Precise Construction of an Antimicrobial Peptide Targeting Bacterial Cell Membranes Derived From Natural Peptides.

We are sharing an important study published in Advanced Science, titled "Precise Construction of an Antimicrobial Peptide Targeting Bacterial Cell Membranes Derived From Natural Peptides."​ This research addresses the challenge of difficult target prediction when using artificial intelligence (AI) algorithms to assist in screening antimicrobial peptides (AMPs). Instead, it adopts a more direct and interpretable strategy of site-directed amino acid mutagenesis. The researchers identified several natural peptide segments from an insect cuticular protein, used one as a template, and rationally designed a novel antimicrobial peptide, P 3-3R-8I, by mutating specific amino acids to arginine (R) and isoleucine (I). This peptide exhibits potent antibacterial activity against various drug-resistant bacteria, including methicillin-resistant Staphylococcus aureus(MRSA). Its uniqueness lies in a dual mechanism of action: it can rapidly penetrate bacterial cell membranes and also enter the cytoplasm to bind bacterial DNA and inhibit its replication. In animal models, P 3-3R-8I effectively promoted the healing of MRSA-infected wounds and, in an MRSA-induced systemic sepsis model, significantly reduced infections in the lungs and spleen while improving survival rates. This work provides a new paradigm for rational, target-specific design of antimicrobial peptides that does not rely on AI.

01 Research Background

Antibiotic resistance has become a major global public health challenge. Antimicrobial peptides (AMPs) are regarded as promising alternatives to combat drug-resistant infections due to their unique membrane-targeting mechanism and low propensity for resistance development. Although AI algorithms have achieved great success in accelerating AMP discovery, the action targets of the AMPs they screen for are often difficult to predict precisely, creating obstacles for subsequent mechanistic studies and rational optimization. Compared to the "black box" predictions of AI, amino acid mutagenesis based on known structure and function is a more traditional but more clearly target-related design strategy. Previously, the research team discovered that an insect cuticular protein possesses antibacterial activity, but its target was unknown, and its large molecular weight limited its bioavailability. Therefore, deriving smaller, target-specific, and more active antimicrobial peptides from this protein and elucidating their structure-activity relationships holds significant scientific value and application potential.

02 Highlights

Employing an "Interpretable" Amino Acid Mutation Strategy Instead of a "Black Box" AI Screen

Addressing the weakness of ambiguous target prediction by AI, the research team eschewed complex algorithmic models and returned to rational chemical and structural biology design. Based on the sequence features of natural peptides, they purposefully replaced neutral or negatively charged residues with positively charged arginine to enhance interaction with negatively charged bacterial membranes. They also replaced some alanines with more hydrophobic isoleucine to increase affinity for membrane lipids. This direct introduction of "functionalized residues" makes the targeting design of the AMP highly interpretable and precise.

Revealing a Unique Dual Bactericidal Mechanism: Membrane Penetration Followed by DNA Binding

The study not only confirmed that P 3-3R-8I can rapidly (within ~2 minutes) penetrate and stably embed into the bacterial membrane but, more importantly, discovered that after entering the cell, it can bind bacterial genomic DNA with high affinity. Electrophoretic mobility shift assays and gel electrophoresis experiments confirmed that this binding effectively inhibits DNA replication. This cascade mechanism of "membrane penetration → DNA binding → replication inhibition" distinguishes it from most AMPs, which rely solely on membrane disruption, providing a new bactericidal pathway and explaining its highly efficient antibacterial activity.

Achieving Potent Inhibition of Multi-Drug Resistant Bacteria and Demonstrating Potential Against MCR⁺ Strains

P 3-3R-8I exhibited excellent antibacterial activity against both Gram-positive and Gram-negative bacteria, including clinically isolated MRSA. Particularly important, the study preliminarily explored its inhibitory effect against strains carrying the mobile colistin resistance gene (mcr-1). Although the effect diminished over time, this points the direction for subsequent peptide sequence optimization targeting such "super-resistant" strains, demonstrating the platform's extensibility.

Validating Efficacy in Two Severe Models: Infected Wound Healing and Systemic Sepsis

Going beyond in vitroantibacterial tests, the study systematically evaluated the therapeutic effects of P 3-3R-8I in an MRSA-infected rat full-thickness skin wound model and a systemic sepsis model. The results showed that it significantly promoted wound healing, accelerated collagen deposition, modulated inflammation-related gene expression, reduced multi-organ infection, lowered systemic inflammatory markers, and improved survival rates. Its overall efficacy was comparable to, and in some aspects superior to, vancomycin, with good in vivosafety, showcasing tremendous clinical translation potential.

03 Results and Discussion

3.1 Rational Design and Successful Construction of Antimicrobial Peptide P 3-3R-8I

Through sequence analysis of the insect cuticular protein OfCPH-2, five natural peptide segments were identified. Based on tryptophan content, hydrophobicity, and the proportion of positively charged residues, P3 and P4 were selected as mutation templates. Proline, serine, and asparagine at positions 7, 8, and 15 of P3 were mutated to arginine, yielding P 3-3R to increase positive charge. Then, alanines at positions 3, 4, 6, 11, 12, 14, 19, and 20 were mutated to isoleucine, yielding P 3-3R-8I to enhance hydrophobicity. Circular dichroism spectroscopy and structural prediction indicated that P 3-3R-8I has a typical β-sheet structure and amphipathic characteristics, consistent with the structural features of most AMPs. t-SNE clustering analysis classified it within known AMP families, confirming successful construction.

3.2 Excellent In Vitro Antibacterial Activity and Membrane Action Characterization

The minimum inhibitory concentration (MIC) of P 3-3R-8I was 100 µM for both MRSA and E. coli, and it could completely kill the bacteria within 8-12 hours. ONPG detection, live/dead bacterial staining, and scanning electron microscopy results all indicated that peptide treatment led to bacterial death and membrane damage. Membrane potential probe experiments and molecular dynamics simulations further confirmed that P 3-3R-8I rapidly interacts strongly with bacterial model membranes (Gram-positive and negative), forming multiple hydrogen bonds and van der Waals forces, and stably embeds into the membrane.

3.3 Unique Intracellular DNA-Targeting Mechanism

Although capable of disrupting membrane integrity, the study found that membrane disruption was not the sole cause of its bactericidal effect. Molecular docking showed that P 3-3R-8I could bind with high affinity to S. aureusand E. coliDNA via hydrophobic interactions and hydrogen bonds. Electrophoretic mobility shift assays directly proved the peptide's binding to bacterial genomic DNA. Most crucially, gel electrophoresis experiments demonstrated that P 3-3R-8I could concentration-dependently inhibit DNA replication in vitro. Together, this reveals its dual bactericidal mechanism: "first penetrating the membrane, then binding to and inhibiting DNA replication."

3.4 Excellent In Vivo Safety, Pharmacokinetics, and Wound Healing Efficacy

In vitroand in vivosafety assessments showed that P 3-3R-8I was non-toxic to mammalian cells, caused no damage to major organs or immune reactions in animals. Its plasma half-life (~5.7 hours) was longer than that of vancomycin (~3.5 hours), and it showed good stability in serum. In the MRSA-infected rat skin wound model, topical application of P 3-3R-8I significantly accelerated wound closure, reduced bacterial load, and promoted granulation tissue and collagen formation. Immunohistochemistry and RNA-seq analysis indicated that its treatment downregulated the inflammatory factor IL-1β and upregulated factors related to tissue development and angiogenesis (e.g., FGF-2, CD31, α-SMA) and Wnt signaling pathway genes, elucidating the molecular mechanisms behind its promotion of infected wound healing.

3.5 Therapeutic Potential in a Systemic Sepsis Model

In an MRSA-induced sepsis model, intraperitoneal injection of P 3-3R-8I increased the survival rate of rats from 20% to 80%, effectively reduced serum levels of TNF-α, IL-1β, and white blood cell/neutrophil counts, and significantly alleviated bacterial infection and histopathological damage in organs like the lungs and spleen, demonstrating its potential to treat severe systemic infections.

04 Conclusion and Future Perspectives

This study successfully employed a rational amino acid mutagenesis strategy, starting from a natural insect protein peptide, to precisely construct a novel antimicrobial peptide, P 3-3R-8I, with a dual mechanism of action. The peptide exerts potent antibacterial effects by rapidly penetrating bacterial membranes and binding to/inhibiting DNA replication. It demonstrated excellent therapeutic efficacy and good safety in models of drug-resistant bacterial wound infection and systemic sepsis.

The core value of this work lies in providing an interpretable, rational design pathway for antimicrobial peptides that does not rely on big data and black-box AI models. By introducing specific functional residues to "encode" membrane targeting and unexpectedly discovering its DNA-targeting capability, the study not only yielded a lead compound but, more importantly, deepened the understanding of the "structure-membrane targeting-intracellular action" relationship. Future research could focus on: ① Further optimizing the sequence based on this dual mechanism to enhance activity against extremely drug-resistant strains like MCR⁺ bacteria; ② Exploring combination strategies of this peptide with other antibiotics to address complex infections; ③ Conducting more in-depth preclinical pharmacokinetic/toxicological studies to advance it towards clinical translation. In summary, this study confirms that "fine-tuned" design based on natural templates and functional residues is a powerful alternative approach for developing novel, highly effective antimicrobial peptides with clearly defined mechanisms.