Results
Antibacterial Module
The microorganisms isolated from the murals
Three species of fungi and six species of bacteria were isolated from the mural samples. Antimicrobial susceptibility testing via the zone of inhibition method was conducted on each microbial strain, followed by simulated conservation tests on mural specimens pre-colonized by these microorganisms.
Bacteriostatic zone experiment
After testing, GRP1 showed varying degrees of inhibitory effects on Bacillus 4-1, Priestia 6-1, Priestia 7-1, Priestia 8-1
a: Bacillus 4-1, b: Priestia 6-1, c: Priestia 7-1, d: Priestia 8-1
The white arrow indicates the experimental group.
After testing, Pn-AMP1 showed varying degrees of inhibitory effects on Scopulariopsis 4-1, Scopulariopsis5-1, Aspergillus 3-1.
a: Aspergillus 3-1, b: Scopulariopsis 4-1, c: Scopulariopsis5-1
Test on mural briquette
After inoculation with BL21- BBa_25XI83ND/ BBa_259HIEMB and mural microbes, no distinct microbial plaques were observed on the simulated mural (Fig 6a). However, murals not inoculated with the antimicrobial engineered bacteria displayed large plaques and exhibited significant damage (Fig 6b), demonstrating that the antimicrobial engineered bacteria can prevent microbial growth on the murals. Murals with plaques were treated with the engineered bacteria. Calcein-AM/PI staining revealed a significant number of viable bacteria on the untreated murals (Fig 6c). However, 24 h after treatment with the engineered bacteria, widespread microbial cell death was observed on the murals (Fig 6d), demonstrating that the antimicrobial engineered bacteria can effectively treat microbial infestations in the murals. Figure 7 shows that the microbial mortality rate on the murals not treated with the antimicrobial engineered bacteria was 30.03%, while that on the murals treated with the engineered bacteria reached 68.23%.
(a) Simultaneous inoculation of antimicrobial engineered bacteria and mural microorganisms; (b) Inoculation of mural microorganisms alone. (c) Calcein-AM/PI staining results after 5 d of inoculation with mural microorganisms; (d) staining results after 5 d of incubation and 1 d of treatment with engineered bacteria. Green represents viable cells, and red represents dead cells.
Pigment Degradation Module
As shown in Fig 8, after incubation with melanin, the absorbance of melanin decreased (Figure 8a), and the color of melanin changed from dark blue to lighter (Fig 8c). The melanin degradation efficiency increased to 59.07%, significantly higher than that of E-CueO (Fig 8b). The CueO-scaffoldin engineered bacteria further enhanced the melanin degradation efficiency of CueO.
(a) UV-visible spectrum of melanin after degradation by CueO-scaffoldin-expressing bacteria; (b) Melanin degradation efficiency by the bacteria; (c) Visual representation of melanin decolorization by CueO-scaffoldin-expressing bacteria.
As shown in Fig 9, after incubation with β-carotene, the absorbance of β-carotene decreased (Fig 9a), and the color of β-carotene changed from dark yellow to light (Fig 9c). The degradation efficiency of β-carotene increased to 49.57%, significantly higher than that of E-DyP (Fig 9b). The DyP-scaffoldin-expressing bacteria further enhanced the degradation efficiency of DyP for β-carotene.
(a) UV-visible spectrum of β-carotene after degradation by DyP-scaffoldin-expressing bacteria; (b) β-carotene degradation efficiency by the bacteria; (c) Visualization of β-carotene decolorization by DyP-scaffoldin-expressing bacteria.
pH-Responsive Switch
To enable our Escherichia coli strain to regulate product synthesis in response to pH variations, we engineered a genetic regulatory module. The Pasr promoter, which exhibits significantly enhanced expression under acidic conditions, was employed. By integrating Pasr with the TetR gene and PtetO3 operator, we constructed a system capable of selectively activating or repressing transcription of specific genes at different pH levels. To evaluate the regulatory performance of this module, fluorescent proteins were incorporated at both termini as transcriptional reporters, with fluorescence intensity serving as the metric for transcriptional control. The engineered bacterial strains were cultured under varying pH conditions, and time-dependent changes in fluorescence intensity of distinct fluorescent proteins were quantified using a microplate reader. The observed increase and decrease in fluorescence intensity directly indicated the activation and repression of fluorescent protein synthesis, respectively.
Eight gradients were set up within the pH range of 4 to 7 for the experiment. The fluorescence intensity of EGFP showed a downward trend over time at pH = 4.25 and pH = 5, and an upward trend at pH = 5.5. The trends were not obvious at other pH values.
Eight gradients were set up within the pH range of 4 to 7 for the experiment. The fluorescence intensity of mCherry showed an upward trend over time at pH values ranging from 4 to 4.75, and a downward trend at pH 5. The trend was not obvious at other pH values.
As shown in Fig 10 and Fig 11, our regulatory elements meet our design goals within the pH range of 4 to 5.5, which can be concluded from the changes in fluorescence intensity.
Suicide Module
To prevent genetic leakage outside the repair materials, we designed a suicide module. In the presence of arabinose, this module expresses the CI protein, which binds to the O1, O2, and O3 sites upstream of the nuclease gene's RBS, inhibiting nuclease expression. In the absence of arabinose, the CI protein cannot activate transcription, allowing the nuclease gene to be expressed and kill the engineered E. coli. We synthesized the suicide module using full gene synthesis, inserted it into the pSB1A3 vector, and transformed it into E. coli (Fig 12). We tested different arabinose concentration gradients, but the number of E. coli colonies did not decrease with reduced arabinose concentrations, nor was there a significant difference compared to the no-arabinose control group. These results indicate that the initially designed suicide module did not meet our requirements.
Five different arabinose concentration gradients were set (0, 0.01%, 0.1%, 0.5%, and 1%). After transforming 1 μL of the suicide plasmid, 100 μL of bacterial suspension was plated per dish. All plates contained ampicillin. The number of E. coli colonies grown in media with different concentrations showed no significant concentration-dependent differences.
To investigate the specific reasons for the failure, we designed a forward primer with a PstI restriction site at the terminus and a reverse primer with an EcoRI restriction site. The nuclease gene carrying the J61100 RBS was cloned individually and ligated into the pUC19 vector using restriction enzyme digestion and ligation. Upon inducing protein expression with IPTG, no significant bactericidal activity was observed (Fig 13).
E. coli transformed with the pUC19 vector plasmid carrying the nuclease gene was induced using 50 mM IPTG. No significant difference was observed between the two groups.
Since the initial nuclease did not achieve the desired bactericidal effect, we sought an alternative suicide gene and identified the protein MazF. MazF is a toxin protein in the E. coli toxin-antitoxin system, functioning as a sequence-specific endoribonuclease that cleaves single-stranded RNA at ACA sequences to induce bacterial suicide. We synthesized the MazF toxin gene carrying an RBS sequence and ligated it into the pET-30a vector. Induction with 50 mM IPTG demonstrated effective bactericidal activity under induction conditions (Fig 14).
E. coli transformed with the pET-30a vector plasmid carrying the MazF gene was induced using 50 mM IPTG. No E. coli growth was observed on the IPTG-containing plates.
After confirming that the MazF toxin protein could successfully kill E. coli, we proceeded to integrate it into the suicide module. We employed restriction enzyme digestion and ligation for modification. The upstream regulatory sequence of the suicide module and the mazF gene were separately cloned, with biobrick prefix and suffix interfaces added to both ends. The upstream regulatory sequence was first ligated into the pSB1A3 vector, followed by restriction enzyme digestion with PstI and EcoRI and ligation of the vector and fragment, ultimately yielding the successfully modified vector. Experimental validation showed that the engineered bacteria exhibited normal growth in the presence of 1% arabinose. However, in the absence of arabinose, the cell death program was successfully executed, and no bacterial colonies grew on the culture medium (Fig 15), indicating that the modified suicide module met our requirements.
Subsequently, to test the survival of the engineered bacteria on the mural surface, we obtained mural samples from Dr. Wu Fasi's laboratory. E. coli carrying EGFP was sprayed onto the mural surface, and the number of surviving cells was observed under a microscope. The experiment demonstrated that the majority of the engineered bacteria failed to survive on the mural surface (Fig 16), aligning with our expectations.
