydroxybutyrate / benzoate) = 0.1675 x mp _B (R ² = 0.9865)

To determine the mass% of intracellular PHB per unit mass of dry cells (w PHB / W dry cell) in the samples, the mass of PHB measured in the sample by the method described above is divided by the dry mass of bacteria in the presence that is known from DCLoo measurement.

B | Results

I. ¹³ CO ₂ reduction tests in a closed reactor

In order to evaluate the ability of the biocatalyst to reduce C0 ₂ , ¹³ C-labeled C0 ₂ is used. The purpose of labeling C0 ₂ is to highlight labeled products from ¹³ C0 ₂ and to distinguish them from those resulting from a simple cell release. To identify these products, the NMR technique is used to analyze the bacterial suspension and its supernatant while the GC-MS technique is used to characterize the reactor headspace. In addition, this marking makes it possible to observe the NMR and GC-MS fingerprints of all the products having one or more ¹³ C which may result from (i) the reduction of ¹³ C0 ₂ and / or (ii) of the fixation. ¹³ C0 ₂ by the cell. The optical density at 600 nm (DCboonm) is measured over time to follow the evolution of the bacterial mass concentration (in dry gels / L). Counts of viable cells are also made to access the bacterial cell concentration (in CFU / mL) and the pH is measured during the reaction. The ammonium ion (NH ₄ ⁺ ) content in the reaction medium is monitored to study the influence of the presence of ammonium on the performance of the C0 ₂ reduction bioprocess.

1. Monitoring assimilation metabolism by NMR and GC-MS

The bacterial suspension of Stenotrophomonas maltophilia is brought into contact with a ¹³ C0 ₂ gaseous atmosphere: atmospheric air (1: 1 v / v). Figure 1 shows the NMR spectra obtained on the complete bacterial suspension (a) initially (start of reaction), (b) after 8 days of reaction and (c) after 35 days of reaction.

From the beginning of the reaction, two peaks are visible: that of ¹³ C0 ₂ (chemical shift d = 124.6 ppm) and that of H ¹³ CO ₃ ^_ (chemical shift d = 160.2 ppm). After 8 days of reaction, a peak with a chemical shift corresponding to that of ¹³ C-formate (d = 171.0 ppm) appears.

To verify that this peak corresponds to the ¹³ C-formate (H ¹³ COO) marked a small quantity of standard ¹³ C-formate is added to the sample (at a final concentration of 10 mg / L). Peak amplitude increased when the standard was added, confirming that this peak is indeed attributable to ¹³ C-formate. No compound other than ¹³ C-formate is detected by NMR, suggesting that ¹³ C-formate results from direct reduction of ¹³ C0 _2. Indeed, ¹³ C-formate has a single carbon, at a lower oxidation state than that of ¹³ C0 ₂ . Whites made without bacteria under the same conditions (that is to say under a mixture of ¹³ C0 ₂ / atmospheric air) do not show any labeled compound, apart from ¹³ C0 ₂ and H ¹³ C0 ₃ , thus confirming the role of the bacteria on the reduction of ¹³ C0 ₂ to ¹³ C-formate.

The reduction kinetics of ¹³ C0 ₂ by the bacterium Stenotrophomonas maltophilia is reproduced and the liquid medium of the reactor is also analyzed at 35 days (Fig. Lc). The peaks corresponding to ¹³ C0 ₂ and H ¹³ C0 ₃ are still visible at 35 days, which reveals that there is no limitation in C0 ₂ . Only the peak corresponding to ¹³ C-formate (d = 171.0 ppm) is still highlighted. This means that the ¹³ C-formate product is not used by the bacteria to form new compounds and ¹³ C0 ₂ is probably final electron acceptor. These new Tests have confirmed that the presence of S. maltophilia bacteria makes it possible to catalyze the ¹³ C0 ₂ reduction reaction in ¹³ C-formate in the liquid phase.

In addition, the GC-MS analysis of the headspace of several closed reactors reveals the presence of labeled methane ( ¹³ CH ₄ ) in a significant amount after 34 days. Under the conditions tested, the content of this gas remains nevertheless less than 1% v / v. Such as ¹³ C- formate, ¹³ CH ₄ is a compound with a single carbon in a lower oxidation state than that of ¹³ C0 _2. This ¹³ CH ₄ is therefore derived from the reduction of ¹³ C0 ₂ , either directly ( ¹³ C0 ₂ ® ¹³ CH ₄ ) or indirectly ( ¹³ C0 ₂ ® ¹³ C-formiate ® ¹³ CH ₄ ). Methane is a particularly interesting product because it can be easily recovered by stripping and used as fuel.

In conclusion, the bacterium S. maltophilia is able to reduce C0 ₂ to formate and methane. In order to quantify the production of ¹³ C-formate in the liquid reaction medium, reduction of the tests of ¹³ C0 ₂ are reproduced in similar conditions, considered as reference conditions. The concentration of ¹³ C-formate is this time determined by GC-MS; in addition, the optical density at 600 nm (DCLoo) and the pH are also monitored.

2. ¹³ CO ₂ reduction tests in a closed reactor under reference conditions

 at. Definition of reference conditions

The reference conditions are defined as follows: initial bacterial concentration of 3.1 ± 0.2 g of dry cell / L, initial gaseous atmosphere composed of a gas mixture ¹³ C0 ₂ : atmospheric air (3: 7 v / v ), an aqueous reaction medium composed of 20 mM phosphates and 5 mM MgCh. Three independent kinetics are conducted in parallel under these reference conditions. b. Follow-up of biomass concentration and pH

The mass concentration of bacteria, denoted [X], is monitored by measuring DCLoo. The correlation giving the DCLoo as a function of the mass concentration of bacteria is given in Section A1]. For all three kinetics monitored, a decrease in the biomass concentration is observed (data not shown). The most likely reason to explain this phenomenon is cell lysis. Indeed, a lack of natural nutrients and an atmosphere rich in C0 ₂ exposes bacterial cells to stressful conditions, which are likely to lead to death and therefore cell lysis of part of the bacterial population. Throughout the kinetics, however, no test shows a total lysis. Indeed, for all experiments, cell mass concentration, initially of 3.1 ± 0.2 g of sèche.L cell ^1, drop over the first 10 days and then stabilize at an average value of 0.5 ± 0.1 g of dry eye / L. This means that only part of the biocatalyst suspension lyses on the first 10 days of kinetics and the mass concentration remains stable.

In addition, enumeration of viable cells in the bacterial suspension after 35 days confirms on the one hand that cell lysis occurs and on the other hand that 15 ± 4 x 10 ⁴ CFU / ml are still viable; the initial concentration being 60 x 10 ⁷ CFU / mL. The concentration of viable cells has therefore decreased significantly but still remains high (of the order of 10 ⁵ CFU / ml). It is hypothesized that this cell lysis may be a way for the bacteria to adapt to its new conditions by releasing organic compounds into the medium by lysing cells (JM Navarro Llorens et al., FEMS Microbiology reviews, 2010, 34). (4): 476-495).

During these kinetics, the pH is also monitored (data not shown). Whatever the experiment, the evolution of the pH is very similar. In fact, the initial pH is fixed at 7.0 ± 0.1 and for all the tests, the pH decreases to an average value of 6.4 ± 0.2 over the first 5 days and then remains constant at this value which corresponds to at the pKa of the acid-base couple C0 ₂ , H ₂ 0 / HC0 ₃ ^_ induced by the dissolution of C0 ₂ in the reaction medium. The pH of the medium is therefore buffered throughout the kinetics. c. Monitoring the production of ¹³ C-formate in the liquid

The production of ¹³ C-formate resulting from the reduction of ¹³ C0 ₂ is quantified by GC-MS and its evolution over the three different kinetics is shown in Figure 2.

Whatever the kinetics, ¹³ C-formate is produced in significant amount. In addition, GC-MS analysis revealed that ¹³ C-formate is found in complete suspensions (i.e. in the presence of bacteria) but not in the respective supernatants. This shows that the production of this ¹³ C-formate is carried out by the honest bacteria remaining in the reaction medium and not by free enzymes derived from cell lysis. The ¹³ C- formate appear after 8 to 10 days, which period may correspond to a latent phase during which the bacterium adapts to its new substrate ¹³ C0 _2. The fact that a high concentration of bacteria (close to 10 ⁵ CFU / mL) is still viable at the end of the kinetics confirms the presence of bacteria maintaining metabolic activity.

¹³ C0 ₂ thus enters the cells, either as C0 ₂ or ¹³ in the form H ₃ ^{_ 13} C0; in fact, the pH stabilizes around the pKa of the pair C0 ₂ , H ₂ 0 / HC0 ₃ ^_ and the two forms therefore coexist in close proportions. Then the cells that survived the reaction conditions and lysis thus use one of their intracellular enzyme system to catalyze the reduction reaction of this C0 ₂ ^{13 13} C-formate. Two families of enzymes, potentially present inside the bacterium S. maltophilia, are able to catalyze this reaction of reduction of C0 ₂ in formate: the Formiate DeHydrogénases (FDHs) and the nitrogenases. At present, these enzymes are first isolated from their native microorganism and then purified before being used (LB Maia et al., Inorganica Chimica Acta, 2017, 455: 350-363, Khadka et al. 2016, Inorg Chem., 55, 8321-8330).

It should be noted that the maximum formate concentrations produced (10 - 20 mg.L ¹ , Figure 2) are lower than the theoretical maximum concentration of ¹³ C-formate that could have been reached if all the ¹³ CO ₂ present in the reactor had been reduced (ie 4.6 gL ¹ ). This calculation is based on (i) the number of moles of ¹³ C0 ₂ introduced into the headspace (obtained by the ideal gas law) and (ii) on a 1: 1 stoichiometry with respect to ¹³ CO ₂ and formate (that is to say, assuming that one mole of ¹³ C0 ₂ provides one mole of C ^13-formate). This suggests that there is no macroscopically limitation the transfer of ¹³ C0 ₂ in the liquid phase containing the bacteria. However, a microscopic limitation of the transport of ¹³ CO ₂ from the liquid core to the biocatalyst is not excluded. Indeed, the presence of free DNA from the cell lysis observed could create significant resistance to mass transfer in the cells. A study has indeed already reported this type of phenomenon for the E. coli bacterium (Castan et al., John Wiler & Sons, Inc. Biotechnol Bioeng 77 (2002) 324-328), which has similarities with the S. maltophilia bacterium, because it is also an aerobic bacterium of Gram-negative bacillus type. In the Castan et al. Study, the presence of free DNA would limit the transfer of dissolved (0 ₂ ) oxygen to the cells. In this case, such a limitation could also happen to the ¹³ C0 ₂ dissolved. A solution to increase the production of formate would be to bring the C0 ₂ to the heart of the solution by a gas distributor. Moreover, the bacterial concentration and the specific C0 ₂ -reductive activity of the cells may also be limiting elements to the reduction reaction. It is therefore possible to play on these parameters to optimize the flow of C0 ₂ reduced. d. Pformiate production at ¹³ C-formate and Fco2 volume flow, reduced by ¹³ C0 ₂

Whatever the kinetics considered (FIG 2), the production of ¹³ C-formate Pformate, in mg ¹³ C-formate (g dry cell) ^{1 1} or mhioI ¹³ C-formate (g dry cell) ^{1 1} , is defined as:

P formate AC / Dί / Cbacterium (Equation 4) with Dΐ: production period (day), AC: difference in ¹³ C-formate concentration over the production period (mg.L ¹ or pmol.L ¹ ),

Cbacteria: initially introduced bacterial concentration (Cbacterium = 3.1 ± 0.2 g dry cell.L ¹ ) to take into account the bacterial supply which it was necessary to implement. Similarly, if one considers that one mole of C ^13-formate produced corresponds to a mole of ¹³ C0 ₂ reduced, then the flow FCO2 ¹³ C0 ₂ reduced mhioΐ ¹³ C02. (G dry cell) ¹ j ¹ will be equal to the production in Pfomüate formate expressed in mhioΐ ¹³ C-formate. (g dry cell) ¹ j ^1. The volume flow Fco ₂ , expressed as ¹³ CO ₂ reduced expressed in mL ¹³ CO 2 (g dry cell) ^{1 1} will be calculated according to the following equation: Fco ₂ , voi = FCO ₂ . v _m (Equation 5) where V _m is the molar volume at 30 ° C of ¹³ C0 ₂ assumed perfect gas (25.19 10 ³ mL.pmol ¹ )

Table 1 details the AC and At values for each of the three kinetics presented above, as well as the ¹³ C-formate (Pfomüate) and the ¹³ C0 ₂ reduced fluxes (F ₂ , vol). Table 1. ¹³ C-formate (Pfomüate) production and ¹³ C0 ₂ reduced flux (Fco ₂ , voi) obtained in a closed reactor

e. ¹³ C-formate production from ¹³ C0 ₂ : reproducibility study

In order to study the reproducibility of this bioprocess, six other independent ¹³ CO ₂ reduction tests were conducted under reference conditions, with bacterial suspensions prepared from different cultures of S. maltophilia. For each of these tests, a significant production of ¹³ C-formate was demonstrated. Applying the calculations in the previous section (Eqs 4 and 5) to the results obtained during these kinetics (6 in total), an average Pfomüate production of 9.1 mhioΐ ¹³ C-formate (g dry cell) results. ^{1 1} (0.4 mg) formate (dry cell) ¹ . ¹ ), with a minimum and maximum respectively of 2.3 and 30 mhioΐ ¹³ C-formate (g dry cell) ¹ .j ¹ (ie 0.1 and 1.4 mg ¹³ C-formate. ) ¹ ./ ¹ ).

In the same way, an average volume flow Fco2, voi ¹³ CO ₂ reduced by 0.24 mL ¹³ CO 2. (g dry cell) ¹ .j ¹ was obtained, with a minimum and maximum respectively of 0.06 and 0.76 mL ¹³ CO 2 (dry corn) ¹ .day ¹ ; these flow differences are certainly related to the differences in physiological state of the bacteria used for the ¹³ C0 ₂ reduction tests. f. Production of ¹³ CH ₄ from ¹³ C0 ₂ : reproducibility study

The production of ¹³ CH ₄ from ¹³ CO ₂ was evaluated at 34 days in 4 independent ¹³ CO ₂ reduction tests conducted under reference conditions, with bacterial suspensions prepared from different cultures of S. maltophilia. . In all cases, a significant production of ¹³ CH ₄ was obtained, with ¹³ CH ₄ contents in the reactor head space varying between 1 and 3% v / v.

3. Effect of various parameters on the reduction in performance of ¹³ CQ ₂

Various parameters that can potentially influence the performance of the CO ₂ reduction reaction are studied. For this, tests are conducted under modified reaction conditions, compared to the reference conditions. A test under reference conditions is always performed simultaneously with the study of the influence of a parameter to determine the impact of the modified parameter. at. Effect of biomass concentration The initial mass concentration of the bacterial suspension used for the ¹³ CO ₂ reduction tests is set at 3.1 ± 0.2 g of dry moisture / L. Indeed, in the culture conditions implemented, this concentration is currently the highest possible to launch different closed reactors and different series of closed reactors. Eventually, the volumes and concentrations of bacteria produced can be increased. A lower concentration (1.6 ± 0.1 g dry cell / L) is tested to verify that the amount of ¹³ C-formate produced is correlated to the amount of cells; the other reaction parameters remaining unchanged. The kinetics is followed on 25 days.

Table 2 summarizes the results obtained. Table 2. Concentrations in initial biomass [Xo] and final biomass [X _f ] (in g of dry cell / L), loss of biomass D [C] at the end of reaction (in%) and final concentrations accumulated in ¹³ C formate resulting from the reduction of ¹³ CO ₂ (in mmol / L and mg / L).

Table 2 shows that the ¹³ C-formate concentration is nearly 20 times higher when the initial concentration of bacteria is doubled. Contrary to what could be expected, the final accumulated ¹³ C-formate concentration is therefore not correlated with the amounts of cells initially introduced. On the other hand, it is interesting to note that by doubling the initial amount of bacteria, the amount of final lysed cells is also doubled (Table 2, difference between [Xo] and [X _f ]). It is therefore possible that the compounds released into the reaction medium by cell lysis can be used by the surviving cells to produce a larger amount of reducing intracellular enzyme C0 _2, which would lead to an increase in production ¹³ C-formate.

In conclusion, the availability of intracellular compounds in the reaction medium appears to be a limiting factor for C0 ₂ reduction performance when the initial bacterial concentration is too low (1.6 g dry weight / L). b. Influence of the nature of the gas mixture

Tests with different gas mixtures are carried out to study the influence of the presence of oxygen (0 ₂ ) and nitrogen (N ₂ ) on the CO ₂ reduction reaction. The ¹³ C-formate concentration is monitored.

Different initial gaseous mixtures containing ¹³ C0 ₂ were tested: (i) the reference mixture of ¹³ C0 _2: atmospheric air (3: 7 v / v), (ii) a mixture of ¹³ C0 _2: N ₂ (3 : 7 v / v) to study the influence of oxygen and (iii) of ¹³ C0 ₂ pure to investigate the effect of nitrogen. These tests were all duplicated twice. The results show that the absence of oxygen and the presence of nitrogen accelerates the appearance of ¹³ C-formate in the medium. Indeed, the production of ¹³ C-formate occurs on average 14 ± 2 days earlier under a mixture of ¹³ C0 ₂ : N ₂ (3: 7 v / v), compared to the ¹³ C0 ₂ mixture: atmospheric air (3: 7 v / v) and ¹³ C0 ₂ pure. On the other hand, whatever the gaseous sky implemented, the maximum ¹³ C-formate concentration accumulated in the medium over a reaction time of 31 days remains of the same order of magnitude (around 0.1 mmol / L). Oxygen is reported to be a potential inhibitor of the activities of nitrogenase and HDD because it can serve as an electron acceptor and lead to the formation of reactive oxygen species (J. Gallon, Trends in Biochemical Science, 1981, 6: pp. 19-23) but it may also affect the synthesis of nitrogenase (Q. Liu et al., Nature Communications, 2015, 6: 5933). The drop in performance observed in the presence of 0 ₂ is not surprising.

Finally, use a mixture ¹³ C0 _2: N ₂ supports the reduction reaction kinetics of ¹³ C0 _2. c. Influence of the composition of the reaction medium

New reduction tests of ¹³ C0 ₂ are made with a reaction medium enriched with ammonium ions (ABSA). MREA contains 2 mmol / L of KCl, 28 mmol / L of NH ₄ Cl, 30 mmol / L of NaHCCL and 5 mmol / L of NaH ₂ PO ₄ . The concentrations of ¹³ C-formate and ammonium ions (NH ₄ ⁺ ) are monitored throughout the reaction (Figure 3).

When ammonium ions are initially present in large quantities, nearly half of the initial amount is consumed over the first 8 days, then the concentration seems to stabilize (Fig. 3. A). Under reference conditions, ammonium ions appear in suspension (Fig. 3. A). This appearance in ammonium ions can be related to two productions: the reduction of N ₂ by the nitrogenase and the deamination of the released proteins in the medium after cell lysis. In addition, these NH ₄ ⁺ ions are not oxidized to nitrates or nitrites, which implies that these NH ₄ ⁺ ions can serve as a source of assimilable nitrogen for the synthesis of enzymes by the surviving bacteria.

Regarding the production of ¹³ C-formate, it is nearly 2.5 times higher at 24 days (ie a concentration of 1.45 mmol / F, that is to say about 65 mg / F) and occurs about 7 days earlier (Fig. 3.B) when the reaction medium is enriched in ammonium. This could be explained in particular by the fact that the excess presence of a source of nitrogen assimilated by the bacterium (that is to say the NH ₄ ⁺ ions) at the beginning of the reaction could have helped the bacterium to over-express its enzymatic system reducing C0 ₂ .

It is therefore also possible to play upstream on the culture conditions of the bacterium S. maltophilia and / or on genetic modifications of the bacterium to over-express the amount of its enzymatic system C0 ₂ -reducing and benefit from an activity specific intrinsic reduction of C0 ₂ more important. d. Study of the reversibility of the CO ₂ reduction reaction

This study showed that S. maltophilia bacteria is able to reduce CO2 to formate. So far, the maximum concentration accumulated in formate is about 1.5 mmol / L (ie 67.5 mg / L). In order to investigate the reversibility of the reaction of this catalyst system, that is to say its ability to oxidize the CO2 formate, a concentration of ¹³ C-sodium formate 1.5 mmol / L was introduced into a suspension of S. maltophilia prepared as for a CO2 reduction test (section A] 3). This suspension is then introduced into an NMR tube and analyzed by NMR (section A] 6). If the reduction equilibrium of ¹³ CC> 2 is reversible, then the oxidation of ¹³ C-formate (H ¹³ COO ^~ ) to ¹³ CC> 2 should be able to occur because ¹³ CC> 2 is initially a minority:

H ¹³ COO ^~ <® ¹³ CC> 2 (Equation 6)

The NMR spectrum obtained after 4 hours is shown in FIG.

Only two peaks are visible, corresponding respectively to ¹³ C-formate ions (d = 171.0 ppm) and ¹³ C-hydrogencarbonate (d = 160.2 ppm). This result confirms that when the ¹³ C- formate is predominantly in front of the ¹³ CC> 2, then the ¹³ C-formate can be oxidized to ¹³ CC> 2, which is solubilized in the hydrogen form (H ₃ C0 ^{~ 13).} Indeed, in the case of this test, the pH of the reaction medium (7.0 ± 0.1) is higher than the pKa of the pair CCL ^ O / HCCL (6.4) because the medium is not buffered by the presence of ¹³ CC> 2. As expected, the reaction catalyzed by S. maltophilia bacteria thus seems to be regulated by the most abundant substrate concentration (CO2 in the direction of reduction and formate in the opposite direction). Moreover, it is interesting to note that the presence of these two NMR peaks only, and thus the absence of other peaks, confirms that the formate produced under these reaction conditions is not used by the bacterium to form the biomass or enzymes (section B] 1.1). Eliminating the formate continuously is therefore an option to promote the CO2 reduction reaction. e. Influence of adding PHB to the reaction medium

The enzymatic reduction of C0 ₂ requires a source of protons and electrons. Since ammonium ions (NH ₄ ⁺ ) are not oxidized to nitrate or nitrite ions, it is therefore unlikely that they serve as electron and proton donors for the CO ₂ reduction reaction. The source of protons and electrons is most certainly accumulated in the cell during the culture phase of the bacterium as an energy reserve. At present, two types of molecules have been identified as being able to play this role: Poly-3-HydroxyButyrate (PHB) which is a natural polymer and lipids. As regards PHB, bacterial cells are indeed capable of depolymerizing PHB in its monomer (3-hydroxybutyrate) and oxidizing this monomer to acetoacetate so as to recover electrons and protons but also a source of carbon ( S. Obruca et al., PLoS One, 2016, 11 (6): e0157778).

To verify the role of the PHB, it is added initially to the bacterial suspension and the CO ₂ reduction reaction is then carried out under the reference conditions (described in section B] 1.2). However, the PHB is a high molecular weight polymer and insoluble in water, which makes it impossible for its transport and assimilation in the cell in its polymerized form. However, the bacterium S. maltophilia is capable of excreting depolymerases that can depolymerize the PHB in its monomer 3-HydroxyButyrate assimilated by the bacterium (S. Wani et al., 3 Biotech, 2016, 6 (2): p. B. Tiwari et al., Bioresource Technology, 2016, 216: 1102-1105).

To determine the amount of PHB to be added, the intracellular PHB contents of the bacterial suspensions used for the tests are first measured by the method detailed in section A] 6. An average mass content of intracellular PHB of 1.0 ± 0, 1% w / w is obtained, corresponding to a PHB concentration of 30 ± 2 mg / L, which is quite common for S. maltophilia species (B. Iqbal et al., Annual Research & Review in Biology, 2016, 9 (5): p.1). A concentration 10 times higher (300 mg / L, ie 0.3 g / L) is then used for the tests to avoid a limitation in PHB.

Two independent series of closed reactors are made from two different bacterial cultures. For each test, kinetics are conducted in parallel under reference conditions (i.e. without addition of PHB). In all the tests, the gaseous atmosphere of the reactors is thus initially composed of a ¹³ C0 ₂ mixture, atmospheric air (3: 7 v / v). For the first series, the follow-up is carried out over 20 days and up to 40 days for the second series. The ¹³ C-formate concentrations as well as the intracellular PHB concentrations are measured during kinetics.

Table 3 presents the results obtained initially then at 20, 30 and 35 days. It is important to note that PHB measurements for kinetics in which PHB is added are not indicated because they are not reproducible, probably due to a lack of homogeneity in the samples taken from the reactors (indeed, PHB is not soluble in water).

Table 3. Concentrations of ¹³ C-formate ([ ¹³ C-formate] qmo 1 / L) and intracellular PHB ([PHBi _ntra ] in mg / L). NM means No Measurable and (-) means Not Measured.

With respect to ¹³ C-formate production, production is much greater when PHB is initially added to the reaction medium. Compared with reference conditions (without addition of PHB), the ¹³ C-formate concentration produced in the presence of PHB is in fact increased by a factor of between 3 and 95 (Table 3, series 1 and 2 respectively). This result shows that the addition of PHB makes it possible to improve the performance of the CO ₂ reduction reaction and suggests that it could be an electron and proton donor for this reaction.

A limitation in PHB under reference conditions could explain why the production of ¹³ C-formate is improved when the medium is enriched in PHB. Indeed, under reference conditions (without addition of PHB), the intracellular PHB concentration decreases from 30 ± 2 to 13 ± 1 mg / L regardless of the kinetics and then remains constant (Table 3, series 1 and 2). For both series, it therefore appears that nearly 20 mg / L of intracellular PHB is consumed by the cells during the reaction. In addition, the limitation in PHB is probably not the only parameter to impact the ¹³ C-formate production. Indeed, under the reference conditions and for the same quantity of PHB consumed after 20 days of reaction, the ¹³ C-formate concentration obtained at 20 days is nearly twice as large for series 1 as for series 2 ( Table 3). This can be explained by the fact that these series were made from two different cultures and that the physiological state of the bacteria was not the same. However, the physiological state may play a role in the intrinsic C0 ₂ -reductive activity of the bacterium but also in its ability to adapt to its new C0 ₂ substrate. In conclusion, the amount of intracellular PHB is a limiting factor for the CO ₂ reduction reaction and for the production of ¹³ C-formate. The PHB is therefore probably a source of electrons and protons for the C0 ₂ reduction reaction.

4. Study of synergistic effects with other bacteria

The bacterium S. maltophilia has been consortiumed with bacteria likely to use formate to study their possible synergies.

Initially, a consortium formed by the bacterium S. maltophilia and the methanotrophic bacterium M. trichosporium OB3b, obtained after culture on methane (according to the protocol described in section A] 1), was implemented under the conditions of reaction. reduction of C0 ₂ detailed in section a] 4. NMR analysis of the reaction medium was conducted initially and after 8 days. The NMR spectra obtained are presented in FIG.

From the beginning (t = 0 days) and throughout the reaction, the ¹³ C0 ₂ resonance peaks (d = 124.6 ppm) and its ionic form H ¹³ C0 ₃ (d = 160.2 ppm) are visible days ( Ligure 5. a). After 8 days of kinetics, a new peak appears (d = 171.0 ppm, Ligure 5. b) corresponding to the standard footprint realized with labeled formate at ¹³ C. This demonstrates the production of ¹³ C-formate under these conditions , at a concentration greater than or equal to the detection limit of formate labeled by NMR analysis, ie 6 mg.L ¹ . The supernatant of the NMR analysis revealed no ¹³ C-labeled formate, which means that the labeled product is intracellular formate, as observed in the kinetics with only S. maltophilia.

No other labeled product has been demonstrated in the NMR spectrum (Ligure 5.b). Trials carried out over 8 days with the methanotrophic bacterium M. trichosporium OB3b alone, whose purity was verified beforehand, showed that this bacterium does not catalyze the reduction of C0 ₂ . Moreover, this bacterium is capable of forming spores that limit the interactions of the bacterium with its environment, which may explain that no synergy between S. maltophilia and M. trichosporium OB3b has been observed. .

This result therefore suggests that the presence of this methanotrophic bacterium in the reaction medium does not interfere with the reduction of CO ₂ and the production of formate by S. maltophilia. For some applications, co-cultivating the S. maltophilia biocatalyst with the M. trichosporium OB3b methanotrophic bacterium can therefore be envisaged to guarantee an optimal carbon balance because methane, of renewable origin, would be the only source of carbon required for cultivation. In a second step, a consortium formed by the bacterium S. maltophilia and the gram-positive microbacterium oxydans bacterium, obtained after culture on LB medium (according to a protocol similar to that described in section A] 1), was implemented in the CO ₂ reduction reaction conditions detailed in section A] 4. An NMR analysis of the reaction medium was carried out at 5, 8 and 12 days (FIG. 6).

Like the initial spectrum obtained with the S. maltophilia / M. trichosporium OB3b consortium, only the peaks of the substrates ( ¹³ C0 ₂ and H ¹³ C0 ₃ ^_ ) are visible at t = 0 days.

After 5 days of reaction, a peak of formate appears (d = 170.9 ppm, Figure 6. a), corresponding to a concentration greater than or equal to 6 mg.L ¹ . Compared with the kinetics performed in pure culture with S. maltophilia, the lag time appears reduced by a minimum of two days (Figures 1 and 6). In addition, the NMR analysis of the supernatant has not revealed the presence of labeled products, this shows that the labeled formate is also produced intra cellulitively.

After 8 days of reaction, the amplitude of the peak of formate increases significantly (by a factor of about 10), which means that the concentration of formate produced is greater than at 5 days (FIGS. 6a and 6). b). In addition, new ¹³ C-labeled intracellular peaks appear, including a peak (d = 182.2 ppm) that may correspond to the footprint of one of the three carbons contained in the lactate, as well as a peak forest in the displacement zone. between 15 and 65 ppm (Figure 6.b).

After 12 days of kinetics, the amplitude of the NMR peak relative to the formate decreases significantly (by a factor of about 10) compared to the peak obtained at 8 days (Figure 6.b and 6.c). The compound that appears to be lactate is still present, in proportions comparable to those obtained at 8 days. On the other hand, the forest of peaks visible in the zone of chemical shifts between 15 - 65 ppm is more intense (Figures 6.b and 6.c) and new compounds labeled with ¹³ C also appear in the zone of chemical shifts between 125-140 ppm (Figure 6.c). This phenomenon is characteristic of a metabolic activity of the cells. As the three carbons lactate are not marked, it is important to emphasize that this marked lactate therefore can not come from a direct reduction of ¹³ C0 _2. This production of labeled lactate can nevertheless result from a binding of ¹³ C0 ₂ and / or ¹³ C-formate on organic molecules already present initially in the cells, as well as the labeled compounds relating to the new peaks (d = 125-140). ppm). A possible explanation for these observations is the establishment of a synergy between the two bacteria present. Indeed, S. maltophilia can reduce the CO ₂ to formate, then excrete the formate produced outside its cell so that M. oxydans can use it and form new compounds necessary for its subsistence such as lactate or other products whose footprints are included in peak forests. Nevertheless, the NMR analysis of the supernatant only showed the signals of the substrates ¹³ C0 ₂ and H ¹³ C0 ₃ but not ¹³ C-formate. A probable hypothesis is that the M. oxydans bacterium captures and introduces the formate produced in its cells as soon as the formate has been excreted by the bacterium S. maltophilia, which would explain the significant decrease in the ¹³ C-formate peak observed at 12 days. (Fig. 6.c). The presence of M. oxydans bacteria would therefore boost the reduction of CO2 by S. maltophilia and thus the production of ¹³ C-formate; This may also explain the decrease in the lag phase for ¹³ C-formate production. Such synergy between S. maltophilia and M. oxydans is quite realistic under these reaction conditions (i.e., under pure CO2) because M. oxydans is facultatively aerobic (Schumann et al., International Journal of Systematic and Evolutionary Microbiology, 1999, 49 (1): 175-177); which is not the case of M. trichosporium OB3b which is aerobic strict (N. Dorina et al., Applied biochemistry and microbiology, 2008, 44 (2): pp. 182-185).

II. CO2 reduction test in semi-closed reactor, assisted by electrolysis 1. Electrochemical device and its principle

Poly-3-Hydroxybutyrate (PHB) polymer is depolymerized to 3-hydroxybutyrate, which is an identified electron and proton donor for the CO2 reduction reaction (Section B] 1.3 e]). Nevertheless, the intracellular stock of PHB is finite. The idea is therefore to provide without limitation and continuously the electrons and protons required for the reaction with an in situ electrolyte device and inexpensive, so that the performance of the CO2 reduction reaction are improved and persist over time. This electrochemical device is therefore intended to replace intracellular donors of protons and electrons. A semi-closed reactor assisted by electrolysis (or bio-electrolyser) was thus implemented to intensify the reaction of reduction of CO2 catalyzed by the bacterium S. maltophilia.

In this 3-electrode device (FIG. 7 and section A) 5), the role of the polarized cathode is to supply the electrons required for the CO2 reduction reaction in a compound whose average carbon oxidation state is lower than that of the compound. in the CO2 molecule (this compound is noted as CO2, reduced) whereas the oxidation of the water at the anode makes it possible on the one hand to provide the protons and on the other hand to close the electrical circuit; the reference electrode allows for its part to maintain a bias voltage (V _po iarisation) constant at the cathode. Conventionally, an Ag / AgCl reference electrode is used. A necessary condition for this device to work is that the bacterium S. maltophilia is electro-active in reduction, that is to say able to exchange electrons with the cathode. So far, however, the bacterium S. maltophilia is known only to exchange electrons in oxidation, that is to say with an anode (K. Venkidusamy and M. Megharaj, Frontiers in microbiology, 2016, 7 : pp. 509-518, H. Xue et al., Sensores and Actuators B: Chemical, 2017, 243: pp. 303-310, J. Shen et al., Bioelectrochemistry, 2017, 114: pp. 1-7). .

In a first step, the bias voltage is chosen on a thermodynamic basis. The oxy-reduction potential of the Acetoacetate / 3-hydroxybutyrate pair being -0.53 V vs. Ag / AgCl at pH 6.4 (PA Loach, Handbook of Biochemistry and Molecular Biology, 1976, 1: p l22), it is therefore necessary initially to impose a greater potential (by absolute value) than this potential (as for example -0.7 V vs Ag / AgCl) so that the bacterium has an energetic interest to use the electrons of the rather than those from oxidation of 3-hydroxybutyrate. Nevertheless, if the stock of intracellular PHB, a source of energy and 3-hydroxybutyrate, is depleted during the reaction by the general metabolism of the bacterium, this constraint on the potential will be lifted and a lower potential (by absolute value ) that -0.53 V vs Ag / AgCl at pH 6.4 can be implemented. Moreover, the retained potential is preferably chosen lower than those of the CO2 / CO2 reducing-oxidant couples, reduced involving CO2 as an oxidant so as to promote the reduction of CO2 at the cathode; the choice of polarization potential also allows to promote the production of a CO2 reduction product, reduced compared to other possible CO2 reduction products (among methane and formate for example). In general, a potential greater than or equal to -0.8 V vs Ag / AgCl (for example -0.7 V vs Ag / AgCl) will allow (i) to limit the reduction of the proton in hydrogen (FL) and (ii) ) to avoid the reduction of phosphates and magnesium ions present in the reaction medium. The minimum potentials below which the proton and CO2 reductions occur can be determined precisely by cyclic voltammetry, under the reaction conditions, to take into account both thermodynamic and kinetic (surge) phenomena; overvoltages depending in particular on the material chosen for the cathode.

According to the literature, the nature of the material can also have a significant impact on the establishment of electronic transfer between the bacterium and the working electrode (M. Rosenbaum et al., Bioresource Technology, 2010, 102 (1): 324-333). The chemical composition and surface structure can induce different responses Electrochemical. Of the many electrode materials available, graphite is porous and has surface roughness. Graphite is therefore likely to promote contact between the bacterium and the electrode, which could promote electronic conduction (Rosenbaum et al., Bioresource Technology, 2010, 102 (1): 324-333). Graphite is therefore chosen as the cathode material for the tests presented. In addition, a platinum anode is retained for these tests in order to reduce the overvoltages at the anode and not to limit the cathodic phenomena. Other anode materials may also be contemplated, such as steel. An anode surface larger than the surface of the cathode is also preferred so as not to limit the reduction reactions at the cathode.

Moreover, a continuous bubbling of pure C0 ₂ is also implemented to saturate the suspension in CO2 and deoxygenate the solution so as to (1) avoid competition with the aerobic metabolic pathway (where the 0 _{2 would} become an electron acceptor at instead of CO ₂ ) and (2) preserve the catalytic activity of reduction of the CO ₂ -reducing enzyme (indeed, oxygen seems to have an inhibitory effect on this enzyme, section B] 1.3 (b). In order to guarantee optimal gas-liquid mass transfer, C0 ₂ is supplied by a porous gas distributor and a volume of liquid (allowing a high WM (Volume of gas per volume of liquid per minute) is set. A high cathode surface / liquid volume ratio is also desirable for concentrating the product formed by CO ₂ reduction at the cathode, a cathode surface / liquid volume ratio of the order of 10 m ² / m ³ is desirable.

2. A first example of a bio-electrolyser powered by pure CO2

Figure 8 shows the evolution of the current measured over time for the first biocerolyser implemented with the bacterium S. maltophilia for the reduction of C0 ₂ . This current is expressed in current density (i.e., in A / m ² ); the current density corresponds to the measured current (in A) relative to the projected cathode area (in m ² ). The test is conducted with a bias voltage of -0.7 V vs Ag / AgCl.

The initial zero current (0.0 ± 0.01 A / m ² ) corresponds to the baseline (Fig 8.a), which is confirmed by the cyclic voltammetry performed at the beginning of kinetics (data not shown). This shows that initially, no element present in the bacterial suspension oxidizes or is reduced.

Then, a small reduction current is observed (up to -0.2 A / m ² at 1.2 days). However, this current returns to the baseline before a significant reduction current appears from the second day (Fig. 8. a); the profile of this current follows a similar pattern to that of microbial growth kinetics (Fig. 8. a). There is no growth microbial visible in the suspension because the optical density (OD ₆ oo) of the bacterial suspension measured initially is 5.6 while it is 2.2 to 5.7 days (136 hours). Nevertheless, this does not exclude the growth of a biofilm on the surface of the cathode; the S. maltophilia bacterium is indeed capable of rapidly forming bio-films on different types of surface (O. Madhi et al., Current Protocols in Microbiology, 2014, 32: Unit-6F.l., 1-6F. ).

In order to experimentally verify the dependence of this reduction stream with CO ₂ , bubbling with an inert noble gas (argon) was carried out with the aim of removing all or part of the dissolved CO ₂ . A first bubbling of argon at a flow rate of 10 mL.min ¹ is carried out for 3 hours and has the effect of reducing the reduction current almost instantaneously by a factor of approximately 2, from -1.4 to -0.8 A / m ² (Fig 8.a and 8.b). Then, any gas flow is cut off and the reduction current remains almost constant, around -0.8 A / m ² (Fig 8.c). The constancy of the reduction current can be explained by the presence of residual CO2 in the bacterial suspension; indeed, CO2 is very soluble in water and the imposed argon flow rate was probably too low to drive out all the dissolved CO2.

When a low CO2 flow rate is reapplied (5 mL.min ¹ ), then the reduction current increases again until it regains its initial stable value, ie -1.4 A / m ² . 8. a and 8.d). Then, the application of a high flow of argon (100 mL.min ¹ ) during 3 ha results in eliminating any reduction current since the current rises to 0.0 A / m ² (Fig 8.e ). These conditions are therefore sufficient to expel all the dissolved CO2 in the suspension and confirm that neither the proton reduction nor that of the phosphates occurs at this polarization potential (-0.7 V vs Ag / AgCl). Finally, the current re-increases when a CO2 flow is performed again (at 25 mL.min ¹ ) and reaches a significant value of -0.3 A / m ² (Fig. 8.f). All of these elements show that the reduction current is well linked to the availability of CO2 in the reaction medium and that the CO2 is the element which is reduced to the cathode.

The reaction that occurs at the cathode is therefore:

CO2 + H ⁺ + 2e- ® HCOO (Equation 7)

This reduction reaction results from the enzymatic activity of the cell which no longer requires electron donors and intracellular protons.

Cyclic voltammetry (CVs) are also performed during the experiment. In particular, a CV realized at 5.7 days confirms the adequacy of the measured current with the imposed polarization potential (data not shown). This voltammetry also shows a diffusion plateau for the reduction current ranging from -0.2 Y to -0.8 V vs Ag / AgCl, which suggests that a lower polarization voltage per absolute value (for example -0.2 V vs Ag / AgCl) may be sufficient to obtain the same reduction current.

In conclusion, this experiment demonstrates the ability of this bio-electrolytic device to substitute for the intracellular electron and proton donors required for the reduction of C0 ₂ catalyzed by S. maltophilia bacteria.

 3. Reproducibility study

In addition to the test presented above, four other independent tests of biocerolyser according to the mode described by the invention have confirmed, by alternating pure CO2 bubbling / pure argon / CCh, that CO2 is reduced to the cathode and that it generates a significant CCh-dependent reduction current (data not shown). During these tests, two polarization potentials were tested: -0.7 V and -0.8 V vs Ag / AgCl and CO2 reduction current densities between -0.5 and -1.5 A / m ² were obtained.

In addition, these tests validated the fact that the CO2 reduction current stabilized after a few days of bubbling under CO2 and that after the bubbling of argon, a bubbling of CO2 allows the current to return to its stable value (reached before bubbling argon). This highlights the robustness of this bio-electrolysis system.

4. An example of a bio-electrolyser fed by a CO2 / CH4 mixture

A complementary bioelectrolyzer test has demonstrated that the presence of methane (CEE) in the gaseous mixture that contains CO2 and that feeds the bioelectrolyser has no influence on the CCh-reducing activity of the S. maltophilia bacteria at the cathode (data not shown). Indeed, CO2 / CH4 bubbling (1: 1 or 1: 2 v / v) allows to maintain a reduction current equivalent to that obtained under pure CO2 under the same conditions. This result is particularly interesting in the context of the treatment of biogas, containing ECE and CO2 in high levels.

The GC-TCD analysis of the input and output gases of the bio-electrolyser made it possible to measure a reduced experimental volume flow of CO2 of 1.3 mL CCh / min for a reactor with a volume of 60 mL (which is equivalent to a flow of 1.9 m ³ CCh / d for a reactor with a volume of 60 L). This flux is nearly 60 times higher than the theoretical reduced CO2 flow (Fco2, voi) computable on the basis of the CO2 reduction current obtained (by Faraday's law). This means that the bacteria adhered to the cathode are not the only ones to participate in the reduction of CO2 and that the planktonic bacteria (in suspension in the liquid of the bio- electrolyzer) contribute to this. A possible fixation of this C0 ₂ by planktonic bacteria is also not excluded.

Finally, it is important to note that the experimental CO2 reduced biocerolyser flow (ie 1.3 mL CCh / min) is 130 times greater than that obtained in a closed reactor (ie 0.6 mL CO2 / J = 0.01 mL CCh / min for an initial bacterial concentration of 2 g dry cells / L, Table 1). This significant increase in the bio-electrolyser can be justified by (1) an improved CO2 mass transfer thanks to the gas supply within the suspension and (2) an electrochemical assistance which allows the bacterium to have a source of electrons and protons inexhaustible.

Claims

1. Process for the recovery of CCh by biological reduction comprising a step of contacting a liquid phase containing the bacterium Stenotrophomonas maltophilia with a gas phase containing CO ₂ under conditions allowing the reduction of CO ₂ . 2. Process for the recovery of CO ₂ by biological reduction according to claim 1, wherein the step of bringing the liquid phase into contact with CO ₂ is carried out in a closed or semi-closed or continuous reactor. 3. Process for the recovery of C0 ₂ by biological reduction according to claim 2, wherein the CO ₂ -containing gas phase is injected at the center of the reactor, for example by a gas distributor. 4. Process for recovering CO2 by biological reduction according to claim 2, wherein the air present in a head space of the reactor is removed, and the gas phase containing CO2 is injected into said head space of the reactor, so placing the CO2 reduction reaction under an atmosphere consisting solely of said gas phase. 5. Process for the recovery of C0 ₂ by biological reduction according to claim 1, wherein the step of bringing the liquid phase into contact with the CO ₂ -containing gas phase is carried out in a semi-closed or continuous reactor, in which the gas phase containing CO2 is provided continuously, preferably by bubbling, inside the reactor. 6. Process for recovering CO2 by biological reduction according to one of the preceding claims, wherein formate and / or methane (CH ₄ ) is produced by reduction of CO2. 7. Process for recovering CO2 by biological reduction according to one of claims 2 to 6, according to which the step of contacting the liquid phase with the CO ₂ -containing gas phase is carried out in the presence of an electron or proton donor, intra or extracellular. 8. Process for recovering CO2 by biological reduction according to one of claims 2 to 7, according to which the reactor contains an electrochemical assistance, preferably of the bioelectrolysis type. 9. Process for recovering CO2 by biological reduction according to the preceding claims, comprising a preliminary phase of co-cultivation of Stenotrophomonas maltophilia with at least one methanotrophic bacterium before the step of contacting the liquid phase with the gas phase. 10. Process for the recovery of C0 ₂ by biological reduction according to one of the preceding claims, wherein the liquid phase also contains at least one microorganism capable of using the formate to produce organic compounds of interest, such as methane. lactate or alcohols, especially C1 / C5. 11. Process for recovering CO2 by biological reduction according to one of the preceding claims, wherein the gaseous phase is a biogas or industrial fumes and / or agricultural C0 ₂ rich. 12. Process for the recovery of C0 ₂ by biological reduction according to one of the preceding claims, wherein the gaseous phase comprises at least 30% by volume of CO ₂ relative to the total volume of the gas phase. 13. Process for the recovery of C0 ₂ by biological reduction according to one of the preceding claims, wherein the gaseous phase comprises between 30% and 100% by volume of C0 ₂ relative to the total volume of the gas phase. 14. Process for the recovery of C0 ₂ by biological reduction according to one of the preceding claims, wherein the liquid phase comprises at least 3 g of dry cells / L of Sîenotrophomonas maltophîlia, at least 10 g of dry cells / L, at least 20 g of dry cells / L, 30 g of dry cells / L, 40 g of dry cells / L, 50 g of dry cells / L, 60 g of dry cells / L, 70 g of dry cells / L, 80 g of dry cells / L, 90 g of dry cells / L, 100 g of dry cells / L. 15. Process for recovering CO2 by biological reduction according to one of the preceding claims, according to which ammonium and / or ammonia are added in the liquid phase and / or ammonia in the gas phase, in particular between 5 mmol / L and 100 mmol / L, preferably between 20 and 30 mmol / L. 16. Process for recovering CO2 by biological reduction according to one of the preceding claims, wherein PHB is added in the liquid phase, preferably between 30 g / L and 3 g / L, in particular about 0.3 g / L. 17. Process for recovering CO2 by biological reduction according to one of the preceding claims, wherein formate is produced, said process comprising an additional step of recovering the formate produced as formate and / or formic acid and / or or organic compounds of interest derived from the use of the formate by a microorganism capable of using said formate to produce organic compounds of interest. 18. A method for treating a biogas or industrial fumes rich in C0 ₂ and / or C0 ₂ rich agricultural comprising a step of reduction of C0 ₂ by biological reduction comprising a step according to which said biogas is brought into contact and / or fumes with a liquid phase containing the bacterium Stenotrophomonas maltophilia under conditions allowing the reduction of C0 ₂ . 19. A process for producing formate from CO ₂ , comprising a step according to which a CO ₂ -containing gas phase is brought into contact with a liquid phase containing the bacterium Stenotrophomonas maltophilia under conditions allowing the reduction of CO ₂ in formate. 20. Process for producing methane from CO ₂ , comprising a step according to which a CO ₂ -containing gas phase is brought into contact with a liquid phase containing the bacterium Stenotrophomonas maltophilia under conditions allowing the reduction of CO ₂ to methane. 2i. Use of a bacterium Stenotrophomonas maltophilia for the production of formate and / or methane by reduction of C0 ₂ .