Discovery of a new family of dieckmann cyclases essential to tetramic acid and pyridone-based natural products biosynthesis

Article abstract of DOI:10.1021/ol5036497

Bioinformatic analyses indicate that TrdC, SlgL, LipX₂, KirHI, and FacHI belong to a group of highly homologous proteins involved in biosynthesis of actinomycete-derived tirandamycin B, streptolydigin, ?±-lipomycin, kirromycin, and factumycin, respectively. However, assignment of their biosynthetic roles has remained elusive. Gene inactivation and complementation, in vitro biochemical assays with synthetic analogues, point mutations, and phylogenetic tree analyses reveal that these proteins represent a new family of Dieckmann cyclases that drive tetramic acid and pyridone scaffold biosynthesis.

Full text of DOI:10.1021/ol5036497

Letter
pubs.acs.org/OrgLett
Discovery of a New Family of Dieckmann Cyclases Essential to
Tetramic Acid and Pyridone-Based Natural Products Biosynthesis
Chun Gui,†,‡,∥ Qinglian Li,†,∥ Xuhua Mo,†,§,∥ Xiangjing Qin,† Junying Ma,† and Jianhua Ju*,†
†CAS Key Laboratory of Tropical Marine Bio-resources and Ecology, Guangdong Key Laboratory of Marine Materia Medica, RNAM
Center for Marine Microbiology, South China Sea Institute of Oceanology, Chinese Academy of Sciences, 164 West Xingang Road,
Guangzhou 510301, China
‡University of Chinese Academy of Sciences, 19 Yuquan Road, Beijing, 110039, China
§Shandong Key Laboratory of Applied Mycology, School of Life Sciences, Qingdao Agricultural University, Qingdao 266109, China
S
* Supporting Information
ABSTRACT: Bioinformatic analyses indicate that TrdC, SlgL, LipX2, KirHI,
and FacHI belong to a group of highly homologous proteins involved in
biosynthesis of actinomycete-derived tirandamycin B, streptolydigin, α-
lipomycin, kirromycin, and factumycin, respectively. However, assignment of
their biosynthetic roles has remained elusive. Gene inactivation and
complementation, in vitro biochemical assays with synthetic analogues, point
mutations, and phylogenetic tree analyses reveal that these proteins represent a
new family of Dieckmann cyclases that drive tetramic acid and pyridone
scaffold biosynthesis.
atural products containing tetramic acid (pyrrolidine-2,4-
dione) and pyridone core scaffolds have been isolated
catalyzing a Dieckmann cyclization in vitro in the bacteria-
derived spiro-tetramates pyrroindomycins (Supporting Infor-
mation, Figure S1, C).12 Conversely, only one strategy thus far
seems to explain pyridone natural products biosynthesis;
oxidative ring expansion from the tetramic acid precursor
catalyzed by a cytochrome P450 enzyme. Examples of this
chemistry include (i) generation of 2-pyridone pretennellin B
from pretennellin A as catalyzed by the CYP450 oxidase
TenA13 and (ii) ApdE-catalyzed generation of aspyridone A
from preaspyridone (Supporting Information, Figure S1, B).14
We report here a new paradigm for biosynthesis of both
tetramic acid and pyridone scaffolds. Specifically, we report the
discovery and characterization of a family of novel Dieckmann
cyclase enzymes, representatives of which generate actino-
mycete-derived tetramic acid and pyridone natural products in
the tirandamycin B (1),15 streptolydigin (2),16 α-lipomycin
(3),17 kirromycin (4),18 and factumycin (5)19 biosynthetic
pathways (Scheme 1), using a uniform strategy.
The tirandamycin gene cluster has been identified from
marine-derived Streptomyces sp. SCSIO 1666 and also from
Streptomyces sp. 307-9.15 Early investigations revealed that
TrdH and TrdK serve as positive and negative regulators of
tirandamycin biosynthesis, respectively.15a Additionally, TrdE is
an atypical glycoside hydrolase responsible for dehydrative
installation of the C-11/C-12 double bond,15c TrdL is a novel
covalently bound FAD-dependent oxidase responsible for
N
from fungi, bacteria, and sponges and display a wide array of
antimicrobial, antitumor, and antiviral activities.1,2 Of the more
than 250 tetramic acid/pyridone-based natural products
discovered to date, fewer than 20 have had their biosynthetic
machineries rigorously characterized. These agents are typically
assembled by hybrid type-I polyketide synthase (PKS) and
nonribosomal peptide synthetase (NRPS) biosynthetic machi-
neries.1−3 Three paradigms have been used to explain the
generation of substituted tetramic acids. These include the
following: (i) the action of a reductive domain (R) within the
hybrid PKS−NRPS megasynthase terminal module as reflected
by the fungal metabolites fusaridione A, cyclopiazonic acid, and
preaspridone A, in which the dissected R domains from FsdS4
(formerly EqiS),5 didomain TR* from CpaS,6 and full length
ApdA7 have been shown in vitro to catalyze Dieckmann
cyclizations en route to the corresponding tetramates
(Supporting Information, Figure S1, A); (ii) the action of a
hybrid iterative PKS−NRPS megasynthase TE bearing module
as highlighted by the bacterial polycyclic tetramic macrolactams
(PTM) HSAF,8 frontalamides,9 and ikarugamycins,10 for which
the terminal TE domain of Orf6 in the HSAF pathway
(possessing both protease and peptide ligase activities)
orchestrates tetramic acid formation (Supporting Information,
Figure S1, B);11 and (iii) chemistry performed by products of a
conserved but phylogenetically distinct two gene cassette
(PyrD3, homologous to various acyltransferase E2 subunits 2-
oxoacid acid dehydrogenase, and PyrD4, belonging to α/β-
hydrolase superfamily) located within the PKS gene region,
Received: December 18, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/ol5036497
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. Conserved Gene Cassettes and Their Putative Roles in the Biosynthetic Pathways for Tirandamycin, Streptolydigin,
α-Lipomycin, Kirromycin, and Factumycin
oxidation of the C-10 hydroxy group to its keto form,15b,e and
TrdI is responsible for C-10 hydroxylation, C-11/C-12
epoxidation, and C-18 hydroxylation.15a,e
apramycin gene cassette using λ-RED recombination technol-
ogy to yield the ΔtrdC mutant. The mutant was identified and
confirmed on the basis of its kanamycin sensitive and
apramycin resistant phenotype and further validated by PCR
(Figure S2, Supporting Information). The ΔtrdC mutant was
fermented, and the resulting broth extracted with butanone;
HPLC analysis of the extract revealed that trdC inactivation
ablated tirandamycin B (1) production. Complementation of
the ΔtrdC mutant in trans was undertaken by cloning trdC into
our modified pSET152AKE vector.20 Introduction of the
modified vector into the ΔtrdC mutant by conjugation yielded
the ΔtrdC:trdC mutant. Fermentation of and subsequent
metabolite analysis for ΔtrdC:trdC revealed restoration of
tirandamycin B production (Figure 1), thus validating the
indispensability of TrdC activity for tirandamycin B biosyn-
thesis. These findings parallel previous studies involving slgL in
the streptolydigin pathway; an engineered ΔslgL mutant unable
to produce streptolydigin was amenable to in trans slgL
complementation that restored streptolydigin production to the
The absence of both TE and R domains within the
tirandamycin PKS−NRPS assembly line has stymied our
understanding of tetramic acid generation during tirandamycin
biosynthesis. Immediately upstream of the NRPS gene (trdD)
resides trdC; both are transcribed in the same direction. We
performed bioinformatics analysis of TrdC and its homologues
in the GenBank database; BLAST results revealed that TrdC
(276 aa) shows homology to SlgL (similarity 53%/identity
37%) residing immediately upstream of the NRPS gene (slgN2)
in the streptolydigin pathway,16 to LipX2 (similarity 55%/
identity 39%) residing immediately upstream of the NRPS gene
lipNrps in the α-lipomycin pathway,17 to KirHI (similarity
58%/identity 45%) residing immediately upstream of the
NRPS gene kirB in the kirromycin pathway,18 and to
KSE_70420 (similarity 55%/identity 41%) residing immedi-
ately upstream of the NRPS gene facHI in the factumycin
pathway.19 Intriguingly, streptolydigin, α-lipomycin, kirromy-
cin, and factumycin all belong to a group of 3-acyltetramic acid/
pyridone natural products isolated from four other actino-
mycetes; their biosynthetic assembly lines have been identified,
but none have been found to possess TE or R domains.
Consequently, we reason that this group of genes, highly
conserved both in terms of sequence homology and localization
within the gene cluster, are responsible for tetramic acid/
pyridone core structure release from the assembly line (Scheme
1).
Figure 1. HPLC analysis of fermentation broth of (I) ΔtrdC mutant
strain; (II) complementary ΔtrdC:trdC strain; (III) Streptomyces sp.
SCSIO1666. 1 represents tirandamycin B.
To determine whether TrdC plays a role in tirandamycin
biosynthesis, we inactivated trdC by gene replacement with the
B
DOI: 10.1021/ol5036497
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
ΔslgL mutant strain.16b These results have also been verified
independently in our laboratory using a different inactivation/
complementation system (Figure S4, Supporting Information).
These data collectively demonstrate that trdC and slgL are
absolutely required for tirandamycin and streptolydigin
biosyntheses, respectively.
reported proteins (Figure 3). Additionally, in vitro biochemical
assays (Figure 2) clearly indicate that KirHI represents a new
We subsequently synthesized biosynthetic intermediate
mimics to evaluate the in vitro biochemical activities of
representatives of this family of enzymes. Acetoacetyl Gly-,
acetoacetyl Ala-,5 and acetoacetyl β-Ala- were each attached as
thioesters to N-acetylcysteamine (NAC); the identities of the
resultant enzyme substrate mimics N-acetoacetyl glycyl-SNAC
(8a), N-acetoacetyl L-alanyl-SNAC (9a), and N-acetoacetyl β-
alanyl-SNAC (10a) were confirmed by MS and H and 13C
1
Figure 3. Unrooted phylogenetical tree of TrdC-type protein, TEs
involved in PTM antibiotics biosynthesis, PyrD4 and PyrD3 involved
in spiro-tetramate and pyrroindomycin synthesis, and R domain
involved in biosynthesis of pyridone compounds originated from fungi.
NMR data. Correspondingly, trdC, slgL, and kirHI were each
cloned into the pET28a(+) vector, overexpressed in Escherichia
coli BL21 (DE3), and their products purified to homogeneity to
yield soluble N-terminus His6-tagged proteins.
Enzymatic reactions with substrate mimics were performed
in the presence of 50 mM Tris−HCl (pH 7.0), 1 mM substrate
(8a, 9a, or 10a), 5 μM enzyme (TrdC, SlgL, or KirHI), and
revealed that TrdC, SlgL, and KirHI rapidly converted 8a and
9a to tetramic acid compounds 8b and 9b, respectively (Figure
2). Conversely, although the conversion of 10a to 10b1/10b2
and unique catalyst for generating 2-pyridone core scaffolds in
bacterially derived pyridone-based natural products. Notably,
the KirHI system contrasts those involved in the biosynthesis of
the 2-pyridone scaffold found in the aspyridones,14 tenellin,13
and demethylbassianin.23
To better understand the reaction mechanism catalyzed by
TrdC and its homologous proteins, we analyzed the secondary
structure of TrdC using HHblit methods.24 To our surprise,
TrdC showed the greatest similarity to a variety of TE domains
(Figure S6, Supporting Information) involved in fatty acid
synthase-,25 PKS-, and NRPS-derived natural products
including fengycin,26 erythromycin,27 and tautomycetin.28 For
the TE-catalyzed reaction, the full-length acyl/peptidyl moieties
are required for transfer to a conserved serine residue affording
the transient oxoester. Additionally, the typical catalytic triad of
Ser-Asp-His characteristic of these TEs has been validated by X-
ray crystallography.26−28 Within the TrdC family of proteins,
and in the same corresponding positions, this conserved motif
is replaced with a Cys88-Asp115-His253 triad. To explore how this
Cys for Ser substitution in the TrdC may affect TrdC function,
we converted Cys88 to Ser88 using site-directed mutagenesis.
The C88S mutant TrdC was overexpressed in E. coli BL21
(DE3), purified, and employed in in vitro enzymatic assays
using 8a as a substrate. HPLC analyses of these reactions
revealed that, contrary to its anticipated hydrolytic activity, the
C88S TrdC retained Dieckmann cyclase activity as ascertained
by comparison to reactions of 8a with native TrdC (Figure S7,
Supporting Information).
Finally, we investigated the biochemical roles of the Cys88-
Asp115-His253 residues composing the postulated catalytic triad
of TrdC by making point mutations and carrying out
subsequent in trans complementations with the ΔtrdC mutant.
In addition, the roles of 13 other conserved (mostly polar)
amino acid residues in TrdC-family homologous proteins were
investigated by mutation to nonpolar and nonreactive residues
(Figure S8, Supporting Information). Consequently, 16 trdC
point mutants were generated and each introduced into the
ΔtrdC mutant strain. Subsequent HPLC-based metabolite
analyses revealed that mutant strains harboring W50F, W70F,
Q105I, E143L, R182L, D211A, W216F, S224A, and D225A
mutations restored tirandamycin B (1) production capabilities
to the ΔtrdC strain; tirandamycin B was absent in similar
analyses of fermentations with strains possessing R29L, D23A,
F34A, C88A, D115A, E192L, and H253F mutations (Figure
Figure 2. In vitro TrdC-, SlgL-, and KirHI-catalyzed transformations of
8a, 9a, and 10a to 8b, 9b and 10b1/10b2, respectively. A and B, at 28
°C for 1 h; C, at 37 °C for 12 h. Control 1, no enzyme added; control
2, trypsin-digested enzymes. 11, dimer of N-acetylcysteamine.
was observed, the efficiency of this enzymatic transformation
was apparently quite low relative to those involving substrate
mimics 8a and 9a (Figure 2). Despite this, these data clearly
demonstrate the ability of TrdC, SlgL, and KirHI to serve as
Dieckmann cyclases. Moreover, these findings imply that these
enzymes drive formation of tetramic acid or pyridone scaffolds
in the corresponding antibiotic biosynthetic pathways.
We next phylogenetically analyzed the TrdC family of
proteins with the NRPS R or R* domains from fungal natural
product biosynthetic pathways.5−7 Similar analyses were carried
out for terminally embedded TE domains in hybrid PKS−
NRPSs for PTM natural product biosynthetic pathways21,22 and
for PyrD3 and PyrD4 within the pyrroindomycin biosynthetic
pathway;12 the ability of these proteins to catalyze Dieckmann
cyclization chemistry has been validated. Notably, TrdC and its
homologues are phylogenetically distinct from these previously
C
DOI: 10.1021/ol5036497
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(9) Blodgett, J. A.; Oh, D. C.; Cao, S.; Currie, C. R.; Kolter, R.;
Clardy, J. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 11692−11697.
(10) Zhang, G.; Zhang, W.; Zhang, Q.; Shi, T.; Ma, L.; Zhu, Y.; Li, S.;
Zhang, H.; Zhao, Y. L.; Shi, R.; Zhang, C. Angew. Chem., Int. Ed. 2014,
53, 4840−4844.
S10, Supporting Information). Thus, the three N-terminal
residues R29, D23, and F34, the three residues C88, D115, and
H253 corresponding to the catalytic triad in the TE domain, as
well as the E192 residue within the TrdC protein are essential
to TrdC activity.
(11) Lou, L.; Chen, H.; Cerny, R. L.; Li, Y.; Shen, Y.; Du, L.
Biochemistry 2012, 51, 4−6.
In summary, we have found that the hypothetical protein
TrdC is essential to tirandamycin biosynthesis. In vitro
biochemical assays with purified proteins demonstrated that
TrdC and its homologues SlgL and KirHI function as
Dieckmann cyclases capable of catalyzing formation of both
tetramic acid and pyridone scaffolds. This is the first report that
the highly conserved new family of proteins represented by
TrdC, SlgL, LipX2, KirHI, and FacHI catalyze formation of
tetramic acid and pyridone moieties during secondary
metabolism by means of Dieckmann cyclization chemistry.
Importantly, the TrdC-type motif is becoming increasingly
prevalent among actinobacteria as reflected by review of the
GenBank database. Many gene clusters displaying this Die-
ckmann cyclase motif remain silent in their host strains (Figure
S11, Supporting Information). Consequently, these homo-
logues might serve as effective genetic markers enabling the
identification of new tetramic acid and/or 2-pyridone-based
natural products using genome mining approaches.
(12) Wu, Q.; Wu, Z.; Qu, X.; Liu, W. J. Am. Chem. Soc. 2012, 134,
17342−17345.
(13) Halo, L. M.; Heneghan, M. N.; Yakasai, A. A.; Song, Z.;
Williams, K.; Bailey, A. M.; Cox, R. J.; Lazarus, C. M.; Simpson, T. J. J.
Am. Chem. Soc. 2008, 130, 17988−17996.
(14) Wasil, Z.; Pahirulzaman, K. A. K.; Butts, C.; Simpson, T. J.;
Lazarus, C. M.; Cox, R. J. Chem. Sci. 2013, 4, 3845−3856.
(15) (a) Mo, X.; Wang, Z.; Wang, B.; Ma, J.; Huang, H.; Tian, X.;
Zhang, S.; Zhang, C.; Ju, J. Biochem. Biophys. Res. Commun. 2011, 18,
341−347. (b) Mo, X.; Huang, H.; Ma, J.; Wang, Z.; Wang, B.; Zhang,
S.; Zhang, C.; Ju, J. Org. Lett. 2011, 13, 2212−2215. (c) Mo, X.; Ma, J.;
Huang, H.; Wang, B.; Song, Y.; Zhang, S.; Zhang, C.; Ju, J. J. Am.
Chem. Soc. 2012, 134, 2844−2847. (d) Carlson, J. C.; Fortman, J. L.;
Anzai, Y.; Li, S.; Burr, D. A.; Sherman, D. H. ChemBioChem 2010, 11,
564−572. (e) Carlson, J. C.; Li, S.; Gunatilleke, S. S.; Anzai, Y.; Burr,
D. A.; Podust, L. M.; Sherman, D. H. Nat. Chem. 2011, 3, 628−633.
(16) (a) Olano, C.; Gom
Lucena, A.; Carbajo, R. J.; Brana, A. F.; Men
́ ́
ez, C.; Perez, M.; Palomino, M.; Pineda-
́
dez, C.; Salas, J. A. Chem.
̃
Biol. 2009, 16, 1031−1044. (b) Gom
́
ez, C.; Olano, C.; Palomino-
Schatzlein, M.; Pineda-Lucena, A.; Carbajo, R. J.; Brana, A. F.; Men
́
dez,
̈
̃
ASSOCIATED CONTENT
* Supporting Information
■
C.; Salas, J. A. J. Antibiot. 2012, 65, 341−348.
S
(17) Bihlmaier, C.; Welle, E.; Hofmann, C.; Welzel, K.; Vente, A.;
Breitling, E.; Muller, M.; Glaser, S.; Bechthold, A. Antimicrob. Agents
̈
Detailed experimental procedures, NMR data, and spectra for
synthetic and enzymatic products. This material is available free
of charge via the Internet at http://pubs.acs.org.
Chemother. 2006, 50, 2113−2121.
(18) Weber, T.; Laiple, K. J.; Pross, E. K.; Textor, A.; Grond, S.;
Welzel, K.; Pelzer, S.; Vente, A.; Wohlleben, W. Chem. Biol. 2008, 15,
175−188.
AUTHOR INFORMATION
Corresponding Author
(19) Thaker, M. N.; García, M.; Koteva, K.; Waglechner, N.;
Sorensen, D.; Medina, R.; Wright, G. D. Med. Chem. Commun. 2012, 3,
1020−1026.
(20) Ma, J.; Wang, Z.; Huang, H.; Zuo, D.; Luo, M.; Wang, B.; Sun,
A.; Cheng, Y.; Zhang, C.; Ju, J. Angew. Chem., Int. Ed. 2011, 50, 7797−
7802.
■
*E-mail: jju@scsio.ac.cn.
Author Contributions
∥These authors contributed equally.
(21) Luo, Y.; Huang, H.; Liang, J.; Wang, M.; Lu, L.; Shao, Z.; Cobb,
R. E.; Zhao, H. Nat. Commun. 2013, 4, 2894.
Notes
(22) Olano, C.; García, I.; Gonzal
Rubio, J.; Sanchez-Hidalgo, M.; Brana, A. F.; Men
Microb. Biotechnol. 2014, 7, 242−256.
́
ez, A.; Rodriguez, M.; Rozas, D.;
The authors declare no competing financial interest.
́
́
dez, C.; Salas, J. A.
̃
ACKNOWLEDGMENTS
■
(23) Heneghan, M. N.; Yakasai, A. A.; Williams, K.; Kadir, K. A.;
Wasil, Z.; Bakeer, W.; Fisch, K. M.; Bailey, A. M.; Simpson, T. J.; Cox,
R. J. Chem. Sci. 2011, 2, 972−979.
This work was supported by MOST (2012AA092104), NSFC
(31300063, 31290233, 81425022), and CAS (XDA11030403).
(24) Remmert, M.; Biegert, A.; Hauser, A.; Soding, J. Nat. Methods
̈
2011, 9, 173−175.
REFERENCES
■
(25) Zhang, W.; Chakravarty, B.; Zheng, F.; Gu, Z.; Wu, H.; Mao, J.;
Wakil, S. J.; Quiocho, F. A. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,
15757−15762.
(1) (a) Royles, B. J. L. Chem. Rev. 1995, 95, 1981−2001.
(b) Schobert, R.; Schlenk, A. Bioorg. Med. Chem. 2008, 16, 4203−
4221. (c) de Silva, E. D.; Geiermann, A. S.; Mitova, M. I.; Kuegler, P.;
Blunt, J. W.; Cole, A. L.; Munro, M. H. J. Nat. Prod. 2009, 72, 477−
479.
(26) Samel, S. A.; Wagner, B.; Marahiel, M. A.; Essen, L. O. J. Mol.
Biol. 2006, 359, 876−889.
(27) Tsai, S. C.; Miercke, L. J.; Krucinski, J.; Gokhale, R.; Chen, J. C.;
Foster, P. G.; Cane, D. E.; Khosla, C.; Stroud, R. M. Proc. Natl. Acad.
Sci. U.S.A. 2001, 98, 14808−14813.
(2) Mo, X.; Li, Q.; Ju, J. RSC Adv. 2014, 4, 50566−50593.
(3) Boettger, D.; Hertweck, C. ChemBioChem 2013, 14, 28−42.
(4) Kakule, T. B.; Sardar, D.; Lin, Z.; Schmidt, E. W. ACS Chem. Biol.
2013, 8, 1549−1557.
(5) Sims, J. W.; Schmidt, E. W. J. Am. Chem. Soc. 2008, 130, 11149−
11155.
(28) Scaglione, J. B.; Akey, D. L.; Sullivan, R.; Kittendorf, J. D.; Rath,
C. M.; Kim, E. S.; Smith, J. L.; Sherman, D. H. Angew. Chem., Int. Ed.
2010, 49, 5726−5730.
(6) Liu, X.; Walsh, C. T. Biochemistry 2009, 48, 11032−11044.
(7) Xu, W.; Cai, X.; Jung, M. E.; Tang, Y. J. Am. Chem. Soc. 2010,
132, 13604−13607.
(8) (a) Yu, F.; Zaleta-Rivera, K.; Zhu, X.; Huffman, J.; Millet, J. C.;
Harris, S. D.; Yuen, G.; Li, X. C.; Du, L. Antimicrob. Agents Chemother.
2007, 51, 64−72. (b) Lou, L.; Qian, G.; Xie, Y.; Hang, J.; Chen, H.;
Zaleta-Rivera, K.; Li, Y.; Shen, Y.; Dussault, P. H.; Liu, F.; Du, L. J. Am.
Chem. Soc. 2011, 133, 643−645.
D
DOI: 10.1021/ol5036497
Org. Lett. XXXX, XXX, XXX−XXX