METHOD FOR PRODUCING ORGANIC ACIDS BY BIOCONVERSION

Abstract

The invention relates to a method for the production of a carboxylic acid from a primary alcohol, said method comprising: a first step of growing a strain of bacterium of the genus Acetobacter capable of selectively oxidizing said alcohol into said carboxylic acid, in an appropriate growth medium comprising glycerol as carbon source, until said bacterium reaches a late exponential growth phase; then a selective oxidation step comprising aerobically culturing said bacterium in a bioconversion reactor containing an appropriate liquid bioconversion medium containing said alcohol and glycerol, for a sufficient time to oxidize said alcohol into said carboxylic acid.

Claims

1. A method for the production of a carboxylic acid from a primary alcohol, comprising: growing a strain of bacterium of the genus Acetobacter capable of selectively oxidizing said alcohol into said carboxylic acid, in an appropriate growth medium comprising glycerol as carbon source, until said bacterium reaches a late exponential growth phase; and selectively oxidizing said alcohol by aerobically culturing said bacterium in a bioconversion reactor comprising an appropriate liquid bioconversion medium comprising said alcohol and glycerol, for a sufficient time to oxidize said alcohol into said carboxylic acid.

2. The method of claim 1, wherein said bioconversion medium contains a concentration of between 5.5 and 20 g/l of glycerol.

3. The method of claim 1, wherein said growth medium contains an initial concentration of between 10 and 20 g/l of glycerol.

4. The method of claim 1, wherein said bioconversion medium contains between 0.5 and 10 g/L of said alcohol.

5. The method of claim 1, wherein said selective oxidation step comprises batch-feeding said alcohol into said bioconversion reactor.

6. The method of claim 1, wherein said alcohol is 1,3-propanediol, 2-methylbutanol, 2-phenylethanol, butanol, 1,4-butanediol, 3-methylbutanol, 1,4-pentanediol or hexanol.

7. The method of claim 1, wherein said selective oxidation step is carried out for at least 4 hours.

8. The method of claim 1, further comprising adjusting the pH of said bioconversion medium between 4 and 5.

9. The method of claim 1, further comprising extracting said carboxylic acid from said bioconversion medium.

10. The method of claim 9, wherein said extraction of the carboxylic acid from said bioconversion medium is carried out continuously.

11. The method of claim 9, wherein said extraction of the carboxylic acid from said bioconversion medium is a liquid-liquid reactive extraction using an amine-containing organic extracting solution.

12. The method of claim 11, wherein said liquid-liquid reactive extraction is carried out outside the bioconversion reactor.

13. The method of claim 12, wherein said liquid-liquid reactive extraction is carried out through a membrane.

14. The method of claim 11, further comprising recovering said carboxylic acid from said extracting solution.

15. The method of claim 1, comprising a preliminary step of bioproducing said alcohol by a micro-organism and a step of feeding the alcohol thus produced into said bioconversion medium without any intermediate purification step.

16. The method of claim 1, wherein said bioconversion medium contains between 2 and 5 g/L of said alcohol.

17. The method of claim 12, wherein said liquid-liquid reactive extraction is carried out in a hollow-fiber membrane contactor.

Description

[0106] The features and advantages of the invention will emerge more clearly in the light of the following examples of implementation, provided for illustrative purposes only and in no way limitative of the invention, with the support of FIGS. 1 to 9, in which:

[0107] FIG. 1 represents a bar graph showing the concentrations of produced 2-methylbutyric acid and residual 2-methylbutanol, a/in the bioconversion medium after 48 h of the selective oxidation step of a method similar to that of the invention with the exception that the bioconversion medium is devoid of glycerol, and b/in the bioconversion medium after 48 h of the selective oxidation step of a method of producing 2-methylbutyric acid from 2-methylbutanol according to the invention using the bacterial strain Acetobactersp. CIP 58.66;

[0108] FIG. 2 represents a bar graph showing, for several alcohols, the concentration of the starting alcohol in the bioconversion medium (TO) and at the end (Tf=48 h) of the selective oxidation step of a method of producing an acid from an alcohol according to the invention, using the bacterial strain Acetobacter sp. CIP 58.66;

[0109] FIG. 3 represents a bar graph showing, for several alcohols, the concentration of the produced acid in the bioconversion medium (TO) and at the end (Tf=48 h) of the selective oxidation step of a method of producing an acid from an alcohol according to the invention, using the bacterial strain Acetobacter sp. CIP 58.66;

[0110] FIG. 4 schematically represents a device for implementing a method according to the invention for the production of a carboxylic acid from an alcohol, said method including an acid extraction step and a back-extraction step;

[0111] FIG. 5 represents graphs showing, as a function of time, during the selective oxidation step of a method of the invention for producing 3-hydroxypropionic acid (3-HP) from 1,3-propanediol by Acetobacter sp. CIP 58.66, the concentration of 3-hydroxypropionic acid respectively in the bioconversion medium, in the extracting solution and in the back-extraction (stripping) solution, in A/ in the first 8 h, and in B/ after the selective oxidation step is over, but the extraction and back-extraction continue, the time being represented in logarithmic scale; in this figure, a/ represents a first 1,3-propanediol addition in the bioconversion medium, b/ represents the start of the extraction step and c/ represents a second 1,3-propanediol addition in the bioconversion medium;

[0112] FIG. 6 represents graphs showing, as a function of time, during experiments of bioconversion of 1,3-propanediol (1,3-PDO) into 3-hydroxypropionic acid (3-HP) by Acetobacter sp. CIP 58.66, a first experiment being carried out without any preliminary cell growth step on glycerol (“without glycerol”, comparative example) and a second experiment being carried out with a preliminary cell growth step on glycerol (“with glycerol”, method according to the invention), in A/, 3-HP production, in B/, biomass (Dry Weight, DW) production, and in C/, 1,3-PDO consumption;

[0113] FIG. 7 represents graphs showing, as a function of time, for an experiment of bioconversion of 2-methylbutanol into 2-methyl butyric acid by a process according to the invention using Acetobacter sp. CIP 58.66, in A/, the optical density at 600 nm (O.D.), pH and glycerol concentration in the bioconversion medium and in B/, the concentrations of 2-methyl butanol and 2-methyl butyric acid in the bioconversion medium;

[0114] FIG. 8 represents a graph showing, as a function of time, for an experiment of bioconversion of 2-phenyl ethanol into 2-phenyl acetic acid by a process according to the invention using Acetobacter sp. CIP 58.66, the concentrations of 2-phenyl ethanol and 2-phenyl acetic acid in the bioconversion medium;

[0115] and FIG. 9 represents a graph showing, as a function of time, for an experiment of bioconversion of 1,4-pentanediol by a process according to the invention using Acetobacter sp. CIP 58.66, the concentrations of 1,4-pentanediol, 4-oxo valeric acid and γ-valerolactone in the bioconversion medium.

Preparation of the Bacterial Strain

[0116] A lyophilisate of Acetobacter sp. CIP 58.66 from the Biological Resource Center of Pasteur Institute (Paris, France) is inoculated into 50 mL of medium comprising 25 g.Math.L.sup.−1 mannitol, 5 g.Math.L.sup.−1 yeast extract, 3 g.Math.L.sup.−1 peptone and a 5.5 mol.Math.L.sup.−1 sulfuric acid solution in appropriate quantity to adjust the pH to 6.5, contained in a 250 mL baffled shake-flask. The culture is incubated 1 week at 25° C. under stirring at 200 rpm. After 1 week, 2 mL of this culture are used to inoculate 100 mL of fresh medium (20 g.Math.L.sup.−1 ethanol, 5 g.Math.L.sup.−1 yeast extract, 3 g.Math.L.sup.−1 peptone; pH adjusted to 6.5) in a 500 mL baffled shake-flask. This second culture is incubated 1 week in the same conditions. Then glycerol is added to the medium (20% (w/v)), as cryoprotectant. The broth is aliquoted in 1 mL tubes and stored at −80° C., with a cell density around 9.5.10.sup.8 cell mL.sup.−1. These cryotubes are used as starters for inoculum preparation for all the experiments described below.

[0117] For each step of the method described below, the medium is sterilized before use by autoclaving for 20 min at 120° C., then stored at 4° C. in a cold room or at room temperature.

[0118] The optical density of the samples is measured at 600 nm. Milli-Q water is used for dilution of the samples when required, to stay within the linearity range of the device.

Example 1—Bioconversion of 2-methylbutanol into 2-methylbutyric acid

Growth Step

[0119] Acetobactersp CIP 58.66 is cultivated in a growth medium containing, per liter, 8.71 g of K.sub.2HPO.sub.4, 5 g of peptone, 3 g of yeast extract and 20 g of glycerol.

[0120] After 24 h the bacterium reaches the late exponential phase (optical density at 600 nm close to 3).

Bioconversion Step

[0121] Bioconversion with Acetobacter sp CIP 58.66 is carried out on a sterile bioconversion medium, in shaken culture, in a 250 ml baffled Erlenmeyer flask containing 25 ml of medium. The medium contains, per liter, 8.71 g of K.sub.2HPO.sub.4, 5 g of peptone, 3 g of yeast extract and either no glycerol or 20 g glycerol. The pH of the medium is adjusted to 6.5 with 5.5 M H.sub.2SO.sub.4. 2-Methyl butanol is sterilized by filtration and added to the sterile medium at the rate of 600 mg/L before inoculation of the bacterium. The medium is seeded at 3% v/v with the bacterial culture obtained at the end of the growth phase.

[0122] A 5 ml sample of the bioconversion medium is taken after 48 h. It is acidified with 300 μl of 1M hydrochloric acid to facilitate the extraction of acids. The medium is centrifuged and then the supernatant (3 ml) is extracted three times with 1 ml of dichloromethane. The dichloromethane extracts are combined and dried up with anhydrous sodium sulfate.

[0123] The extract is analysed by GC-MS (gas chromatography—mass spectrometry) for the presence of 2-methylbutyric acid and 2-methylbutanol, using two different columns.

[0124] The alcohol was analysed with a 6890 Agilent GC-MS equipped with an automatic sampler (injecting 1 μl CH.sub.2Cl.sub.2 sample extract), The samples were injected in Splitless mode, with ultrapure helium as carrier gas at an initial flow rate of 1.2 mL/min, (Split open after 1 min, 30 ml/min) on a 60 m long BPX 70 capillary column (0.25 mm diameter with a 0.25 μm phase thickness). Oven was programmed from 35 to 60° C. (20° C./min) then from 60 to 240° C. (3° C./min) and set for 5 min when 240° C. was reached. Detection was made with a mass spectrometer quadrupole (Source 230° C., 70 MeV). The alcohol was identified from injection of standards on the basis of its retention time and mass spectra. An external standard curve was made from standard solutions made from the initial culture medium extracted and analysed as the samples. Total Ion Current was used for quantification.

[0125] The acid was also analysed with the same GC-MS device but with 30 m long FFAP capillary column (0.25 mm diameter with a 0.25 μm phase thickness). Oven was programmed from 40 to 110° C. (3° C./min) then from 110 to 250° C. (7° C./min). Detection was made with the mass spectrometer as described for the alcohol. Identification and quantification were made as for the alcohol.

[0126] The results obtained are shown in FIG. 1.

[0127] It is observed that in the absence of glycerol in the fermentation medium the bioconversion to 2-methylbutyric acid is low (42+/−10 mg/L of acid in the bioconversion medium) and the precursor alcohol is not completely consumed, as there remains 273+/−28 mg/L of residual 2-methylbutanol in the bioconversion medium, i.e. almost half of the initial concentration. The molar yield is low (26.5%).

[0128] In the presence of glycerol in the bioconversion medium, about 15 times more 2-methylbutyric acid are produced (593+/−21 mg/L) and there is no more than 1 mg/L of 2-methylbutanol left in the bioconversion medium. The molar conversion yield is close to 100% (apart from measurement errors, 107% measured).

[0129] These results demonstrate that, surprisingly, the molar yield of the bioconversion reaction is much higher when the bioconversion medium is supplied with glycerol, compared to when it is devoid of glycerol.

[0130] Furthermore, the mean productivity of the method is also much higher when the bioconversion medium is supplied with glycerol, compared to when it is devoid of glycerol: for a 48 hours reaction in a bioconversion medium containing glycerol, a mean productivity of 13.1 mg.Math.l.sup.−1.Math.h.sup.−1 is measured; whereas a mean productivity of only 1.58 mg.Math.l.sup.−1.Math.h.sup.−1 is measured fora 55 hours bioconversion in a bioconversion medium devoid of glycerol.

Example 2—Bioconversion of Other Alcohols/Diols

Growth Step

[0131] The growth medium contains, per liter, 8.71 g of K.sub.2HPO.sub.4, 5 g of peptone, 3 g of yeast extract, 20 g of glycerol and deionised water (qs 1 L). The pH of the medium is adjusted to 6.5 with sulfuric acid 5.5 M.

[0132] The growth step is carried out in 500 ml baffled flasks filled with 10% by volume of growth medium and inoculated with 1 cryotube containing the bacterium. The flasks are closed with carded cotton and incubated at 30° C. for 48 h, under shaking at 200 rpm. After 48 h the cells have reached late exponential growth phase (optical density at 600 nm close to 2.5). 3 ml of the bacterial preculture are collected for the subsequent selective oxidation step of the method.

Selective Oxidation Step

[0133] The bioconversion medium is the same as the growth medium, with or without 20 g/L of glycerol.

[0134] The following alcohols are tested: 2-phenyl ethanol, n-butanol, 2-methyl butanol, 3-methyl butanol, 1,4-pentanediol, hexanol.

[0135] Solutions of these alcohols at a concentration of 7.5 g/L, are prepared as follows. 0.375 g of each alcohol is weighed into 50 mL volumetric flasks using a precision balance and glass pipettes. The flasks are then filled up to the mark with Milli-Q water, mixed by manual stirring, and their contents are transferred to sterile 180 mL plastic bottles. The sterility of these solutions is ensured by filtration in 50 ml glass vials, which have been previously autoclaved, using 20 ml syringes and sterile filters having pores with a 0.22 μm diameter.

[0136] The selective oxidation step is carried out in 250 ml baffled vials filled to 10% by volume of the bioconversion medium, seeded with 3 ml of bacterial preculture and fed with 2 ml of alcohol solution.

[0137] The incubation conditions are the same as in the growth step.

[0138] To ensure that there is no contamination, bacterial aliquots are regularly sampled and laid down on solid growth medium contained in sterile Petri dishes of 9 cm diameter. The composition of the solid growth medium is identical to the one of the liquid growth medium but it additionally contains 20 g.Math.L.sup.−1 of agar. A 100 μL sample of the culture to be tested at the desired dilution (10.sup.−6, 10.sup.−7, or 10.sup.−8) is placed on the solid growth medium, and distributed using a sterile rake. The dishes are closed and placed in an oven at 30° C. for a few days, checking them regularly to monitor growth and to ensure the absence of contamination.

[0139] After 48 hours of culturing, 5 ml samples are collected from the bioconversion medium and acidified with 300 μL of 1 M HCl, to lower the pH to about 2. The cell pellet is separated by centrifugation at 13,400 G at 15° C. for 5 min. The supernatant containing the molecules of interest is recovered.

[0140] Extraction is carried out by adding 1 mL of CH.sub.2Cl.sub.2 to the recovered supernatant, followed by vortexing for 10 s. The organic phase is recovered with a glass pipette in 4 mL vials. This operation is performed 3 times per sample to maximize the extraction yield. The CH.sub.2Cl.sub.2 phase is then dehydrated with anhydrous sodium sulfate.

[0141] Each sample thus obtained is analysed by GC-MS (Gas Chromatography coupled with Mass Spectrometry) using the same materials and methods as described in example 1.

[0142] To allow a quantitative assay, 3 external standards comprising the expected compounds at different known concentrations (500, 250, 125 and 62.5 mg/L) are prepared and analysed in parallel with the samples. This makes it possible to correlate the surfaces of the peaks obtained with the respective concentrations of the different molecules.

[0143] The constituents separated by chromatography are fragmented and ionized by electronic impact at 70 eV, separated according to their m/z ratio using a quadrupole and finally detected using a high energy conversion dynode. The total electric current proportional to the abundance of ions selected by the quadrupole is then measured by the machine.

[0144] These experiments were carried out in triplicate.

[0145] The results obtained when glycerol was present in the bioconversion medium are shown on FIG. 2 (concentration of alcohol in the bioconversion medium at the beginning and after 48 h of the selective oxidation step) and on FIG. 3 (concentration of produced organic acid in the bioconversion medium at the start and after 48 h of the selective oxidation step).

[0146] For 2-phenylethanol, butanol, 3-methylbutanol, 2-methylbutanol and hexanol, a very significant bacterial growth was observed. In each corresponding culture, the added alcohol was exhausted after 48 h of culture. 90% of 1,4-pentanediol was consumed after 48 h of culture.

[0147] Depending on the alcoholic precursor, the concentration of produced acid was between 0.6 and 0.8 g/l, and the productivity was between 12 and 15 mg.Math.l.sup.−1.Math.h.sup.−1.

Example 3—Complete Integrated Process

Device

[0148] The device used for this experiment in shown in FIG. 4.

[0149] It comprises: [0150] a bioconversion reactor 10 for containing the bioconversion medium 104, equipped with stirring means 101, heating means 102 and aerating means 103; [0151] a first membrane contactor 20 containing a hollow fibers membrane. The hollow fibers are contained in an external shell, around a central baffle (not shown on the figure); [0152] means for circulating bioconversion medium 104 from the bioconversion reactor 10 through the shell compartment of the first membrane contactor 20 and back to the bioconversion reactor 10, said means comprising: a first pipe 201 for circulating said liquid from the bioconversion reactor to a first end 21 of the first membrane contactor 20, a second pipe 202 for circulating said bioconversion medium from a second end 22 of the first membrane contactor 20 opposite the first end 21, to the bioconversion reactor, and a first pump 203 for flowing said liquid in the first pipe 201 and in the second pipe 202; [0153] means for circulating a first liquid solution 24, in countercurrent with respect to the bioconversion medium 104, from a first flask 204 containing said first liquid solution 24, in the hollow fibers of the membrane of the first membrane contactor 20, and back to the first flask 204. Said means comprise: a third pipe 205 for circulating said first liquid solution 24 from the first flask 204 to the second end 22 of the first membrane contactor 20, a fourth pipe 206 for circulating said first liquid solution 24 from the first end 21 of the first membrane contactor 20 to the first flask 204, and a second pump 207 for flowing said first liquid solution 24 in the third pipe 205 and in the fourth pipe 206. In the figure, a dotted line indicated by reference 23 symbolizes the separation between the phases in the first membrane contactor; [0154] a second membrane contactor 30 containing a hollow fibers membrane. The hollow fibers are contained in an external shell, around a central baffle (not shown on the figure); [0155] means for circulating the first liquid solution 24 from the first flask 204 through the shell compartment of the second membrane contactor 30 and back to the first flask 204, said means comprising: a fifth pipe 301 for circulating said first liquid solution 24 from the first flask 204 to a first end 31 of the second membrane contactor 30, a sixth pipe 302 for circulating said first liquid solution 24 from a second end 32 of the second membrane contactor 30 opposite the first end 31, to the first flask 204, and a third pump 303 for flowing said first liquid solution 24 in the fifth pipe 301 and in the sixth pipe 302; [0156] and means for circulating a second liquid solution 34, in countercurrent with respect to the first liquid solution 24, from a second flask 304 containing said second liquid solution 34, through the hollow fibers of membrane of the second membrane contactor 30, and back to the second flask 304, said means comprising: a seventh pipe 305 for circulating said second liquid solution 34 from the second flask 304 to the second end 32 of the second membrane contactor 30, an eighth pipe 306 for circulating said second liquid solution 34 from the first end 31 of the second membrane contactor 30 to the second flask 304, and a fourth pump 307 for flowing said second liquid solution 34 in the seventh pipe 305 and in the eighth pipe 306. In the figure, a dotted line indicated by reference 33 symbolizes the separation between the phases in the second membrane contactor.

[0157] The bioconversion reactor 10 is a 3.6 L Labfors 4 bioreactor (Infors), with its associated software (Iris v.5), for data acquisition and process control. The stirring means 101 equipping it are a single Rushton turbine for broth stirring. The aerating means 103, for providing air into the bioconversion medium contained in the reactor 10, comprise a mass air flowmeter. The heating means 102 are formed of a double wall surrounding the reactor 10, in which hot water circulates.

[0158] The first hollow fibres membrane contactor 20 and the second hollow fibres membrane contactor 30 are identical. They include a 2.5×8 Liqui-Cel® module containing X50 fibres (Membrana, USA). Characteristics of the module are as follows: material: polypropylene, internal diameter 58.4 mm, internal length 203 mm, number of fibres 9800, contact area 0.4 m.sup.2. The characteristics of the fibers are as follows: material: polypropylene, internal diameter 220 μm, external diameter 300 μm, effective length 146 mm, wall thickness 40 μm, porosity 40%, average pore size 0.03 μm.

Primary Growth Step on Glycerol

[0159] Biocatalyst production is performed in batch mode, with glycerol as growth substrate, in the reactor 10. An initial volume of 1.2 L of a growth medium B2 containing 5 g/L of yeast extract, 3 g/L of peptone, 10 g/L of glycerol, of pH 5.0, is inoculated with the bacterial pre-culture, so that the initial biomass concentration is 0.02 cell dry weight per liter. Due to the inoculum addition, initial pH rises from 5.0 to 5.2. During bacterial growth, pH is left uncontrolled, but agitation speed (from 100 to 800 rpm) and airflow rate (from 1 to 4 NL.Math.min.sup.−1) are automatically controlled in order to maintain the partial oxygen pressure in the reactor (pO.sub.2) above 40%. The temperature of the growth medium is 30° C.

[0160] In all the experiment, the Cell dry weight (CDW) is determined by optical density (OD) measurement, and according to the formula: CDW=0.59×OD.

[0161] When the late exponential phase is reached the cell dry weight is around 0.88 (+/−0.05) g/L, corresponding to an optical density at 600 nm of around 1.49.

Fed-Batch Selective Oxidation Step

[0162] When late exponential phase is reached, after 32 h of growth step, 6 mL of filter-sterilized 1,3-propanediol (98% w/v) is added to the medium in order to trigger bioconversion, as indicated by reference 11 on FIG. 4. The bioconversion medium 104 thus obtained contains 5 g/l of 1,3-propanediol and 8.6 g/l of glycerol, that has not been consumed during the growth step.

[0163] The selective oxidation step is carried out at 30° C. with partial oxygen pressure (pO.sub.2) controlled at 40%. The bacterial cells produce 3-hydroxypropionic acid by oxidizing 1,3-propanediol.

Extraction and Back-Extraction Steps

[0164] At the end of the growth step on glycerol the pH of the medium is close to 7. An acidification of the bioconversion medium occurs after the 1,3-propanediol addition, due to production by the bacterium of 3-hydroxypropionic acid. The bioconversion medium 104 is brought in contact with the extracting solution only after pH has reached 4.6, which is close to the pKa of 3-hydroxypropionic acid (4.51). The extraction is started 0.45 h after the 1,3-propanediol addition. No base is added for pH control, and pH varies freely according to production and extraction rates.

[0165] The extracting solution (first liquid solution 24) consists of 20% (v/v) of didodecylmethylamine (DDMA) diluted in 40% (v/v) dodecanol and 40% (v/v) dodecane. The corresponding partition coefficient, i.e. the ratio, at equilibrium, between the 3-hydroxypropionic acid concentration in the organic phase (extracting solution 24) and in the aqueous phase (bioconversion medium 104) and viscosity are determined as described in the publication of Sanchez-Castaneda et al. 2020, Journal of Chemical Technology and Biotechnology 95:1046-1056 and are equal to 0.78 and 4.7 mPa.Math.s, respectively.

[0166] For back-extraction, a second liquid solution 34, also called stripping solution, containing NaOH 0.5 mol.Math.L.sup.−1 in water, is used.

[0167] The method is implemented as follows.

[0168] Before each experiment, the membrane contactors 20, 30 have been washed by circulating a 60% (v/v) isopropanol solution for 2 hours through the fibres and the shell side, then drained and rinsed with sterilized Reverse Osmosis (RO) water and finally dried overnight by flushing compressed air.

[0169] The bioconversion medium 104 is continuously circulated at a flow rate of 450 ml/min in the first pipe 201 in the direction indicated by arrow 41 on FIG. 4, through the shell of the first membrane contactor 20, then in the second pipe 202, in the direction indicated by arrow 42, back to the bioconversion reactor 10. At the same time, the extracting solution 24 is continuously circulated at a flow rate of 430 ml/min from the first flask 204, in the third pipe 205 in the direction indicated by arrow 43 on FIG. 3, in the hollow fibers of the first membrane contactor 20, in countercurrent flow with respect to the bioconversion medium 104, then in the fourth pipe 206, in the direction indicated by arrow 44, back to the first flask 204.

[0170] The central baffle of the first membrane contactor 20 forces a radial flow in the shell compartment, between the fibres that are woven together. The membrane being hydrophobic, the pores are filled with the organic phase (extracting solution 24), which is therefore in direct contact with the aqueous phase (bioconversion medium 104).

[0171] 3-hydroxypropionic acid is thereby extracted from the bioconversion medium 104 into the extracting solution 24, through an organic acid-base interaction. The extracting solution 24 gets gradually loaded with this acid. Glycerol and 1,3-propanediol are not affected by the extraction step, neither are the bacterial cells contained in the bioconversion medium 104.

[0172] For recovery of the 3-hydroxypropionic acid from the extracting solution 24, a back-extraction is performed in the second membrane contactor 30.

[0173] To this end, 1 L of the second liquid solution 34 is continuously circulated inside the fibers of the second membrane contactor 30, while the extracting solution 24 is continuously circulated in the shell of the second membrane contactor 30.

[0174] More precisely, the extracting solution 24 is continuously circulated at a flow rate of 430 ml/min in the fifth pipe 301 in the direction indicated by arrow 45 on FIG. 4, through the shell of the second membrane contactor 30, then in the sixth pipe 302, in the direction indicated by arrow 46, back to the first flask 204. At the same time, the stripping solution 34 is continuously circulated at a flow rate of 450 ml/min from the second flask 304, in the seventh pipe 305 in the direction indicated by arrow 47 on FIG. 4, in the hollow fibers of the second membrane contactor 30, in countercurrent flow with respect to the extracting solution 24, then in the eighth pipe 306, in the direction indicated by arrow 48, back to the second flask 304.

[0175] 3-hydroxypropionic acid is thereby gradually transferred from the extracting solution 24 to the aqueous stripping solution 34, in the form of a sodium salt.

[0176] After 4.6 h of bioconversion, a new 6 mL addition of 1,3-propanediol in the bioconversion medium is performed.

[0177] The selective oxidation step is carried out for 8 h. Extraction and back-extraction are continued for a further 100 h.

[0178] Samples are taken from the bioconversion medium, the extracting solution and the stripping (back-extraction) solution at different times, for components quantification. Biomass evolution is also monitored in the bioconversion reactor, by off-line optical density measurements.

Results

[0179] Concentrations of 3-hydroxypropionic acid, 1,3-propanediol and glycerol are quantified through HPLC (high-performance liquid chromatography). The samples are prepared differently for aqueous (bioconversion medium and stripping phase) and organic (extracting phase) samples, as follows: [0180] bioconversion medium samples: proteins are precipitated with trichloroacetic acid, then the samples are centrifuged and only supernatants are recovered. The samples are then filtrated with a nylon filter (0.22 μm pore diameter); [0181] extracting phase samples: these samples are first back-extracted overnight with an equal volume of NaOH (0.5 mol/L), then the aqueous phase is recovered and filtered (nylon filter, 0.22 μm pore diameter); [0182] stripping phase samples: these samples are filtered (nylon filter, 0.22 μm pore diameter).

[0183] All samples are analyzed on an Aminex® HPX-87H exchange column and a refractive index detector, with H.sub.2SO.sub.4 as mobile phase. For 3-hydroxypropanal, the analysis conditions are 35° C., H.sub.2SO.sub.4 at 0.6 mL/min and 5 mmol/L. For 3-hydroxypropionic acid, 1,3-propanediol and glycerol the conditions are: 65° C., H.sub.2SO.sub.4 at 0.4 mL/min and 0.5 mmol/L.

[0184] At the end of the growth step, after 32 h of growth, Cell dry weight (CDW) concentration reaches 0.98+/−0.12 g.sub.CDW L.sup.−1, with a corresponding yield of 0.42+/−0.12 g.sub.CDW.Math.g.sup.−1.sub.glycerol. Only 21.1+/−3.3% of the initial glycerol was consumed. No by-product of glycerol consumption, such as organic acids, can be detected in the medium.

[0185] During the selective oxidation step, both 1,3-propanediol inputs, at 0 h and 4.6 h of the bioconversion step, can be associated with a new exponential growth phase, in all replicates. Numbers of generations (Ng) are similar after both inputs, and overall, CDW increases from 1.01+/−0.06 to 2.75+/−0.07 g.sub.CDW.Math.L.sup.−1. The pH decreases from 6.9 to 4.2, and it decreases even further, until 3.9, after the second 1,3-propanediol addition.

[0186] During the selective oxidation step, glycerol is further consumed, but never fully depleted: after 25 h, 37.9+/−2.9% of the glycerol initially supplied was consumed, instead of 21.1+/−3.3% at the beginning of the selective oxidation step.

[0187] While glycerol is likely the main growth substrate, the presence of 1,3-propanediol in concentrations under 10 g.Math.L.sup.−1 is also shown to enhance the growth of Acetobacter sp. CIP 58.66 on this complex medium.

[0188] The results obtained in the bioconversion reactor are shown in table 1.

TABLE-US-00001 TABLE 1 Performance of a method of the invention for producing 3-hydroxypropionic acid from 1,3-propanediol 1.sup.st alcohol addition 2.sup.nd alcohol addition Time (h) 0 4.46 4.61 7.63 Growth phase 1.sup.st exponential 2.sup.nd exponential CDW (g/L) 1.01 1.57 1.56 1.99 Maximal specific growth rate 0.17 0.08 Total acid production (g) 0 6.05 6.65 13.00 Acid yield (mol.sub.acid/mol.sub.alcohol) — 0.83 0.90 0.91 Maximal specific acid 2.36 2.58 productivity (g.sub.acid/(g.sub.CDW/h)) Average specific acid 1.11 1.12 productivity (g.sub.acid/(g.sub.CDW/h))

[0189] The physiological state of Acetobacter sp. CIP 58.66 is monitored during the selective oxidation step, by dual fluorescent staining and flow cytometry analysis. For each sample, cells can be divided into three sub-populations: (i) enzymatically active cells, (ii) altered cells, that are still enzymatically active, but whose membranes are deteriorated, and (iii) dead cells. During the first 4 h of the selective oxidation step, the proportion of enzymatically active cells remains between 75 and 87%, while dead cells represent between 11 and 23% of the overall population. After 4 h of the selective oxidation step, the proportion of enzymatically active cells decreases during approximately 2 h, before reaching values between 46 and 55% at the end of the selective oxidation step. Overall, the biocompatibility of the system is very satisfactory, since half of the bacterial population is still active after 7.63 h of the selective oxidation step, when 1,3-propanediol is almost fully depleted. Interestingly, between 18 and 25 h after the selective oxidation step starts, the proportion of enzymatically active cells remains between 14 and 17%. This demonstrates that this process setup is suitable for continuous extractive bioconversion, for at least 25 h.

[0190] For both 1,3-propanediol additions, 3-hydroxypropionic acid is the main product, with an overall yield of 0.91+/−0.06 mol.sub.acid.Math.mol.sup.−1 alcohol. The maximal specific productivities (q.sub.acid,max) that are reached for each successive phases are similar. A slight and transient accumulation of the corresponding intermediate product 3-hydroxypropanal is detected around 1 h after each alcohol input, without exceeding 0.20+/−0.11 and 0.24+/−0.14 g.Math.L.sup.−1, respectively.

[0191] 3-hydroxypropionic acid distribution among the bioconversion medium, the extracting solution and the stripping solution, is shown in FIG. 5.

[0192] As can be seen on this figure, during the selective oxidation step extraction starts slowly, leading to a progressive 3-hydroxypropionic acid accumulation in the extracting solution. The extraction rate increases progressively with 3-hydroxypropionic acid production, and back-extraction in the stripping solution starts later: 3-hydroxypropionic acid is detected in the stripping solution only after 1.7 h. Back-extraction rate increases as acid accumulates in the extracting solution, eventually exceeding the extraction rate, as shown by the decrease in 3-hydroxypropionic acid in the extraction solution (in B in the figure). After 8 h of bioconversion, all the added 1,3-propanediol is consumed and 3-hydroxypropionic acid production stops. From then on, 3-hydroxypropionic acid concentration decreases both in the bioconversion reactor and in the extracting solution, while the concentration in the stripping solution increases. The system is left running in order to recover all the produced 3-hydroxypropionic acid. After 114 h, nearly all the 3-hydroxypropionic acid is recovered in the stripping solution, thus achieving therein a final concentration in 3-hydroxypropionic acid of 12.63+/−0.50 g.Math.L.sup.1, while almost no 3-hydroxypropionic acid is detected in the bioconversion medium and in the extracting solution.

[0193] Higher 3-hydroxy propionic acid concentrations can be reached easily by reducing the back-extraction phase volume:

[0194] It is observed that the overall effect of the pertraction system of the invention is well tolerated by Acetobacter sp. CIP 58.66 during bioconversion, and that 3-hydroxypropionic acid can be produced with high yield by the method of the invention, leading to recovery of this acid with a high purity degree in the stripping solution.

[0195] The mean productivity, for the first 8 hours of the selective oxidation step, is 1.53 g/(L.Math.h), with respect to the working volume of the reactor.

Example 4—Comparison of Production of 3-hydroxypropionic acid (3-HP) from 1,3-propanediol (1,3-PDO) with or without the First Step of Preliminary Growth on glycerol

[0196] Direct Fed-Batch Bioconversion (without a Preliminary Growth on Glycerol)

[0197] Single-step fed-batch experiments are carried out in a 3.6 L Labfors 4 bioreactor. 1 L of medium (5 g/L yeast extract, 3 g/L peptone, 5 g/L 1,3-PDO, pH=5.0) is sterilized inside the bioreactor, at 120° C., during 20 min. The temperature is controlled at 30° C. all along the experiment, and the partial dissolved 02 pressure (pO.sub.2) is kept above 40% by automatic control of stirring speed (between 100 and 800 rpm) and air flow rate (between 1 and 4 NL.Math.min.sup.−1). The medium is inoculated using the preculture of Acetobacter sp. CIP 58.66 so that the initial cell dry weight (DW) is around 0.02 g.sub.DW L.sup.−1. The pH value is maintained at 5.0 by automatic addition of an equimolar solution of ammonium hydroxide (NH.sub.4OH) and 1,3-PDO, at 6 mol.Math.L.sup.−1, thus also serving as continuous substrate input. Samples are regularly withdrawn for optical density (OD) measurements and metabolites quantification.

Sequential 3-HP Production Process (with a Preliminary Growth on Glycerol)

Primary Growth on Glycerol

[0198] Biocatalyst production is performed in batch mode, with glycerol as growth substrate, in a 3.6 L Labfors 4 bioreactor. An initial 1 L volume of medium (5 g/L yeast extract, 3 g/L peptone, 10 g/L glycerol, pH=5.0) is inoculated with the pre-culture of Acetobacter sp. CIP 58.66 so that the initial biomass concentration reaches 0.02 g.sub.DW L.sup.−1. Due to the inoculum addition, initial pH rises from 5.0 to 5.2. During this step, pH is left uncontrolled, but agitation speed (from 100 to 800 rpm, Rushton turbine) and airflow rate (from 1 to 4 NL.Math.min.sup.−1) are automatically controlled in order to maintain pO.sub.2 above 40%.

Fed-Batch Bioconversion

[0199] When late exponential phase is reached (at 32 h), pH is adjusted to 5.0 using H.sub.2SO.sub.4 (5.5 mol.Math.L.sup.−1). 5 mL of filter-sterilised 1,3-PDO are then added to the medium in order to trigger the bioconversion, and the equimolar solution of NH.sub.4OH and 1,3-PDO is plugged to the bioreactor, for both pH control and 1,3-PDO feeding purposes. Samples are regularly taken during growth phase and the bioconversion phase, for optical density (OD) measurements and metabolites quantification.

Results

[0200] The results obtained are shown in FIG. 6, which compares 3-HP (A/) and biomass (Dry Weight, DW) (B/) production, as well as 1,3-PDO consumption (C/) in presence and in absence of a preliminary growth step on glycerol (“with glycerol”, and “without glycerol”, respectively).

[0201] In absence of a primary growth phase on glycerol, a slight cell growth is observed with a final DW of 0.09±0.03 g.sub.DW and a maximal growth rate of 0.25±0.03 h.sup.−1. Bioconversion of 1,3-PDO into 3-HP occurs during growth of Acetobacter sp. CIP 58.66 in fed-batch mode, even though it remains limited. Even though 1,3-PDO is kept around 5 g/L, its overall consumption remains low (3.2±0.2 g), and final 3-HP titres are also low (2.5±0.8 g/L). The final 3-HP yield is 0.28±0.01 mol/mol.

[0202] On the contrary, when biomass of Acetobacter sp. CIP 58.66 is first produced from 10 g/L glycerol, higher cell densities are obtained. Late exponential phase is reached by 31 h of culture, with a 0.90±0.09 g/L DW concentration. Furthermore, by 31 h, glycerol consumption remains low (21% of initial glycerol) and no by-product can be detected. Because no buffer solution is added to the medium, pH rises up to 6.9. This rise in pH may be the cause for growth slowing down before full substrate depletion. Once the bioconversion is triggered at 32 h by adding 1,3-PDO, a secondary exponential growth phase can be identified, without any latency. DW concentration further increases up to 2.08±0.04 g.sub.DW/L. Glycerol is further consumed during 1,3-PDO oxidation: by 57 h, 42% of initial glycerol has been consumed, instead of 21% at the beginning of bioconversion. It can thus be hypothesized that glycerol serves as carbon source for the strain's growth.

[0203] Consumption of 1,3-PDO begins as soon as it is added to the medium after the primary growth on glycerol. At 57 h, a total of 51.45±6.63 g of 1,3-PDO has been consumed and converted almost stoichiometrically into 61.58±5.88 g of 3-HP.

Example 5—Analysis Over Time

[0204] Bioconversion of the following substrates was carried out by a method according to the invention: 2-methyl butanol (into 2-methyl butyric acid), 3-methyl butanol (into 3-methyl butyric acid), 2-phenyl ethanol (into 2-phenyl acetic acid), 1,4-pentanediol (into 4-hydroxy-valeric acid then γ-valerolactone after cyclisation) and fusel oil. Fusel oil is an industrially available substrate obtained from ethanol distilleries and rich in 2-methyl butanol and 3-methyl butanol.

Growth Step

[0205] Acetobacter sp CIP 58.66 is cultivated in a growth medium containing, per liter, 10 g of K.sub.2HPO.sub.4, 6 g of casein peptone, 3 g of yeast extract and 20 g of glycerol. The pH of this medium has been beforehand adjusted to 6.5 by adding a few drops of 5.5 M H.sub.2SO.sub.4, then sterilized.

[0206] After 24 h the bacterium reaches the late exponential phase (optical density at 600 nm close to 3).

Bioconversion Step

[0207] Bioconversion with Acetobacter sp CIP 58.66 is carried out on the same sterile bioconversion medium, in shaken culture, in a 500 ml baffled Erlenmeyer flask containing 50 ml of medium.

[0208] Concentrated solutions of the substrates are sterilized by filtration on 0.22 μm filters before being added to the medium (5 ml/flask).

[0209] The medium is seeded at 10% v/v with the bacterial culture obtained at the end of the growth phase. All cultures are agitated (200 rpm) and the temperature controlled (30° C.).

[0210] Samples are taken over time to monitor growth, pH, glycerol, substrates and products concentrations by CPG/MS.

[0211] For GC/MS analyses, the samples (5 ml) are acidified with 300 μl of 1M hydrochloric acid and distributed into two Eppendorf tubes (2 ml/tube). The sample is centrifuged (12000 g for 5 min at 15° C.). A 1.8 ml fraction is taken from each of the two tubes and these fractions are combined for extraction. The 3.6 ml of culture thus treated are extracted successively 3 times with 1 ml of CH.sub.2Cl.sub.2. The two phases (aqueous sample and dichloromethane) are mixed and stirred (vortex 30 s) and the dichloromethane is recovered and then dried with anhydrous Na.sub.2SO.sub.4. The extracts (1 μl) thus obtained are injected (splitless/split) into the gas chromatograph (GC) equipped with an FFAP column, to which an appropriate temperature gradient is applied. The compounds are identified by the retention time of standards of the same structure and their mass spectra. The quantifications are carried out from a range of concentrations of compounds in the culture medium extracted by the same method. The quantification is carried out using these concentration ranges and the signals of the total ionic flux, coming from the mass spectrometer.

[0212] The results obtained for 2-methyl butanol are shown in FIG. 7, for the monitoring over time of optical density at 600 nm, pH and glycerol concentration in the bioconversion medium (in A/) and the kinetics of bioconversion of 2-methyl butanol into 2-methyl butyric acid (in B/).

[0213] Under these conditions, the bioconversion reaches, between the 8.sup.th hour and the 24.sup.th hour, an average production rate of 53 mg.Math.L.sup.−1.Math.h.sup.−1 with an average molar yield of 1.02.

[0214] The results obtained for 2-phenyl ethanol, in terms of concentrations of 2-phenyl ethanol (substrate) and 2-phenyl acetic acid (product) in the bioconversion medium, are shown in FIG. 8.

[0215] The results obtained for 1,4-pentanediol, in terms of concentrations of 1,4-pentanediol (substrate) and γ-valerolactone (final product), in the bioconversion medium, are shown in FIG. 9. Small and increasing amount of 4-oxo valeric acid is also formed showing that the strain was also able to oxidize the secondary alcohol function of the diol but with a much slower yield.

[0216] The metabolic intermediate 4-hydroxy valeric acid was not identified but the occurrence of the production of γ-valerolactone and 4-oxo valeric acid simultaneously to the disappearance of 1,4 pentanediol clearly evidence the oxidation of the diol. The concentration of 4-oxo valeric acid is expressed as an equivalent γ-valerolactone (g/l) (quantification made on the basis of the calibration curve of the γ-valerolactone)).

[0217] Table 2 below shows the best performances obtained for bioconversion of 2-phenyl ethanol into 2-phenyl acetic acid, 1,4-pentanediol into γ-valerolactone and the components of fusel oil: 2-methyl butanol into 2-methyl butyric acid and 3-methyl butanol into 3-methyl butyric acid. The values indicated in this table are mean values of three different experiments.

TABLE-US-00002 TABLE 2 CC at CC at ΔS/ΔT or T0 T1 ΔT T0 T1 ΔP/ΔT (min) (min) (min) (g/l) (g/l) (mg .Math. l.sup.−1 .Math. h.sup.−1) Compound 2-phenyl- 0 48.7 48.7 1.48 0.46 −21 ethanol Phenyl acetic 0 48.7 48.7 0.050 1.59 +31.6 acid 1,4 8.22 23.9 15.7 1.27 0 −81 pentanediol γ-valerolactone 8.22 23.9 15.7 0.11 0.65 +34 Fusel oil 2 and 3-methyl 0 24 24 0.40 0 −16.6 butanol 2-methyl 0 24 24 0 0.0119 +0.5 butyric acid 3-methyl 0 24 24 0 0.0509 +2.1 butyric acid T0: start of the period during which the cultures show the best bioconversion performance; T1: end of the period during which the cultures show the best bioconversion performance; ΔT = T1 − T0; CC = concentration; ΔS = “substrate concentration at T1” − “substrate concentration at T0”; ΔP = “product concentration at T1” − “product concentration at T0”

[0218] All the above results demonstrate successful bioconversion by the method according to the invention of the primary alcohols substrates into the desired carboxylic acid products, even in the case of fusel oil, a complex mixture containing two different primary alcohols.