Screening of electron shuttles to facilitate glycerol utilization by A. succinogenes NJ113
Firstly, glycerol utilization by A. succinogenes NJ113 was investigated in sealed bottles supplemented with three commonly used electron shuttles: neutral red, riboflavin and methylene blue. As shown in Figure 2, the highest dry cell weight (DCW) and glycerol consumption were achieved by supplemented with 0.1 mM NR. The DCW reached 0.89 g/L, and 3.01 g/L of glycerol was consumed with 2.17 g/L succinate accumulated after 72 h (Figure 2a). Other two electron shuttles also showed the ability to promote glycerol utilization of NJ113. However, the glycerol consumption and the succinate production with riboflavin were lower than those supplemented with NR (Figure 2b). Figure 2c showed that methylene blue was toxic to NJ113 even though concentration was below 0.1 mM.
Considering that 0.1 mM NR might be insufficient but NR higher than 0.1 mM was toxic to NJ113 (Figure. 2), anaerobic fermentation was further carried out in MFC to oxidize the reduced NR. Figure 3 showed the cyclic voltammetry curve in MFC system with or without NR. The oxidation peak (-0.51 V) and reduction peak (-0.63 V) were consistent with previous report that midpoint of redox potential of NR was -0.56 V (V.S. Ag/AgCl) [13]. This indicated that NR can be adopted as an electron shuttle in this glycerol fermentation medium with unknown electron acceptor, likely derived from the yeast extract. As a result, glycerol consumption was increased from 3.01 g/L to 4.94 g/L in MFC (Figure 4), which might be due to the supplement of sufficient oxidized NR. However, the glycerol consumption rate was still very low. Thus, fermentations were further carried out in MFC with a 0.2 V potential applied in anode electrode to investigate the effects of promoted anode oxidized reaction. As a result, the glycerol consumption was increased to 8.52 g/L. However, the yield of succinate was reduced from 0.64 g/g to 0.38 g/g with more accumulation of formate.
Mutation and selection of acid-resistant A. succinogenes
Glycerol utilization by A. succinogenes NJ113 can be promoted by accelerating electron transfer rate with anode potential, but the carbon flux was redirected to relative oxidized byproducts. Pervious research showed that A. succinogenes gained a better pH tolerance by increasing the content of unsaturated fatty acids. Thus, we assumed that this mutant may have improved ability of transmembrane transport of lipophilic neutral red. Here, we adopted ARTP strategy to select acid resistant mutants. Firstly, A. succinogenes NJ113 was treated by ARTP mutagenesis for 30 s according to the lethal curve (Figure 1S). After several rounds of ARTP treatment and acid stress test, five mutants of A. succinogenes NJ113 were obtained which could grow on the acidic agar plates (pH 5.8). To investigate the growth and metabolism performance, all five mutants were tested by anaerobic fermentation in sealed bottles at pH 5.8 with 10 g/L glucose. As shown in Table 1, four mutants (JF1311, JF1313, JF1315 and JF1319) had better performance of glucose utilization and succinate production compared to that of A. succinogenes NJ113. Among them, strain JF1315 produced the highest 5.93 g/L succinate with the yield of 0.55 g/g, which is 83.33% higher than that of parent strain (0.30 g/g) under similar condition. It indicated that A. succinogenes JF1315 had the best acid-resistant ability.
As speculated before, the acid tolerant JF1315 might have improved transmembrane transport of NR. Thus, we further conducted the fermentations in sealed bottles with A. succinogenes JF1315 using glycerol as sole carbon source. As summarized in Figure 5, the DCW and glycerol consumption of A. succinogenes JF1315 were 1.68 g/L and 6.07 g/L, increased by 88.8% and 101.6% compared with A. succinogenes NJ113, respectively. The enhanced ability of glycerol utilization indicated that the mutant JF1315 had improved bidirectional transportation of NR, although the rate of glycerol consumption was still relatively low and 3.93 g/L glycerol was remaining in the broth after 72 h.
Enhanced glycerol utilization and succinate synthesis by acid-resistant A. succinogenes in MFC
In order to gain better glycerol utilization and succinate synthesis, mutant JF1315 was further investigated in MFC, and fermentations with different initial glycerol concentrations were carried out. In MFC, glycerol utilization and succinate synthesis were both improved significantly (As shown in Figure 6). Under low concentration of glycerol (less than 10 g/L), the glycerol was depleted and 5.21 g/L of succinate (0.83 g/g glycerol) was produced with small amounts of by-products (Figure 6d, 6e). In addition, only 5.33 g/L glycerol was consumed when the external resistance was taken away. It indicated effective transfer of electrons in MFC played an important role in glycerol utilization. When the concentration of glycerol was up to 30 g/L, 23.92 g/L succinate accumulated with a yield of 0.88 g/g and the glycerol could be depleted. However, when the glycerol concentration was further increased to 60 g/L, the remaining glycerol was more than 20 g/L and the succinate yield was decreased to 0.57 g/g, which might be due to the remarkable accumulation of 15.28 g/L formate and 5.03 g/L acetate (Figure 6d, 6e).
Cell growth was also improved significantly in MFC system with <30 g/L initial glycerol (Figure 6c), and the DCW of JF1315 could achieve 2.11 g/L with increment of 93.58% and 25.60% compared with NJ113 in MFC (1.09 g/L) and JF1315 in sealed bottles (1.68 g/L), respectively. However, when initial concentration of glycerol was up to around 60 g/L, cell growth was inhibited at the late stage of anaerobic fermentation, during which high concentration of formate and acetate were accumulated.
MFC performance and power output
In order to evaluate the performance of MFC system, the polarization curve was made by varying the external resistance. With the increment of the external resistance, cell voltage kept increasing and finally reached 425.5 mV (Figure 7a), whereas the power output only increased with lower resistance and dropped sharply along with the increment of external resistance. A maximum power output of 348.6 mW was achieved with a current of 2.7 mA and a voltage of 128.0 mV at 47.0 Ω external resistance (Figure 7b).
As shown in Figure 8, after A. succinogenes NJ113 and JF1315 were inoculated in MFC, cell potential increased to the peak voltage after a short start-up time and then a sustainable power output was generated for at least 48 h with <30g/L glycerol. The peak values were 172.6 mV and 266.9 mV in A. succinogenes NJ113 and JF1315 with 10 g/L glycerol, respectively. For A. succinogenes JF1315, relative constant and high value above 300 mV was obtained for at least 48 h with 30 g/L glycerol. However, potential gradually dropped after the peak value of 607.6 mV when initial glycerol increased up to 60 g/L.
Compared with the long-term power output with 30 g/L glycerol, the stable phase of cell potential did not last long and dropped sharply after 30 h with A. succinogenes JF1315 in the presence of 60 g/L glycerol (Figure 8). It indicated that the cell activity of JF1315 decreased with high concentration of glycerol. In MFC fed with 60 g/L glycerol, more than 15 g/L formate was generated, which might be the main reason leading to the stagnation of cell growth and the decrease of power output.