In the 2024 project, we constructed engineered A. ferrooxidans strains with elevated levels of c-di-GMP by overexpressing c-di-GMP synthases (Figure 1-1). In subsequent studies conducted by our research group on the engineered bacterial strains constructed during the 2024 project, transcriptomic analysis revealed that the overexpression of DGC AFE_0053 (BBa_K5323000) led to significant up regulation of electron transport-related genes in A. ferrooxidans (Figure 1-2).

A. ferrooxidans thrives in highly acidic environments (pH 1.5–2.5) by oxidizing ferrous iron (Fe²⁺) to obtain energy, a process in which electron transfer plays a central role. Iron, being a multivalent metal, can exist in solution as either Fe²⁺ or Fe³⁺, making it an effective electron shuttle. It has been reported that A. ferrooxidans utilizes soluble iron species as electron shuttles^[1].

Based on these findings, we propose to construct a microbial electrochemical system using A. ferrooxidans as the chassis organism to explore its electrochemical applications.

Current-time curves were recorded, and the current density was calculated by normalizing the current to the working electrode area (1 cm × 3 cm=3 cm2). The resulting current density-time profile is presented in Figure 1-3. To further investigate the role of iron as an electron shuttle, we also examined the system's behavior under varying initial concentrations of ferrous ions, as summarized in the same figure.

We measured the intracellular c-di-GMP levels in the electron transfer module engineered strains and recorded the current-time curves. The current density was calculated based on the working electrode area. The current density-time profile is presented in Figure 1-4.

Figure 1-4 illustrates the current density trends of the engineered p0053 group and the pYDT group over time. Overall, both groups exhibited a declining trend in current density (toward more negative values), but the decrease was markedly more pronounced in the p0053 group. By approximately 300 hours, the current density of the p0053 group approached -20 A/m², whereas the pYDT group remained above -15 A/m². These results indicate that, over extended operation, the p0053 strain demonstrates a significantly faster and greater decline in current density compared to the control, reflecting its enhanced electron utilization capability. Collectively, under identical conditions, the p0053 engineered strain likely possesses stronger electron transfer capacity.

We further performed a Student's t-test using R to statistically evaluate these differences. The results of the t-test are presented in Figure 1-5 (A). Additionally, a corresponding p-value plot was generated and is shown in Figure 1-5 (B). As shown in Figure 1-5(A), a statistically significant difference in current density was observed between the engineered p0053 group and the pYDT control group (denoted by, *** p < 0.001). Specifically, the p0053 group exhibited an average current density of -7.53 ± 7.01 A/m², compared to -5.40 ± 4.50 A/m² for the pYDT group. The lower (more negative) mean current density, along with greater variability (standard deviation) in the p0053 group, indicates not only a significant divergence in current output between the two groups but also higher data dispersion within the engineered strain. Figure 1-5(B) displays the p-values from statistical significance analysis of current density differences between the engineered p0053 and pYDT groups over time, presented as -log₁₀(p-value). During the initial phase of the experiment (approximately 0–250 hours), the -log₁₀(p-value) remained close to zero, indicating no significant difference between the two groups. However, after approximately 250 hours, the p-value decreased sharply (reflected by a sharp increase in -log₁₀(p-value)), reaching an extremely high level of statistical significance ( *** p < 0.001). This trend aligns with the patterns observed in the previous two figures, further confirming that significant differences in current density between the groups emerged and became statistically robust during the later stage of the experiment.

During the initial phase of the experiment, no significant difference in current density was observed between the engineered p0053 group and the pYDT control group, as indicated by the corresponding p-value curve showing non-significance in this period. This phenomenon can be largely attributed to the stages of bacterial proliferation and initial biofilm formation, during which differences in electron transfer capacity had not yet been fully manifested.

Over time, as the biofilm matured and bacterial density reached a stable level, the p0053 group—benefiting from higher intracellular c-di-GMP concentrations—exhibited enhanced electron transfer capability, leading to a rapid decline (more negative shift) in current density. In contrast, the pYDT group, with weaker electron transfer capacity, showed a more gradual decay. Thus, upon biofilm maturation, the disparity in electron transfer efficiency became the dominant factor governing current density dynamics, with the p0053 strain demonstrating significantly higher electron transfer efficiency in the later stage, resulting in a pronounced difference in current density compared to the control.

In summary, we have successfully established a microbial electrochemical system based on A. ferrooxidans and constructed an efficient electron transfer module.

To promote efficient glycerol synthesis in engineered bacteria and enable successful glycerol efflux, we selected and introduced three gene fragments into the engineered bacteria, namely Fission yeast polyol transporter 1 (Fps1), Glycerol-3-phosphate dehydrogenase (NAD(+)) (Gpd1), and glycerol-1-phosphatase (Gpp2), which are derived from Saccharomyces cerevisiae S288C. A sequence encoding the GGGGS amino acid peptide was inserted between Gpd1 and Gpp2, forming a Gpd1-Gpp2 fusion protein to facilitate intracellular glycerol synthesis in the engineered bacteria. fps1 encodes a glycerol channel protein, which enables glycerol efflux. During the engineering phase, we constructed a recombinant plasmid pFps1-fus. This plasmid uses the IPTG-inducible Ptac promoter to regulate the expression of the Fps1-fus fusion protein.

To evaluate whether these genes can successfully synthesize glycerol and enable glycerol efflux, we first conducted studies using E. coli BL21 as the model strain. We set up different IPTG concentrations (0 mM, 0.1 mM, 0.5 mM, and 1 mM), cultured the strain in M9 medium, and plotted the growth curve.

As shown in Figure 2-1, when IPTG concentration was 0 mM, there was no effect on the growth of either strain, providing a control for other experimental groups. At different IPTG concentrations, the growth of the pYDT strain was generally better than that of the pFps1-fus strain. This indicates that IPTG inhibits the growth of the engineered bacteria, and the inhibitory effect becomes more pronounced as the IPTG concentration increases. Comprehensive comparison showed that when IPTG concentration was 0.5 mM, it was not only beneficial for the expression of the target genes but also favorable for the growth of the engineered bacteria.

To further investigate glycerol production capacity and the impact of IPTG, thereby identifying the optimal induction level, we quantified glycerol yields in both strains^[2]. The corresponding glycerol standard curve is shown in Figure 2-2.

Glycerol production under different IPTG concentrations was calculated based on the standard curve. Experimental results demonstrated that the highest glycerol yield was achieved in the engineered strain at 0.5 mM IPTG, showing a statistically significant increase over the empty vector control. Peak glycerol accumulation consistently occurred during the late logarithmic phase.

As shown in Figures 2-3A and 2-3B, glycerol production in the experimental group exceeded that of the empty vector group across conditions. Specifically, at an IPTG concentration of 0.5 mM, the engineered strain produced significantly more glycerol. This can be attributed to the substantial metabolic burden imposed by higher IPTG concentrations, which forces cells to divert more energy towards maintenance rather than product synthesis.

Furthermore, comparison of glycerol yields at different time points (Figures 2-3C and 2-3D) revealed higher production at 11 hours compared to 23 hours. This pattern is explained by the fact that around 11 hours—corresponding to the late logarithmic phase—bacterial colonies had largely completed active growth and had extensively consumed available nutrients. Subsequently, growth rates declined, and the remaining nutrients in the culture were likely only sufficient to support basic cellular activities, leaving no surplus for additional glycerol synthesis.

The Calvin-Benson-Bassham (CBB) cycle is the primary carbon fixation pathway in A. ferrooxidans. Among the enzymes involved, phosphoribulokinase (PRK) is a key enzyme of the CBB cycle. Therefore, overexpressing PRK can improve the efficiency of the CBB cycle, thereby enhancing the carbon fixation capacity of A. ferrooxidans. To achieve this goal, we aimed to screen the optimal gene fragment between PCC6301pg_RS05100 and ACIR131C_RS0121050. Accordingly, we constructed recombinant vectors pYDT_RS05100 and pYDT_RS0121050, respectively. Both vectors use an IPTG-inducible promoter to regulate the expression of the PRK-encoding gene sequence derived from E. coli BL21.

To evaluate which of pYDT_RS05100 or pYDT_RS0121050 exhibits higher PRK activity—thereby accelerating the CBB cycle and improving carbon fixation efficiency in A. ferrooxidans—we initially used E. coli BL21 as the model organism. First, we investigated suitable culture conditions and assessed the growth curves in LB medium supplemented with IPTG at induction concentrations of 0 mM, 0.1 mM, 0.5 mM, and 1 mM, respectively.

As shown in Figure 3-1, under different IPTG concentrations, the growth curves of the pYDT strain showed no significant differences. This indicates that IPTG itself has no obvious effect on the growth or physiological state of the empty vector strain (i.e., the pYDT strain). In contrast, the pRS0121050 strain exhibited the best growth when the IPTG concentration was 0 mM, and its growth capacity gradually decreased as the IPTG concentration increased. This result demonstrates that higher IPTG concentrations lead to increased expression levels of phosphoribulokinase (PRK); however, the high expression of PRK imposes a metabolic burden on the cells, thereby inhibiting bacterial growth.

As shown in Figure 3-2, even under different IPTG concentrations, the growth curves of E. coli BL21 harboring pYDT and pRS05100 both indicated that the growth rate of pRS05100 was slower than that of pYDT. Furthermore, with the increase in IPTG concentration, the growth inhibitory effect was further exacerbated in a dose-dependent manner, with higher induction levels resulting in slower growth rates.

Based on these findings, we conclude that the PRK expression efficiency of the recombinant vector pYDT_RS0121050 is higher than that of pYDT_RS05100.