Biology subjects

D.T. Torres, M.

Publications and source records attributed to D.T. Torres, M..

3 recordsLinked to original sources

bioRxiv2025

Atom-level backbone engineering preserves peptide function while enhancing stability

Peptide therapeutics offer unmatched potency and selectivity but are limited by rapid proteolysis and poor pharmacokinetics. Backbone engineering provides a rational approach to enhance stability while preserving function, yet direct comparisons across strategies remain scarce. Using bradykinin as a model, we systematically evaluated four backbone modifications--D-amino acid substitution, N-methylation, -methylation, and azapeptide incorporation. Each modification produced distinct outcomes in synthesis, conformation, proteolytic stability, and receptor pharmacology. While D- and N-methyl substitutions yielded high stability, they compromised receptor binding and in vivo function. In contrast, the azapeptide analogue maintained native-like affinity and physiological activity while achieving an enhanced stability profile. These findings highlight the need to balance stability and function in peptide design and position azapeptides as an underexplored class with strong therapeutic potential. More broadly, this study establishes a framework for systematic, data-driven peptide design and optimization.

biochemistry↗

bioRxiv2025

A deep reinforcement learning platform for antibiotic discovery

Antimicrobial resistance (AMR) is projected to cause up to 10 million deaths annually by 2050, underscoring the urgent need for new antibiotics. Here we present ApexAmphion, a deep-learning framework for de novo design of antibiotics that couples a 6.4-billion-parameter protein language model with reinforcement learning. The model is first fine-tuned on curated peptide data to capture antimicrobial sequence regularities, then optimised with proximal policy optimization against a composite reward that combines predictions from a learned minimum inhibitory concentration (MIC) classifier with differentiable physicochemical objectives. In vitro evaluation of 100 designed peptides showed low MIC values (nanomolar range in some cases) for all candidates (100% hit rate). Moreover, 99 our of 100 compounds exhibited broad-spectrum antimicrobial activity against at least two clinically relevant bacteria. The lead molecules killed bacteria primarily by potently targeting the cytoplasmic membrane. By unifying generation, scoring and multi-objective optimization with deep reinforcement learning in a single pipeline, our approach rapidly produces diverse, potent candidates, offering a scalable route to peptide antibiotics and a platform for iterative steering toward potency and developability within hours.

molecular biology↗

bioRxiv2025

Molecular response to the non-lytic peptide bac7 (1-35) triggers disruption of Klebsiella pneumoniae biofilm

Klebsiella pneumoniae is becoming increasingly difficult to treat as multidrug-resistant (MDR) strains become more prevalent. The formation of biofilm heightens this threat by embedding bacterial cells in a polysaccharide-rich matrix that limits antibiotic penetration. Here we dissect the anti-biofilm bovine host-defense cathelicidin peptide fragment bac7 (1-35), exploring its anti-biofilm mechanism, evaluating its ability to curb dissemination of hypervirulent K. pneumoniae, and testing its breadth of activity against diverse clinical isolates. Transcriptomic profiling revealed that bac7(1-35) simultaneously compromises the bacterial membrane and inhibits ribosomal function, a dual assault that precipitates rapid biofilm collapse and blocks bacterial spread. Further, the peptide eradicated biofilms produced by the strongest MDR clinical isolates in the Multidrug-Resistant Organism Repository and Surveillance Network (MRSN) diversity panel. Although bac7 (1-35) kills bacterial cells via a cytosolic mechanism, membrane interaction profiles varied among MRSN isolates, correlating with differential peptide translocation. In a delayed-treatment murine skin-abscess model, bac7 (1-35) halted in vivo dissemination of the hypervirulent strain NTUH-K2044. Collectively, these results delineate a multifaceted mode of action for bac7 (1-35) and underscore its therapeutic promise against biofilm-associated MDR K. pneumoniae infections. Author SummaryKlebsiella pneumoniae is a top-priority pathogen for new therapies, with many strains already approaching pan-drug resistant status. Biofilm formation further complicates treatment, yet biofilm-active therapeutics have not reached the clinic, in part because we still lack a detailed understanding of how to disrupt these impenetrable structures. Antimicrobial peptides are promising candidates and have shown biofilm-disruption potential. Here we provide mechanistic insight into how a host defense peptide dismantles pre-formed K. pneumoniae biofilms. We find that the peptides dual targeting of bacterial membranes and ribosomes triggers dispersal from the biofilm state and concomitantly downregulates factors required for surface attachment and extracellular matrix production. This mechanism involves a protein that, to our knowledge, has not been characterized in K. pneumoniae. Our findings reveal a switch that can be leveraged to reprogram biofilm maintenance toward dispersal in K. pneumoniae, advancing the path to peptide-based antibiofilm therapeutics.

microbiology↗