Mining of Cryptic Secondary Metabolism in Aspergillus

Aspergillus species are ubiquitous filamentous fungi that can produce industrially important enzymes and a broad spectrum of structurally diversified secondary metabolites including pharmaceutical products like statins and echinocandin-type antibiotics. With the rapid evolvement of sequencing technologies, whole-genome sequences of Aspergillus species are becoming available. The access to the genomic data has made genome mining guided discovery of secondary metabolites (SMs) and linking the known and new products to the related biosynthetic gene clusters (BGCs) possible. However, only a limited number of Aspergillus species have been well studied for chemical discovery purposes. A plethora of new compounds still awaits discovery from the less known species or rare species which represent an untouched natural product arsenal. This PhD project focuses on two poorly studied species A. californicus and A. homomorphus.

The first aim of this PhD study has been to explore new secondary metabolites from the rare species A. californicus for antibacterial and cytotoxic screening. A. californicus was first isolated from Chamise chaparral (Adenostoma fasciculatum) soil in California in 1978. This asexual reproducing and poor sporulating species belongs to section Cavernicolarum of subgenus Nidulantes but resembles typical section Usti species because of its long light brown conidiophores. With only one isolate, this rare species has not been chemically investigated before. Through multiple chromatographic purification steps of the organic extracts, we were able to obtain fifteen compounds including ten previously unknown ones (Chapter 2.1). The structures of the new compounds were elucidated by a combination of techniques such as mass spectrometry, nuclear magnetic resonance, optical rotation, infrared spectroscopy, and electronic circular dichroism. Based on the biosynthetic origins, these compounds were categorized as eleven polyketides, two nonribosomal peptides, one sesterterpene, and one polyketide-nonribosomal peptide hybrid. Among the polyketides, nine octaketides were closely related and thus were proposed to be biosynthesized from the same pathway. The octaketide calidiol A showed moderate activities against methicillin-resistant Staphylococcus aureus
MB5393 with a minimum inhibitory concentration of 48 µg/ml (Chapter 2.1.1 and Appendix 1). Two other polyketides were elucidated as naphthyl derivatives (Chapter 2.1.2 and Appendix 2). The two nonribosomal peptides bear a unique oxepine-pyrimidinone-ketopiperazine (OPK) scaffold that has only been reported from fungal sources so far, particularly 70% from Aspergillus species (Chapter 2.1.3 and Appendixes 3 - 4).

The second aim of this study was to investigate the biosynthetic pathways of newly reported metabolites from this project as well as known structures with uncommon biosynthetic origins (Chapter 2.2). Genome mining tools like antiSMASH were used to predict the BGCs for the selected products. The hybrid compound calipyridone A is comprised of a 2-pyridone moiety and an indole substructure, the latter originating from a tryptophan residue. Its biosynthetic pathway was studied by engineering the host strain A. californicus with CRISPR-Cas9, which was the first genetic manipulating work on this non-model species. The results indicated that the 2-pyridone moiety was formed through aldol condensation without enzymatic ring expansion catalyzed by P450s, distinguishing calipyridone A from the biosynthesis of other 2-pyridone analogs. Besides,
we successfully discovered two new precursors of calipyridone A using a molecular networking strategy and MSMS fragmentation analysis (Chapter 2.2.1 and Appendix 5). A. homomorphus is a protoheterothallic reproducing fungus belonging to section Nigri of subgenus Circumdati. Biosynthetic investigation of homopyrones from this species showed their synthesis to be of mixed biosynthetic origin starting with cinnamoyl-CoA as the starter unit, which was likely a product of uncommon phenylalanine ammonia-lyase in fungi, followed by three extension reactions facilitated by the polyketide synthase to give the aromatic α-pyrones (Chapter 2.2.2 and Appendix 6). Similar types of products are very commonly found in plants but rarely in filamentous fungi. Besides, we also present our effort on attempting to elucidate the biosynthetic pathway of the two OPK compounds, as well as linking of 5 genes, which were initially misproposed as the BGC for the two OPK products, to their encoding metabolites (Chapter 2.2.3 and Appendix 7).

Altogether, this PhD thesis has cast new insights into the secondary metabolism in Aspergillus, in particular in two less-studied species A. californicus and A. homomorphus. Regarding the secondary metabolite discovery, we have characterized fifteen compounds from A. californicus including ten new ones. These compounds include polyketides (PKs), nonribosomal peptides (NRPs), sesterterpene, as well as two unusual hybrid types of natural products. Furthermore, in the biosynthesis investigation, we have succeeded in characterizing the biosynthetic pathway of calipyridone A in A. californicus representing a rare 2-pyridone formation without a P450-catalyzed ring expansion from a tetramic acid intermediate. Besides, the biosynthesis of two yellow homopyrones in A. homomorphus has been demonstrated to start with an unusual cinnamoyl-CoA unit which is rare in fungi. Importantly, our work represents the first genetic manipulations in both A. californicus and A. homomorphus, and thus as a huge benefit sets the scene for further biochemical investigations in these species.