Edited by: Sabine Kleinsteuber, Helmholtz Centre for Environmental Research—UFZ, Germany
Reviewed by: Amy Michele Grunden, North Carolina State University, USA; Dr. S. Venkata Mohan, Council of Scientific and Industrial Research-Indian Institute of Chemical Technology, India; Hanno Richter, Cornell University, USA
*Correspondence: Florian Oswald
Stefan Dörsam
This article was submitted to Microbiotechnology, Ecotoxicology and Bioremediation, a section of the journal Frontiers in Microbiology
†Florian Oswald and Stefan Dörsam are co-first authors and have contributed equally to the paper.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Synthesis gas (syngas) fermentation using acetogenic bacteria is an approach for production of bulk chemicals like acetate, ethanol, butanol, or 2,3-butandiol avoiding the fuel vs. food debate by using carbon monoxide, carbon dioxide, and hydrogen from gasification of biomass or industrial waste gases. Suffering from energetic limitations, yields of C4-molecules produced by syngas fermentation are quite low compared with ABE fermentation using sugars as a substrate. On the other hand, fungal production of malic acid has high yields of product per gram metabolized substrate but is currently limited to sugar containing substrates. In this study, it was possible to show that
Nowadays most bulk chemicals are still based on fossil fuels like crude oil and natural gas. It is consensus that, due to dwindling resources and climate change, it is necessary to develop sustainable methods for the production of industrially relevant chemicals. Suitable candidates for these demands are various dicarboxylic acids because of their suitability to be used for the synthesis of various polymers, as was summarized by Lee et al. (
Neither are acetogenic bacteria able to produce dicarboxylic acids like malic acid or fumaric acid, as their product spectrum is limited due to energetic reasons, nor are
If not stated differently all chemicals were purchased from Carl-Roth (Germany) or Sigma-Aldrich (Germany).The organism used for the syngas fermentation part of the study was
The
Fermentations were carried out in Minifors bench-top stirred tank reactors from Infors-HT (Switzerland) with a total volume of 2.5 L and a liquid volume of 1.5 L leaving a 1 L headspace. Figure
For anaerobic syngas fermentation with
For
Measurement of the off-gas composition of the syngas fermentation was conducted via a GC-2010 Plus AT gas chromatograph (GC) system (Shimadzu, Japan) equipped with a customized column setup using a ShinCarbon ST 80/100 Column (2 m × 0.53 mm ID, Restek, Germany) and a Rtx-1 capillary column (1 μm, 30 m × 0.25 mm ID, Restek, Germany). The installed detector was a thermal conductivity detector with helium as the carrier gas. Column flow rate was 3 mL/min and oven temperature was kept at 40 °C for 3 min followed by a ramp of 35 °C/min. Total analysis time was 7.5 min. The off-gas line of every bioreactor was connected to the GC using a stream selector which, after each measurement, automatically selected the next bioreactor and thus enabled for automated off-gas analysis. Since the syngas in this work contained nitrogen and
Using Equation (1), the ideal gas law and conditions in the lab (
Equation (3) calculates the amount of substance balance (
Since there is no other sink or source for H2, CO, and CO2 other than the metabolism of
To get the actual amount of CO that has gone into products and biomass it was necessary to account for CO that has been converted to CO2 by
The complete derivation of equation (4) can be found in data sheet
Liquid samples of 2 mL were collected every 2 h during the day. No samples were taken at night. Before the collection of a single liquid sample 3 mL of reactor broth were taken and discarded to account for the dead volume of the sampling line. Cell concentrations were determined using an Ultrospec1100pro spectrophotometer (Amersham Bioscience, USA) at a wavelength of 600 nm. Therefore, the optical density (OD) of 1 mL of a liquid sample was measured, then cells were removed via centrifugation at 16100 × g for 10 min and OD of the supernatant was measured. The difference of both values gave the OD of the sample. This procedure was necessary because OD values of the supernatant changed during fermentation. At measured OD > 0.45 the OD exceeded the linear range of the OD/cell mass relation and samples had to be diluted using 9 g/L NaCl solution. For correlation between OD and bio dry mass (BDM) the BDM at the end of the syngas fermentation was determined in duplicates. Therefore, 60 mL of fermentation broth were collected and transferred into pre-weight, dry screwing cap reaction tubes (30 mL each). The tubes were centrifuged at 4816 × g and 4 °C for 15 min. The supernatant was discarded and pellets were washed two times with a 9 g/L NaCl solution. The tubes with the washed pellets were dried at 60 °C for 48 h before they were weighted again and the BDM was calculated. The quotient of BDM and OD at the end of the syngas fermentation gave the BDM/OD correlation coefficient of 0.139 ± 0.041 g/L. Liquid samples were centrifuged to remove cells and the supernatant was stored frozen at −20 °C and used for further off-line analysis.
Measurement of fructose was done using an enzymatic D-fructose/D-glucose assay of Roche Yellow line (Hoffmann-La Roche, Switzerland) following the instructions delivered with the assay. Concentrations of ethanol and acetic acid of samples containing fructose were also measured with respective enzymatic assays from Roche Yellow line following their instructions. Samples which were collected after complete consumption of leftover fructose from the pre-cultures were analyzed for ethanol and acetic acid using a 6890N GC (Agilent, USA) equipped with auto-sampler, ROTICAP-FFAP capillary column (0.5 μm, 30 m × 0.32 mm ID, Carl-Roth, Germany) and flame ionization detector. Carrier gas was helium with a pressure of 1 bar and split ratio was 7.5:1. Analytical standard mixture consisted of 5 mM ethanol, 5 mM sodium acetate, and 9.09 mM isobutanol in 0.18 M HCl. Samples were prepared by acidifying 500 μL of sample with 50 μL internal standard solution consisting of 100 mM isobutanol in 2 M HCl. Analysis was conducted by injecting 1 μL of sample or standard. The temperature profile of the column oven started with initial 60 °C for 2 min followed by a temperature ramp of 10 °C/min up to an end temperature of 180 °C. Total analysis time was 20 min.
The concentrations of malic and acetic acid during cultivation with
Since malic acid is produced by
Syngas fermentation |
2.5 | 0.76 ± 0.29 | 15.27 ± 1.68 | 0.57 ± 0.58 | 0.39 ± 0.06 | 0.75 ± 0.04 | 0.77 ± 0.05 | 0.01 ± 0.04 | 0.68 ± 0.05 |
NH4-reduced |
0.33 | 0.45 ± 0.14 | 17.08 ± 2.28 | 1.14 ± 0.77 | N/A | 0.64 ± 0.05 | 0.74 ± 0.04 | −0.09±0.13 | 0.67 ± 0.04 |
After 96 h of fermentation on syngas using syngas fermentation medium (see above) with 2.5 g/L ammonia chloride,
Large quantities of organic acids are produced by certain fungi generally under nitrogen limiting conditions and a simultaneous excess of carbon source. For the production of malic acid with
Organic acid production medium |
Glucose 109 | – | 47.84 ± 3.49 | 0.8 |
Organic acid production medium | Glucose 109 | 0.533 g/L Cysteine | 44.2 ± 5.85 | 0.64 |
Organic acid production medium | Glucose 109 | 0.533 g/L Sodium sulfide | 54.04 ± 14.16 | 0.65 |
Organic acid production medium | Acetic acid 50 | Exchange of carbon source | 8.62 ± 1.15 | 0.28 |
Organic acid production medium | Acetic acid, ethanol 33.33, 16.66 | Exchange of carbon source | 11.68 ± 1.27 | 0.55 |
Organic acid production medium | Ethanol 50 | Exchange of carbon source | 0 | 0 |
Syngas fermentation medium |
Acetic acid 50 | Exchange of carbon source | 2.69 ± 0.81 | 0.09 |
Syngas fermentation medium | Acetic acid 50 | Exchange of carbon source, without ammonium | 4.11 ± 0.50 | 0.37 |
Syngas fermentation medium: fermented | Acetic acid 9.80 ± 0.21 | Removal of |
0 | 0 |
Syngas fermentation medium: fermented | Acetic acid 15.84 ± 1.55 | Reduced ammonium | 4.34 ± 0.10 | 0.27 |
Syngas fermentation medium: fermented | Acetic acid 8.88 ± 3.42 | Reduced ammonium Removal of |
0 | 0 |
Influence of major medium components on malic acid production was evaluated in shake flask experiments and fermentation experiments in bioreactors as indicated. The biggest differences of the established process were the carbon source and the nitrogen concentration. It could be shown that acetic acid is an appropriate carbon source for malic acid production in established malic acid production medium with an
The most important byproduct of syngas fermentation is ethanol. Various concentrations of both, acetate and ethanol, could be achieved in different fermentations. To analyze the influence on or the suitability of these molecules as substrates for malic acid production, ethanol was added to production medium. With a yield of 0.55 g/g the acetic acid/ethanol-approach reached a final product concentration of 11.68 ± 1.27 g/L (Table
It could be shown that the reduction agents are not problematic and acetic acid is an appropriate carbon source for malic acid production. The syngas fermentation medium, which has a significant different composition compared to the malic acid production medium, was a further challenge. Especially the initial ammonium concentration proved to be problematic as ammonium was not completely consumed during syngas fermentation, so that considerable amounts of ammonium remained for the subsequent fungal fermentation medium. Therefore, to prove if syngas fermentation medium is in general suitable for malic acid production, fungal cultivations in shake flasks with normal concentration and without nitrogen were conducted. Syngas fermentation medium was mixed, autoclaved and enriched with acetic acid as carbon source. Normal ammonium concentration leads to a final malic acid concentration of 2.69 ± 0.81 g/L with a yield of 0.09 g/g. If no nitrogen source was added 4.11 ± 0.50 g/L and a yield of 0.37 g/g could be achieved (Table
However, in the sequential mixed culture fermentation, the syngas fermentation medium might possibly undergo unknown modifications as a result of
Because initial shake flask experiments were promising, the sequential mixed culture approach was tested under realistic conditions, i.e., syngas fermentation followed by fungal fermentation without medium removal and/or delay in between.
For the main experiment, NH4-reduced medium was used for syngas fermentation to ensure nitrogen limited conditions after 96 h. Syngas was delivered into the broth with a starting rate of 20 mL/min and was increased after 42 h to 25 mL/min. Figure
In the beginning of the experiment, fructose concentration and amount of carbon monoxide in the off-gas decreased constantly until fructose was not detectable anymore at 19.5 h. Biomass concentration continuously increased until 49 h and stayed at 0.3 g/L for the rest of the fermentation. Acetate and Ethanol concentrations increased to maximum mean values at the end of the syngas fermentation of 15.9 g/L and 2.0 g/L, respectively. Similar to the decrease of carbon monoxide in the off-gas, carbon dioxide increased up to a local maximum of 0.2 mmol/min after 17.3 h of cultivation, then dropped to an average of 0.07 mmol/min when hydrogen consumption started. Hydrogen off-gas values stayed as low as 0.005 mmol/min and slightly increased when the rate of ingoing syngas was increased. After about 47 h the hydrogen content in the off-gas started to increase. In contrast to hydrogen and carbon dioxide, carbon monoxide values in the off-gas stayed low until 83.0 h when they started to increase until the end of the fermentation. The decreases of gas flow rates at 71.0 h were due to reduction of the ingoing gas stream to 20 mL/min.
For better illustration, the consumed amount of substance (
The amount of consumed carbon monoxide increased continuously from inoculation and reached an average maximum of 0.84 mol/L. This equals to 79.6 % of the total CO that went into the bioreactor (dotted yellow line). The amount of consumed hydrogen started to increase considerably after 18.5 h and went up to 0.77 mol or 73.6 % of total hydrogen (dotted red line). Similar to the increase of carbon dioxide in the off-gas in Figure
Directly following the syngas fermentation the reactor was changed to fungal fermentation as stated above. Microbial biomass was not removed.
The sequential mixed culture was accomplished in three replicates in the described fermentation setup in a bioreactor. Results for malic acid fermentation are shown in Figure
In two of the three bioreactor runs, malic acid production was detected. In one bioreactor, acetic acid was partly metabolized, but no product was formed. In reactor
For the sequential mixed culture fermentation from syngas to malic acid, the main challenges were the requirements of the involved microorganisms in terms of reactor set-up, medium composition and product synthesis. Optimizing product yield and productivity for a certain process usually addresses the needs of the organism involved. Since sequential mixed culture fermentation uses at least two different organisms the key aspect for sequential mixed culture fermentation is either a medium compromise for both organisms or the compatibility of the first (optimized) medium for the second organism in terms of product synthesis. Furthermore, the second organism has to be able to use the product of the first process as a carbon source. The combination of both aspects must be fulfilled to achieve an optimal value added chain from syngas to malic acid.
For the first time, this study shows that
Cultivation of
Comparing the fungal fermentation on acetic acid in the two different media, it could be seen that a reduction of the nitrogen concentration is mandatory. If ammonium is omitted in syngas fermentation medium a similar yield was achieved for malic acid production medium (
During the first 18 h of syngas fermentation
Up to the time point when hydrogen consumption started to decrease, acetate was continuously produced. This point also marked the beginning of ethanol formation. In case the decrease of hydrogen consumption was due to the increase in the gas sparging rate and therefore the carbon monoxide supply, we reduced the flow rate of syngas to stabilize the uptake rates again. This brought a temporary improvement but hydrogen consumption rate decreased for the rest of the syngas fermentation. The observed occurrence of sudden decrease in hydrogen consumption is a known phenomenon when cultivating
For the main experiment, average acetate concentration after 96 h was about 1 g per liter lower and standard deviation was 0.5 g per liter higher compared to preliminary bioreactor experiments. The differences in acetate concentration between the three bioreactors might be due to different rates of decreasing hydrogen consumption. In addition, off-gas composition of the bioreactors for H2, CO, and CO2 showed increasing deviations after the reduction of the ingoing gas stream.
The malic acid production in the three bioreactors with
Although sequential mixed cultures have been used for centuries in food industry, e.g., sake production, applications for production of value added chemicals is rare. It could be shown, that this kind of biotechnological process is suitable for the production of low price chemicals like single cell oils for biofuel production (Hu et al.,
We could successfully show that production of high-value L-malate from syngas is possible. Further increase of yield is feasible as the process medium was neither optimized for acetic acid production nor for malic acid production and only wild type strains of
FO: Substantial performance of the experiments with
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This work by SD was supported by a grant from the Ministry of Science, Research, and the Arts of Baden-Württemberg Az.: 33-7533-10-5/75B as part of the BBW ForWerts Graduate Program. Work of FO was supported by a grand from the Ministry of Science, Research, and the Arts of Baden-Württemberg Az.: 33-7533-6-195/7/1 and Az.: 33-7533-6-195/7/9 as part of the BW2 Graduate Program. We want to thank the group of Prof. Clemens Posten, Section for Bioprocess Engineering of the Institute of Process Engineering in Life Sciences for conducting the ion chromatography measurements for determination of
The Supplementary Material for this article can be found online at: