De Novo Production of Metabolites by Fungal Co-culture of

pure strain culture but were present only in the co-culture (red dot in Figure 2B). ...... Ernest Oppong-Danquah , Delphine Parrot , Martina Blüm...
0 downloads 0 Views 4MB Size
Article pubs.acs.org/jnp

De Novo Production of Metabolites by Fungal Co-culture of Trichophyton rubrum and Bionectria ochroleuca Samuel Bertrand,† Olivier Schumpp,‡ Nadine Bohni,† Michel Monod,§ Katia Gindro,‡ and Jean-Luc Wolfender*,† †

School of Pharmaceutical Sciences, EPGL, University of Geneva, University of Lausanne, Quai Ernest-Ansermet 30, CH-1211 Geneva 4, Switzerland ‡ Swiss Federal Research Station Agroscope Changins-Wädenswil, Route de Duillier, P.O. Box 1012, CH-1260 Nyon, Switzerland § Department of Dermatology and Venereology, Laboratory of Mycology, Centre Hospitalier Universitaire Vaudois, CH-1011 Lausanne, Switzerland S Supporting Information *

ABSTRACT: The co-cultivation of fungi has recently been described as a promising strategy to induce the production of novel metabolites through possible gene activation. A large screening of fungal co-cultures in solid media has identified an unusual long-distance growth inhibition between Trichophyton rubrum and Bionectria ochroleuca. To study metabolite induction in this particular fungal interaction, differential LCMS-based metabolomics was performed on pure strain cultures and on their co-cultures. The comparison of the resulting fingerprints highlighted five de novo induced compounds, which were purified using software-oriented semipreparative HPLC-MS. One metabolite was successfully identified as 4″hydroxysulfoxy-2,2″-dimethylthielavin P (a substituted trimer of 3,5-dimethylorsellinic acid). The nonsulfated form, as well as three other related compounds, were found in the pure strain culture of B. ochroleuca.

N

tions.4−6,8,9,11 Moreover, stress-induced molecules exhibited specific antimicrobial,12−17 anticancer,18,19 and phytotoxic activities.14 To detect the induction of metabolite biosynthesis in microorganism co-cultures within the complex fungal metabolome, sensitive high-resolution (HR) techniques mainly based on mass spectrometry are required. Recently induction phenomena of protein and chemical substances in solid culture were demonstrated in bacterial co-culture through state-of-theart MALDI-MS20−22 or nanoDESI-MS23 imaging approaches. To further identify induced secondary metabolites and assess their bioactivity, analytical strategies based on LC were employed, which allow for targeted microisolation of selected NPs.24 Therefore, metabolite-profiling technology, such as ultra-high-pressure liquid chromatography coupled to electrospray ionization and time-of-flight mass spectrometry (UHPLC-TOF-MS), provided a fast and efficient determination of metabolome modification in complex matrices such as microbial extracts.25,26 When applied to fungal co-cultures, this approach allowed an efficient comparison of pure strain cultures versus their corresponding co-culture metabolome fingerprints to highlight de novo induced biomarkers14,27 and

atural products (NPs) are a historical source of valuable lead medicinal compounds,1 of which microbial compounds represent an important part. However, the attractiveness of NPs is diminished because of the difficulties involved in working with complex mixtures2 and the continual rediscovery of the same bioactive chemical structures in pharmacological screening despite the existence of dereplication processes.3 To generate chemical diversity, several approaches such as nontargeted metabolic engineering, epigenetic modification or elicitation, and the production of unnatural−natural scaffolds have recently been developed.4−7 In the case of microorganisms the induction of novel metabolites can be achieved by the activation of cryptic biosynthetic pathways dedicated to the production of secondary metabolites.4−6,8,9 Such approaches take into account the tremendous metabolite diversity revealed by genome sequencing programs and aim to reduce the continual rediscovery of NPs. One of these strategies consists of the co-cultivation of two microorganisms in a single confined environment to induce the production of new natural products via possible interspecies crosstalk.6,8 Co-cultivation of bacteria or fungi exploits the fact that in their environment (such as soil, rhizospheres, plants, mucosal membranes, and guts) microorganisms are in constant interaction.10 These interactions lead to the activation of complex regulatory mechanisms, which results in the biosynthesis of highly diverse NPs such as pheromones, defense molecules, and metabolites involved in symbiotic associa© 2013 American Chemical Society and American Society of Pharmacognosy

Received: March 27, 2013 Published: June 4, 2013 1157

dx.doi.org/10.1021/np400258f | J. Nat. Prod. 2013, 76, 1157−1165

Journal of Natural Products

Article

solid culture were generated to provide enough material for purification of the induced metabolites. The long-distance interaction morphological pattern was reproduced in all replicates. Identification of Induced Metabolites. Because chemical induction was reported to occur mainly in the confrontation zone, the solid media of that specific zone was excised for further chemical analyses. The results were compared to those obtained from the cultures of the corresponding pure strains grown in parallel under the same conditions. The excised pieces of agar medium were extracted by sonication using a dichloromethane−methanol−water mixture according to our previously reported protocol.27 The extracts were then subjected to reverse-phase solid-phase extraction to remove highly nonpolar compounds and reduce carryover effects in LC-MS. Each extract was subjected to UHPLC-TOF-MS fingerprinting.27 This procedure took advantage of the high resolution and fast separation capacities of UHPLC, as well as the HR detection offered by the TOF-MS detector.34 The UHPLC-TOF-MS fingerprints generated 2D ion maps (tR × m/z), in which all features associated with fungal metabolites were resolved in both the LC and MS dimension with high repeatability. Figure 2A shows the negative ionization (NI) 2D ion maps obtained for the culture of the pure strains and the co-culture. All features (tR × m/z) detected in the fingerprints of all replicates were extracted using an automated peak picking procedure that was optimized to retrieve only relevant chromatogram ion traces with a relatively low signal-to-noise threshold to keep as much relevant information as possible.35 Because the goal was to localize and identify de novo induced compounds, intensities of signals from metabolites present in the co-cultures were kept to compare with the intensities of the corresponding pure strains. Only the most intense peaks (peak area >2000 counts·min) that occurred at least eight times in the nine replicates were considered. All of these features were displayed in reconstituted 2D ion maps (Figure 2B). This representation of fingerprinting data enabled a clear distinction of features that were not present in the pure strain culture but were present only in the co-culture (red dot in Figure 2B). These features were verified directly in the raw data by the extraction of their corresponding selected ion traces (Figure 2C). On the basis of this data treatment, from more than 600 features recorded in the initial fingerprints, only five compounds that displayed intense MS signals and were induced de novo in the zone of interaction were highlighted and considered for further metabolite identification (Figure 2, Figure S1 and Table S1). Dereplication of the Induced Metabolites. To obtain preliminary information about the selected metabolites, a full process of dereplication was undertaken. The Dictionary of Natural Products36 (DNP) was used to search for positive matches of each of the selected masses.27 The high mass accuracy molecular weights were deduced by taking into account the possible presence of the most probable adducts in the LC-MS conditions used ([M − H]− and [M + HCO2]− in NI).25 To improve the chances of finding a matching candidate, a relatively large mass-accuracy tolerance (