Cyanobacteria from Caribbean coral reefs produce a group of natural substances that effectively inhibit digestive enzymes. This was discovered by an international research collaboration led by Marburg-based pharmacist Professor Dr. Raphael Reher and Tübingen-based biochemist Dr. Daniel Petras, using a newly developed combination of methods to systematically identify the large number of potential active pharmaceutical agents from complex environmental samples. The team reports its findings in the scientific journal Nature Communications.
Microorganisms produce numerous metabolites whose usefulness for human purposes remains to be discovered. To determine the mode of action of such substances and their molecular structure, they must first be obtained in pure form. "Together with the complete structural elucidation, this process usually takes months to years," explains senior author Daniel Petras, who heads a Junior Research Group at the Cluster of Excellence "Controlling Microbes to Fight Infections" (CMFI) at the University of Tübingen. "This time-consuming process represents a significant bottleneck for the systematic search for new cures from natural products," adds first author Raphael Reher, who recently began teaching pharmaceutical biology and biotechnology at Marburg's Department of Pharmacy.
To accelerate the identification of pharmaceutically promising substances, the scientists involved combined several advanced methods. In this way, the team succeeded in combining two goals: On the one hand, the effective separation of molecules leads to sorting apart the components of the crude extract; on the other hand, the interactions of binding partners can be traced.
To this end, the research group made particular use of a novel approach in mass spectrometry: Whereas in mass spectrometry proteins are usually decomposed into their constituent parts, native mass spectrometry is concerned precisely with preserving the spatial structure of the protein, because this is what determines its function.
Petras and Reher developed the experimental setup while they were still postdoctoral fellows together at the University of California San Diego in the United States. "The project is a good example of how complex cross-border research collaboration should work," Petras points out. "I'm very proud of our international team collaborated across disciplines, such as natural products research, analytical chemistry and informatics, to make the study possible."
"The fact that we have now been able to establish and further develop the technology here, thanks to the excellent infrastructure and new collaborations in Tübingen and Marburg, is an important point. This has led to the successful completion of the study and will be the basis for our future research," adds Reher.
The team's first application was to look for new inhibitors of proteases - enzymes that break down proteins. "Such protease inhibitors are used to treat cancer and viral infections such as SARS-CoV-2, HIV and hepatitis C, as well as diabetes and high blood pressure," explains Reher.
The research group chose cyanobacteria from the sea, which form biofilms, as a source of bioactive natural compounds. The team used the protein-splitting enzyme chymotrypsin as a target, against which they sought inhibitors from the bacterial community. "Using our approach, we immediately identified 30 compounds that bind to chymotrypsin," Reher reports. When checked against specific databases, it turned out that most of the molecules found were previously unknown. This led to the targeted isolation and structural elucidation of a family of new, highly effective inhibitors. "We call these rivulariapeptolides after the group of bacteria from which the substances originate," Petras tells us.
As the authors write, they believe their approach will facilitate the search for a variety of molecular interactions in complex mixtures. "This will not only benefit drug discovery and ecological studies, but could also provide training data for learning methods based on artificial intelligence."
In addition to the research groups of Reher and Petras and other scientists from the universities of Marburg and Tübingen, researchers from Spain, Puerto Rico, China and the United States participated in the research. The German Research Foundation (DFG) and other funders provided financial support to participating scientists.
Participating CMFI Principal Investigators:
Reher R, Aron AT, Fajtová P, Stincone P, Wagner B, Pérez-Lorente AI, Liu C, Shalom IYB, Bittremieux W, Wang M, Jeong K, Matos-Hernandez ML, Alexander KL, Caro-Diaz EJ, Naman CB, Scanlan JHW, Hochban PMM, Diederich WE, Molina-Santiago C, Romero D, Selim KA, Sass P, Brötz-Oesterhelt H, Hughes CC, Dorrestein PC, O’Donoghue AJ, Gerwick WH, Petras D, Native metabolomics identifies the rivulariapeptolide family of protease inhibitors. Nature Communications 13, 4619. (2022)
(Source: Press Release University of Marburg/Translation Leon Kokkoliadis)