Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Sep 15.
Published in final edited form as: J Am Chem Soc. 2010 Sep 15;132(36):12733–12740. doi: 10.1021/ja1049302

Genome-based Characterization of Two Prenylation Steps in the Assembly of the Stephacidin and Notoamide Anticancer Agents in a Marine-derived Aspergillus sp

Yousong Ding 1, Jeffrey R de Wet 2, James Cavalcoli 2, Shengying Li 1, Thomas J Greshock 3, Kenneth A Miller 3, Jennifer M Finefield 3, James D Sunderhaus 3, Timothy McAfoos 3, Sachiko Tsukamoto 4, Robert M Williams 3,*, David H Sherman 1,*
PMCID: PMC2941195  NIHMSID: NIHMS230649  PMID: 20722388

Abstract

Stephacidin and notoamide natural products belong to a group of prenylated indole alkaloids containing a core bicyclo[2.2.2]diazaoctane ring system. These bioactive fungal secondary metabolites have a range of unusual structural and stereochemical features but their biosynthesis has remained uncharacterized. Herein, we report the first biosynthetic gene cluster for this class of fungal alkaloids based on whole genome sequencing of a marine-derived Aspergillus sp. Two central pathway enzymes catalyzing both normal and reverse prenyltransfer reactions were characterized in detail. Our results establish the early steps for creation of the prenylated indole alkaloid structure and suggest a scheme for the biosynthesis of stephacidin and notoamide metabolites. The work provides the first genetic and biochemical insights for understanding the structural diversity of this important family of fungal alkaloids.

INTRODUCTION

Structurally complex fungal-derived natural products account for a significant number of clinical therapeutics for treatment of human and animal diseases1. Due to emerging appreciation for the high level of biodiversity within this group of eukaryotes, an increasing number of natural products have been isolated from fungal sources and screened for bioactive secondary metabolites2. Recently, a family of fungal-derived prenylated alkaloids has attracted increasing interest for its remarkably diverse bioactivities including insecticidal, antitumor, anthelmintic, calmodulin inhibitory, and antibacterial properties, and intriguing structural features. These natural products are comprised of l-tryptophan, a second cyclic amino acid residue, and one or two isoprene units (Scheme 1)3. The isolation and characterization of two key biosynthetic intermediates, preparaherquamide (1) and premalbrancheamide (2) in the biosynthesis of paraherquamides (3) and malbrancheamides (4)4, 5, respectively, suggest that two amino acid residues are oxidatively condensed to generate cyclo-l-tryptophan-l-proline analog 5 or 6. The tryptophanyl subunit of the dipeptide is subsequently prenylated in a reverse manner to generate compound 7 or 8 (Scheme 1). The bicyclo[2.2.2]diazaoctane core in 1 and 2 possibly arises from an intramolecular Diels-Alder (IMDA) reaction after oxidizing 7 or 8 to form a putative pyrazine-derived azadienophile. However, the detailed understanding of assembly and modification of these biosynthetic building blocks remain highly obscure.

Scheme 1.

Scheme 1

Biosynthetic subunits and putative route to paraherquamide (3) and malbrancheamide (4). Molecules in the boxes have been validated as the building blocks or biosynthetic intermediates based on precursor incorporation studies.

Recently, a group of new prenylated indole alkaloids, the notoamides (A-E, 9-13), were isolated from a marine-derived Aspergillus sp. (Fig. 1a)6, 7. Interestingly, stephacidin A (14) and deoxybrevianamide E (15) were purified from the same fungal strain, indicating the possible role of 15 as a common biosynthetic intermediate7. In 2006, a bimodular non-ribosomal peptide synthetase (NRPS) gene (ftmA) was mined from an A. fumigatus genome sequence, and its heterologous expression led to accumulation of the cyclo-l-tryptophan-l-proline product brevianamide F (16)8. We reasoned that an NRPS with a function coincident with FtmA would be expected for the notoamide biosynthetic pathway , where 16 in this marine-derived Aspergillus sp. is elaborated in an alternative manner compared to A. fumigatus9 that mediates biosynthesis of fumitremorgin. Herein, we report the identification of the first gene cluster for the biosynthesis of this family of unique prenylated alkaloids based on genome mining and biochemical analysis. These studies include a detailed characterization of the elusive deoxybrevianamide E synthase (e.g. reverse prenyltransferase) as well as a second normal prenyltransferase that provide new insights into assembly of the structurally and biologically diverse class of bicyclo[2.2.2]diazaoctane-derived natural products.

Figure 1.

Figure 1

Genetic studies of fungal alkaloids produced in the marine-derived fungus Aspergillus sp. (a). Selected fungal alkaloids isolated from the marine-derived Aspergillus sp. Compound 16 was not reported in the fungal strain but was expected as the direct precursor of compound 15. (b) The notoamide biosynthetic gene cluster derived from complete sequencing and bioinformatic mining of Aspergillus sp. MF297-2 genome.

RESULTS

Localization and analysis of the notoamide gene cluster from a marine-derived Aspergillus sp. through genome mining

The genome of stephacidin- and notoamide-producing marine-derived Aspergillus sp. MF297-2 was sequenced to ~ 15 times coverage of the average published Aspergillus genome size (32.5 Mb) using Roche 454FLX technology (unpublished data). An open reading frame (orf) named notE (Fig. 1b) was identified from the genome sequence using ftmA to probe for homologous genes8. NotE is a presumed bimodular NRPS with adenylation (A)-thiolation (T)-condensation (C)-A-T-C domain organization, and shares 47% amino acid sequence identity with FtmA (see Supplementary Table 1 online). In addition to notE, eighteen other genes were identified in a 42456-bp region of the chromosome encompassed by four overlapping genome assembly nodes (Fig. 1b). At the left end of the gene cluster, the product of orf1 was predicted to be the N-terminus of a capsule polysaccharide biosynthesis protein involved in a primary metabolic pathway. At the right end of the gene cluster, a protein encoded by notR showed 38% sequence identity to the AflJ aflatoxin pathway transcriptional co-activator10. Bioinformatic analysis indicated that NotB and NotI show high similarity to FAD-dependent monooxygenases while NotD is a presumed flavin-dependent oxidoreductase. NotG and NotH show high sequence similarity to fungal CYP450s, both of which might be involved in the formation of the isoprene-derived pyran ring (Scheme 2). Furthermore, NotN and NotO are predicted to function as a dehydrogenase and a short-chain dehydrogenase/reductase, respectively. The notK gene encodes a putative efflux pump, which might specify excretion of alkaloid products from the cell. As with NotR, NotL shares high protein sequence similarity to AflR while NotA is a predicted biosynthetic pathway transcriptional repressor11. These regulators offer opportunities to understand notoamide pathway gene expression, and the potential to manipulate fungal alkaloid production in this unique marine-derived Aspergillus sp12. NotC and NotF, two predicted aromatic prenyltransferases, presumably catalyze the two key prenylation reactions including a first reverse prenyltransfer step leading to 15. NotC shows 50% sequence identity to FtmH (also called FtmPT2) in A. fumigatus while NotF shows the highest identity (40%) to a putative dimethylallyl tryptophan synthase (EER24759) in Coccidioides posadasii13. However, putative functions of products encoded by notJ, notM, notP, and notQ remain unknown based on bioinformatics analysis.

Determination of NotF as the deoxybrevianamide E synthase

We first examined the role of NotF in notoamide biosynthesis. Its cDNA was prepared by removing the 72-bp intron using an overlapping PCR strategy (Supplementary Table 2 online). The recombinant enzyme was purified with Ni-NTA resin to about 90% purity, and its native protein status was determined as an oligomer with an observed molecular weight of 292 kDa (53.6 kDa as the calculated monomeric size) in gel filtration (Supplementary Fig. 1 online).

Next, the function of NotF was tested with doubly 13C-labelled brevianamide F (17, Supplementary Fig. 2 online). The product 18 exhibited the same retention time (17.38 min) as authentic 15 but its [M+H]+ ion was 354.19, bearing the expected 2.00-Da shift from [15+H]+ (352.19) (Fig. 2a). In MS2 analysis, 18 was fragmented in the same manner as 15 and the m/z differences (1 or 2 Da) of three major fragments (m/z values at 199.14, 286.17, and 298.14 for 18) in two MS2 spectra came from the two labeled 13C atoms in 17 (Supplementary Fig. 3 online). These results demonstrate that NotF is the deoxybrevianamide E 15 synthase and catalyzes the key step at C-2 of the indole ring leading to the bicyclo[2.2.2]diazaoctane core during biosynthesis of many fungal alkaloids within this family. In contrast, 16 in A. fumigatus is ultimately converted to fumitremorgin following normal prenylation at C-2 by FtmB9.

Figure 2.

Figure 2

Determination of NotF and NotC prenyltransferase activities. (a). Identification of NotF product (18) by LC-MS/MS analysis as described in Methods. (b). Identification of NotC reaction product (25) by LC-MS/MS analysis as described in Methods. The product (25) was further characterized by 1H and 13C NMR analysis. (c). Investigation of key residues in the reaction of the reverse prenyltransferase NotF by site-directed mutagenesis. Data shown are means ± s.d. from two independent experiments.

As a next step, the substrate selectivity of NotF was investigated with l-Trp, 17, cyclo-(l-Phe-l-Pro) 19, cyclo-(l-Trp-l-Trp) 20, and cyclo-(l-Trp-l-Tyr) 21 (Supplementary Fig. 2 online). No prenylated products were detected for any of these unnatural substrates by LC-MS analysis. This result provides strong evidence for the early timing and high selectivity of the NotF-catalyzed reaction in the alkaloid biosynthetic pathway (Scheme 2a). Moreover, the structural similarities among 16, 19, 20, and 21 suggested that both amino acid residues in 16 are critical for selective interactions between the substrate and the NotF reverse prenyltransferase.

Scheme 2.

Scheme 2

The putative biosynthetic pathway for stephacidin A (14) and notoamide natural products. (a). The early stage in stephacidin and notaomide biosynthesis. 25 serves as the common precursor to 13 and 14. 13 is then converted into 11 and 12. Substrates used in NotC studies were labeled in blue. (b). Notoamide A (9) and B (10) are possibly derived from 14. The solid arrows represent reactions that have been confirmed with bioinformatic analysis, biochemical analysis, or precursor incorporation experiments, while the dashed arrows indicate proposed biosynthetic steps. The red symbol X indicates the reaction is not supported by the current study.

Determination of NotC as the 6-hydroxy-7-prenyl-deoxybrevianamide E synthase

The role of NotC, the second predicted prenyltransferase from the marine-derived Aspergillus sp. MF297-2, was also investigated for the biosynthesis of stephacidin and notoamide. Its cDNA was similarly generated by an overlapping PCR strategy and was expressed in E. coli (Supplementary Table 2 online). The recombinant protein was purified with a single Ni-NTA affinity column, and its native protein status was determined as a monomer with an observed molecular weight of 61 kDa (51.1 kDa as the calculated monomeric size) in gel filtration (Supplementary Fig. 1 online).

Stephacidin A (14) is a central advanced intermediate featuring a pyran ring but lacking a spiro centering ring system, implicating the mode of assembly and timing of the bicyclo[2.2.2]diazaoctane family of fungal alkaloids7, 14. We propose that 14 is produced from deoxybrevianamide E (15) in a series of reactions, including hydroxylation at C-6, followed by normal prenylation at C-7, oxidation of the dioxopiperazine ring, IMDA and ring closure to the pyran (Scheme 2a). This hypothesis and the order of these reactions were examined by the determination of NotC activity with four structurally-related putative substrates, 15, 22, 23, and 24 (Scheme 2a). Compounds 22 and 24 were synthesized according to Supplementary Scheme 1 and Supplementary Scheme 2 online, respectively. NotC showed high selectivity toward 6-hydroxy-deoxybrevianamide E (22) with three additional substrates failing to be converted to products by the enzyme (Fig. 2b). The m/z value of the product was 436.17, the same as that of the singly prenylated 22 (calculated MW of 435.25), in MS analysis. Also, three major fragment ions in the MS2 spectrum of the enzyme product exhibited m/z values at 280.20, 368.17, and 380.22 (Supplementary Fig. 4 online). The difference between these and the major ions in MS2 spectrum of 15 is 84-Da, which is consistent with the MW sum of one oxygen atom and one isoprene unit linked to the indole ring of 15. We also chemically synthesized notoamide S (6-hydroxy-7-prenyldeoxybrevianamide E, 25)15 as the authentic standard and compared the authentic substance with the NotC product by LC and 1H NMR analyses. Both compounds exhibit the same LC retention time (Supplementary Fig. 5 online) and displayed identical 1H NMR spectra (Supplementary Fig. 6 online). The NotC product was further confirmed to be notoamide S (25)15 by comparison with 13 (Supplementary Table 3 online). Chemical shifts of 1H and 13C NMR spectra between the product and 13 were essentially identical with the exception of C-25 and C-276. The significant differences at these two positions reflect the double bond position in the attached isoprene unit and also indicate that the pyran is not formed in 25. These results demonstrated that NotC catalyzes a normal prenyltransfer reaction at C-7 of the indole aromatic ring system in 22. Moreover, analysis of NotC substrate selectivity suggested that the biosynthetic pathway of 13 and 14 might not proceed through substrates 23 and 24 (Scheme 2a). Instead, 15 is first hydroxylated at C-6 of the indole ring, and the product 22 is subsequently prenylated at C-7 by NotC to generate 25, a biosynthetic precursor of both 13 and 14. Accordingly, notoamide J was isolated from the culture of this marine-derive fungus, and contains only one C-6 hydroxy group16. The precise path from 25 to 14 requires further investigation, although we have demonstrated that stephacidin A (14) does not arise directly from notoamide E (13), suggesting that generation of the pyran follows formation of the bicycl[2.2.2]diazaoctane core6.

Biochemical characterization of NotF and NotC

Both NotF and NotC tolerated a broad range of temperature (4 to 42 °C for NotF while 16 to 42 °C for NotC) and pH (6.0 to 9.0) conditions for catalysis (Supplementary Fig. 7 online). Enzyme activity was independent of divalent cation, although addition of 5 mM Mg2+, Ca2+ and Mn2+ slightly enhanced catalysis (about 100-120 %) (Supplementary Fig. 8 online). Significant reduction of enzyme activity (2% - 35%) was observed with Cu2+, Zn2+, Fe2+, or Sn2+ (5 mM). Unlike previous observations with a normal aromatic prenyltransferase (MaPT)17 and CloQ18, EDTA caused only minor effects on NotF and NotC activity (remaining 90-95%), possibly indicating that the active-site pocket of both prenyltransferases might be less exposed to solvent. However, the exact mechanism for non-chelating inhibition of this group of metal-independent enzymes by EDTA remains unclear.

The reactions of both NotF and NotC followed Michaelis-Menton kinetics (Supplementary Fig. 9 online). The Km and Vmax values for doubly 13C-labeled brevianamide F (17) in the NotF reactions were 4.33±0.43 μM and 0.89±0.02 μM/min, respectively, giving a maximal turnover value of 19.1±0.4 min−1. Similarly, the enzyme Km and Vmax values for dimethylallyl diphosphate (DMAPP) were 1.31±0.22 μM and 1.18±0.03 μM/min, respectively. Its maximal turnover value (25.3±0.6 min−1) and enzyme catalytic efficiency value (19.31 μM−1• min−1) were slightly higher than those of 17 (kcat/Km = 4.41 μM−1 Compared to FtmB using 16 as a substrate in a normal prenylation step (Km = 55 μM, kcat/Km = 6.08 μM−1 · min−1), NotF showed more restricted substrate selectivity and higher substrate binding affinity with a similar enzyme catalytic efficiency19. In addition, we also determined kinetic parameters for the NotC normal prenyltransferase. The Km and Vmax values for 22 in the NotC reactions were 2.64 ±0.33 μM and 1.30 ±0.04 μM/min, respectively, while for DMAPP these values were determined to be 1.89 ±0.20 μM and 1.45 ± 0.03 μM/min, respectively. Similar to NotF, the maximal turnover value (67.4 ± 1.4 min−1) and enzyme catalytic efficiency value (35.66 μM−1 · min−1) of NotC were also slightly higher than those of 22 (60.5 ± 1.9 min−1, 22.92 μM−1 · min−1, respectively).

In this study, three key amino acid residues located in the NotF predicted reaction pocket were mutated to probe the reaction mechanism of this reverse aromatic prenyltransferase (Supplementary Fig. 10 online). R108 was predicted to be a substitute for the divalent metal ion, and to interact with the DMAPP pyrophosphate group. Generation of the R108H and R108G mutants (Supplementary Fig. 1a online) resulted in proteins with less than 2% catalytic activity, confirming the vital role of R108 for effective catalytic function (Fig. 2c). Another highly conserved key residue for substrate binding in NotF is E108, which may form a H-bond with N-H in the 16 indole ring system20. Both E108D and E108G mutants lost at least 92% of their activity (Fig. 2c). This result suggests that a specific side chain length in this acidic amino acid residue is important for H-bond formation. Recently, the crystal structure of FgaPT2, a normal fungal aromatic prenyltransferase, revealed the presence of a defined network consisting of five Tyr residues to prevent the DMAPP-derived carbocation from reacting adventitiously with nucleophiles20. All of these residues are conserved in NotF except Y413, which is replaced with W424 (Supplementary Fig. 10 online). Although W424Y still retained about 25% of its activity for production of 18, the corresponding W424G lesion resulted in >98% loss of catalytic activity, validating the importance of this analogous aromatic network in the notoamide reverse prenyltransferase reaction (Fig. 2c).

Proposed biosynthetic pathway for stephacidin and notoamide biosynthesis in marine-derived Aspergillus sp. MF297-2

Based on biochemical characterization of two prenyltransferases and predicted biochemical function of related gene products in the isolated gene cluster, a putative notoamide biosynthetic pathway is proposed (Scheme 2). Briefly, we presume that 16 is produced from l-Trp and l-Pro by the NotE NRPS, and is subsequently reverse prenylated at C-2 by NotF to produce 15. In the next step, the indole ring is hydroxylated at C-6, which is likely catalyzed by one of the two not pathway-encoded P450s. NotC is then responsible for normal prenylation at C-7 of 22 to produce 25. From this intermediate, notoamide E (13) is generated following ring closure to the pyran possibly controlled by the second P450, and may be converted into notoamide C (11) and notoamide D (12)6. The direct connection from 13 to 14 was not observed in double 13C-labeled precursor incorporation experiments, whereby 13 was previously envisioned to be converted by an oxidase to generate the pyrazine-derived dienophile followed by IMDA to directly produce stephacidin A (14) (Scheme 2a)6. The enzyme(s) that promote bicyclo[2.2.2]diazaoctane formation remain unknown. Recently, a flavin-dependent oxidase in solanapyrone biosynthesis was shown to catalyze both oxidation and subsequent Diels-Alder cycloaddition reactions21, indicating that the predicted oxidoreductases (e.g. NotB, NotD, and NotI) are possible candidates for catalyzing the IMDA reaction in notoamide biosynthesis. Alternative to the pathway through 13, 14 may be produced from 25 after of the step-wise oxidation, IMDA, and ring closure (Scheme 2a). We propose that 14 is regiospecifically hydroxylated by a monooxygenase, possibly NotB or NotI, for subsequent pinacol-like rearrangement to produce notoamide B (10). Such an intriguing rearrangement reaction has not been previously observed in a natural product biosynthetic pathway, and a putative mechanism is proposed (Supplementary Fig. 11 online). A rare N-hydroxylation reaction is required to generate the final notoamide A product (9) (Scheme 2b). The biosynthetic scheme proposed here provides an initial understanding of the assembly and modification of biosynthetic building blocks for this important group of bioactive prenylated fungal alkaloids.

DISCUSSION

The advent of next-generation sequencing has provided tremendous opportunities to identify novel natural products and their biosynthetic pathways through genome mining of bacterial, fungal22 and plant23 genomes. This approach has proved increasingly important in current natural product biosynthesis studies with the availability of increasing numbers of microbial genome sequences. For example, many orphan biosynthetic gene clusters were identified from the genome sequence of well-studied Streptomyces coelicolor A3(2), and novel natural products such as 2-alkyl-4-hydroxymethylfuran-3-carboxylic acids were uncovered with the guidance of genome mining24, 25. In this study, we employed an allied approach to study the biosynthesis of prenylated indole alkaloids and also to initiate an understanding of fungal genetic evolution and adaptation to different environmental niches. The entire genome of the notoamide-producing marine-derived Aspergillus sp. was sequenced at ~ 15 X coverage, enabling notoamide gene cluster identification through sequence file database comparison to ftmA (Fig. 1). With this gene cluster it is now possible to pursue studies toward a complete understanding of the assembly, tailoring and regulation of this family of bioactive fungal alkaloids and to develop them further as medicinal agents.

Fungal aromatic prenyltransferases have attracted increasing interest because of their important role in the biosynthesis of natural products and potential applications in drug development. Currently, more than 100 putative indole prenyltransferases have been revealed by BLAST searches in the public genome database26. Biochemical investigation of over ten recombinant enzymes in this group finds that catalytic functions are independent of divalent metal ions26, 27. In these reactions, an isoprene unit can be transferred onto different positions of the indole ring system27. Moreover, in a normal prenylation reaction, DMAPP alkylates an aromatic substrate through its C1′ atom via an SN2 displacement, while the C3′ position is involved in the reverse prenyltransfer reaction via an SN2′ displacement. Remarkably, these enzymes can utilize a series of structurally similar analogs as their aromatic substrates17, 27, 28. In contrast to other characterized enzymes in this group, NotF (reverse) and NotC (normal) prenyltransferases showed highly restricted substrate specificities. NotF specifically prenylates 16 (at C-2), confirming its role as the elusive deoxybrevianamide E synthase, while only 22 is utilized by NotC (C-7 alkylation) in the biosynthesis of stephacidin and notoamides (Fig. 2). The mechanism of the reverse prenyltransferase was also probed by site-directed mutagenesis to understand the reaction of this group of enzymes17, 20. Future structural studies of NotF compared to FgaPT220 is expected to add further information about regio- and stereospecificity of the reverse and normal prenyltransfer reactions. We expect this analysis will likely illuminate the lack of facial selectivity previously observed for the reverse prenylation step29, and facilitate expansion of the enzyme substrate range, and efficiency. More importantly, the combined studies of NotF and NotC provide direct evidence to establish a biosynthetic scheme for this family of bioactive prenylated fungal alkaloids. Finally, the high in vitro catalytic efficiencies of recombinant NotF and NotC suggests their potential value as biocatalysts for chemoenzymatic production of bioactive fungal alkaloid analogs in drug development.

Identification and characterization of the notoamide gene cluster also provides the initial basis to understand the formation of three pairs of antipodal natural products derived from a marine-derived and terrestrial Aspergillus sp.30, 31. In the marine-derive fungal strain, (−)-notoamide B (10), (−)-versicolamide B (38), and (+)-stephacidin A (14) are produced7, 30, while their antipodal counterparts, 39, 40, and 41, respectively, are isolated from the terrestrial A. versicolor NRRL 25660 strain (Supplementary Fig. 12 online)31. Based on the putative notoamide biosynthetic pathway, we propose that formation of 14 and 41 might be controlled by the IMDA reaction. Subsequently, 10 and 39 are possibly derived from 14 and 41, respectively in these two distinct fungal strains. It remains unclear whether generation of (−) and (+)-versicolamide B (38 and 40) occur in the pathway through stephacidin A due to their opposite stereogenic centers at C-6 (Supplementary Fig. 12 online)30. Instead, 13 might be converted into 11, subsequently producing 38 in the following IMDA reaction in the marine-derived fungus30. The detailed biochemical characterization of biosynthetic enzymes from both fungal strains is in progress and will shed more light on the biosynthesis of these unique antipodal natural products.

Identification of biocatalysts from fungal alkaloid biosynthetic pathways may also enable production of natural products and their analogs through heterologous expression and metabolic engineering32-34. It is estimated that > 99 % of microorganisms in the environment fail to grow in the laboratory while the potential to find pharmaceutically important natural products from fungal sources remains vastly underexplored. Introducing natural product gene clusters into more technically- and industrially-amenable microorganisms such as E. coli and yeast represents an attractive way to obtain suitable quantities of natural products and to identity novel leads in drug discovery and development programs34. Moreover, a microorganism can be further optimized for the efficient production of a target metabolite using traditional mutation and selection methods, as well as new tools from systems biology and synthetic biology35. Identification of the notoamide gene cluster provides such an opportunity to produce bioactive fungal alkaloids and analogs thereof through pathway engineering and heterologous expression.

METHODS

Materials and Strains

Authentic deoxybrevianamide E (15), doubly 13C-labelled brevianamide F (17), and keto-premalbrancheamide (23) were synthesized following previously published procedures4, 36. Standard methods for DNA isolation and manipulation were performed as described by Sambrook et al37. Genomic DNA from Aspergillus MF297-2 was isolated with MasterPure Yeast DNA Purification kit (Epicentre Biotechnologies) as described in the manual. The GenBank accession numbers for notC, notF, and the complete assembled not gene cluster are GU564534, GU564535, and HM622670, respectively. E. coli DH5α was used for cloning and plasmid harvesting while E. coli BL21 CodonPlus-(DE3)-RIPL was used for protein overexpression.

Expression and purification of NotC and NotF

Details about preparation of notC and notF cDNAs, and of notF mutant DNAs were included in Supplementary Methods and Supplementary Table 2 online. The expressed enzymes were purified with a single Ni-NTA column (Supplementary Methods online). As determined by SDS-PAGE analysis, the purity of proteins was more than 90%. The native status of proteins was determined by gel filtration (Supplementary Fig. 1 online).

Determination of enzyme activities

Compounds 22 and 24 were chemically synthesized to examine NotC activity (Supplementary Methods online). The 100-μl reaction mixture contained 0.5 μg NotF, its mutants, or NotC, 5 mM MgCl2, 0.1 mM 17 (NotF or its mutants) or 22 (NotC), and 0.15 mM DMAPP in the reaction buffer (50 mM Tris-Cl, pH 7.5, 10 % glycerol, and 3 mM β-mercaptoethanol). The reaction was initiated by adding enzyme after pre-warming the other components at room temperature for 1 min. After mixing well and briefly centrifuging, the reactions were further incubated at room temperature for 45-60 min and stopped with 10 μl of 1.5 M trichloroacetic acid. The mixtures were mixed and centrifuged at 13000 g for 5 min. An aliquot of 100-μl solution was subjected to HPLC coupled with an XBridge C18 column (5μm, 4.6 × 250 mm), at a wavelength of 222 nm. Solvent B (acetonitrile in 0.1 % TFA) was increased from 30 % to 40 % for 5 min and then increased to 80 % over 20 min for the detection of products. LC-MS2 analysis was performed by using a ThermoFinnigan LTQ linear ion-trap instrument equipped with electrospray source and Surveyor HPLC system at room temperature. Separations were performed with an XBridge C18 (3.5 m, 2.1 × 150mm) column at a flow rate of 200 l/min with solvent A (water with 0.1 % formic acid) and solvent B (acetonitrile with 0.1% formic acid). Solvent B was kept at 2 % in solvent A for 4 min and then was gradually increased to 90 % over 16 min. After washed with 90 % solvent B for 2 min, the column was further re-equilibrated with 2 % solvent B for 10 min. The spectra were record in positive ion mode. Product 25 was further characterized with 1H and 13C NMR analysis (Supplementary Methods online).

Kinetics analysis

The 100-μl reaction mixture contained 0.25 μg NotF or 0.11 μg NotC and 5 mM MgCl2 in the reaction buffer. Details about the experiment procedures were included in Supplementary Methods online. All experiments were performed in duplicate. The data were fit to the MichaelisMenton equation in Prism 4.0 (GraphPad Software).

Supplementary Material

1_si_001

Table 1.

Features of the not gene products

Protein Size
bp/aa
Exon Function Relative
(identity/similarity [%])
Accession
number
Orf1 731/224 1-113, 173-731 partial
polysaccharide
synthase
Capsule polysaccharide
biosynthesis protein from
Aspergillus fumigatus (43/63)
XP_748327
NotA 1199/339 1293-1568, 1643-1971,
2023-2179, 2234-2491
negative
regulator
NmrA family protein from
Ajellomyces capsulatus (45/64)
EEH03447
NotB 1344/401 3486-4079, 4141-4487,
4568-4829
FAD binding
domain protein
FAD binding domain protein from
A. clavatus (44/63)
XP_001268514
NotC 1350/427 5819-6996, 7066-7168 prenyl-
transferase
FtmH from A. fumigatus (50/66) BAH24002
NotD 2025/621 8012-8294, 8389-8927,
8996-10036
oxidoreductase oxidoreductase from Microsporum
canis (40/59)
EEQ33235
NotE 6723/2241 10787-17509 NRPS FtmA from A. fumigatus (47/67) XP_747187
NotF 1431/453 17924-18053, 18126-
19354
prenyl-
transferase
tryptophan dimethylallyltransferase
from Coccidioides posadasii
(40/62)
EER24759
NotG 1901/544 19899-20086, 20171-
20272, 20337-20635,
20689-20810, 20879-
21799
P450 cytochrome P450 from A.
fumigatus (62/75)
XP_747185
NotH 1836/502 22422-22668, 22734-
22822, 22897-22996,
23060-23128, 23187-
23765, 23836-24257
P450 cytochrome P450 from
Neosartorya fischeri (47/65)
XP_001261652
NotI 1423/434 24803-24962, 25021-
25939, 26003-26225
FAD binding
domain protein
FAD binding domain protein from
A. clavatus (44/63)
XP_001268514
NotJ 1113/371 26390-27502 unknown hypothetical protein from
Salinispora arenicola (52/65)
YP_001537335
NotK 1851/564 28771-29141, 29196-
29569, 29620-30389,
30445-30621
efflux pump MFS transporter from N. fischeri
(87/93)
XP_001265322
NotL 1455/484 31789-33243 transcriptional
activator
C6 zinc finger domain protein from
N. fischeri (53/62)
XP_001265321
NotM 1266/402 33816-34597
34654-35080
unknown hypothetical protein from
Talaromyces stipitatus (74/82)
XP_002482929
NotN 1126/340 35192-35244, 35299-
35895, 35948-36317
dehydrogenase alcohol dehydrogenase from
Penicillium marneffei (60/76)
XP_002147947
NotO 993/331 36520-37512 short-chain
dehydrogenases
/reductase
hypothetical protein from Nectria
haematococca (66/80)
EEU36425
NotP 1020/322 37770-37930, 37985-
38789
unknown metallo-β-lactamase domain
protein from T.
stipitatus (80/88)
XP_002482927
NotQ 569/152 39871-40059, 40120-
40316, 40370-40439
unknown hypothetical protein from T.
stipitatus (88/94)
XP_002482928
NotR 1517/461 40514-41140, 41212-
41727, 41791-42030
Transcriptional
co-activator
hypothetical protein from P.
marneffei (45/61)
XP_002144868

Acknowledgments

We thank Dr. Patricia Cruz Lopez for assistance with NMR analysis. This work was supported by NIH grant CA070375 (R.M.W. and D.H.S), the H. W. Vahlteich Professorship (D.H.S) and by a University of Michigan Rackham Predoctoral Fellowship (Y. D.).

Footnotes

Supplementary Methods online for remaining methods, including genome sequencing and assembly, assay for enzyme metal independence, detailed synthetic procedures of 22 and 24, determination of enzyme optimal conditions, characterization of 25, antipodal fungal prenylated alkaloids, and the complete list of authors in reference 25. This material is available free of charge via the Internet at http://pubs.acs.org.

REFERENCES

  • 1.Keller NP, Turner G, Bennett JW. Nat Rev Microbiol. 2005;3(12):937–47. doi: 10.1038/nrmicro1286. [DOI] [PubMed] [Google Scholar]
  • 2.Pelaez F. Biological activities of fungal metabolites. In: An Z, editor. Handbook of Industrial Mycology. Marcel Dekker; New York: 2005. pp. 49–922. [Google Scholar]
  • 3.Williams RM, Cox RJ. Acc Chem Res. 2003;36(2):127–39. doi: 10.1021/ar020229e. [DOI] [PubMed] [Google Scholar]
  • 4.Ding Y, Greshock TJ, Miller KA, Sherman DH, Williams RM. Org Lett. 2008;10(21):4863–6. doi: 10.1021/ol8019633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Stocking EM, Sanz-Cervera JF, Williams RM. Angew Chem Int Ed Engl. 2001;40(7):1296–1298. doi: 10.1002/1521-3773(20010401)40:7<1296::aid-anie1296>3.0.co;2-g. [DOI] [PubMed] [Google Scholar]
  • 6.Tsukamoto S, Kato H, Greshock TJ, Hirota H, Ohta T, Williams RM. J Am Chem Soc. 2009;131(11):3834–5. doi: 10.1021/ja810029b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Kato H, Yoshida T, Tokue T, Nojiri Y, Hirota H, Ohta T, Williams RM, Tsukamoto S. Angew Chem Int Ed Engl. 2007;46(13):2254–6. doi: 10.1002/anie.200604381. [DOI] [PubMed] [Google Scholar]
  • 8.Maiya S, Grundmann A, Li SM, Turner G. Chembiochem. 2006;7(7):1062–9. doi: 10.1002/cbic.200600003. [DOI] [PubMed] [Google Scholar]
  • 9.Grundmann A, Li SM. Microbiology. 2005;151(Pt 7):2199–207. doi: 10.1099/mic.0.27962-0. [DOI] [PubMed] [Google Scholar]
  • 10.Meyers DM, Obrian G, Du WL, Bhatnagar D, Payne GA. Appl Environ Microbiol. 1998;64(10):3713–7. doi: 10.1128/aem.64.10.3713-3717.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Woloshuk CP, Foutz KR, Brewer JF, Bhatnagar D, Cleveland TE, Payne GA. Appl Environ Microbiol. 1994;60(7):2408–14. doi: 10.1128/aem.60.7.2408-2414.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Flaherty JE, Payne GA. Appl Environ Microbiol. 1997;63(10):3995–4000. doi: 10.1128/aem.63.10.3995-4000.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Grundmann A, Kuznetsova T, Afiyatullov S, Li SM. Chembiochem. 2008;9(13):2059–63. doi: 10.1002/cbic.200800240. [DOI] [PubMed] [Google Scholar]
  • 14.Qian-Cutrone J, Huang S, Shu YZ, Vyas D, Fairchild C, Menendez A, Krampitz K, Dalterio R, Klohr SE, Gao Q. J Am Chem Soc. 2002;124(49):14556–7. doi: 10.1021/ja028538n. [DOI] [PubMed] [Google Scholar]
  • 15.McAfoos Timothy J., L. S, Tsukamoto Sachiko, Sherman David H., Williams Robert M. Heterocycles. 2010;82(1) doi: 10.3987/COM-10-S(E)19. DOI: 10.3987/COM-10-S(E)19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Tsukamoto S, Kato H, Samizo M, Nojiri Y, Onuki H, Hirota H, Ohta T. J Nat Prod. 2008;71(12):2064–7. doi: 10.1021/np800471y. [DOI] [PubMed] [Google Scholar]
  • 17.Ding Y, Williams RM, Sherman DH. J Biol Chem. 2008;283(23):16068–76. doi: 10.1074/jbc.M801991200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Pojer F, Wemakor E, Kammerer B, Chen H, Walsh CT, Li SM, Heide L. Proc Natl Acad Sci U S A. 2003;100(5):2316–21. doi: 10.1073/pnas.0337708100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Zou H, Zheng X, Li SM. J Nat Prod. 2009;72(1):44–52. doi: 10.1021/np800501m. [DOI] [PubMed] [Google Scholar]
  • 20.Metzger U, Schall C, Zocher G, Unsold I, Stec E, Li SM, Heide L, Stehle T. Proc Natl Acad Sci U S A. 2009;106(34):14309–14. doi: 10.1073/pnas.0904897106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Kasahara K, Miyamoto T, Fujimoto T, Oguri H, Tokiwano T, Oikawa H, Ebizuka Y, Fujii I. Chembiochem. 2010;11(9):1245–52. doi: 10.1002/cbic.201000173. [DOI] [PubMed] [Google Scholar]
  • 22.Corre C, Challis GL. Nat Prod Rep. 2009;26(8):977–86. doi: 10.1039/b713024b. [DOI] [PubMed] [Google Scholar]
  • 23.Lodeiro S, Xiong Q, Wilson WK, Kolesnikova MD, Onak CS, Matsuda SP. J Am Chem Soc. 2007;129(36):11213–22. doi: 10.1021/ja073133u. [DOI] [PubMed] [Google Scholar]
  • 24.Corre C, Song L, O'Rourke S, Chater KF, Challis GL. Proc Natl Acad Sci U S A. 2008;105(45):17510–5. doi: 10.1073/pnas.0805530105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Bentley SD, et al. Nature. 2002;417(6885):141–7. doi: 10.1038/417141a. [DOI] [PubMed] [Google Scholar]
  • 26.Li SM. Appl Microbiol Biotechnol. 2009;84(4):631–9. doi: 10.1007/s00253-009-2128-z. [DOI] [PubMed] [Google Scholar]
  • 27.Steffan N, Grundmann A, Yin WB, Kremer A, Li SM. Curr Med Chem. 2009;16(2):218–31. doi: 10.2174/092986709787002772. [DOI] [PubMed] [Google Scholar]
  • 28.Balibar CJ, Howard-Jones AR, Walsh CT. Nat Chem Biol. 2007;3(9):584–92. doi: 10.1038/nchembio.2007.20. [DOI] [PubMed] [Google Scholar]
  • 29.Stocking Emily M., S.-C. JF, Williams Robert M. Angewandte Chemie International Edition. 1999;38(6):786–789. doi: 10.1002/(SICI)1521-3773(19990315)38:6<786::AID-ANIE786>3.0.CO;2-7. [DOI] [PubMed] [Google Scholar]
  • 30.Tsukamoto S, Kawabata T, Kato H, Greshock TJ, Hirota H, Ohta T, Williams RM. Org Lett. 2009;11(6):1297–300. doi: 10.1021/ol900071c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Greshock TJ, Grubbs AW, Jiao P, Wicklow DT, Gloer JB, Williams RM. Angew Chem Int Ed Engl. 2008;47(19):3573–7. doi: 10.1002/anie.200800106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Chemler JA, Koffas MA. Curr Opin Biotechnol. 2008;19(6):597–605. doi: 10.1016/j.copbio.2008.10.011. [DOI] [PubMed] [Google Scholar]
  • 33.Wenzel SC, Muller R. Curr Opin Biotechnol. 2005;16(6):594–606. doi: 10.1016/j.copbio.2005.10.001. [DOI] [PubMed] [Google Scholar]
  • 34.Zhang H, Wang Y, Pfeifer BA. Mol Pharm. 2008;5(2):212–25. doi: 10.1021/mp7001329. [DOI] [PubMed] [Google Scholar]
  • 35.Lee SY, Kim HU, Park JH, Park JM, Kim TY. Drug Discov Today. 2009;14(1-2):78–88. doi: 10.1016/j.drudis.2008.08.004. [DOI] [PubMed] [Google Scholar]
  • 36.Miller KA, Welch TR, Greshock TJ, Ding Y, Sherman DH, Williams RM. J Org Chem. 2008;73(8):3116–9. doi: 10.1021/jo800116y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Sambrook J, Russel DW. Molecular Cloning - A Laboratory Manual. 3rd ed. Cold Spring Harbor Laboratory Press; New York: 2001. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1_si_001

RESOURCES