Biosynthetic Origin of the Hydroxamic Acid Moiety of Trichostatin A

May 12, 2017 - Trichostatin A (TSA) is widely used in the field of epigenetics because it potently inhibits histone deacetylase (HDAC). In-depth studi...
1 downloads 0 Views 898KB Size
Communication pubs.acs.org/JACS

Biosynthetic Origin of the Hydroxamic Acid Moiety of Trichostatin A: Identification of Unprecedented Enzymatic Machinery Involved in Hydroxylamine Transfer Kei Kudo,† Taro Ozaki,†,§ Kazuo Shin-ya,‡ Makoto Nishiyama,† and Tomohisa Kuzuyama*,† †

Biotechnology Research Center, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan National Institute of Advanced Industrial Science and Technology, 2-4-7 Aomi, Koto-ku, Tokyo 135-0064, Japan



S Supporting Information *

oxygenase to give a hydroxylated amine, followed by acetylation or formylation of the resulting hydroxylated amine to give the hydroxamic acid.11,12 Fosmidomycin or FR-900098 biosynthesis exemplifies this pathway involving acetylation or formylation, respectively (Supplementary Figure 1a).12 However, the biosynthesis of a terminal hydroxamic acid group is unknown. In this context, trichostatin A (TSA, 1) (Figure 1) is a suitable molecular target for elucidation of an unprecedented biosynthetic mechanism for the terminal hydroxamic acid moiety. TSA was first reported as an antifungal antibiotic in 197613 and rediscovered as a potent histone deacetylase (HDAC) inhibitor in 1990.14 Crystallographic studies have demonstrated that the hydroxamic acid group of TSA is crucial to the mechanism of HDAC inhibition because of its ability to chelate the zinc(II) ion in the active site of HDAC.15 Furthermore, the aliphatic moiety of TSA is suitable for the placement of the hydroxamic acid group near the active site of HDAC to enable efficient coordination of the zinc(II) ion by the hydroxamic acid moiety. Because of this property, TSA is generally utilized as a reagent in epigenetics research, and many studies on TSA have referred to biological activities.16−19 However, no biosynthetic studies on TSA have been reported since its discovery. In this paper, we describe the whole biosynthetic pathway of TSA, especially shedding light on the enzymatic machinery required for the formation of the terminal hydroxamic acid, which involves a nonproteinogenic amino acid, L-glutamic acid γ-monohydroxamate (GluHx), that is unprecedented in natural product biosynthesis. To identify the whole gene cluster responsible for TSA biosynthesis, we first performed genome scanning of a draft genome sequence of Streptomyces sp. RM72, a producer of TSA and several TSA analogues (Supplementary Figure 2).20 Draft genome sequencing yielded 8.0 Mb of DNA sequences divided into 288 scaffolds. According to the chemical structure of TSA, we hypothesized that the TSA skeleton is biosynthesized by a type-I polyketide synthase (PKS) and that the pab genes responsible for the biosynthesis of p-aminobenzoic acid (PABA), which is a possible starter unit for the PKS, constitute a biosynthetic gene cluster with the PKS genes. Using these genes as queries, we located a putative gene cluster for TSA biosynthesis spanning over 28.7 kb (Figure 2a). To clarify the gene cluster responsible for TSA biosynthesis, we performed a

ABSTRACT: Trichostatin A (TSA) is widely used in the field of epigenetics because it potently inhibits histone deacetylase (HDAC). In-depth studies have revealed that the hydroxamic acid group in TSA chelates the zinc(II) ion in the active site of HDAC to realize the inhibitory activity. Here we report the first identification of a complete TSA biosynthetic gene cluster from Streptomyces sp. RM72 and the heterologous production of TSA in Streptomyces albus. Biochemical analyses unambiguously demonstrate that unprecedented biosynthetic machinery catalyzes the direct transfer of hydroxylamine from a nonproteinogenic amino acid, L-glutamic acid γ-monohydroxamate, to the carboxylic acid group of trichostatic acid to form the hydroxamic acid moiety of TSA. The present study establishes the biosynthetic pathway of TSA, paving the way toward understanding the biosynthesis of other hydroxamic acidcontaining natural products.

N

atural products that contain a hydroxamic acid moiety often exhibit inhibitory effects against various enzymes due to the important role of this moiety as an excellent ligand for metal ions.1−4 Therefore, hydroxamic acid-containing natural products are promising drug lead compounds.5 The hydroxamic acid moiety is often found in the N-hydroxy peptides of siderophores6 and rarely found in the terminal groups of molecules (Figure 1).7−10 The biosynthetic pathway of the hydroxamic acid moiety in siderophores involves the oxidation of a primary amine by a FAD-dependent mono-

Figure 1. Structures of trichostatin A (1) and other naturally occurring terminal hydroxamic acids. © 2017 American Chemical Society

Received: March 1, 2017 Published: May 12, 2017 6799

DOI: 10.1021/jacs.7b02071 J. Am. Chem. Soc. 2017, 139, 6799−6802

Communication

Journal of the American Chemical Society

Next, we sequenced and analyzed the inserted DNA of cos10-4. The inserted DNA had a length of 41 194 bp and contained 26 open reading frames (ORFs), including the genes for type-I PKSs, PabAB, and PabC (Supplementary Table 1 and Figure 2a). Bioinformatics analysis revealed three PKS genes (tsnB1−3) harboring three ketoacyl synthase domains that are responsible for three decarboxylative condensations in the chain elongation process, which utilizes two methylmalonylCoA and one malonyl-CoA as building blocks. The structure of the tailoring domains, ketoreductase and dehydratase, in the PKS genes is reasonable for the biosynthesis of the TSA skeleton. In contrast, the putative loading module in TsnB1 is unusually composed of only an acyl carrier protein (ACP). We suspect that the loading of the starter unit onto the unusual PKS may involve CoA ligase encoded by tsnB12. Taken together, the bioinformatics data on the tsnB gene cluster for TSA biosynthesis suggest that CoA ligase (tsnB12), the PKSs (tsnB1−3), and methyltransferase (tsnB8) synthesize trichostatic acid (TS acid, 2),22 a putative intermediate of TSA, from PABA. In the biosynthesis, TsnB8 likely catalyzes two methyl transfer reactions to form dimethyl-PABA or dimethyl-PABAACP (Figure 2c). Considering that both tsnB4 and tsnB5 are transcriptional regulators and that tsnB10, tsnB11, tsnB13, and tsnB14 are likely responsible for supplying PABA, we hypothesized that the remaining tsnB genes, tsnB6, tsnB7, and tsnB9, may be involved in the formation of the hydroxamic acid group of TSA (Figure 2c). TsnB9 has 38% sequence identity and 56% sequence similarity to asparagine synthetase (AsnB) of Bacillus subtilis. AsnB is a class-II amidotransferase, which is a member of the N-terminal nucleophile hydrolase family.23 The N-terminal domain of AsnB hydrolyzes glutamine to produce glutamic acid and ammonia, and the C-terminal domain then catalyzes a condensation reaction between aspartic acid and the resulting ammonia to produce asparagine in an ATP-dependent manner (Supplementary Figure 3a).24 Curiously, it has been reported that Escherichia coli AsnB promiscuously accepts GluHx as well as Gln in vitro.25 Similar N- and C-terminal domains are found in TsnB9. This similarity suggests that TsnB9 may catalyze the formation of an amide bond in TSA in the presence of ATP. TsnB7 has 24% sequence identity to AurF, a member of the ferritin-like superfamily that catalyzes the oxidation of PABA using molecular oxygen to give p-nitrobenzoic acid (Supplementary Figure 1c), which is a precursor for aureothin biosynthesis.26,27 The oxidation mechanism catalyzed by AurF via p-(hydroxyamino)benzoic acid suggests that TsnB7 may function as a monooxygenase for the nitrogen atom in an amide bond (Supplementary Figure 1b). Finally, TsnB6 is a hypothetical protein without any known conserved domains. However, we hypothesized that TsnB6 also participates in TSA biosynthesis because the tsnB6−9 genes likely form a functional operon. To test our hypotheses, we fed TS acid to the S. albus transformants carrying the plasmids harboring tsnB6, tsnB7, and/or tsnB9. TS acid was converted into TSA only in the S. albus::pSE-tsnB9-7-6 culture (Supplementary Figure 4). This result indicates that the tsnB6, tsnB7, and tsnB9 genes are all indispensable for the conversion of TS acid to TSA in vivo. Then, to clarify the catalytic functions of each gene product, we overexpressed TsnB6, TsnB7, and TsnB9 in E. coli as Histagged proteins and purified each protein to near homogeneity (Supplementary Figure 5). First, the enzyme activity of TsnB9

Figure 2. Heterologous expression of the biosynthetic gene cluster for 1 and proposed biosynthetic pathway. (a) Gene organization of cos104. The arrows depicted in rigid lines (tsnB1−14; GenBank accession number LC217606) are the predicted biosynthetic gene cluster for 1, and the arrows in light gray (orf1−12) are presumed to have no relationship. The predicted functions of the genes are summarized as shown at the bottom. (b) UHPLC analysis of culture extracts from S. albus::cos10-4 (I), Streptomyces sp. RM72 (parent strain) (II), and S. albus::pKU465cos (negative control) and MS spectra of 1 (m/z 303.1696 [M + H]+ in (I) and m/z 303.1701 [M + H]+ in (II); calcd for [M (C17H22N2O3) + H]+ of 1: 303.1703). (c) Proposed biosynthesis of 1.

heterologous expression experiment. We constructed a cosmid library of Streptomyces sp. RM72 using the cosmid vector pKU465cos21 and screened the library for the cosmid containing the putative gene cluster for TSA biosynthesis. A positive cosmid, named cos10-4, was integrated into the genome of Streptomyces albus G153 using the phiC31 integration site. The resultant transformant S. albus::cos10-4 was cultured for heterologous expression in TSA production medium for Streptomyces sp. RM72. The heterologous production of TSA by S. albus::cos10-4 was confirmed using UHPLC and LC/MS, indicating that cos10-4 contains whole genes responsible for the biosynthesis of TSA (Figure 2b). 6800

DOI: 10.1021/jacs.7b02071 J. Am. Chem. Soc. 2017, 139, 6799−6802

Communication

Journal of the American Chemical Society

of AurF, we hypothesized that Gln could be oxidized by TsnB7 to give GluHx and then incubated TsnB7 with Gln. UHPLC analysis of the TsnB7-catalyzed reaction mixture revealed that Gln was oxidized when TsnB6 was added to the reaction mixture (Figure 3b). Thus, TsnB7 catalyzed the oxidation of the nitrogen atom in the amide bond of Gln with the support of Fe(NH4)2(SO4)2, NADH, and phenazine methosulfate as cofactors as well as TsnB6. This result shows that the formation of GluHx is totally dependent on both enzymes. Interestingly, a negligible amount of GluHx was detected in the absence of TsnB6, which suggests that TsnB7 is the true catalyst for the oxidation of Gln and that TsnB6 has a supportive function in this reaction, although the function of TsnB6 remains elusive. Because TsnB6 enhances the production of GluHx in vitro, we suspect that TsnB6 functions in the regeneration of the diiron(II/II) center of TsnB7. Through the biosynthetic study of TSA, we identified TsnB7 as a novel enzyme that synthesizes GluHx, which is an unprecedented amide donor for natural product biosynthesis. TsnB7 catalyzes a single oxidation of nitrogen to produce a hydroxamic acid group in GluHx, whereas AurF and its homologues catalyze a four-electron oxidation of nitrogen.29−32 Thus, TsnB7 is a novel N-oxygenase that catalyzes two-electron oxidation. Another significant discovery is the identification of TsnB9, which catalyzes hydroxylamine transfer. Because TsnB9 significantly prefers GluHx to Gln as a substrate, we concluded that TsnB9 functions as a bona fide “hydroxyamidotransferase”. In contrast, the previously characterized asparagine synthetase AsnB promiscuously accepts Gln and GluHx.25 Thus, TsnB9 is the first enzyme in the secondary metabolism that specifically catalyzes hydroxylamine transfer from GluHx to the carboxylic acid group of an acceptor molecule. By analogy with the reaction catalyzed by the well-characterized AsnB, presumably the N-terminal domain of TsnB9 hydrolyzes GluHx to produce hydroxylamine and Glu, and the C-terminal domain then catalyzes a condensation reaction between the resulting hydroxylamine and the tentatively adenylated carboxylic acid group of TS acid (Supplementary Figure 3b), thus forming a new amide bond to give TSA. The sequence alignment of the selected amidotransferases indicates that active-site residues are all conserved, except for Ser-77 in TsnB9, which is substituted by Asn, Thr, or Cys in some amidotransferases (Supplementary Figure 10). Therefore, Ser-77 in TsnB9 may partially be involved in substrate specificity. Insights into the determinant residue(s) on the substrate specificity of TsnB9 would require mutational and structural analysis. In conclusion, we identified a complete TSA biosynthetic gene cluster from Streptomyces sp. RM72 and established the biosynthetic pathway of TSA. Most notably, we first identified the machinery for terminal hydroxamic acid biosynthesis. We unambiguously demonstrated that TsnB7, assisted by TsnB6, synthesizes GluHx and that TsnB9 then transfers hydroxylamine from GluHx to a carboxylic acid to form the hydroxamic acid in TSA biosynthesis. The amino acid sequences of TsnB9, TsnB7, and TsnB6 may lead to the discovery of novel hydroxamic acid-containing natural products through genome mining. In addition, the catalytic capability of TsnB9 hydroxyamidotransferase makes it a valuable addition to the biosynthetic toolbox that may find applications in the construction of new secondary metabolites.

was assayed using TS acid against two putative amide donors: Gln (3), which is a common substrate for class-II amidotransferases, and GluHx (4), which was an unprecedented amide donor for natural product biosynthesis at the time. If Gln is the physiological substrate of TsnB9, the product would be FL-657C,28 which should be further oxidized by TsnB7 or an unknown enzyme to give TSA. Meanwhile, if GluHx is used by TsnB9 as a substrate, the reaction product would presumably be TSA. UHPLC analysis of the TsnB9catalyzed reaction mixtures revealed that TSA and AMP formed with almost complete consumption of TS acid only in the presence of both GluHx and ATP after 15 min of incubation (Figure 3a and Supplementary Figures 6−8). Meanwhile, when

Figure 3. In vitro assays with the recombinant TsnB6, TsnB7, and TsnB9 proteins. (a) HPLC analysis of the TsnB9 reaction mixtures. Reaction schemes for both amino donors are shown. (b) HPLC analysis of the TsnB6 and TsnB7 reaction mixture. 3 was oxidized in the presence of TsnB7 to give 4. TsnB6 appeared to enhance the oxidation.

Gln was used as the amide donor, no detectable product formed after incubation for 15 min. Prolonged incubation (120 min) with Gln produced a trace amount of FL-657C (Supplementary Figure 9). However, the specific activity (8.6 × 10−4 μmol min−1 mg−1) was much lower than that in the case of GluHx (1.0 × 10−1 μmol min−1 mg−1). These results indicate that TsnB9 is an essential and specialized enzyme for the transfer of the hydroxylamine group from GluHx to the carboxylic acid group of TS acid in an ATP-dependent manner to produce TSA. Next, we investigated the origin of GluHx, which is the physiological substrate of TsnB9. Since TsnB7 is a homologue 6801

DOI: 10.1021/jacs.7b02071 J. Am. Chem. Soc. 2017, 139, 6799−6802

Communication

Journal of the American Chemical Society



(20) Hosoya, T.; Hirokawa, T.; Takagi, M.; Shin-ya, K. J. Nat. Prod. 2012, 75, 285. (21) Komatsu, M.; Uchiyama, T.; Omura, S.; Cane, D. E.; Ikeda, H. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 2646. (22) Morioka, H.; Ishihara, M.; Takezawa, M.; Hirayama, K.; Suzuki, E.; Komoda, Y.; Shibai, H. Agric. Biol. Chem. 1985, 49, 1365. (23) Smith, J. L. Biochem. Soc. Trans. 1995, 23, 894. (24) Richards, N. G.; Schuster, S. M. Adv. Enzymol. Relat. Areas Mol. Biol. 2006, 72, 145. (25) Boehlein, S. K.; Schuster, S. M.; Richards, N. G. J. Biochemistry 1996, 35, 3031. (26) He, J.; Hertweck, C. J. Am. Chem. Soc. 2004, 126, 3694. (27) Choi, Y. S.; Zhang, H.; Brunzelle, J. S.; Nair, S. K.; Zhao, H. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 6858. (28) Ishihara, M.; Morioka, H.; Shibai, H.; Komada, Y. Jpn. Kokai Tokkyo Koho JP 1989299264A 19891204, 1989. (29) Platter, E.; Lawson, M.; Marsh, C.; Sazinsky, M. H. Arch. Biochem. Biophys. 2011, 508, 39. (30) Indest, K. J.; Eberly, J. O.; Hancock, D. E. J. Gen. Appl. Microbiol. 2015, 61, 217. (31) Knoot, C. J.; Kovaleva, E. G.; Lipscomb, J. D. JBIC, J. Biol. Inorg. Chem. 2016, 21, 589. (32) Li, N.; Korboukh, V. K.; Krebs, C.; Bollinger, J. M. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 15722.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b02071. Methods and supplementary table and figures (PDF)



AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Makoto Nishiyama: 0000-0001-8143-8052 Tomohisa Kuzuyama: 0000-0002-7221-5858 Present Address §

T.O.: Graduate School of Science, Hokkaido University, Sapporo 060-0810, Japan. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the Japan Agency for Medical Research and Development (AMED) (to K.S. and T.K.) and JSPS KAKENHI (Grant 16H06453 to T.K.).



REFERENCES

(1) Hashimoto, S.; Murai, H.; Nitta, K.; Fujie, A.; Okuhara, M.; Kohsaka, M.; Imanaka, H. J. Antibiot. 1990, 43, 281. (2) Chen, D. Z.; Patel, D. V.; Hackbarth, C. J.; Wang, W.; Dreyer, G.; Young, D. C.; Margolis, P. S.; Wu, C.; Ni, Z. J.; Trias, J.; White, R. J.; Yuan, Z. Biochemistry 2000, 39, 1256. (3) Tanzawa, K.; Ishii, M.; Ogita, T.; Shimada, K. J. Antibiot. 1992, 45, 1733. (4) Bertrand, S.; Hélesbeux, J.-J.; Larcher, G.; Duval, O. Mini-Rev. Med. Chem. 2013, 13, 1311. (5) Saban, N.; Bujak, M. Cancer Chemother. Pharmacol. 2009, 64, 213. (6) Hider, R. C.; Kong, X. Nat. Prod. Rep. 2010, 27, 637. (7) Hashimoto, S.; Murai, H.; Ezaki, M.; Morikawa, N.; Hatanaka, H.; Okuhara, M.; Kohsaka, M.; Imanaka, H. J. Antibiot. 1990, 43, 29. (8) Gordon, J. J.; Devlin, J. P.; East, A. J.; Ollis, W. D.; Sutherland, I. O.; Wright, D. E.; Ninet, L. J. Chem. Soc., Perkin Trans. 1 1975, 9, 819. (9) Wang, F.; Tan, J.-W.; Liu, J.-K. Helv. Chim. Acta 2004, 87, 1912. (10) Yotsu-yamashita, M.; Kim, Y. H.; Dudley, S. C., Jr; Choudhary, G.; Pfahnl, A.; Oshima, Y.; Daly, J. W. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 4346. (11) Oves-Costales, D.; Kadi, N.; Challis, G. L. Chem. Commun. 2009, 43, 6530. (12) Johannes, T. W.; DeSieno, M. A.; Griffin, B. M.; Thomas, P. M.; Kelleher, N. L.; Metcalf, W. W.; Zhao, H. Chem. Biol. 2010, 17, 57. (13) Tsuji, N.; Kobayashi, M.; Nagashima, K.; Wakisaka, Y.; Koizumi, K. J. Antibiot. 1976, 29, 1. (14) Yoshida, M.; Kijima, M.; Akita, M.; Beppu, T. J. Biol. Chem. 1990, 265, 17174. (15) Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.; Rifkind, R. A.; Marks, P. A.; Breslow, R.; Pavletich, N. P. Nature 1999, 401, 188. (16) Miyake, Y.; Keusch, J. J.; Wang, L.; Saito, M.; Hess, D.; Wang, X.; Melancon, B. J.; Helquist, P.; Gut, H.; Matthias, P. Nat. Chem. Biol. 2016, 12, 748. (17) Thangavel, J.; Samanta, S.; Rajasingh, S.; Barani, B.; Xuan, Y.-T.; Dawn, B.; Rajasingh, J. J. Cell Sci. 2015, 128, 3094. (18) Furumai, R.; Ito, A.; Ogawa, K.; Maeda, S.; Saito, A.; Nishino, N.; Horinouchi, S.; Yoshida, M. Cancer Sci. 2011, 102, 1081. (19) Avila, A. M.; Burnett, B. G.; Taye, A. A.; Gabanella, F.; Knight, M. A.; Hartenstein, P.; Cizman, Z.; Di Prospero, N. A.; Pellizzoni, L.; Fischbeck, K. H.; Sumner, C. J. J. Clin. Invest. 2007, 117, 659. 6802

DOI: 10.1021/jacs.7b02071 J. Am. Chem. Soc. 2017, 139, 6799−6802