The rise of multidrug-resistant (MDR) Gram-negative bacteria poses a significant threat to global health. These bacteria employ various resistance mechanisms, including efflux pumps, impermeable membranes, and antibiotic-modifying enzymes, rendering many traditional antibiotics ineffective. Existing membrane-targeting antibiotics, like colistin and polymyxin B, are increasingly facing resistance and have narrow therapeutic indices with significant toxicity. Antimicrobial peptides (AMPs) represent an alternative approach, offering a potentially vast source of novel antibiotics. Arenicins, AMPs isolated from the lugworm *Arenicola marina*, demonstrate potent in vitro activity against MDR Gram-negative bacteria but also exhibit cytotoxicity and hemolysis. This study aimed to leverage structure-based design to optimize arenicin-3, enhancing its Gram-negative activity and in vivo efficacy while minimizing its toxicity.

Previous research has highlighted the challenges in developing effective peptide antibiotics, emphasizing the need for strategies to overcome limitations such as cytotoxicity and resistance. Studies on the physicochemical properties of small-molecule antibiotics have established guidelines for designing compounds that effectively penetrate and persist within Gram-negative bacteria. These guidelines include considerations of size, polarity, globularity, charge, and structural rigidity. While numerous AMPs show promise in vitro, their translation to preclinical models and clinical development has been limited. The study by Yang et al. showed specific binding of an arenicin analog (NZ17074) to lipopolysaccharide (LPS), suggesting a potential mechanism of action. However, this study sought to explore if a refined structure-based design approach could overcome these limitations for arenicin-3.

The researchers determined the NMR solution structure of arenicin-3, revealing a slightly twisted β-hairpin configuration with amphipathic properties. Structure-guided design led to the synthesis of several arenicin-3 analogs, including AA139. The antimicrobial activity of these analogs was assessed using broth microdilution (BMD) assays against a panel of Gram-negative and Gram-positive bacteria and yeasts, including MDR and extensively drug-resistant (XDR) clinical isolates. Cytotoxicity and hemolytic activity were evaluated using cell viability and hemolysis assays. Membrane binding and permeabilization were investigated using surface plasmon resonance (SPR), 1-N-phenylnaphthylamine (NPN), DiSC3(5), and SYTOX Green assays. The effect of arenicin-3 on ATP release was examined using a luminescent microbial cell viability assay. The potential for resistance development was evaluated using spontaneous mutation frequency assays and serial passage experiments. In vivo efficacy was assessed in murine models of peritonitis, pneumonia, and urinary tract infection (UTI). Pharmacokinetic (PK) and toxicokinetic (TK) properties were characterized in mice, cynomolgus monkeys, and minipigs using intravenous and aerosolized administration. Transposon-directed insertion sequencing (TraDIS) was employed to identify genes involved in bacterial adaptation to arenicin-3 exposure. Whole-genome sequencing was used to analyze genetic changes in *E. coli* isolates after serial passage with arenicin-3.

Structure-based optimization of arenicin-3 yielded AA139, a peptide with significantly improved activity and reduced toxicity. AA139 demonstrated broad-spectrum in vitro activity against MDR and XDR Gram-negative bacteria, including those resistant to last-resort antibiotics, with MIC90 values ranging from 1.0 µg/mL (*E. coli*) to 8.0 µg/mL (*P. aeruginosa*). AA139 showed low cytotoxicity and hemolysis, with CC50 and HC10 values significantly higher than MICs. Mechanism of action studies suggested that AA139 targets bacterial membranes, inducing permeability and ATP release. TraDIS analysis revealed the *mla* operon (involved in phospholipid transport) as a potential target of arenicin-3. Serial passage experiments showed low rates of spontaneous resistance development for AA139 compared to colistin, with resistance potentially arising from mutations in *mlaC*. In vivo studies demonstrated the efficacy of AA139 in murine models of peritonitis, pneumonia, and UTI. The ED50 in the peritonitis model was 1.85 mg/kg in peritoneal fluid and 1.55 mg/kg in blood. In the pneumonia model, aerosolized AA139 led to a substantial reduction in *P. aeruginosa* burden. AA139 showed a favorable PK profile across species, with primarily renal clearance, and a NOAEL dose 10-fold higher than the efficacious dose in infection models. There was no evidence of cross-resistance between AA139 and colistin.

The findings demonstrate the successful application of structure-guided design to improve the therapeutic index of an AMP. AA139 overcomes the limitations of the progenitor peptide by maintaining potent antimicrobial activity while reducing cytotoxicity and hemolysis. The low rate of spontaneous resistance development suggests that AA139 may offer a significant advantage over current last-resort antibiotics. The identification of the *mla* operon as a potential target provides valuable insights into the mechanism of action and could guide further optimization efforts. The favorable PK/TK profile and efficacy in multiple infection models indicate AA139's promise as a lead candidate for the development of a new class of antibiotics to treat MDR Gram-negative bacterial infections.

This study successfully optimized the AMP arenicin-3, resulting in AA139, a broad-spectrum antibiotic with efficacy against MDR and XDR Gram-negative bacteria and reduced toxicity. The low resistance profile, favorable PK/TK properties, and in vivo efficacy suggest AA139's high potential as a lead compound for drug development. Future research could focus on further optimization to enhance potency and improve its pharmacokinetic profile for clinical application. Investigation of the precise mechanism of action involving the *mla* operon is also warranted.

The study primarily focused on Gram-negative bacteria; the effectiveness of AA139 against other types of pathogens requires further investigation. The murine infection models, while useful, may not fully recapitulate the complexity of human infections. Further studies in larger animal models are necessary before clinical trials. The exact mechanism of action by which AA139 impacts phospholipid transport needs more detailed investigation.