A Quantitative Model to Estimate Drug Resistance in Pathogens

Area of Science:

• Medical Mycology
• Computational Biology
• Drug Discovery

Background:

• Pneumocystis pneumonia (PCP) is an opportunistic infection caused by Pneumocystis fungi.
• Current treatments like trimethoprim/sulfamethoxazole target folic acid biosynthesis enzymes (DHFR, DHPS).
• Emerging resistance of *Pneumocystis jirovecii* to these drugs is a growing concern due to enzyme mutations.

Purpose of the Study:

• To develop a quantitative model for predicting drug binding affinity changes to mutated enzymes.
• To assess the model's accuracy and generalizability across different drug-target pairs and organisms.

Main Methods:

• Utilized evolutionary information and amino acid covariance analysis.
• Developed a predictive model for inhibitor binding affinity (Ki, IC50) to mutated proteins.
• Validated the model using experimental data for *Pneumocystis jirovecii* and *Staphylococcus aureus*.

Main Results:

• The developed model accurately predicts changes in inhibitor binding affinity upon enzyme mutations.
• Predicted binding affinity changes strongly correlate with experimentally measured data.
• The model demonstrated high performance across various drug/target pairs and organisms, including PjDHFR/TMP, PjDHPS/SMX, and SaDHFR/TMP.

Conclusions:

• The predictive model is a valuable tool for evaluating drug resistance in emerging pathogen variants.
• The model's approach can be extended to other drug targets and microbial species.
• This work contributes to understanding and combating antimicrobial resistance in opportunistic infections.