RESUMO
Background: Pyrazinamide is one of four first-line antibiotics used to treat tuberculosis; however, antibiotic susceptibility testing for pyrazinamide is challenging. Resistance to pyrazinamide is primarily driven by genetic variation in pncA, encoding an enzyme that converts pyrazinamide into its active form. Methods: We curated a dataset of 664 non-redundant, missense amino acid mutations in PncA with associated high-confidence phenotypes from published studies and then trained three different machine-learning models to predict pyrazinamide resistance. All models had access to a range of protein structural-, chemical- and sequence-based features. Results: The best model, a gradient-boosted decision tree, achieved a sensitivity of 80.2% and a specificity of 76.9% on the hold-out test dataset. The clinical performance of the models was then estimated by predicting the binary pyrazinamide resistance phenotype of 4027 samples harbouring 367 unique missense mutations in pncA derived from 24â231 clinical isolates. Conclusions: This work demonstrates how machine learning can enhance the sensitivity/specificity of pyrazinamide resistance prediction in genetics-based clinical microbiology workflows, highlights novel mutations for future biochemical investigation, and is a proof of concept for using this approach in other drugs.
RESUMO
A major challenge in creating universal influenza vaccines is to focus immune responses away from the immunodominant, variable head region of hemagglutinin (HA-head) and toward the evolutionarily conserved stem region (HA-stem). Here we introduce an approach to control antigen orientation via site-specific insertion of aspartate residues that facilitates antigen binding to alum. We demonstrate the generalizability of this approach with antigens from Ebola, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and influenza viruses and observe enhanced neutralizing antibody responses in all cases. We then reorient an H2 HA in an 'upside-down' configuration to increase the exposure and immunogenicity of HA-stem. The reoriented H2 HA (reoH2HA) on alum induced stem-directed antibodies that cross-react with both group 1 and group 2 influenza A subtypes. Electron microscopy polyclonal epitope mapping (EMPEM) revealed that reoH2HA (group 1) elicits cross-reactive antibodies targeting group 2 HA-stems. Our results highlight antigen reorientation as a generalizable approach for designing epitope-focused vaccines.
RESUMO
Antifolates, folate analogs that inhibit vitamin B9 (folic acid)-using cellular enzymes, have been used over several decades for the treatment of cancer and inflammatory diseases. Cellular uptake of the antifolates in clinical use occurs primarily via widely expressed facilitative membrane transporters. More recently, human folate receptors (FRs), high affinity receptors that transport folate via endocytosis, have been proposed as targets for the specific delivery of new classes of antifolates or folate conjugates to tumors or sites of inflammation. The development of specific, FR-targeted antifolates would be accelerated if additional biophysical data, particularly structural models of the receptors, were available. Here we describe six distinct crystallographic models that provide insight into biological trafficking of FRs and distinct binding modes of folate and antifolates to these receptors. From comparison of the structures, we delineate discrete structural conformations representative of key stages in the endocytic trafficking of FRs and propose models for pH-dependent conformational changes. Additionally, we describe the molecular details of human FR in complex with three clinically prevalent antifolates, pemetrexed (also Alimta), aminopterin, and methotrexate. On the whole, our data form the basis for rapid design and implementation of unique, FR-targeted, folate-based drugs for the treatment of cancer and inflammatory diseases.
