Introduction

The utilization of syngas in biological processes is of interest, since they are often waste streams from large industries, such as the steel sector [1,2,3]. Usually, the syngas is composed of carbon monoxide (CO), carbon dioxide (CO2), and hydrogen (H2). As potential microbial catalysts for gas fermentation, several bacterial strains have been investigated in recent decades, such as Clostridium ljungdahlii, C. autoethanogenum, C. carboxidivorans P7, C. ragsdalei, Butyribacterium methylotrophicum, and Eubacterium limosum [3,4,5,6,7,8,9,10]. Both CO and H2 can be utilized as energy sources during their growth, and they can be used to produce several products such as C2−C6 alcohols from syngas [11,12,13].

For the effective utilization of gases, many studies have been performed to enhance microbial growth and product formation by changing medium components and concentrations [8, 9, 14,15,16,17,18]. Meanwhile, a few studies have been conducted to reduce the cost of the medium by changing the buffer solution [19], replacing yeast extract with corn steep liquor, cotton seed extract, trypticase, and biochar [20,21,22,23,24,25]. In addition, agents, such as methyl viologen [17], neutral red [26], viologen dyes [27], and sodium sulfide [28, 29], were successfully used to improve the concentration of end-products. Previously, Panneerselvam et al. [17] studied the effects of various reducing agents on syngas fermentation by C. ragsdalei, and found that benzyl viologen caused cell death, and no ethanol production was detected; 0.1 mM methyl viologen promoted ethanol production compared to the control containing no reducing agent, while the addition of 0.2 or 0.3 mM methyl viologen delayed the production of ethanol compared with 0.1 mM methyl viologen; neutral red neither promoted nor detracted from ethanol production when compared to the control [17]. Oliveira et al. [10] found that a continuous sodium sulfide feed can increase ethanol production more than threefold by C. ragsdalei. Chandgude et al. [30] found that cysteine (Cys), ascorbic acid, and dithiothreitol had a different mechanism of action than conventional reducing agents such as viologens and neutral red. For example, Cys itself affects the distribution among the biochemicals, and thus affects the fermentation product distribution [31]. However, most of the initially added sulfur from Cys was stripped out within the first day of the batch process (first half of the exponential growth phase of C. ragsdalei) [10]. Meanwhile, it was also reported that cysteine was not a limiting factor for cell growth of C. ljungdahlii since its supplementation did not have a noticeable impact on product formation (i.e., acetic acid and ethanol) or overall gas consumption [32]. Bizarrely, in the culture medium for the production of alcohols from the syngas using C. ljungdahlii, a large amount of Cys or Cys-HCl was added [12, 20]. Thus, it need to further evaluate the effect of Cys on biochemical production using C. ljungdahlii.

C. ljungdahlii is capable of assimilating gaseous carbon sources such as pure CO or CO2/H2 [3, 12], and sugars such as fructose and sucrose [33, 34] for the production of biochemicals (i.e., ethanol, acetate, and 2,3-butanediol). Among the products, 2, 3-butanediol is a chemical platform used in several applications, such as polymers, cosmetics, fuels, and medicines [35, 36]. Less than 10% of current studies have focused on nutritional supplementation to improve 2, 3-butanediol production, and the redox balance achieved by reducing agents can affect the activity of key enzymes in 2, 3-butanediol metabolism [37]. However, the effect of Cys on the production of the biochemicals using C. ljungdahlii is still unclear. The addition of reducing agents has improved end-product formation in anaerobic fermentation processes [17, 30], where most enzymes preferentially utilize NADH as their cofactor [30, 38,39,40]. The reducing agents used can affect the parameters of redox potential (Orp), ATP, and cofactors, such as NADH. In turn, the affected parameters influence the expression of certain genes and activities of specific enzymes [41]. Recently, it was observed that the genome-wide transcriptional profile of gas-fermenting C. ljungdahlii differed significantly in the presence of sugars and C1 gases [42]. Therefore, we speculate first that the production of the biochemicals (i.e., ethanol, acetate, and 2,3-butanediol) may be improved with the addition of Cys. Furthermore, we conducted an in-depth investigation of the effect of L-Cys on the production of the biochemicals.

Materials and Methods

Microorganism, Media, and Cultivation Conditions

C. ljungdahlii DSM 13528 was purchased from the DSMZ (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany) and conserved by freezing mid-exponential phase cultures at −80 °C with 20% glycerol for long-term storage. The modified DSMZ 879 medium with the following composition was used (per liter) [12, 33]: 1.0 g NH4Cl, 0.1 g KCl, 0.2 g MgSO4·7 H2O, 0.8 g NaCl, 0.02 g CaCl2·2 H2O, 0.1 g KH2PO4, 2.5 mg Na2WO4·2 H2O, 1.0 g NaHCO3, 1.0 g Cys-HCl·H2O, 1 g yeast extract, 5.0 g fructose, 0.5 g Cys, 0.5 mg resazurin, 10 mL trace element solution, and 10 mL vitamin solution. Trace element solution contains 2.0 g nitrilotriacetic acid, 1.3 g MnCl2·H2O, 0.4 g FeSO4·7 H2O, 0.2 g CoCl2·7 H2O, 0.2 g ZnSO4·7 H2O, 0.2 g Na2MoO4·2 H2O, 0.02 g NiCl2·6 H2O, and 0.1 g Na2SeO3·5 H2O in 1 L distilled water. Vitamin solution per liter involves 2 mg biotin, 2 mg folic acid, 10 mg pyridoxine-HCl, 25 mg thiamine-HCl·2 H2O, 5 mg riboflavin, 5 mg nicotinic acid, 5 mg D-Ca-pantothenate, 0.1 mg vitamin B12, 5 mg ρ-aminobenzoic acid, and 5 mg lipoic acid in 1 L distilled water. The modified DSMZ 879 medium was used in all the fermentation experiments, and the medium was assembled in anaerobic chamber. With a constant pressure of 0.8 bar, the headspace of gas mixture was H2: CO2, 60:40 which was the same as reported in the literatures [12]. Analytical grade chemicals used in the medium were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). The medium was assembled in anaerobic chamber (Ruskinn technology Ltd., Sony Technology center, Pencoed, Bridgend Mid Glamorgan, UK). After autoclaving, FeSO4, vitamins, Cys-HCl, and NaHCO3 were added using syringe with a 0.2-μm filter. In all fermentation experiments, the seed culture of the strain DSM 13528 was inoculated by the 7.5 mL freezing mid-exponential phase cultures. Then, to obtain the seed culture, a 250-mL screw-cap bottle with a 75 mL working volume of modified DSMZ 879 medium was cultured at 37°C for 2 days in a rotary shaker (HYG-A, Taicang Experimental Equipment Factory, China) at 150 rpm. Batch fermentations were performed in 250-mL screw-cap bottles under the addition of 4 g/L CaCO3 with a working volume of 75 mL. The gas in the headspace was substituted by the syngas as required with a pressure of 0.8 Bar. Then, for bioreactor culture, the seed culture broth (150 mL) was transferred to a 2.7 L bioreactor (BioFio®110, New Brunswick Scientific, San Francisco, USA) with a 1500 mL working volume. The temperature and stirring speed in the bioreactor were kept at 37°C and 200 rpm, respectively. Fermentation was carried out under the completely closed exhaust pipe case (i.e., no syngas was escaped from the bioreactor), and the syngas in the headspace of the bioreactor was kept at 0.8 Bar with the syngas that entered the bioreactor through a microflowmeter. Meanwhile, no matter the experiments were carried out in bioreactor or screw cap flasks, the syngas in the headspace was replaced every 1 day.

Analytical Methods

A total of 5 mL samples were withdrawn from the culture for cell density monitoring and products analysis. The concentrations of ethanol, acetic acid, and 2,3-butanediol were measured by a HPLC apparatus (LC-20AT, Shimadzu, Kyoto, Japan) equipped with an Aminex HPX-87H ion exclusion column and refractive index detector. The process was performed at a temperature of 50 °C, and a flow rate of 0.6 mL/min with 5 mmol/L H2SO4 as the moving phase. The growth of C. ljungdahlii was monitored by using a UH5300 spectrophotometer (Hitachi high-tech science corporation, Tokyo, Japan) to measure the optical densities at 600 nm. In the screw-cap bottle conditions, the samples were analyzed after culturing for 7 days, and the excess CaCO3 was removed by the addition of 1 M HCl before the measurement of biomass. In the bioreactor conditions, the samples were analyzed every 12 h. According to the genome analysis [33], genes involved in the ethanol production (adhE1 and aor1), acetic acid production (ack), and carbon fixation (metF, fold and fdh) were analyzed by Majorbio (Beijing, China). Analysis of genome-wide differential message RNA (mRNA) expression provides us with greater insights into biological pathways and molecular mechanisms that regulate cell fate and development. Cell pellets from cultures in the bioreactor were collected by centrifugation at 10,000×g under 4 °C for 10 min at 72 h, frozen in liquid nitrogen immediately, and stored at −80 °C. The mRNA isolation and high-throughput mRNA sequencing (RNA-Seq) were performed by Majorbio (Beijing, China). Total RNA was extracted using the TruSeqTM Stranded Total RNA Library Prep Kit (Ambion, Santa Clara, CA, USA) following the manufacturer’s protocol. RNA integrity was evaluated using the Agilent 2100 Bio-analyzer (Agilent Technologies, Santa Clara, CA, USA). The samples with RNA Integrity Number (RIN) ≥ 7 were subjected to subsequent analysis. The libraries were constructed using the TruSeq Stranded mRNA LTSample Prep Kit (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions. Then, these libraries were sequenced on the Illumina sequencing platform (HiSeqTM 2500) and 150-bp/125-bp paired-end reads were generated. Based on reads per kilobase of transcript per million mapped reads (RPKM) normalization, the gene expression profiles were analyzed. The differential genes were analyzed using Bioconductor edgeR (V3.4.6); information was from Clusters of Orthologous Groups (COG, https://www.ncbi.nlm.nih.gov/research/cog/api/cog/accessed on 5 January 2022) and Kyoto Encyclopedia of Genes and Genomes (KEGG, http://www.genome.jp/kegg/ accessed on 5 January 2022). Meanwhile, KEGG annotation results were derived from the KAAS (KEGG Automatic Annotation Server). Statistical analysis of the different experimental groups was conducted by subjecting the experimental data to one-way analysis of variance (ANOVA) using OriginPro 2018 software (Origin Lab Corporation, Northampton, MA, USA) at a 95% confidence level. The data presented in the figures are the average values with error bars.

Results and Discussion

The Effect of Cys and Sodium Sulfide on the Production of Biochemicals

In the modified DSMZ 879 medium, 1.0 g/L Cys-HCl×H2O (i.e., 0.69 g/L Cys) and 0.5 g/L Cys were added [12, 33]. However, the effect of Cys on the production of biochemicals (i.e., 2,3-butanediol, ethanol, and acetic acid) is still unclear. Thus, to evaluate the effect of Cys on the production of the biochemicals, different concentrations of Cys were first supplemented at the beginning of fermentation in screw cap flasks. It should be indicated here that the concentration of Cys used in the modified DSMZ 879 medium was the sum of Cys from Cys-HCl×H2O and Cys where the ratio of Cys-HCl×H2O to Cys was 2:1 (w/w). As shown in Fig. 1a, without the addition of Cys, the final 2,3-butanediol production was 0.69 g/L with a yield of YP/X (i.e., product (g)/dry cell weight (g)) =258.4. When the Cys concentration was below 0.24 g/L, 2,3-butanediol production showed little difference; when Cys concentration was beyond 0.24 g/L, 2,3-butanediol production increased with the enhanced concentration of Cys, and the maximal 2,3-butanediol concentration (1.5 g/L) with a yield of YP/X=694.4 was obtained under the addition of 1.7 g/L Cys which was increased 2.17 times compared to the no-addition of Cys case. For acetic acid production, when Cys concentration was below 0.24 g/L, acetic acid concentration decreased sharply from 2.82 to 0.64 g/L with the enhanced concentration of Cys; when Cys concentration was beyond 0.24 g/L, acetic acid production increased and the maximal acetic acid only reached 1.29 g/L under the addition of 1.7 g/L Cys, which was decreased by 54.3% compared to the no-addition of Cys case. For ethanol production, when Cys concentration was below 0.24 g/L, ethanol concentration increased sharply from 0.051 to 0.57 g/L with the enhanced concentration of Cys; when Cys concentration was between 0.24 and 1.2 g/L, ethanol production decreased; when Cys concentration was beyond 1.2 g/L, ethanol production remained almost constant. For cell growth, when the Cys concentration was below 0.24 g/L, cell growth increased with the enhanced concentration of Cys; then, after the Cys concentration was above 0.24 g/L, the cell concentration decreased. These results suggested that the Cys showed significant effects on the production of biochemicals from the syngas H2/CO2. Similarly, it was reported that Cys itself affected the distribution among the biochemicals [31].

Fig. 1
figure 1

The effect of cysteine (a) and sodium sulfide (b) on the production of biochemicals. The fermentations were carried out in 250-mL screw-cap bottles under 40% CO2 plus 60% H2. In (a), different concentrations of Cys were first supplemented at the beginning of fermentation. In (b), without the addition of Cys, different concentrations of Na2S were first supplemented at the beginning of fermentation. Data are given as mean ± SD, n = 3

Full size image

In addition, it was assumed that Cys was used as a reducing agent, and the production of the biochemicals responded to the different reducing values. To verify this assumption, the effect of another reducing agent, sodium sulfide (Na2S) on the production of 2,3-butanediol was investigated. As shown in Fig. 1b, when the Na2S concentration was below 0.4 g/L, 2,3-butanediol production increased slightly; when the Na2S concentration was beyond 0.4 g/L, 2,3-butanediol production decreased with the enhanced concentration of Na2S, and the maximal 2,3-butanediol concentration (0.89 g/L) was obtained under the addition of 0.4 g/L Na2S. For the production of acetic acid, when the Na2S concentration was below 0.4 g/L, acetic acid concentration decreased sharply from 2.82 g/L to 0.51 g/L with the enhanced concentration of Na2S; when the Na2S concentration was beyond 0.4 g/L, acetic acid production increased and the maximal acetic acid only reached 0.91 g/L under the addition of 1.6 g/L Na2S. For ethanol production, when the Na2S concentration was below 0.8 g/L, the ethanol concentration increased sharply from 0.051 to 0.94 g/L with the enhanced concentration of Cys; when the Na2S concentration was beyond 0.8 g/L, ethanol production decreased. In addition, it should indicated that the excess CaCO3 was removed by the addition of 1 M HCl before the measurement of biomass in screw-cap bottle culture, which resulted that the values of OD in Fig. 1b cannot really reflect the cell growth, since some black precipitated materials, such as FeS, were formed during the acidization. These results also suggested that Na2S had significant effects on the production of the biochemicals from H2/CO2 syngas. In detail, the reducing power or element S can slightly promote the production of 2,3-butanediol and ethanol, while it decreased the production of acetic acid. Although Cys and Na2S could both be used as reducing agents, their effects on the production of the biochemicals were significantly different, especially Cys was better than Na2S for the production of 2,3-butanediol. Thus, besides Cys was used as a reducing agent, it may be used as a metabolic regulator or an extra carbon source. If Cys was used as an extra carbon source, the amino acids having the equal concentration of C may be used to replace the Cys. If the Cys was used as a S donor, the amino acids having the element S may be used to replace the Cys. If Cys was used as a metabolic regulator, the cysteine may be used to replace the Cys. However, we still cannot clearly identify these factors. Therefore, in the following sections, the effect of Cys on the production of the biochemicals will be further studied.

The Effect of the Elements in Cys on the Production of Biochemicals

To further verify the elements in Cys on the production of biochemicals, the effect of different amino acids on the production of biochemicals was investigated (Fig. 2a). The amino acids used included two categories: sulfur-containing amino acids (i.e., methionine, cystine, and Cys) and sulfur-free amino acids (i.e., glycine, phenylalanine and arginine). There was 29.46 mmol/L C from the amino acids used, which was the same concentration of C from 1.0 g/L Cys-HCl×H2O and 0.5 g/L Cys in the modified DSMZ 879 medium. It should be indicated here that the Cys-HCl×H2O and Cys were removed from the DSMZ 879 medium in this section (Fig. 2a). Under equal C and S from Cys and cystine (Cys-Cys ), the concentration of 2,3-butanediol was 1.54 g/L and 1.51 g/L with a YP/X of 523.8 and 474.8, respectively. Meanwhile, the 2,3-butanediol production was higher using Cys than that using other amino acids. However, compared to those using Cys-Cys, the concentration of acetic acid (i.e., 0.289 g/L with a YP/X of 98.3) using Cys was decreased, while the concentration of ethanol (i.e., 0.587 g/L with a YP/X of 199.7) was enhanced. In addition, under the same C and the lower S in methionine (Met) compared to Cys or Cys-Cys, the concentration of 2,3-butanediol was 0.41 g/L with a YP/X of 160.8 which was significantly decreased, while the productions of acetic acid and ethanol were enhanced. Moreover, compared to the addition of Cys, 2,3-butanediol production was significantly decreased, and the productions of acetic acid and ethanol were enhanced under the addition of sulfur-free amino acids (i.e., glycine, phenylalanine, and arginine). Furthermore, compared to methionine (Met) with less S, the productions of 2,3-butanediol, acetic acid, and ethanol showed no similar trends to those under Cys. For the cell growth, the concentrations of cell were a little higher under the addition of sulfur-free amino acids compared to those under the addition of sulfur-containing amino acids. In detail, the concentrations of cell were similar among the sulfur-containing amino acids, and between Cys-cys and Cys, while the Met gave the lowest concentration of cell. Thus, we can conclude that the different amino acids showed different effects on the production of biochemicals, and the Cys-Cys or Cys was appropriate to high 2,3-butanediol production; besides the reducing power introduced by Cys, the element S also participated in regulating metabolic processes for the production of biochemicals, which was similar to the results in Fig. 1a.

Fig. 2
figure 2

The effect of the elements in cysteine on the biochemicals production. In (a), different amino acids were added without extra reducing agents. In (b), different amino acids were added under the equal Na2S of 1.6 g/L. There was equal 29.46 mmol/L C of the different amino acids in (a) and (b). The fermentations were carried out in 250-mL screw-cap bottles under 40% CO2 plus 60% H2. Data are given as mean ± SD, n = 3

Full size image

To further verify the element S in Cys on the production of the biochemicals, different amino acids were added in the initial fermentation process under the addition of 1.6 g/L Na2S. As showed in Fig. 2b, the 2,3-butanediol production was higher using Cys-Cys or Cys than that using other amino acids, which was similar to that in Fig. 2a. Meanwhile, the 2,3-butanediol concentration was slightly higher using Cys-Cys than using Cys, which was different from that in Fig. 2a. However, for the addition of sulfur-free amino acids (i.e., glycine, phenylalanine, and arginine), the 2,3-butanediol concentration was not enhanced, as predicted, compared to those in Fig. 2a, although the S and reducing power were both introduced in the broth. This phenomenon was consistent with that in Fig. 1, which suggested that the element S from Na2S could not be used to replace the element S from Cys for high 2,3-butanediol production, and the element S from Cys was more suitable for the production of 2,3-butanediol. Compared with the concentrations of ethanol in Fig. 2a, ethanol production decreased regardless of the addition of sulfur-free amino acids or sulfur-containing amino acids when Na2S was introduced in the broth (Fig. 2b). However, acetic acid production showed the reverse trend to ethanol production under the same conditions (Fig. 2b). These results indicated that the metabolic process for the production of the biochemicals could be regulated by the element S regardless of the reducing power which was similar to those reported by Oliveira et al. [10]. Meanwhile, these results were different from those reported in Hu et al. [28], who found that an increased concentration of sulfide would lead to a more negative redox potential and consequently resulted in improved ethanol production. In view of the results in Figs. 1 and 2, it can be concluded that the Cys itself served as a metabolic regulator in the production of the biochemicals, which offered the element S, reducing power and extra C/H donor. In addition, the extra H in Cys may serve as an extra energy source in the production of the biochemicals since H2 can be utilized as an energy source during the autotrophic growth of C. ljungdahlii [12]. Thus, the gas composition under the addition of Cys may affect the production of biochemicals, which will be evaluated in the following section.

The Effect of Gas Composition on the Production of Biochemicals

To investigate the gas composition on the production of the biochemicals, batch cultures were further performed in a 2.7-L bioreactor with a working volume of 1.5 L under the addition of 1.2 g/L Cys. When the syngas only contained CO2 (Fig. 3a), the maximal concentrations of 2,3-butanediol, ethanol, and acetic acid were 3.16, 0.65, and 0.42 g/L with a Yp/x of 3427.0, 705.0, and 455.5, respectively. The cell grew to its maximal concentration at 60 h with an OD600 of 1.21, and then it began to decrease until the end of fermentation at 168 h with an OD600 of 0.66. When the syngas contained 40% CO2 and 60% H2 (Fig. 3b), the maximal concentrations of 2,3-butanediol, ethanol, and acetic acid were 2.57, 0.61, and 0.49 g/L with a Yp/x of 1404.3, 333.3, and 267.8, respectively. The cells grew to their maximal concentration at 60 h with an OD600 of 1.07, and then it remained approximately constant until the end of fermentation at 168 h. Compared with those in Fig. 3a, although the maximal concentrations of ethanol and acetic acid showed little difference between the two cases, the final concentrations of cell and acetic acid in Fig. 3b were increased by 53.1% and 45.5%, respectively. Compared with those in Fig. 3a, although the maximal concentrations of cells in Fig. 3b was decreased by 11.6%, the finial concentration of 2,3-butanediol was similar, while the final concentration of ethanol was decreased by 41.3%. Moreover, the variation trends in the whole fermentation process for the production of 2,3-butanediol, ethanol, and acetic acid were the same, that is, the productions of 2,3-butanediol and ethanol decreased until the end of fermentation after reaching their maximal concentrations, while the production of acetic acid increased until the end of fermentation. Compared with the results in Fig. 3a, it can be concluded that the introduced H2 in the syngas mainly decreased 2,3-butanediol production. In addition, it also indicated that the 2,3-butanediol was only produced during growth on fructose and not during growth on gaseous carbon sources. However, the concentrations of C and H were different between Fig. 3a and b. Thus, it needed to further study the effect of Cys on the production of 2,3-butanediol.

Fig. 3
figure 3

The effect of gas composition on the biochemicals production. The time courses refer to 100% CO2 (a) and 40% CO2 plus 60% H2 (b). The fermentations were carried out in a 2.7-L bioreactor. Data are given as mean ± SD, n = 2

Full size image

Moreover, without the addition of Cys and sodium sulfide (Fig. 4a), the maximal productions of 2,3-butanediol, ethanol, and acetic acid were 0.42, 0.72, and 1.17 g/L with a Yp/x of 276.3, 473.63, and 769.7, respectively. With the addition of sodium sulfide (Fig. 4b), the maximal productions of 2,3-butanediol, ethanol, and acetic acid were 0.36, 0.86, and 1.29 g/L with a Yp/x of 112.1, 267.9, and 401.9, respectively. With the addition of Cys (Fig. 4c), the maximal productions of 2,3-butanediol, ethanol, and acetic acid were 1.56, 0.64, and 0.81 g/L with a Yp/x of 559.1, 229.3, and 290.3, respectively. These results indicated again that the addition of Cys could enhance the production of 2,3-butanediol, which was similar to the results in Figs. 1 and 2. Moreover, the exogenous Cys slightly decreased the acetic acid production (Fig. 4b), which were similar to the results in Fig. 1a. However, the exogenous S or reducing power from the addition of Na2S slightly increased the acetic acid production (Fig. 4c), which were different to the results in Fig. 1b. Under the addition of Cys, the above results in the bioreactor (Figs. 3 and 4) showed that the concentration of 2,3-butanediol and ethanol decreased after reaching their maximum concentration. However, as showed in Fig. 4, when the 2,3-butanediol concentration was higher, the produced 2,3-butanediol was much more easily assimilated by the strain DSM 13528 after the fructose was depleted. Thus, it was speculated that the three products themselves can be mutual transformation, which should be studied. Firstly, different concentrations of exogenous acetic acid were added in the initial fermentation at the same concentration of Cys (Fig. 5a) since acetate formation can produce ATP [12]. It was found that 2,3-butanediol production increased with the increasing acetic acid concentration, while when the acetic acid concentration was beyond 160 mM, 2,3-butanediol production decreased. However, ethanol production showed the reverse trend to 2,3-butanediol production. In the whole fermentation process, a small amount of the added exogenous acetic acid was assimilated by strain DSM 13528. Thus, 2,3-butanediol can be synthesized from the added acetic acid (Fig. 4), which may be used as an additional carbon source or to produce ATP [12]. Secondly, the effect of exogenous ATP on the production of biochemicals was further evaluated (Fig. 5b). It was found that 2,3-butanediol production increased with increasing ATP concentration below 0.3 g/L ATP, while acetic acid production showed the reverse trend. When the ATP concentration was beyond 0.3 g/L, 2,3-butanediol production was not further increased, and the acetic acid concentration was also not further decreased. In addition, when 0.3 g/L ATP was supplied, the intracellular concentration of ATP showed no much difference compared to the case without the addition of ATP. Thus, a higher concentration of ATP can inhibit the acetic acid production (Fig. 5b), which will increase the 2,3-butanediol production at the same concentration of carbon source (i.e., CO2). Moreover, 2,3-butanediol formation requires NADPH as a cofactor during gas fermentation of C. ljungdahlii [12, 33]. Thus, it indicated that the reducing power and S donator were only two factors of Cys on 2,3-butanediol production, which were also verified by the results in the bioreactor (Figs. 3 and 4). Moreover, the genes of C. ljungdahlii response to the addition of Cys should be further evaluated.

Fig. 4
figure 4

The effect of cysteine and sodium sulfide on the biochemicals production in a 2.7-L bioreactor. The time courses refer to a without the addition of cysteine and sodium sulfide (i.e., Fig. 4a = control), b with the addition of 1.2 g/L cysteine (i.e., Fig. 4b =cys), and c with the addition of 1.6 g/L Na2S (i.e., Fig. 4c = Na2S). The fermentations were carried out in a 2.7 L bioreactor under 40% CO2 plus 60% H2. Data are given as mean ± SD, n = 2

Full size image
Fig. 5
figure 5

The effect of acetic acid (a), ATP (b), and the Orp (c) on the biochemicals production. Batch fermentations for (c) were the same as those in Fig. 4a and b. In (a) or (b), the fermentations were carried out in 250-mL screw-cap bottles under 40% CO2 plus 60% H2. Data are given as mean ± SD, n = 3 in (a) or (b), and n=2 in (c)

Full size image

Unraveling the Effect of Cys on Production of Biochemicals

A high concentration of 2,3-butanediol can be produced under the addition of Cys (Fig. 4b) compared with the case without the addition of Cys (Fig. 4a). Then, the Orp between the two cases was further investigated (Fig. 5c). It was found that Orp was higher under the addition of Cys, since the biosynthesis process of 2,3-butanediol can consume NADPH. Thus, we proposed that the reducing power introduced in the broth by the added Cys was not the reason for the high production of 2,3-butanediol. To verify this hypothesis, sodium sulfide was used to replace some Cys (Fig. 6). It was found that 2,3-butanediol production decreased when Cys was replaced by sodium sulfide. At the same concentration of Cys, the 2,3-butanediol production decreased when the concentration of sodium sulfide was increased. Previously, the addition of Cys into the hydrogen production system accelerated the formation of ethanol-type fermentation and enhanced the hydrogen production by creating an optimal low Orp environment, and by increasing the biomass growth simultaneously [43]. The plausible provision of more electrons into the culture in the presence of Cys-HCl enhanced ethanol production (48%) and the ethanol to acetate production ratio (24%) compared to the cells cultivated in standard growth medium recommended by ATCC [44]. When sodium sulfide was added to the broth during syngas fermentation using Clostridial bacteria denoted as P11, the increased concentration of sulfide led to a more negative redox potential and consequently resulted in improved ethanol production [28]. However, 2,3-butanediol production showed different responses to Cys and sodium sulfide, and Cys could not be replaced by some sodium sulfide (Fig. 6). Thus, the reducing power introduced in the broth by the added Cys was not the sole reason for the high production of 2,3-butanediol, and we proposed that the Cys shifted the metabolic pathways toward the production of 2,3-butanediol was another main reason.

Fig. 6
figure 6

The effect of cysteine plus sodium sulfide on the biochemical production. The M0 means that 1.2 g/L cysteine (i.e., 1.0 g/L cysteine-HCl ×H2O and 0.5 g/L cysteine) was added in the fermentation medium. The M1, M2, and M3 means that 0.75, 0.5, and 0.25 g/L cysteine were added in the fermentation medium under the addition of 1.6 g/L sodium sulfide, respectively. The M4, M5, and M6 means that 0.75, 0.5, and 0.25 g/L cysteine were added in the fermentation medium under the addition of 0.4 g/L sodium sulfide, respectively. The fermentations were carried out in 250-mL screw-cap bottles under 40% CO2 plus 60% H2. Data are given as mean ± SD, n=3

Full size image

Furthermore, according to the genome analysis [33, 45], genes involved in the ethanol production (adhE1 and aor1), acetic acid production (ack), carbon fixation (metF, fold, and fdh), and acetoin reductase were further analyzed at 72 h (Table 1), since the fermentation processes were more similar between the control (i.e., no-addition of Cys and sodium sulfide) and the case with the addition of Cys at 72 h (Fig. 4). The gene expression of fdh that is involved in the methyl sub-unit synthesis from CO2 had significant alternation during the fermentation of H2/CO2 after the fructose was depleted. As major enzymes involved in the same gene cluster for the acetate synthesis pathway, ack was only slightly down-regulated. However, the ethanol production-related gene dhE1 were significantly upregulated, and the acetoin reductase-related gene showed no much difference. Therefore, the irreplaceability of Cys on the production of 2,3-butanediol was both caused by its utilization as a reducing agent and its effect on the metabolic pathway flux.

Table 1 Gene expression profiles during the fermentation with H2 and CO2
Full size table

Conclusion

In this work, the element S and reducing power all could significantly affect the production of the biochemicals, and Cys was better than sodium sulfide for the production of 2,3-butanediol. Compared to the no-addition of Cys case, the 2,3-butanediol production was enhanced 2.17 times under the addition of 1.7 g/L Cys. Meanwhile, the gene expression profiles indicated that the fdh and dhE1 were significantly upregulated under the addition of Cys case, which involved in pathways of the CO2 fixation and ethanol production. Therefore, the irreplaceability of Cys on the production of biochemicals was both caused by its utilization as a reducing agent and its effect on the metabolic pathway.