Structural Elucidation by Composition Formula Predictor Software Using MSn Data

Importance of the Topic

Determining the precise molecular formula and structure of pharmaceutical compounds and their impurities is critical for drug safety, efficacy, and regulatory compliance. Combining high-resolution multistage MS data with advanced software tools can greatly accelerate this process while reducing manual interpretation time.

Study Goals and Overview

The study aimed to evaluate a hybrid ion trap–time-of-flight mass spectrometer in conjunction with Composition Formula Predictor software to rapidly assign chemical formulas and structures. Erythromycin A produced by Saccharopolyspora erythraea was used as a model system, including characterization of related impurities.

Methodology and Instrumentation

Erythromycin was dissolved in methanol and injected onto a reversed-phase C18 column under isocratic conditions. Full scan MS and MS2 spectra were acquired in negative electrospray mode. Data were processed using Composition Formula Predictor to propose molecular formulas based on mass accuracy, isotope distribution, double-bond equivalence, hydrogen-to-carbon ratio, and the nitrogen rule.

Used Instrumentation

• Shimadzu LCMS-IT-TOF hybrid ion trap time-of-flight mass spectrometer
• Shimadzu Prominence SIL-20AC autosampler and LC-20AD pumps
• SPD-20A UV detector (200 nm)

Key Results and Discussion

The base ion at m/z 571.3588 was assigned to C37H68N2O13 (erythromycin A) with a mass error of 2.9 ppm. Fragmentation patterns corresponded to sequential losses of the D-desosamine (Area A), L-cladinose (Area B), and the 14-membered lactone ring (Area C). Impurities at m/z 733.4439, 763.4581, and 783.4421 exhibited MS2 spectra consistent with erythromycin A oxime derivatives. An impurity at m/z 761.4375 showed unique fragmentation, indicating a structurally distinct compound that was likely an external contaminant.

Benefits and Practical Applications

Integrating MSn data with predictive software enhances the speed and reliability of formula assignment and structural interpretation, supporting efficient impurity profiling and quality control in pharmaceutical analysis.

Future Trends and Potential Applications

• Automated workflows combining spectral prediction and machine learning
• Broader application to complex natural products and metabolomics
• Real-time structure elucidation in process analytical technology

Conclusion

This approach demonstrates effective use of high-resolution multistage mass spectrometry and formula prediction software for rapid structural elucidation, improving analytical throughput and confidence in impurity identification.