TL;DR
Marine microbes are a treasure trove of novel drug compounds, but finding them is like searching for a needle in a haystack. This study uses metabolomics (LC-HRMS and NMR) combined with multivariate analysis (PCA) to rapidly screen bacterial extracts from marine invertebrates, identifying promising strains for drug discovery much faster than traditional methods.
What Problem Does This Paper Address?
The discovery of new drugs from marine natural products is hampered by several challenges. Marine invertebrates, while rich sources of bioactive compounds, are often difficult to cultivate and unsustainable to harvest in large quantities. The actual source of many of these compounds is believed to be symbiotic bacteria living within these invertebrates. However, traditional methods for identifying these bacteria and their valuable secondary metabolites are time-consuming. These methods often involve extensive literature searches, genetic screening, and bioactivity testing, followed by laborious bioassay-guided isolation. This process can be inefficient, especially since bacteria that look similar genetically or morphologically can produce vastly different chemical compounds. The authors aim to accelerate this process by using metabolomic tools to quickly assess the chemical diversity of bacterial extracts and prioritize strains with novel and potentially bioactive metabolites.
Methods
The core of this study lies in a multi-faceted analytical workflow designed for rapid strain selection.
- Metabolomic Profiling: Liquid Chromatography-High Resolution Mass Spectrometry (LC-HRMS) was employed to generate detailed chemical fingerprints of 77 bacterial extracts. This technique provides accurate mass-to-charge ratio (m/z) data, allowing for the identification of a wide range of small molecules.
- Data Processing and Clean-up: A custom-developed Excel macro was crucial for processing the large datasets. This macro handled tasks like removing features originating from culture media, combining data from positive and negative ionization modes, and aligning chromatograms.
- Dereplication: An in-house Excel macro, integrated with the AntiMarin database, was used to identify known metabolites within the bacterial extracts. This process, called dereplication, helps to distinguish novel compounds from those already described.
- Multivariate Analysis: Principal Component Analysis (PCA) was applied to the LC-HRMS data. PCA is an unsupervised statistical method that helps to visualize and identify patterns in complex datasets. It was used here to group bacterial strains based on their chemical profiles and to identify outlier strains that exhibited distinct metabolomes.
- Structural Elucidation: Nuclear Magnetic Resonance (NMR) spectroscopy, specifically ¹H and COSY experiments, was used to obtain detailed structural information for selected compounds, confirming functional groups and spin systems, and aiding in the identification of novel metabolites.
Datasets
The study utilized extracts from 77 bacterial isolates obtained from cold water marine invertebrates collected from Orkney, Scotland. These isolates were cultured on six different media (M1, ISP2, oligotrophic, Luria agar, marine agar, and R2A). The chemical profiles of these extracts formed the primary dataset for analysis.
Key Insights and Ideas
- Metabolomics as a Shortcut: The study powerfully demonstrates that metabolomics can significantly fast-track the discovery of novel natural products by providing a chemical snapshot of bacterial extracts. This bypasses the need for extensive bioactivity screening and isolation for every single extract.
- PCA for Chemical Diversity: Principal Component Analysis proved to be an effective tool for visualizing and differentiating bacterial strains based on their complex chemical profiles. Outliers in the PCA plot represent strains with unique or diverse secondary metabolomes, making them prime candidates for further investigation.
- Beyond 16S rRNA: The research highlights that even strains with very similar 16S rRNA sequences can produce distinct secondary metabolites, underscoring the importance of chemical profiling over solely genetic information for drug discovery.
- Combinatorial Approach: The integration of LC-HRMS, multivariate analysis, and dereplication into a comprehensive workflow is a key innovation, allowing for efficient prioritization of strains with both chemical novelty and potential bioactivity.
Key Results
- Strain Prioritization: The study successfully identified three bacterial strains – Bacillus sp. 4117, Rhodococcus sp. ZS402, and Vibrio splendidus strain LGP32 – as having particularly interesting secondary metabolomes and bioactivities, making them suitable for scale-up and further drug discovery efforts.
- Metabolite Identification: Through dereplication, several known metabolites, including pumilacidins (surfactins) from Bacillus sp. 4115, oxyplicacetin from Vibrio splendidus strain LGP32, and xestoaminol C from Rhodococcus sp. ZS402, were identified.
- Novelty Indicated: A significant percentage of detected features remained unidentified after dereplication (e.g., 72% in Rhodococcus sp. ZS402), suggesting the presence of novel compounds within these bacterial isolates.
- Workflow Efficiency: The developed workflow enabled the rapid analysis and comparison of 77 bacterial extracts, significantly reducing the time and effort required for strain selection.
Limitations and Future Work
- Database Limitations: The accuracy of dereplication is inherently limited by the comprehensiveness of existing natural product databases. The study acknowledges that a significant portion of detected features remained unidentified, indicating the need for continuous database expansion.
- Isomer Separation: While LC-HRMS provides high resolution, differentiating between structural isomers that elute at the same retention time can still be challenging.
- Instability: The study noted that further investigation into Psychrobacter sp. was halted due to observed instability in metabolite production after freezing and thawing, highlighting the need to assess strain stability for long-term drug discovery programs.
- Future Work: The authors suggest that this metabolomics-driven approach can be further refined and applied to a broader range of marine microbial symbionts. The identified novel compounds warrant further isolation, structural elucidation, and biological evaluation to assess their therapeutic potential. The integration of bioactivity data with metabolomic profiles is crucial for guiding drug discovery efforts.
Summary updated