Engineering Success

Abstract

With the increasing frequency of human–pet interactions, the prevention and control of zoonotic pathogens has become a critical issue. To address this, we constructed and optimized a broad-spectrum antimicrobial peptide, AMP-TB2, with the aim of developing a safe and efficient antibacterial spray specifically for pets. The project underwent five rounds of the DBTL engineering cycle.

First, we achieved heterologous expression of the candidate antimicrobial peptide Ulink-AMP in E. coli BL21(DE3) using a SUMO peptide, and designed three mutants through site-directed mutagenesis. We then introduced an anionic peptide fusion strategy, which effectively alleviated host toxicity while improving both yield and purification efficiency. Minimum inhibitory concentration (MIC) assays and agar plate inhibition tests confirmed that AMP-TB2 exhibited the strongest antibacterial activity against both Gram-negative (E. coli) and Gram-positive (B. subtilis) bacteria.

Molecular docking further revealed that AMP-TB2 maintained stable binding to key target proteins of Staphylococcus aureus, Campylobacter, and Salmonella, validating its broad-spectrum antibacterial potential. In thermal stability tests, AMP-TB2 retained high activity against Gram-positive bacteria but showed reduced efficacy against Gram-negative strains, indicating that its application conditions require further optimization. To address this, we proposed a strategy of freeze-drying and solvent-separated storage, ultimately leading to the improved design of a dual-chamber spray bottle and small packaging system.

Overall, this project demonstrates a complete engineering workflow from molecular design and functional validation to hardware implementation, establishing the feasibility and application prospects of AMP-TB2 as the core component of a pet antibacterial spray.

Part 1:Engineering and Optimization of AMP Expression Strains

Design 1

Using a deep learning model and PeptideRanker, our laboratory previously screened and validated the candidate antimicrobial peptide Ulink-AMP from the AMPsphere database. Building on this foundation, we aimed to achieve its heterologous expression and purification in E. coli BL21(DE3).

Since small peptides often suffer from low yield, instability, and potential toxicity to the host during heterologous expression, we adopted the pET-28a-SUMO vector for constructing the expression system. The SUMO fusion tag enhances protein solubility and stability while mitigating host toxicity, and is therefore considered an effective optimization strategy [1].

In addition, to further enhance activity, we performed semi-rational design based on protein modeling and molecular docking results, targeting key amino acid residues. Three site-directed mutants were constructed: AMP-TB1 (A42F), AMP-TB2 (G40A), and AMP-TB3 (G40A & A42F), as detailed in the Model section.

The three mutants were constructed based on pET-28a-Ulink-AMP using whole-plasmid inverse PCR.

Build 1

The Ulink-AMP gene fragment and the linearized vector pET-28a-SUMO were amplified using a high-fidelity PCR enzyme. The products were then assembled via Gibson assembly and transformed into E. coli BL21(DE3) competent cells, followed by plating on LB agar containing antibiotics for selection.

After single colonies were picked, colony PCR was performed to preliminarily verify whether the mutant band corresponded to the expected size of 686 bp. The preliminary results were correct, and the successfully verified clones were subsequently expanded, plasmids were extracted, and the samples were sent to a biotechnology company for sequencing.

After confirming successful plasmid construction, a pair of complementary primers carrying the desired mutation bases was designed for each target amino acid codon. The mutation site was introduced in the middle of the primers, with sufficient homologous arms retained on both sides to ensure recombination efficiency. Using the plasmid containing the original antimicrobial peptide gene as the template, PCR amplification was performed to obtain linearized mutant fragments, followed by DpnI digestion to remove the template plasmid. The purified PCR products were then circularized in vitro using a homologous recombinase to efficiently generate the recombinant plasmids.

The recombinant products were transformed into competent cells and plated. Single colonies were picked, and colony PCR was performed to preliminarily verify the expected band size. Successfully verified clones were sent for sequencing, which confirmed that the target amino acid sites had been successfully mutated without introducing any unintended mutations.

Test 1

The wild-type and three mutant strains were each inoculated into 5 mL of LB medium for overnight culture, and then subcultured at a 1:100 ratio into fresh 50 mL LB medium. When the cultures reached an OD600 of approximately 0.6–0.8, IPTG was added to a final concentration of 0.1 mM for induction. Cells were harvested after 20 h of cultivation at 20 °C, followed by lysis through ultrasonic disruption. Protein samples were then purified using Ni²⁺ affinity chromatography.

SDS-PAGE analysis of both crude extracts and purified products revealed a target band at approximately 18.8 kDa in all four engineered strains, confirming successful expression of the antimicrobial peptides. However, the overall protein yield appeared low based on band intensity, suggesting that reliance on the SUMO tag alone was insufficient to achieve optimal expression and purification efficiency.

Learn 1

The results of Cycle 1 demonstrated that Ulink-AMP and its mutants could be successfully expressed in E. coli BL21(DE3), but the overall protein yield was low, and the purified bands were suboptimal. This suggested that the direct expression of antimicrobial peptides may exert toxicity on the host, thereby suppressing expression efficiency. In addition, short peptides are prone to intracellular degradation, further compromising their stability.

Literature review revealed that such issues are commonly encountered in the heterologous expression of antimicrobial peptides [2]. Previous studies have shown that the introduction of an anionic peptide can alleviate host toxicity through charge-balancing effects, while also improving the stability and yield of AMPs [3]. This finding provided a clear direction for the design of our next engineering cycle.

Design 2

To address the low yield and poor purification efficiency observed in Cycle 1, we planned to improve the stability and solubility of the antimicrobial peptide through a fusion expression strategy. Previous studies have shown that AMPs often exhibit reduced expression levels in the host due to their toxic effects on cell membranes, while their short peptide structures are also prone to rapid degradation by intracellular proteases [2]. To overcome these challenges, we designed the introduction of an anionic antioxidant peptide (EELDNALN) derived from porcine myofibrillar protein, fused in tandem with the target antimicrobial peptide Ulink-AMP via a flexible linker (EDPNG) [3]. This strategy was expected to mitigate host cell toxicity through charge-neutralizing effects, enhance the structural stability of the fusion protein, and thereby improve soluble expression and purification efficiency.

Build 2

To implement the fusion expression strategy, we constructed the plasmid pET-28a-eel-ulink-AMP using whole-plasmid inverse PCR, while simultaneously introducing the three designed mutations. In primer design, the sequences of EELDNALN and EDPNG were incorporated into the central region, with sufficient homologous arms retained at both ends to ensure recombination efficiency.

After PCR amplification, the products were digested, ligated, and transformed into competent cells. Following overnight cultivation, single colonies were picked for colony PCR. The results showed bands of the expected size, and the successfully verified clones were subsequently sent for sequencing to confirm the accuracy of the inserted sequences.

Test 2

After induction and purification of the four successfully constructed engineered strains, the expression of the fusion proteins was analyzed by SDS-PAGE. A single clear target band (~19.6 kDa) was observed within the 15–25 kDa range, with band intensities significantly stronger than those from the single-expression strategy in Cycle 1. These results demonstrate that the introduction of the anionic peptide fusion strategy markedly improved both the expression level and purification efficiency of the antimicrobial peptides.

Learn 2

SDS-PAGE results demonstrated that the fusion expression system effectively improved the expression level and purification efficiency of antimicrobial peptides in E. coli. Compared with the single-expression results in Cycle 1, the introduction of the anionic peptide EELDNALN significantly alleviated host toxicity, while the bands appeared clearer and more distinct, indicating enhanced protein stability. This validated that the charge-neutralization strategy reported in the literature was also applicable in our system.

However, improving expression and purification efficiency was only the first step. It remained necessary to determine whether the engineered antimicrobial peptides retained or even enhanced their expected functional activity, as well as to evaluate their stability under practical application conditions. Therefore, in the next engineering cycle, we focused on testing the antibacterial efficacy of the peptides and assessing their stability under different conditions.

Part 2:Verification of the Antibacterial Activity of Engineered AMPs

Design 3

In Cycle 2, we significantly improved the expression and purification efficiency of antimicrobial peptides. The next critical question was whether these engineered peptides retained or enhanced their expected biological functions. To systematically evaluate their antibacterial effects, we designed two parallel validation assays: minimum inhibitory concentration (MIC) determination and agar diffusion tests.

According to CLSI standards, MIC provides a quantitative measure of the ability of antimicrobial peptides to inhibit bacterial growth and is an important indicator for assessing antibacterial activity [4].

However, due to safety regulations, we were not permitted to conduct antibacterial tests using common zoonotic pathogens. Under the guidance of Dr. Duan, an expert in synthetic biology, we selected Escherichia coli BL21(DE3) and Bacillus subtilis ATCC 6633 as representative test strains. These two organisms serve as models for Gram-negative and Gram-positive bacteria, respectively, thereby enabling a comprehensive evaluation of the broad-spectrum antibacterial potential of our engineered peptides.

At the same time, the agar diffusion assay provides a more visual representation of the antibacterial effects. By observing and measuring the diameters of inhibition zones on solid media, we can directly compare the inhibitory performance of different peptides and identify the most effective candidate for subsequent studies.

Build 3

We performed a two-fold serial dilution of the original antimicrobial peptide and its three mutants, covering a concentration range of 20–1280 mg/mL. Fresh bacterial colonies were picked and cultured overnight at 37 °C with shaking. The cultures were then inoculated into fresh medium at a 1% ratio and grown to the logarithmic phase. Subsequently, the bacterial suspension was diluted to 1×10⁵ CFU/mL, dispensed into 96-well plates, and treated with different concentrations of antimicrobial peptides. PBS was used as the negative control, and ddH₂O served as the blank control. Three biological replicates were set for each concentration.

To visually assess the antibacterial ability of the peptides, we simultaneously established an agar diffusion assay system. Escherichia coli BL21(DE3) and Bacillus subtilis ATCC 6633 were streaked onto LB agar plates and incubated overnight at 37 °C. Single colonies were then inoculated into LB liquid medium and cultured overnight at 37 °C with shaking at 180 rpm. The resulting cultures were adjusted to an OD600 of 0.1 and further diluted 1,000-fold to prepare bacterial suspensions for the inhibition assay.

Sterile cotton swabs were used to evenly spread the diluted bacterial suspension across the surface of LB agar plates. Then, 100 μL of antimicrobial peptide solution was added to each plate, with sterile water as the negative control and a silver ion antibacterial agent as the positive control. All plates were incubated statically at 37 °C for 16 hours to allow the formation of inhibition zones.

Test 3

After overnight incubation at 37 °C, bacterial growth was assessed by visual inspection. The results showed that all antimicrobial peptide samples exhibited clear inhibitory effects against both test strains. Among them, AMP-TB2 displayed the lowest MIC value, demonstrating stronger antibacterial activity against both E. coli and B. subtilis compared to the original sequence and other mutants. These findings indicate that AMP-TB2 not only achieved the best performance in expression and purification, but also exhibited the highest potential in terms of practical antibacterial activity.

In the agar diffusion assay, all antimicrobial peptides at a concentration of 150 mg/mL produced distinct inhibition zones against both bacterial strains, further confirming their broad-spectrum antibacterial activity. Comparative analysis revealed that AMP-TB2 exhibited the strongest inhibition, with a zone diameter of 13.12 ± 0.45 mm against B. subtilis, which was significantly larger than that of the original peptide and other mutants.

By comparing the antibacterial effects of different peptides, we directly observed that AMP-TB2 exhibited the strongest inhibitory activity against both bacterial strains, with significantly larger inhibition zones than the other mutants, indicating superior antibacterial efficiency. Taken together, these findings demonstrate that our optimization system not only improved the expression and purification performance of the antimicrobial peptides but also significantly enhanced their biological functionality, thereby confirming AMP-TB2 as the most promising candidate for subsequent functional validation and stability studies.

Learn 3

Through MIC and agar diffusion assays, we confirmed that Ulink-AMP and its mutants exhibited antibacterial activity against both Gram-negative (E. coli) and Gram-positive (B. subtilis) bacteria. Among them, AMP-TB2 consistently demonstrated superior performance, with MIC values and inhibition zone diameters significantly better than those of the original peptide and other mutants.

However, E. coli and B. subtilis can only serve as representative models of Gram-negative and Gram-positive bacteria, and do not fully reflect the complexity of zoonotic pathogens. To further validate the broad-spectrum antibacterial potential of AMP-TB2, we plan to conduct molecular docking experiments in the next engineering cycle, selecting Staphylococcus aureus, Salmonella, and Campylobacter as targets. These analyses will allow us to evaluate the mechanisms of action and antibacterial potential of AMP-TB2 against clinically relevant zoonotic pathogens.

Design 4

Due to the biosafety regulations for high school teams this year, we were not allowed to perform inhibition assays using bacterial strains outside the approved whitelist. Therefore, to further explore the potential mechanisms of antimicrobial peptides against other common zoonotic pathogens, we employed protein modeling and molecular docking approaches to simulate their antibacterial interactions.

To further validate the broad-spectrum potential of AMP-TB2, we selected three common zoonotic pathogens as research subjects: Staphylococcus aureus, Campylobacter, and Salmonella. Based on the functional roles of their key pathogenic proteins, we chose penicillin-binding protein (PBP), N-acetyltransferase PseH, and a glycosyltransferase from Salmonella as docking targets, in order to evaluate whether AMP-TB2 could form stable interactions with their active regions.

Build 4

We first predicted the three-dimensional structure of AMP-TB2 using AlphaFold3 to obtain a reliable conformation at the atomic level. Subsequently, the crystal structures of the three key target proteins from zoonotic pathogens were retrieved from the PDB database. After standard preprocessing steps—including removal of water molecules, removal of ligands, and addition of hydrogen atoms—AMP-TB2 was prepared as the ligand and docked with the three targets using AutoDock Vina. By comparing binding energies and screening docking conformations, we were able to identify the potential binding sites of AMP-TB2 with different pathogenic proteins and evaluate its binding stability.

Test 4

Molecular docking results showed that AMP-TB2 was able to form stable interactions with the key target proteins of all three zoonotic pathogens. In Staphylococcus aureus PBP, AMP-TB2 established hydrogen bonds and hydrophobic interactions with multiple residues in the active pocket, exhibiting strong binding affinity. In Campylobacter PseH and Salmonella glycosyltransferase, AMP-TB2 similarly entered the functional sites and interacted with critical residues, maintaining a stable conformation and suggesting potential interference with their normal catalytic processes. Overall, the docking results were consistent with our previous antibacterial assays, further supporting the broad-spectrum antibacterial potential of AMP-TB2. See Model section for details.

Learn 4

From our earlier MIC and agar diffusion assays, we confirmed that AMP-TB2 exhibited significant antibacterial activity against the Gram-negative strain E. coli and the Gram-positive strain B. subtilis, providing initial evidence of its inhibitory effects across both major bacterial classes.

In this round of molecular docking experiments, we further extended the study to three common zoonotic pathogens: Staphylococcus aureus, Campylobacter, and Salmonella. The results showed that AMP-TB2 formed multiple hydrogen bonds and hydrophobic interactions within the active sites of their key target proteins, with stable conformations and relatively low binding energies. These findings suggest that AMP-TB2 may interfere with bacterial growth and pathogenicity by blocking cell wall synthesis, disrupting flagellar modification, or inhibiting glycosyltransferase activity—consistent with the antibacterial effects observed in our earlier assays.

Taken together, AMP-TB2 demonstrated consistent activity and binding characteristics in both representative model strains and clinically relevant zoonotic pathogens, indicating a generalizable mechanism of action. This combined evidence—spanning Gram-positive bacteria, Gram-negative bacteria, and zoonotic pathogens—further reinforces the broad-spectrum antibacterial potential of AMP-TB2.

Part 3 Stability Test of AMP-TB2

Design 5

In the previous cycles, we validated the construction, optimization, antibacterial activity, and molecular docking mechanism of AMP-TB2, all of which demonstrated its significant broad-spectrum potential as a core candidate peptide. However, antibacterial efficacy alone is far from sufficient for practical applications. The stability of AMP-TB2 under environmental conditions is equally crucial. As small peptides, antimicrobial peptides are prone to inactivation by factors such as temperature, pH, and proteases. For the future development of a spray formulation, temperature stability is particularly important, since the product must remain active during room-temperature storage, transportation, and repeated opening and use by consumers.

Therefore, in this round of experiments, we designed a temperature stability assay, in which AMP-TB2 was stored under different temperature conditions and its antibacterial activity was tested over time. This experiment directly addresses whether AMP-TB2 can maintain effectiveness under realistic application scenarios, while also providing experimental evidence for determining storage requirements and guiding improvements in spray packaging.

Build 5

Purified AMP-TB2 was prepared at a concentration of 150 mg/mL and stored at 4 °C, 20 °C, and 37 °C. To simulate practical storage and usage scenarios, two time points were set at 7 and 14 days. After storage under these conditions, the samples were retrieved and tested for antibacterial activity against B. subtilis (Gram-positive) and E. coli (Gram-negative). By comparing changes in antibacterial activity across different temperature and time groups, the temperature stability of AMP-TB2 could be evaluated.

Test 5

The results showed that AMP-TB2 retained strong antibacterial activity against B. subtilis after 7 and 14 days of storage at 4 °C, 20 °C, and 37 °C, indicating that its activity against Gram-positive bacteria remained stable under these common storage temperatures. In contrast, its antibacterial effect on E. coli was weaker and further declined with extended storage time. These findings suggest that the stability of AMP-TB2 may vary across bacterial strains, with more persistent inhibitory activity observed against Gram-positive bacteria.

Learn 5

The results of the temperature stability assay showed that AMP-TB2 maintained strong antibacterial activity against B. subtilis after storage at 4 °C, 20 °C, and 37 °C for 7–14 days, whereas its activity against E. coli declined significantly. This outcome was consistent with earlier experiments: AMP-TB2 exhibited stronger and more stable inhibition against Gram-positive bacteria, while showing certain limitations against Gram-negative strains. Combined with the molecular docking results, it can be inferred that AMP-TB2 has higher affinity in its binding mode to cell wall–associated proteins of Gram-positive bacteria, thereby sustaining activity under varying temperature and time conditions.

From an application perspective, this finding is of considerable importance. A spray formulation must remain stable during storage and transportation at ambient temperature. The fact that AMP-TB2 retained robust activity against Gram-positive bacteria at 20 °C provides practical feasibility for its use as the core component of pet hygiene products. At the same time, its weaker and less stable performance against Gram-negative bacteria highlights the need to further optimize storage and usage strategies in product design.

To address this, we proposed a new solution: preparing AMP-TB2 as a freeze-dried powder and storing it separately from the solvent. Users would reconstitute the peptide immediately before use, thereby minimizing the risk of inactivation associated with long-term storage in liquid form. This concept directly informed our hardware design improvements—by introducing a dual-chamber spray bottle and small-packaging strategy, we enable convenient separation and mixing of freeze-dried peptide and solvent, ensuring both stability and effectiveness of the product. See details in the HP section.

References

• [1]
  Panavas, Tadas et al. “SUMO fusion technology for enhanced protein production in prokaryotic and eukaryotic expression systems.” Methods in molecular biology (Clifton, N.J.) vol. 497 (2009): 303–17. doi:10.1007/978-1-59745-566-4_20
• [2]
  Roldán-Tapia, Marisol et al. “Streptomyces as Overexpression System for Heterologous Production of an Antimicrobial Peptide.” Protein and peptide letters vol. 24,6 (2017): 483–488. doi:10.2174/0929866524666170208154327
• [3]
  Kim, Ha-Kun et al. “Expression of the cationic antimicrobial peptide lactoferricin fused with the anionic peptide in Escherichia coli.” Applied microbiology and biotechnology vol. 72,2 (2006): 330–8. doi:10.1007/s00253-005-0266-5
• [4]
  Kowalska-Krochmal, Beata, and Ruth Dudek-Wicher. “The Minimum Inhibitory Concentration of Antibiotics: Methods, Interpretation, Clinical Relevance.” Pathogens (Basel, Switzerland) vol. 10,2 165. 4 Feb. 2021, doi:10.3390/pathogens10020165