Talarolide A, a Cyclic Heptapeptide Hydroxamate from an Australian

Apr 6, 2017 - *E-mail: [email protected]. ... secondary metabolite media dependence in an Australian marine tunicate-associated fungus, Talaromyces sp...
0 downloads 0 Views 1MB Size
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Letter pubs.acs.org/OrgLett

Talarolide A, a Cyclic Heptapeptide Hydroxamate from an Australian Marine Tunicate-Associated Fungus, Talaromyces sp. (CMB-TU011) Pradeep Dewapriya,‡ Pritesh Prasad,‡ Rakesh Damodar, Angela A. Salim, and Robert J. Capon* Institute for Molecular Bioscience, The University of Queensland, St. Lucia, QLD 4072, Australia S Supporting Information *

ABSTRACT: A miniaturized 24-well plate microbioreactor approach was used to explore secondary metabolite media dependence in an Australian marine tunicate-associated fungus, Talaromyces sp. (CMB TU011). Detailed chemical investigations of an antifungal M1-saline cultivation yielded talarolide A (1), only the second reported natural cyclic peptide hydroxamate, and the first from a fungus. The antifungal properties of the M1-saline extract were attributed to the known diterpene glycoside sordarin (2). Structure elucidation of 1 and 2 was achieved by detailed spectroscopic analysis, with amino acid configurations in 1 assigned by the C3 and C18 Marfey’s methods, and L-Ala and D-Ala regiochemistry by the recently reported 2D C3 Marfey’s method.

A

s part of our investigations into the secondary metabolites of Australian microbes, we examined a fungal strain CMB-TU011 isolated from an unidentified tunicate collected from intertidal waters off Tweed Heads, NSW, Australia. Genomic analysis and morphological examination identified this strain as a Talaromyces sp., a ubiquitous genus found in both terrestrial and marine ecosystems, including estuarine sediments, salt marshes, mangrove wetlands, and deep-sea sediments, known to produce a diverse array of secondary metabolites.1,2 To explore the secondary metabolite potential of CMBTU011 we employed a 24-well plate microbioreactor (Figure S3), generating extracts across a range of solid-phase media. The resulting in situ EtOAc extracts were analyzed for secondary metabolite production by high-performance liquid chromatography with diode-array detection and electrospray ionization mass spectrometry (HPLC-DAD-ESIMS) and evaluated for growth inhibitory activity against bacterial, fungal, and mammalian cells. With differential metabolite profiles and biological activity across extracts, our attention was drawn to antifungal M1-saline and YES media extracts that exhibited HPLC peaks eluting at 8.7 (1, m/z 718) and 10.2 (2, m/z 515) min, which were not evident in the antifungal inactive tryptic soy agar (TSA) media extract (Figure 2). To investigate these metabolites, we subjected 50 × 25-day 3.3% saline M1 agar plate cultivations to solvent extraction and partitioning, followed by reversed phase chromatography, to yield 1 and 2 (Figure 1). The antifungal activity of the M1-saline extract was attributed to the known fungal metabolite sordarin (2), with the structure confirmed by detailed spectroscopic analysis and literature comparisons (Supporting Information, SI).3 First reported in 1971 from Sordaria araneosa (Order Sordariales),4 2 attracted considerable attention due to its potent antifungal activity and mechanism of action as a selective inhibitor of fungal protein synthesis elongation © 2017 American Chemical Society

Figure 1. Metabolites talarolide A (1) and sordarin (2).

factor 2.5 Despite its original source, a 2009 study on the taxonomic distribution of 2 concluded that production is actually uncommon among the orders Sordariales, Eurotiales, and Microascales, and more prevalent within the Xylariales.6 Our isolation of 2 is the first recorded account from a Talaromyces sp. (Order Eurotiales). By contrast, the cometabolite 1 did not exhibit antifungal activity, but did prove to be a new cyclic heptapeptide rich in D-amino acids and incorporating a rare hydroxamate residue. The structure elucidation of 1 was achieved by de novo spectroscopic analysis, and a combination of C3, C18, and 2D C3 Marfey’s analyses, as outlined below. Received: March 2, 2017 Published: April 6, 2017 2046

DOI: 10.1021/acs.orglett.7b00638 Org. Lett. 2017, 19, 2046−2049

Letter

Organic Letters

Figure 2. HPLC-DAD (210 nm) chromatograms of Talaromyces sp. (CMB-TU011) cultivated in different media.

Figure 4. C3 Marfey’s analysis of 1. (a) C3 HPLC-DAD (340 nm) chromatogram revealing D-FDAA amino acid derivatives. (b−f) C3 HPLC-MS-SIE chromatograms for D-FDAA derivatives of authentic standards (broken lines) and the acid hydrolysate of 1 (shaded peaks), with the inset (c) C18 HPLC-MS-SIE chromatogram. Traces confirm that 1 incorporates (b) N-Me-L-Tyr (SIE m/z 448), (c) NMe-D-Ala (SIE m/z 356), (d) L-Ala and D-Ala (SIE m/z 342), (e) D-allo-Ile (SIE m/z 384), and (f) N-Me-D-Leu (SIE m/z 398). Note, the D-FDAA derivative of N-OH-Gly was not detected.

Establishing the regiochemistry of enantiomeric amino acid residues within natural product peptides can prove challenging, and often relies on either partial or total synthesis of multiple isomers.7 To establish the regiochemistry of L-Ala and D-Ala residues in 1, we employed the recently described 2D C3 Marfey’s method.7 HPLC-DADESIMS analysis of a partial hydrolysate of 1 (150 μg) derivatized with Nα-(2,4-dinitro-5-fluorophenyl)-D-alaninamide (D-FDAA) revealed a peak (i) attributed to the dipeptide D-FDAA-D-allo-Ile-Ala ([M + H]+, m/z 455.0, Figure 6, green highlight). Analytical reversed phase HPLC fractionation of this derivatized hydrolysate yielded pure (i), which was subsequently subjected to acid hydrolysis and C3 Marfey’s analysis7 to confirm the presence of D-FDAA-D-Ala (inset in Figure 6b). This analysis permits the unambiguous positioning of L-Ala and D-Ala residues in 1. To the best of our knowledge, talarolide A (1) is only the second cyclic peptide (other than diketopiperazines) to be reported from the genus Talaromyces, with the first being talaromins A and B reported in 2013 from an endophytic strain of Talaromyces wortmannii.8 More significantly, talarolide A (1) is also only the second reported example of a nonribosomal peptide synthase (NRPS) derived cyclic peptide incorporating an internal hydroxamate linkage, with the first being pargamicin reported in 2008 from an Amycolatopsis sp.9 In our hands talarolide A (1) did not inhibit growth of human embryonic kidney (HEK-293) or colorectal (SW-620) adenocarcinoma cells, or the fungus Candida albicans (ATCC 90028), or the Gram-negative bacteria Escherichia coli (ATCC 11775) and Pseudomonas

Figure 3. Diagnostic 2D NMR (DMSO-d6) correlations for 1.

HRESI(+)MS analysis of 1 afforded a sodium adduct ([M + Na]+) consistent with a molecular formula (C35H55N7O9, Δppm +0.4) requiring 12 double bond equivalents (DBE). The 1D NMR (DMSO-d6) data for 1 (Table 1) exhibited characteristic resonances for a p-disubstituted benzene and seven carbonyls, accounting for 11 DBE, while diagnostic 2D NMR correlations allowed for partial assembly of a cyclic heptapeptide planar structure (Figure 3). C3 and C18 Marfey’s7 analysis permitted identification of six of the proposed amino acid residues as N-Me-L-Tyr, D-allo-Ile, NMe-D-Ala, N-Me-D-Leu, L-Ala, and D-Ala (Figure 4). The remaining C2H3NO2 fragment was attributed to an N-OHGly residue, as confirmed by NMR data, which revealed an isolated C(O)-CH2-NOH [δH 4.75, d (17.1 Hz) and 3.76, d (17.1 Hz); δC 50.2] exhibiting HMBC correlations to the carbonyl of an adjacent N-Me-D-Leu, with an additional diagnostic HMBC from NH of an adjacent Ala residue (Figure 3). Although the 2D NMR data did not permit unambiguous assignment of the amino acid sequence, this was readily confirmed by MS-MS fragmentations (Figure 5), albeit without differentiation of L-Ala and D-Ala regiochemistry. 2047

DOI: 10.1021/acs.orglett.7b00638 Org. Lett. 2017, 19, 2046−2049

Letter

Organic Letters Table 1. NMR (600 MHz, DMSO-d6) Data for Talarolide A (1) δH, mult, (J in HZ) I: D-allo-Ile 1 2 3 4 5 6 N-H II: D-Ala 1 2 3 N-H III: N-Me-D-Leu 1 2 3 4 5 6 N-Me IV: N−OH-Gly 1 2 N−OH V: L-Ala 1 2 3 N-H VI: N-Me-D-Ala 1 2 3 N-Me VII: N-Me-L-Tyr 1 2 3 4 5/9 6/8 7 7-OH N-Me a

4.72, m 1.95, m a 1.42, m b 1.07, m 0.94, dd (7.3, 7.3) 0.81, d (6.9) 7.24, d (9.6)

4.49, qd (6.8, 4.1) 1.19, d (6.8) 8.66, d (4.1)

5.05, dd (11.8, 3.9) a 1.79, ddd (14.4, 10.3, 3.9) b 1.58, ddd (14.4, 11.8, 3.9) 1.37, m 0.77, d (6.5) 0.88, d (6.5) 3.00, s

a 4.75, d (17.1) b 3.76, d (17.1) 9.31, s

4.34, qd (7.1, 5.4) 1.12, d (7.1) 8.87, d (5.4)

4.71, q (6.5) 0.49, d (6.5) 2.70, s

4.80, dd (10.5, 4.9) a 2.84, dd (14.3, 10.5) b 2.60, dd (14.3, 4.9) 6.93, d (8.4) 6.64, d (8.4) 9.21, s 2.66, s

δC 169.5 53.7 38.5 26.2 12.0 13.7

174.1 45.2 15.7

167.3 54.5 36.0 24.4 21.0 23.3 31.0 172.0 50.2

171.1 45.8 14.9

169.8 46.7 15.1 28.6a 168.2 56.6 34.1 126.6 130.8 114.8 155.9 28.6a

COSY

N-H 2, 4a, 4b, 6 3, 4b, 5 3, 4a, 5 4a, 4b 3 2

HMBC

3, 6 2, 2, 3, 2, 1,

3, 6, 5 3, 6, 5 4 3, 4 1(N-Me-L-Tyr)

3, N-H 2 2

3 1, 2

3a, 3b 2, 3b, 4 2, 3a, 4 3a, 3b, 5, 6 4 4

3, N-Me, 1(D-Ala)

2b 2a

1(N-Me-D-Leu) 1(N-Me-D-Leu)

3, N-H 2 2

1, 3 2, 1 2, 3, 1(Gly)

3 2

1, 3, N-Me, 1(L-Ala) 1, 2 2, 1(L-Ala)

3a, 3b 2, 3b 2, 3a

1, 3, N-Me, 1(N-Me-D-Ala) 4, 5, 9 4, 5, 9

6, 8 5, 9

3, 7, 9/5 4, 8/6

5 3, 6 3, 4, 5 2, 1(D-Ala)

6/8, 7 2, 1(N-Me-D-Ala)

Overlapping resonances.

aeruginosa (ATCC 10145), or the Gram-positive bacteria Staphylococcus aureus (ATCC 9144 and ATCC 25923) and Bacillus subtilis (ATCC 6633 and ATCC 6051) (IC50 > 30 μM), suggestive of a more specialized ecological purpose (i.e., not merely a cytotoxin). Of note, the high level of ROESY correlations to the NOH moiety in 1 suggests it is centrally oriented within the cyclic peptide ring. This raises the prospect that through strategic H-bonding the N-OH may play a pivotal role in

regulating cyclic peptide tertiary structure. As the tertiary structure of cyclic peptides is typically overlooked in the natural products science, it is interesting to speculate on the structural and biological consequences of strategic insertion of N-OH moieties into other bioactive cyclic peptide scaffolds. Further investigations into the biosynthesis and pharmacology of 1 and other cyclic peptide co-metabolites encoded within the genome of Talaromyces sp. (CMBTU011) are in progress. 2048

DOI: 10.1021/acs.orglett.7b00638 Org. Lett. 2017, 19, 2046−2049

Letter

Organic Letters

Figure 5. Diagnostic MS−MS fragmentations for talarolide A (1).

All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS R.D., P.P., and P.D. acknowledge the University of Queensland for Postgraduate Research Scholarships. We thank P. Kalansuriya (UQ), Z. Khalil (UQ), and M. Quezada (UQ) for bioassays support and A. Jones (UQ) for MS−MS support. This research was funded in part by the Institute for Molecular Bioscience, the University of Queensland, and the Australian Research Council (LP120100088).

Figure 6. 2D C3 Marfey’s analysis of 1 derivatized with D-FDAA. (a) UV (340 nm) of partial hydrolysate of 1 (* = residual DFDAA). (b) SIE revealing (i) D-FDAA-D-allo-Ile-Ala (m/z 455, green peak). Inset broken lines correspond to SIE chromatograms for pure D-FDAA standards of D-allo-Ile and D-Ile (m/z 384), and DAla (m/z 342). Inset shaded peaks correspond to SIE chromatograms for D-FDAA derivatized amino acid residues from the hydrolysate (i).





(1) Zhai, M.-M.; Li, J.; Jiang, C.-X.; Shi, Y.-P.; Di, D.-L.; Crews, P.; Wu, Q.-X. Nat. Prod. Bioprospect. 2016, 6, 1−24. (2) Nicoletti, R.; Trincone, A. Mar. Drugs 2016, 14, 37. (3) Chiba, S.; Kitamura, M.; Narasaka, K. J. Am. Chem. Soc. 2006, 128, 6931−6937. (4) Hauser, D.; Sigg, H. P. Helv. Chim. Acta 1971, 54, 1178−1190. (5) Domínguez, J. M.; Kelly, V. A.; Kinsman, O. S.; Marriott, M. S.; de Las Heras, F. G.; Martín, J. J. Antimicrob. Agents Chemother. 1998, 42, 2274−2278. (6) Vicente, F.; Basilio, A.; Platas, G.; Collado, J.; Bills, G. F.; González Del Val, A.; Martín, J.; Tormo, J. R.; Harris, G. H.; Zink, D. L.; Justice, M.; Nielsen Kahn, J.; Peláez, F. Mycol. Res. 2009, 113, 754−770. (7) Vijayasarathy, S.; Prasad, P.; Fremlin, L. J.; Ratnayake, R.; Salim, A. A.; Khalil, Z.; Capon, R. J. J. Nat. Prod. 2016, 79, 421− 427. (8) Bara, R.; Aly, A. H.; Wray, V.; Lin, W.; Proksch, P.; Debbab, A. Tetrahedron Lett. 2013, 54, 1686−1689. (9) Igarashi, M.; Sawa, R.; Kinoshita, N.; Hashizume, H.; Nakagawa, N.; Homma, Y.; Nishimura, Y.; Akamatsu, Y. J. Antibiot. 2008, 61, 387−393.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.7b00638. General experimental, fungal isolation and taxonomy, compounds isolation and identification, NMR data and NMR spectra (PDF)



REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Robert J. Capon: 0000-0002-8341-7754 Author Contributions ‡

These authors contributed equally to this work. The manuscript was written through contributions of all authors. 2049

DOI: 10.1021/acs.orglett.7b00638 Org. Lett. 2017, 19, 2046−2049