Comprehensive 2D-Quantitative Structure-Activity Relationship Study on Monobactam Analogues Against Gram-Negative...

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Antimicrobial Drug Discovery

Background:

• Multidrug-resistant gram-negative bacteria (GNB) pose a severe threat to human health.
• Monobactam antibiotics show efficacy against these resistant infections.

Purpose of the Study:

• To develop a predictive two-dimensional quantitative structure-activity relationship (2D-QSAR) model for rational design of novel monobactams.
• To target infections caused by multidrug-resistant (MDR) gram-negative bacteria (GNB), specifically *Escherichia coli* (Eco) and *Klebsiella pneumoniae* (Kpn).

Main Methods:

• Utilized kernel partial least squares regression (KPLS) algorithm.
• Constructed a 2D-QSAR model based on 65 known monobactams against *Eco* and *Kpn* strains.
• Validated the model using 34 external monobactams from the lab.

Main Results:

• Achieved high performance for *Eco* and *Kpn* KPLS models with RMSE: 0.681/0.596, R²: 0.946/0.882, Q²: 0.922/0.877, and RMSU: 0.625/0.593.
• External validation showed strong correlation between experimental and predicted values (R²: 0.878 for *Eco*, 0.871 for *Kpn*).

Conclusions:

• The developed and verified 2D-QSAR models offer a robust strategy for designing new monobactams.
• These novel monobactams are intended to be effective against MDR gram-negative bacterial infections.
• The study facilitates the rational design of next-generation antibiotics to combat antimicrobial resistance.