Over-expressing NadA quinolinate synthase in Escherichia coli enhances the bioelectrochemistry in microbial fuel cells

The microbial fuel cell (MFC) is a bioelectrochemical reactor that generates electricity from organic substrates using electricity-producing microorganisms (Hampton et al., 2021; Grzegorz et al., 2019). It attracted a lot of attention because of its potential applications in dealing with the energy crisis and sewage treatment (Liu et al., 2020; Li et al., 2018; Uwe et al., 2003; Yousefi et al., 2021). However, low power generating efficiency is still a key bottleneck that needs to be addressed (Zhang et al., 2017; Bruce, 2009). It has been indicated that power generation can be improved by optimizing cell structures, the development of electrode materials, and most fundamentally enhancing the electricity-producing ability of microorganisms (Ren, 2017).

Screening of newly identified microorganisms, such as electrochemically active bacteria (EAB), is a widely used method for researchers (Iskander et al., 2017; Liu and Yin, 2017). However, genetic modification of EAB to enhance bioelectricity generation would be more feasible. Both the increased number of releasable electrons and the improved efficiency of transferring released electrons to the anodes have been described previously (Lital, 2010; Luo et al., 2018). Through random transposon insertion mutations in S. oneidensis MR-1 which promoted bioanabolism of cell-surface polysaccharides, Kouzuma et al. generated approximately 50% more current than using the wild-type strain (Kouzuma et al., 2010). Bretschger et al. made a series of deletion mutants in mtrA, mtrB, mtrC, gspD, gspG, etc., and studied consequent effects in MFCs (Bretschger et al., 2007). Berríos-Rivera et al. reported that over-expression of pncB elevated the total NAD levels (Berríos-Rivera et al., 2002a,b). These findings also suggest that enhanced metabolic pathway or ATP synthesis contribute to the electricity-producing efficiency of EAB.

Cofactors play crucial roles in regulating energy metabolism. Nicotinamide adenine dinucleotide (NAD⁺), with its reduced form NADH, are essential cofactors as electron carriers involved in cellular metabolic process and energy production (Minhas et al., 2019; Berríos-Rivera et al., 2002a,b). Accelerating anabolism of NAD by over-expressing the Nad quinolinate synthase (NadA) could enhance the bioelectrical properties of EAB (Belenky et al., 2007). NadA has recently been cloned and could regulate the NAD biosynthesis process indirectly by catalyzing the depletion of L-aspartic acid (Akira et al., 2006; Liang et al., 2019). As a result, we supposed that increased expression of nadA could induce an enlarged NAD (H/⁺) pool and consequently enhance the electricity production of EAB.

In this study, we constructed a plasmid vector bearing the wild-type nadA gene from Escherichia coli. Over-expression of nadA in E. coli BL21 (DE3) strongly increased the voltage in regular MFC devices with aspartic acid as the substrate. Our study provides evidence for an effective approach through genetic modification of EAB to enhance the electricity producing ability of MFCs.

The 1044-bp wild-type nadA gene was cloned from E. coli BL21 using polymerase chain reaction (PCR) and then we followed routine genetic techniques for construction of over-expressing plasmids. Sequencing analysis proved the consistency during the vector recombinant process (data now shown). Recombinant plasmid vectors bearing the wild-type nadA gene were transformed into E. coli BL21.

Gene expression was induced with 50 mg/ml IPTG at 37°C for about 8 h. Bacteria were collected, then the total mRNA and proteins were extracted, respectively. The mRNA level of nadA was detected through RT-PCR. A significant increase in nadA cDNA concentration was found compared with bacteria transformed with blank vectors (Fig. 1A). In addition, an increased nadA protein level was shown after gel electrophoresis (SDS-PAGE) following Coomassie Brilliant Blue staining (thick band indicated with an arrow in Fig. 1B).

Because over-expression of exogeneous proteins may cause a burden on the growth status, we detected the OD₆₀₀ of cultured medium with bacteria. Over-expressing NadA protein slightly suppressed the amplification of E. coli in Luria-Bertani (LB) medium within 14 h (Fig. 1C).

As described previously (Liao et al., 2019), nadA functions in a salvage pathway and catalyzed the synthesis of NAD. An increased NAD (H/⁺) pool may enhance the metabolism of aspartic acid in EAB. We next analyzed the aspartic acid concentrations in anodic chambers with the WST-8 method after 12 days incubation. As Fig. 1D shows, the normalized NADH content was clearly augmented (*P=0.03). The NAD⁺ and NAD (H/⁺) pool also increased but were not statistically significant (P-values were 0.1 and 0.05, respectively).

We hypothesized that over-expressing nadA could increase the NAD (H/⁺) pool. As a result, the electron carrier would be increased, which could facilitate energy metabolism. In order to evaluate the electricity producing efficiency, we set up MFC devices and cultured nadA over-expressed E. coli in the anodic chambers. After a 2-week incubation, freshly prepared medium was added, and the voltage outputs were collected for about 2 weeks until the peak values were obtained.

In E. coli, aspartic acid can be used as the substrate to synthesize NAD (H/⁺), and to generate energy through the tricarboxylic acid cycle (TCA). First, we used 1 g/l aspartic acid as the substrate following a previous report (Nishimura and Kisumi, 1984). Compared with the wild-type strain, which only produced limited voltage output (around 20 mV), E. coli with increased nadA levels significantly enhanced the electricity producing efficiency as indicated in Fig. 2. We plotted the voltage output curves and peak column diagram (Fig. 2A) during these days. There was a more than two times (52.46 mV) increase compared with the control group. Next, we tried a higher concentration of aspartic acid (5 g/l) to see whether the performance of the MFCs would be better. Indeed, we found a more than three times elevation of the peak voltage output (70.81 mV) (Fig. 2B).

To better analyze the overall performance of MFCs, we then recorded the electrochemical performance with an electrochemical workstation. As shown in Fig. 3A and B, the peak value of power density was 0.29 μW/cm² for NadA group, which was 20.83% higher than that of the control group (0.24 μW/cm²). In Fig. 3C, the voltage-current curves from two cycles of recording were overlaped, indicating a relatively stable electrochemical performance of the NadA group.

With the hypothesis that increased expression of the cofactor NAD⁺/NADH could enhanced the electricity producing ability of EAB, we over-expressed nadA gene in E. coli BL21 (DE3). We found an increased concentration of NAD⁺ and slightly increased NADH. After setting up the MFC devices, we recorded remarkably enhanced peak voltage output in parallel with over-expression of nadA. In addition, the power density was increased accordingly.

Multiple strategies were tested to improve the performance of MFCs. Elevating the EAB's metabolism was proved to be effective (Schröder, 2007; Förster et al., 2003). It strengthened coulombic efficiency of MFCs with energy efficiency of 77.6% based on electron transfer. Meanwhile, over-expression of nadA gene in E. coli may promote the stability of Fe-S cluster (Rousset et al., 2008; Chan et al., 2012) thus enhancing the electrochemical activity of genetically modified strain with a steady increase in NADH. Findings in this work were consistent with a previous report that increasing the concentration of cofactors in Pseudomonas aeruginosa, such as NAD⁺/NADH, could enhance the electricity production of MFCs (Yong et al., 2014). Our data further suggested the effectiveness of genetic modification of metabolic pathway in EAB to enhance the electricity producing performance.

Two-chamber MFC devices were assembled following reported papers (Liu et al., 2006; Arkatkar et al., 2021). Anodic and cathodic chambers were separated by proton exchange membrane (Dupont, Beijing, China). The volume of each chamber was 100 cm³. Carbon clothes (5 cm×5 cm) were utilized in both the anodic and cathodic chambers. All MFC devices were performed at room temperature.

Anode buffer: L-aspartic acid (1, 5 g/l), ammonium chloride (0.19 g/l), sodium chloride (0.5 g/l), calcium chloride dihydrate (0.03 g/l), magnesium sulfate heptahydrate (0.03 g/l), sodium hydrogen carbonate (1 g/l), potassium phosphate monobasic (5 g/l), potassium phosphate dibasic (3.86 g/l), trace elements (12.5 ml/l).

Cathode buffer: sodium chloride (0.5 g/l), calcium chloride dihydrate (0.03 g/l), magnesium sulfate heptahydrate (0.03 g/l), sodium hydrogen carbonate (1 g/L), potassium phosphate monobasic (5 g/l), potassium phosphate dibasic (3.86 g/l), sodium nitrate (0.3 g/l), trace elements (12.5 ml/l).

Trace elements: nitrilotriacetic acid (1.5 g/l), magnesium sulfate heptahydrate (6.14 g/l), manganous sulfate (0.5 g/l), sodium chloride (1 g/l), iron (II) sulfate heptahydrate (0.1 g/l), calcium chloride dihydrate (0.1 g/l), cobaltous chloride (0.1 g/l), zinc chloride (0.13 g/l), copper (II) sulfate pentahydrate (0.01 g/l), aluminium potassium sulfate dodecahydrate (0.01 g/l), boric acid (0.01 g/l), sodium molybdate (0.03 g/l).

The E. coli series Top10 and BL21 (DE3) were purchased from Tiangen Co., Ltd (Beijing, China). pET-28 (a+) plasmid vector was a lab stock. Primers (nadA Primer 1:5′-CGCGGATCCATGAGCGTAATGTTTGATCCAGACA-3′; nadA Primer 2: 5′-CCGTCTCGAGTCCACGTAGTGTAGCCGCAAAA-3′; cDNA upstream: 5′-ATGGCGCGCTTCGGTGCAAA-3′; cDNA downstream: 5′-CTCTGATGTGGCAATGTTTTCGCAGCAGC-3′) were preserved in −20°C. All strains and plasmids were preserved in −80°C. E. coli were cultured with routine procedures at 37°C in LB medium.

The influences on growth status of E. coli by over-expressing nadA was indicated by the cell concentrations, which were detected by measuring the OD₆₀₀ values with ultraviolet spectrophotometry. The strains were cultured oscillatory with 220 rpm at 37°C in LB medium for about 14 h.

NAD (H/⁺) content was examined following the instruction of the determination kit. Briefly, the NAD (H/⁺) and the protein samples were extracted from bacterial after 5 h. The protein contents were determined with BCA protein standard curve (BCA Protein Assay Kit, Solarbio, Beijing, China). The NAD (H/⁺) concentrations were detected with NAD⁺/NADH Assay Kit (Beyotime, Beijing, China) and then normalized by the protein concentrations.

E. coli strains were imported into MFCs. Voltage outputs were collected for about 12 days by an EM9636 voltage collector (Zhongtai, Beijing, China) at a particular resistance (1000 Ω) connected to MFC device in different conditions (1 g/l or 5 g/l aspartic acid) (Nishimura and Kisumi, 1984). The plotted voltage output curves and averaged output peaks were used to analyze the electricity producing efficiency.

Power density and voltage-current curve were measured following the published paper. The 7-day power density (5 g/l aspartic acid) was recorded by a CH1660E electrochemical workstation (Chenhua, Shanghai, China) with scanning parameters were set as init E (0.1 V), final E (0.01 V), scan rate (0.01 V/s), run time (30 s), sensitivity (1×10⁻⁴ A/V). Scanning parameters for voltage-current curve were init E (0.28 V), final E (0.05 V), scan rate (0.01 V/s), run time (60 s) and two cycles.

Data were averaged from three independent experiments.

Results were expressed as the mean±s.e.m. t-test with SPSS 19.0 software was used to perform the statistical analysis. *P<0.05, **P<0.01, and ***P<0.001.

We thank Dr. Lei Zhang and Dr. Zhao Yang (Beijing University of Chemical Technology) for suggestions on experiments and data analysis and the Molecular Diagnostic Technology Innovation Center of BUCT for supporting the key equipments.