Pathways to improved polyene antimicrobial agents (PIPA)

Natural products are molecules made by plants and microorganisms that have inspired the development of many important pharmaceuticals that we rely on today. For example, soil bacteria, such as Streptomyces from the Actinobacteria family, are particularly prolific in producing antimicrobial agents, which can kill bacterial, fungal, and other microbial pathogens. Many of these natural antimicrobial agents have been widely used to treat life-threatening infectious diseases. However, existing antimicrobial drugs are becoming increasingly ineffective due to antimicrobial resistance (AMR), with microbial pathogens rapidly evolving ways to evade the effects of these compounds. Fungal infections can be particularly problematic, with conditions such aspergillosis and cryptococcosis resulting in millions of life-threatening infections every year. Only very recently, it became apparent that many COVID-19 patients (particularly in India) died of a secondary infection, caused by the "black fungus" mucormycosis. Unfortunately, there are only a small number of antifungal drugs available and very few promising drug candidates in development. Moreover, many antifungals currently in use are largely ineffective against emerging multidrug-resistant fungal pathogens such as Candida auris which pose a major threat to global health. Currently the most effective and widely used antifungals are the polyenes amphotericin B, nystatin and pimaricin, which are all WHO essential medicines. Polyenes are complex natural products derived from Actinobacteria, with similar macrocyclic structures. They show excellent broad spectrum antifungal activity and, unlike other antifungal drugs, resistance to polyenes is less widespread. Despite possessing favourable properties, polyenes suffer from low solubility and exhibit toxicity. This is because polyenes bind to and disrupt the cell wall (membrane) of fungal pathogens, but can also bind to the human cell membrane, leading to toxic effects. Polyene derivatives with reduced toxicity and increased solubility have previously been prepared by chemical synthesis. However, this typically requires laborious and expensive multistep synthetic procedures, which are unsustainable, polluting and too costly to scale-up for manufacture.

Recently, we have sequenced the genomes of several Actinobacteria and discovered genes encoding enzymes (catalysts) that are required for the biosynthesis (assembly) of novel polyenes. We were able to isolate and determine the structure of new polyenes. In addition, we obtained preliminary characterisation for some "tailoring" enzymes that add key functionality during the latter stages of polyene biosynthesis. In this project, we aim to further characterise the new tailoring enzymes, testing them with natural substrates (biosynthetic intermediates) as well as precursors from other pathways to known polyenes (amphotericin B etc.). We will also obtain structures or models which we can use to mutate (engineer) the tailoring enzymes to broaden their substrate scope and facilitate catalysis of alternative tailoring reactions. We then aim to combine the different tailoring enzymes in reactions to create a diverse library of polyene derivatives, which will be tested for antifungal activity, toxicity, and solubility. This will provide a detailed structure-activity relationship (SAR), enabling us to establish which combination of tailoring reactions provides polyenes with the best properties. Initially, we will use isolated enzymes (in vitro) to produce new polyenes for testing. However, we will also develop in vivo engineering approaches to create Actinobacterial strains that can produce the best polyene derivatives in a single-step fermentation. These bio-based approaches, particularly in vivo fermentation, can provide much more sustainable, efficient, and cost-effective routes to the improved polyene antifungal agents that we urgently need to combat the emerging drug-resistant fungal pathogens.

Many essential pharmaceuticals are derived from natural products, including the polyene antifungals amphotericin B, nystatin and pimaricin (all WHO essential medicines). These complex macrolides are assembled by polyketide synthase enzymes, with tailoring enzymes adding key functionality. Although polyenes have excellent antifungal activity, they display significant toxicity and low solubility. Improved polyene derivatives have been synthesised, but this requires multiple steps, extensive use of protecting groups and deleterious reagents/solvents, which is not viable for scalable manufacture. Bioengineering methods providing sustainable and cheaper routes to polyene variants with improved activity and reduced toxicity would be highly desirable. Recently we discovered several novel polyene tailoring enzymes, an Fe(II)/aKG-dependent hydroxylase, amidotransferases and glycosyltransferases (GTs), that are capable of diversifying polyene scaffolds. Here, we aim to further investigate and engineer the new tailoring enzymes, providing more sustainable single-step routes to antifungal agents that are urgently needed to overcome antimicrobial resistance. Initially we will focus on engineering hydroxylases that can accept a broader range of polyene substrates and catalyse alternative halogenation or azidation reactions. We will engineer amidotransferases to improve activity towards alternative glutamine co-substrates producing polyene amides, hydrazides and hydroxamic acids that can be further derivatised. We will use GTs to introduce a wide range of sugars to increase solubility and modulate bioactivity in combination with other tailoring enzymes. We already have X-ray structures for the hydroxylase and aim to obtain structures or models for other selected tailoring enzymes to guide engineering. Polyenes produced in vitro will be subject to biological assays (MICs & toxicity). Finally, we will establish methods for more efficient in vivo production of the best polyenes.