Inventa: A computational tool to discover structural novelty in natural extracts libraries

Chemical characteristics vectors map the chemical space of natural biomes from untargeted mass spectrometry data

Untargeted metabolomics can comprehensively map the chemical space of a biome, but is limited by low annotation rates (< 10%). We used chemical characteristics vectors, consisting of molecular fingerprints or chemical compound classes, predicted from mass spectrometry data, to characterize compounds and samples. These chemical characteristics vectors (CCVs) estimate the fraction of compounds with specific chemical properties in a sample. Unlike the aligned MS1 data with intensity information, CCVs incorporate the chemical properties of compounds, allowing chemical annotation to be used for sample comparison. Thus, we identified compound classes differentiating biomes, such as ethers which are enriched in environmental biomes, while steroids enriched in animal host-related biomes. In biomes with greater variability, CCVs revealed key clustering compound classes, such as organonitrogen compounds in animal distal gut and lipids in animal secretions. CCVs thus enhance the interpretation of untargeted metabolomic data, providing a quantifiable and generalizable understanding of the chemical space of natural biomes.

Self-supervised learning of molecular representations from millions of tandem mass spectra using DreaMS

Characterizing biological and environmental samples at a molecular level primarily uses tandem mass spectroscopy (MS/MS), yet the interpretation of tandem mass spectra from untargeted metabolomics experiments remains a challenge. Existing computational methods for predictions from mass spectra rely on limited spectral libraries and on hard-coded human expertise. Here we introduce a transformer-based neural network pre-trained in a self-supervised way on millions of unannotated tandem mass spectra from our GNPS Experimental Mass Spectra (GeMS) dataset mined from the MassIVE GNPS repository. We show that pre-training our model to predict masked spectral peaks and chromatographic retention orders leads to the emergence of rich representations of molecular structures, which we named Deep Representations Empowering the Annotation of Mass Spectra (DreaMS). Further fine-tuning the neural network yields state-of-the-art performance across a variety of tasks. We make our new dataset and model available to the community and release the DreaMS Atlas-a molecular network of 201 million MS/MS spectra constructed using DreaMS annotations.

Strategies for Actinobacteria Isolation, Cultivation, and Metabolite Production that Are Biologically Important

Novel antimicrobial agents are urgently needed to combat antimicrobial resistance from multidrug-resistant organisms. Actinobacteria are key sources of bioactive metabolites with diverse biological activities. Despite their contributions to drug discovery, the process from strain identification to drug manufacturing faces many challenges, especially the rediscovery of known compounds. Recent technological and scientific advancements have accelerated drug development. Efforts to isolate and screen rare actinobacterial species could yield novel bioactive compounds. This review summarizes techniques for selectively isolating rare actinobacteria, improving bioactive metabolite production, and discovering potential strains. Notably, new genomic strategies and new discoveries regarding spectroscopic signature-based bioactive natural products containing specific structural motifs are also discussed. Furthermore, this review updates the compounds derived from rare actinobacteria and their biological applications.

Toward an integrated omics approach for plant biosynthetic pathway discovery in the age of AI

Elucidating plant biosynthetic pathways is key to advancing a sustainable bioeconomy by enabling access to complex natural products through synthetic biology. Despite progress from genomic, transcriptomic, and metabolomic approaches, much multiomics data remain underutilized. This review highlights state-of-the-art multiomics strategies for discovering plant biosynthetic pathways, addressing challenges in data acquisition and interpretation with emerging computational tools. We propose an integrated workflow combining molecular networking, reaction pair analysis, and gene expression patterns to enhance data utilization. Additionally, artificial intelligence (AI)-driven approaches promise to revolutionize pathway discovery by streamlining data analysis and validation. Integrating multiomics data, chemical insights, and advanced algorithms can accelerate understanding of plant metabolism and bioengineering valuable natural products efficiently.

High-Resolution Isolation of Natural Products: Efficient Combination of Dryload Injection and HPLC Gradient Transfer

The isolation of pure compounds from complex extracts is a crucial step in natural products (NPs) research. Historically, this process has been recognized to be slow and laborious. However, significant advancements have been made in isolation methods. Efficient separation conditions can now be efficiently determined at the analytical scale using high- or ultra-high-performance liquid chromatography and transferred to the preparative scale by chromatographic calculations. This ensures consistent selectivity at both scales, offering precise separation predictions. High-resolution chromatographic conditions at the preparative scale can be achieved through optimized sample preparation and dry load sample introduction. Monitoring chromatographic traces issued from ultraviolet (UV), mass spectrometry (MS), and/or universal detectors such as evaporative light-scattering detectors (ELSD) can precisely guide the isolation or trigger the collection of specific NPs with diverse structural scaffolds. These approaches can be applied across different scales depending on the amounts of NPs targeted for isolation. This chapter presents a detailed description of an isolation protocol, using the aforementioned methodologies.

Advanced high-resolution chromatographic strategies for efficient isolation of natural products from complex biological matrices: from metabolite profiling to pure chemical entities

The isolation of pure compounds from extracts represents a key step common to all investigations of natural product (NP) research. Isolation methods have gone through a remarkable evolution. Current approaches combine powerful metabolite profiling methods for compounds annotation with omics mining results and/or bioassay for bioactive NPs/biomarkers priorisation. Targeted isolation of prioritized NPs is performed using high-resolution chromatographic methods that closely match those used for analytical profiling. Considerable progress has been made by the introduction of innovative stationary phases providing remarkable selectivity for efficient NPs isolation. Today, efficient separation conditions determined at the analytical scale using high- or ultra-high-performance liquid chromatography can be optimized via HPLC modelling software and efficiently transferred to the semi-preparative scale by chromatographic calculation. This ensures similar selectivity at both the analytical and preparative scales and provides a precise separation prediction. High-resolution conditions at the preparative scale can notably be granted using optimized sample preparation and dry load sample introduction. Monitoring by ultraviolet, mass spectrometry, and or universal systems such as evaporative light scattering detectors and nuclear magnetic resonance allows to precisely guide the isolation or trigger the collection of specific NPs with different structural scaffolds. Such approaches can be applied at different scales depending on the amounts of NPs to be isolated. This review will showcase recent research to highlight both the potential and constraints of using these cutting-edge technologies for the isolation of plant and microorganism metabolites. Several strategies involving their application will be examined and critically discussed.

Graphical abstract

Comprehensive mass spectrometric metabolomic profiling of a chemically diverse collection of plants of the Celastraceae family

Natural products exhibit interesting structural features and significant biological activities. The discovery of new bioactive molecules is a complex process that requires high-quality metabolite profiling data to properly target the isolation of compounds of interest and enable their complete structural characterization. The same metabolite profiling data can also be used to better understand chemotaxonomic links between species. This Data Descriptor details a dataset resulting from the untargeted liquid chromatography-mass spectrometry metabolite profiling of 76 natural extracts of the Celastraceae family. The spectral annotation results and related chemical and taxonomic metadata are shared, along with proposed examples of data reuse. This data can be further studied by researchers exploring the chemical diversity of natural products. This can serve as a reference sample set for deep metabolome investigation of this chemically rich plant family.

Integration of Wnt-inhibitory activity and structural novelty scoring results to uncover novel bioactive natural products: new Bicyclo[3.3.1]non-3-ene-2,9-diones from the leaves of Hymenocardia punctata

In natural products (NPs) research, methods for the efficient prioritization of natural extracts (NEs) are key for discovering novel bioactive NPs. In this study a biodiverse collection of 1,600 NEs, previously analyzed by UHPLC-HRMS2 metabolite profiling was screened for Wnt pathway regulation. The results of the biological screening drove the selection of a subset of 30 non-toxic NEs with an inhibitory IC50 ≤ 5 μg/mL. To increase the chance of finding structurally novel bioactive NPs, Inventa, a computational tool for automated scoring of NEs based on structural novelty was used to mine the HRMS2 analysis and dereplication results. After this, four out of the 30 bioactive NEs were shortlisted by this approach. The most promising sample was the ethyl acetate extract of the leaves of Hymenocardia punctata (Phyllanthaceae). Further phytochemical investigations of this species resulted in the isolation of three known prenylated flavones (3, 5, 7) and ten novel bicyclo[3.3.1]non-3-ene-2,9-diones (1, 2, 4, 6, 8-13), named Hymenotamayonins. Assessment of the Wnt inhibitory activity of these compounds revealed that two prenylated flavones and three novel bicyclic compounds showed interesting activity without apparent cytotoxicity. This study highlights the potential of combining Inventa's structural novelty scores with biological screening results to effectively discover novel bioactive NPs in large NE collections.

Differential Volatile Organic Compound Expression in the Interaction of Daldinia eschscholtzii and Mycena citricolor

Fungi exhibit a wide range of ecological guilds, but those that live within the inner tissues of plants (also known as endophytes) are particularly relevant due to the benefits they sometimes provide to their hosts, such as herbivory deterrence, disease protection, and growth promotion. Recently, endophytes have gained interest as potential biocontrol agents against crop pathogens, for example, coffee plants (Coffea arabica). Published results from research performed in our laboratory showed that endophytic fungi isolated from wild Rubiaceae plants were effective in reducing the effects of the American leaf spot of coffee (Mycena citricolor). One of these isolates (GU11N) from the plant Randia grandifolia was identified as Daldinia eschscholtzii (Xylariales). Its antagonism mechanisms, effects, and chemistry against M. citricolor were investigated by analyzing its volatile profile alone and in the presence of the pathogen in contactless and dual culture assays. The experimental design involved direct sampling of agar plugs in vials for headspace (HS) and headspace solid-phase microextraction (HS-SPME) gas chromatography-mass spectrometry (GC-MS) analysis. Additionally, we used ultrahigh-performance liquid chromatography coupled to high-resolution mass spectrometry (UHPLC-HRMS/MS) to identify nonvolatile compounds from organic extracts of the mycelia involved in the interaction. Results showed that more volatile compounds were identified using HS-SPME (39 components) than those by the HS technique (13 components), sharing only 12 compounds. Statistical tests suggest that D. eschscholtzii inhibited the growth of M. citricolor through the release of VOCs containing a combination of 1,8-dimethoxynapththalene and terpene compounds affecting M. citricolor pseudopilei. The damaging effects of 1,8-dimethoxynaphthalene were corroborated in an in vitro test against M. citricolor pseudopilei; scanning electron microscopy (SEM) photographs confirmed structural damage. After analyzing the UHPLC-HRMS/MS data, a predominance of fatty acid derivatives was found among the putatively identified compounds. However, a considerable proportion of features (37.3%) remained unannotated. In conclusion, our study suggests that D. eschscholtzii has potential as a biocontrol agent against M. citricolor and that 1,8-dimethoxynaphthalene contributes to the observed damage to the pathogen's reproductive structures.

Prioritization of Microorganisms Isolated from the Indian Ocean Sponge Scopalina hapalia Based on Metabolomic Diversity and Biological Activity for the Discovery of Natural Products

Despite considerable advances in medicine and technology, humanity still faces many deadly diseases such as cancer and malaria. In order to find appropriate treatments, the discovery of new bioactive substances is essential. Therefore, research is now turning to less frequently explored habitats with exceptional biodiversity such as the marine environment. Many studies have demonstrated the therapeutic potential of bioactive compounds from marine macro- and microorganisms. In this study, nine microbial strains isolated from an Indian Ocean sponge, Scopalina hapalia, were screened for their chemical potential. The isolates belong to different phyla, some of which are already known for their production of secondary metabolites, such as the actinobacteria. This article aims at describing the selection method used to identify the most promising microorganisms in the field of active metabolites production. The method is based on the combination of their biological and chemical screening, coupled with the use of bioinformatic tools. The dereplication of microbial extracts and the creation of a molecular network revealed the presence of known bioactive molecules such as staurosporin, erythromycin and chaetoglobosins. Molecular network exploration indicated the possible presence of novel compounds in clusters of interest. The biological activities targeted in the study were cytotoxicity against the HCT-116 and MDA-MB-231 cell lines and antiplasmodial activity against Plasmodium falciparum 3D7. Chaetomium globosum SH-123 and Salinispora arenicola SH-78 strains actually showed remarkable cytotoxic and antiplasmodial activities, while Micromonospora fluostatini SH-82 demonstrated promising antiplasmodial effects. The ranking of the microorganisms as a result of the different screening steps allowed the selection of a promising strain, Micromonospora fluostatini SH-82, as a premium candidate for the discovery of new drugs.