Building Natural Product Libraries Using Quantitative Clade-Based and Chemical Clustering Strategies

Rationally Minimizing Natural Product Libraries Using Mass Spectrometry

Natural product libraries are crucial to drug development, but large libraries drastically increase the time and cost during initial high throughput screens. Here, we developed a method that leverages liquid chromatography-tandem mass spectrometry spectral similarity to dramatically reduce library size, with minimal bioactive loss. This method offers a broadly applicable strategy for accelerated drug discovery with cost reductions, which enable implementation in resource-limited settings.

Persephacin Is a Broad-Spectrum Antifungal Aureobasidin Metabolite That Overcomes Intrinsic Resistance in Aspergillus fumigatus

Fungi pose a persistent threat to humankind with worrying indications that emerging and re-emerging pathogens (e.g., Candida auris, Coccidioides spp., drug-resistant Aspergilli, and more) exhibit resistance to the limited number of approved antifungals. To address this problem, our team is exploring endophytic fungi as a resource for the discovery of new antifungal natural products. The rationale behind this decision is based on evidence that endophytes engage with plants in mutualistic relationships wherein some fungi actively participate by producing chemical defense measures that suppress pathogenic microorganisms. To improve the odds of bioactive metabolite discovery, we developed a new hands-free laser-cutting system capable of generating >50 plant samples per minute that, in turn, enabled our team to prepare and screen large numbers of endophytic fungi. One of the fungal isolates obtained in this way was identified as an Elsinoë sp. that produced a unique aureobasidin analogue, persephacin (1). Some distinctive features of 1 are the absence of both phenylalanine residues combined with the incorporation of a novel amino acid residue, persephanine (9). Compound 1 exhibits potent antifungal effects against a large number of pathogenic yeast (including several clinical C. auris strains), as well as phylogenetically diverse filamentous fungi (e.g., Aspergillus fumigatus). In an ex vivo eye infection model, compound 1 outperformed standard-of-care treatments demonstrating the ability to suppress fluconazole-resistant Candida albicans and A. fumigatus at a concentration (0.1% solution) well below the clinically recommended levels used for fluconazole and natamycin (2% and 5% solutions, respectively). In 3D tissue models for acute dermal and ocular safety, 1 was found to be nontoxic and nonirritating at concentrations required to elicit antifungal activity. Natural product 1 appears to be a promising candidate for further investigation as a broad-spectrum antifungal capable of controlling a range of pathogens that negatively impact human, animal, and plant health.

Location-specific signatures of Crohn's disease at a multi-omics scale

BACKGROUND

Crohn's disease (CD), an inflammatory bowel disease (IBD) subtype, results from pathologic interactions between host cells and its resident gut microbes. CD manifests in both isolated disease locations (ileum or colon) or a combination of locations (ileocolonic). To date, a comprehensive understanding of how isolated CD subtypes influence molecular profiles remains outstanding. To address this, we sought to define CD location signatures by leveraging a large cross-sectional feature set captured from the stool of over 200 IBD patients and healthy controls using metaproteomics, shotgun metagenomics, 16S rRNA sequencing, metabolomic profiling, and host genetics paired with clinical endoscopic assessments.

RESULTS

Neither metagenomic nor host genetics alone distinguished CD location subtypes. In contrast, ileal and colonic CD were distinguished using mass spectrometry-based methods (metabolomics or metaproteomics) or a combined multi-omic feature set. This multi-omic feature set revealed colonic CD was strongly associated with neutrophil-related proteins. Additionally, colonic CD displayed a disease-severity-related association with Bacteroides vulgatus. Colonic CD and ulcerative colitis profiles harbored strikingly similar feature enrichments compared to ileal CD, including neutrophil-related protein enrichments. Compared to colonic CD, ileal CD profiles displayed increased primary and secondary bile acid levels and concomitant shifts in taxa with noted sensitivities such as Faecalibacterium prausnitzii or affinities for bile acid-rich environments, including Gammaproteobacteria and Blautia sp. Having shown robust molecular and microbial distinctions tied to CD locations, we leveraged these profiles to generate location-specific disease severity biomarkers that surpass the performance of Calprotectin.

CONCLUSIONS

When compared using multi-omics features, colonic- and ileal-isolated CD subtypes display striking differences that suggest separate location-specific pathologies. Colonic CD's strong similarity to ulcerative colitis, including neutrophil and Bacteroides vulgatus involvement, is also evidence of a shared pathology for colonic-isolated IBD subtypes, while ileal CD maintains a unique, bile acid-driven profile. More broadly, this study demonstrates the power of multi-omics approaches for IBD biomarker discovery and elucidating the underlying biology. Video Abstract.

Assessing Microbial Metabolic and Biological Diversity to Inform Natural Product Library Assembly

The pressing need for novel chemical matter to support bioactive compound discovery has led natural product researchers to explore a wide range of source organisms and environments. One of the implicit guiding principles behind those efforts is the notion that sampling different environments is critical to accessing unique natural products. This idea was tested by comparing fungi from disparate biomes: aquatic sediments from Lake Michigan (USA) and terrestrial samples taken from the surrounding soils. Matched sets of Penicillium brevicompactum, Penicillium expansum, and Penicillium oxalicum from the two source environments were compared, revealing modest differences in physiological performance and chemical output. Analysis of LC-MS/MS-derived molecular feature data showed no source-dependent differences in chemical richness. High levels of scaffold homogeneity were also observed with 78-83% of scaffolds shared among the terrestrial and aquatic Penicillium spp. isolates. A comparison of the culturable fungi from the two biomes indicated that certain genera were more strongly associated with aquatic sediments (e.g., Trichoderma, Pseudeurotium, Cladosporium, and Preussia) versus the surrounding terrestrial environment (e.g., Fusarium, Pseudogymnoascus, Humicola, and Acremonium). Taken together, these results suggest that focusing efforts on sampling the microbial resources that are unique to an environment may have a more pronounced effect on enhancing the sought-after natural product diversity needed for chemical discovery and screening collections.