Microbial Databases for Rapid Screening and Dereplication of Microbial Natural Products Using UPLC-QTof-MS Coupled to Novel Informatics Platform

Importance of the Topic

Natural product discovery from microbial sources often generates complex mixtures containing known and unknown metabolites. Rapid dereplication—identifying previously characterized compounds—streamlines drug discovery and metabolomics by focusing efforts on novel entities. The integration of high-resolution mass spectrometry with informatics libraries addresses a key bottleneck: confident compound identification in complex extracts.

Objectives and Study Overview

This study aimed to establish a high-throughput workflow for automated screening and dereplication of marine microbial natural products. Custom and public microbial databases were combined with UPLC-QTof-MS and the UNIFI scientific information platform to accelerate compound annotation directly from crude extracts and fractions.

Methodology

Marine sediment-derived microbes were cultured with XAD-16 resin, extracted in methanol/dichloromethane (1:1), and fractionated on an ACQUITY UPLC BEH C18 column (2.1 × 50 mm, 1.8 µm) at 50 °C. A 4.5 min gradient using 0.1% formic acid in water and acetonitrile (0.8 mL/min) was applied. Data-independent acquisition (DIA) alternated low-energy scans for precursor masses (m/z 50–1800) with high-energy scans for fragment ions in both positive and negative ESI modes.

Used Instrumentation

• Waters ACQUITY UPLC with BEH C18 column
• Waters QTof mass spectrometer
• UNIFI Scientific Information System Software
• Custom microbial natural product library (439 compounds)
• Online libraries: ChemSpider, MarinLit, Marine Drugs, NIST

Main Results and Discussion

A UNIFI library was built by importing .mol files and metadata (formula, retention time, fragments) for 439 known microbial compounds. The platform matched precursor masses and in silico predicted fragments against experimental DIA data. For example, deferoxamine B was confidently annotated by matching 14 high-energy fragments and sub-ppm mass error. Unmatched peaks were processed by the UNIFI discovery toolset: elemental composition calculation, ChemSpider searches (including MarinLit), and fragment matching. An unknown feature at m/z 796.5204 was proposed by connecting 16 matched fragments to a candidate structure.

Benefits and Practical Applications

• Rapid dereplication reduces time and resource expenditure in natural product screening
• High confidence in compound annotation through combined precursor and fragment matching
• Flexible integration of custom and public databases for diverse microbial sources
• Automated workflows suitable for QA/QC, metabolomics, and drug discovery

Future Trends and Applications

Emerging opportunities include expansion of spectral libraries with collision cross section data, integration of ion mobility separations, machine-learning–driven fragment prediction, and direct linking of metabolomic and genomic databases to accelerate discovery of novel bioactive metabolites.

Conclusion

The coupling of high-resolution UPLC-QTof-MS with a customizable informatics platform and extensive microbial databases enables rapid, automated dereplication and discovery workflows. The approach enhances confidence in compound identification and directs efforts toward novel metabolites in complex natural product research.

References

• Allard P.M. et al. Integration of Molecular Networking and In-Silico MS/MS Fragmentation for Natural Products Dereplication. Anal. Chem. 2016;88(6):3317–3323.
• Kurita L.K. and Linington R.G. Connecting Phenotype and Chemotype: High-Content Discovery Strategies for Natural Product Research. J. Nat. Prod. 2015;78:587–596.
• Mullin L. et al. Building a UNIFI Scientific Library for HRMS Screening Experiments. Waters Application Note 720004927EN, 2014.
• ChemSpider. http://www.chemspider.com/ (accessed 2018).