Due to its un-conductivity, the polysaccharide in S. oneidensis MR-1 biofilm matrix reduced the extracellular electron transfer (EET) in S. oneidensis MR-1 biofilm. Thus, we deleted polysaccharide biosynthesis-associated gene ptpA from S. oneidensis MR-1 genome. Then, the conductivity of ptpA gene knockout strain ΔptpA and wild-type (WT) biofilms grown on anodes in microbial fuel cells (MFCs) were investigated. The ΔptpA produced a significantly higher maximum voltage output (153.39 ± 10.27 mV) than the WT biofilm whose peak value was 116.85 ± 8.75 mV (Figure 1A). The performance of MFCs with both strains was further evaluated through the measurement of the polarization curves and the power density curves. As shown in Figure 2B, the slopes of the polarization curve of the MFCs with ΔptpA was much smaller than that with the WT strain (Figure 2B); the maximum power density (MPD) was also improved from 45.96 mW m-2 (WT) to 74.85 mW m-2 (ΔptpA) (Figure 2C). Collectively, the results suggest that the deletion of ΔptpA enhances EET in electrode-attached S. oneidensis biofilms.
Figure 1. (A) Voltage output curves, (B) polarization and (C) power density curves of MFCs with the S. oneidensis WT and ΔptpA strain.
To test if this conductive biofilm can be used as sensitive elements to improve the sensitivity of electroactive biofilm (EAB)-based sensor, we comparatively analyzed concentration-response curves of MFCs inoculated with ΔptpA and WT via As3+ shocks. EAB-based sensor senses analysts by converting metabolic changes to easily detectable electrical signals. Thus, we firstly tested the inhibitory effect of arsenic toxicity on bacterial growth. WT and ΔptpA were cultured in arsenic-containing medium (sodium arsenite). It was found that a concentration of 100 mg/L of arsenic significantly inhibited the growth of S. oneidensis MR-1 (Figure 2), so this concentration was subsequently diluted for toxicity sensor testing in MFC systems.
Figure 2. The growth curves of S. oneidensis WT and ΔptpA strain in the medium containing (A) 50 mg/L, (B) 100 mg/L and (C) 500 mg/L As3+.
After MFCs generated stable voltages, As3+ with different concentrations ranging from 0 to 100 mg L-1 was fed into the systems. The voltage outputs of MFCs after As3+ shocks were shown in Figure 3. As3+ shocks contributed to the sharply decreased voltages for both ΔptpA and WT MFCs in 15 mins when As3+ concentrations were above 50 mg/L (Figure 3A and B), whereas only ΔptpA MFCs showed a significant decline in voltage outputs when As3+ concentration was lowed to 25 mg/L (Figure 3C) . These indicate that the EAB-based sensors with ΔptpA biofilms can detect lower concentrations of As3+ than ones with WT biofilms. However, the presence of other heavy metals, such as Cd2+ (10 mg/L) tested in our project, also resulted in a decline in voltage (Figure 3D). It is indictive of un-specificity of our biosensor.
Figure 3. The decreased voltages of MFCs with S. oneidensis WT and ΔptpA biofilms under exposure to As3+ with the concentrations including 100 (A), 50 (B), 25 (C) mg L-1 and Cd2+ (D) with the concentration of 10 mg L-1.
To construct an EAB-based biosensor for specific detection of As3+, we introduced an arsenic responsive transcription system composed of the transcription regulator arsR and its corresponding promoter Pars. To test whether the arsenic responsive transcription system can work in S. oneidensis MR-1, we constructed Part 1 (Figure 4A) in which the green fluorescent protein (gfp) as a reporter gene was placed at the downstream of the arsenic responsive transcription system. The MR-1/Part 1, in which S. oneidensis MR-1 WT contains the gene circuit Part 1, produced fluorescence signals in the presence of As3+ (Figure 4B and C). The fluorescence increased with an increase in As3+ concentrations (Figure 4B), whereas the increased fluorescence was not observed when Cd2+ was present in the system (Figure 4C). Collectively, the results indicate the feasibility of using arsR- Pars system to construct a S. oneidensis MR-1 biosensor for specific detection of As3+.
Figure 4. (A) Gene circuit contains an arsenic responsive transcription arsR- Pars system and a reporter gene gfp. (B) Dose-response fluorescence curve of an arsenic responsive transcription system to different concentrations of As3+. (C) The response of arsenic responsive transcription system to As3+ and Cd2+. (NS: No significance. * 0.01< P < 0.05.** P < 0.01)
Previous studies showed that the deletion of MtrC or CymA, which are important cytochromes for extracellular electron transfer (EET) pathway in S. oneidensis MR-1, significantly disrupted the ability of S. oneidensis MR-1 to reduce Fe(Ⅲ) or anode in MFCs. To convert the presence of As3+ to the electrical signals, we used the arsenic responsive transcription system to control the expression of MtrC or CymA. As shown in Figure 5A, we constructed gene circuits Part 3 and Part 4. Then, Part 3 and Part 4 were transformed into ΔcymA/ptpA (double deletion strain) and ΔmtrC/ptpA (mtrC and ptpA double deletion strain) , respectively. The resulting strains were named as ΔcymAptpA /CymA and ΔmtrCptpA/MtrC. In these two strains, the presence of arsenic can induce the expression of MtrC or CymA to restore the EET of S. oneidensis MR-1, which couples arsenic concentration in the environment with the ability of S. oneidensis MR-1 for Fe(Ⅲ) or anode reduction. We conducted the Fe(Ⅲ) reduction experiment and found that the Fe(Ⅲ) reduction rates of ΔmtrCptpA/MtrC and ΔcymAptpA /CymA exhibited faster trend with the increase in As3+ concentration (Figure 5B and C). Moreover, when MFCs reached a maximum voltages, we added As3+ with the concentration of 100 μM and observed an increased in voltages of ΔmtrCptpA/MtrC and ΔcymAptpA /CymA MFCs. All results suggested that As3+ induced the expression of mtrC and cymA to recover EET of ΔmtrCptpA/MtrC and ΔcymAptpA /CymA, respectively.
Figure 5. (A) The schematic illustration of Part 3 and Part 4 in which the arsenic responsive transcription arsR - Pars system controls the expression of EET-associated gene cymA and mtrC, respectively. The Fe(Ⅲ) reduction of ΔmtrCptpA/MtrC (B) and ΔcymAptpA /CymA (C) incubated with different concentrations of As3+. The voltage of ΔmtrCptpA/MtrC (D) and ΔcymAptpA /CymA (E) MFCs fed 50 and 100 μM As3+ at 25 h.
To further improve the sensitivity of our biosensor, we introduced a positive feedback loop using the LuxR autoregulatory elements to arsenic EAB-based biosensor. As Figure 6A shows, the transcriptional activator, a variant of LuxR, was used to replace mtrC or cymA in Part 3 and Part 4, and it was regulated by the arsR-Pars circuit, while mtrC or cymA together with a second luxR was placed under the promoter PluxI, which was activated by LuxR. When arsenic is present, it activates the expression of the first LuxR, which turns on the expression of mtrC (or cymA) and the second LuxR. The second LuxR activates its own expression as well as that of mtrC (or cymA) and forms a positive feedback loop to enhance the output signal. The gene circuits were named as Part 8 and Part 9. Then, the Part 8 and Part 9 were transformed into ΔmtrCptpA and ΔcymAptpA, resulting in ΔmtrCptpA/MtrC(+) and ΔcymAptpA/CymA(+), respectively. We conducted the Fe(Ⅲ) reduction experiments and found that the Fe(Ⅲ) reduction rates of ΔmtrCptpA/MtrC(+) and ΔcymAptpA/CymA(+) were enhanced with an increase in As3+ concentrations (Figure B and C). It indicated that As3+ can induce the expression of cymA and mtrC. Comparation of the Fe(Ⅲ) reduction rates between ΔmtrCptpA/MtrC without the amplifier system (+) and ΔmtrCptpA/MtrC(+) with the amplifier system showed that Fe(Ⅲ) reduction rates of ΔmtrCptpA/MtrC(+) were significantly faster than those of ΔmtrCptpA/MtrC when 10 and 100 μM As3+ were added in the system (Figure 6D). Similar result was observed for ΔcymAptpA/CymA and ΔcymAptpA/CymA(+) strains (Figure 6E). All results suggest that the response of the positive feedback amplifier biosensor to As(Ⅲ) was faster, and the expression of MtrC and CymA were higher than those of the ones without positive feedback.
Figure 6. (A) The schematic illustration of Part 8 and Part 9 that are As(Ⅲ) biosensors with an amplifier circuit. The Fe(Ⅲ) reduction of ΔmtrCptpA/MtrC(+) (B) and ΔcymAptpA/CymA(+) (C) at different concentrations of As3+. The Fe(Ⅲ) reduction rate of ΔmtrCptpA/MtrC and ΔmtrCptpA/MtrC(+) for 26 h (D), as well as ΔcymAptpA/CymA and ΔcymAptpA/CymA(+) (E) with different concentrations of As3+ for 31 h. A two-sided Student’s t test was used in (D) and (E) to analyze the statistical significance (NS: No significance. * 0.01< P < 0.05).
To test whether ΔmtrCptpA/MtrC(+) and ΔcymAptpA/CymA(+) biofilms as a bio-element can improve the sensitivity of EAB-based biosensors, As3+ with the concentrations of 50 and 100 μM was fed into the systems after MFCs generated stable voltages. We found that the presence of As3+ (50 and 100 μM) in the system increased the voltages of both ΔmtrCptpA/MtrC(+) and ΔcymAptpA/CymA(+) MFCs (Figure 7A and B). We compared the enhanced rate (ER) of voltages of ΔmtrCptpA/MtrC and ΔmtrCptpA/MtrC(+), and found that the enhanced rates of voltages of ΔmtrCptpA/MtrC(+) with an amplifier circuit were significantly higher than those of ΔmtrCptpA/MtrC when the concentrations of As3+ were 50 and 100 μM (Figure 7C). ER is used to evaluate the voltage promotion rate of the MFC. It is calculated as follows: ER(%) = (Vm - V0)/V0 × 100% where V0 is the maximum stable voltage before feeding arsenic, and Vm is the maximum voltage inducted. We observed the similar results for the ΔcymAptpA/CymA and ΔcymAptpA/CymA(+) strains. The results show that amplifier system improved the performance of EAB-based sensors.
Figure 7. The voltage of ΔmtrCptpA/MtrC (+) (A) and ΔcymAptpA /CymA(+) (B) MFCs fed 50 and 100 μM As3+ at 25 h. The enhanced rate of voltages of ΔmtrCptpA/MtrC and ΔmtrCptpA/MtrC (+) (C), as well as ΔcymAptpA /CymA and ΔcymAptpA /CymA(+) (D). A two-sided Student's t test was used in (C) and (D) to analyze the statistical significance . ( ** P < 0.01).
