Cytotoxic tetramic acid derivative produced by a plant type-III polyketide synthase

Article abstract of DOI:10.1021/ja2006737

The tetramic acid (2,4-pyrrolidinedione) scaffold has been recognized as an important structural feature because of its mycotoxic, antibacterial, antiviral, and antioxidant activities. This important class of natural products is reportedly produced by the type-I polyketide synthase/nonribosomal peptide synthetase (PKS/NRPS) hybrid megaenzyme systems. In contrast, the benzalacetone synthase (BAS) from Rheum palmatum is a structurally simple, plant-specific type-III PKS that catalyzes the one-step decarboxylative condensation of malonyl-CoA with 4-coumaroyl-CoA. The type-III PKS exhibits unusually broad substrate specificity and notable catalytic versatility. Here we report that R. palmatum BAS efficiently produces a series of unnatural, novel tetramic acid derivatives by the condensation of malonyl-CoA with aminoacyl-CoA thioesters chemically synthesized from l- and d-amino acids. Remarkably, the novel tetramic acid dimer d-5 formed from d-phenylalanoyl-CoA showed moderate antiproliferative activity against murine leukemia P388 cells.

Full text of DOI:10.1021/ja2006737

COMMUNICATION
pubs.acs.org/JACS
Cytotoxic Tetramic Acid Derivative Produced by a Plant Type-III
Polyketide Synthase
Toshiyuki Wakimoto, Takahiro Mori, Hiroyuki Morita, and Ikuro Abe*
Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
S Supporting Information
b
ABSTRACT: The tetramic acid (2,4-pyrrolidinedione)
scaffold has been recognized as an important structural
feature because of its mycotoxic, antibacterial, antiviral,
and antioxidant activities. This important class of natural
products is reportedly produced by the type-I polyketide
synthase/nonribosomal peptide synthetase (PKS/NRPS)
hybrid megaenzyme systems. In contrast, the benzalacetone
synthase (BAS) from Rheum palmatum is a structurally
simple, plant-specific type-III PKS that catalyzes the one-
step decarboxylative condensation of malonyl-CoA with
4-coumaroyl-CoA. The type-III PKS exhibits unusually
Figure 1. Representative tetramic acid-containing natural products.
broad substrate specificity and notable catalytic versatility.
Here we report that R. palmatum BAS efficiently produces a
series of unnatural, novel tetramic acid derivatives by the
condensation of malonyl-CoA with aminoacyl-CoA thio-
esters chemically synthesized from L- and D-amino acids.
Remarkably, the novel tetramic acid dimer D-5 formed from
D-phenylalanoyl-CoA showed moderate antiproliferative
activity against murine leukemia P388 cells.
and accept a variety of nonphysiological substrates.9,10 The catalytic
versatility is extremely useful for the production of unnatural novel
polyketides by precursor-directed biosynthesis.10,11
We previously reported that the benzalacetone synthase
(BAS) from rhubarb (Rheum palmatum) is a unique plant-
specific type-III PKS that catalyzes the one-step decarboxylative
condensation of malonyl-CoA with 4-coumaroyl-CoA as a starter
substrate, producing the C6ꢀC4 skeleton of a diketide benzal-
acetone (Scheme 1A).12 The diketide-forming activity is attrib-
uted to the characteristic substitution of the CHS’s conserved
active-site residue Phe215 with Leu in R. palmatum BAS
(numbering according to Medicago sativa CHS2).13 Indeed,
our X-ray crystallographic analysis of R. palmatum BAS at 1.8
Å resolution revealed that the F215L substitution causes not only
conformational changes at the active-site cavity but also a 2-fold
increase in the surface area of the active-site entrance of BAS
relative to that of M. sativa CHS.14 This widening of the active-
site entrance contributes to the unique substrate and product
specificities of R. palmatum BAS and allows it to accept a wide
range of exogenous substrates, including bulky N-methylanthra-
niloyl-CoA as a starter molecule.15,16
The tetramic acid (2,4-pyrrolidinedione) scaffold is reportedly
produced by the modular and iterative type-I polyketide
synthase/nonribosomal peptide synthetase (PKS/NRPS) hybrid
megaenzyme systems. This process involves initial amide bond
formation by coupling of an activated amino acid with an
polyketide moiety and subsequent cyclization by the Dieckmann
condensation.17ꢀ23 In contrast, it was anticipated that if the
structurally simple, promiscuous type-III PKS enzyme were to
accept aminoacyl-CoA as a starter substrate and catalyze the
condensation with malonyl-CoA to produce an aminoacylꢀ
polyketide intermediate, it would then be possible to generate
yrrolidin(on)e alkaloids, including tetramic acids, are quite
P
important because of their wide distribution in terrestrial
and marine organisms.1,2 They are common structural
scaffolds found in biologically active natural products such as
cylindramide,3 aurantosides,4 reutericyclin,5 and discodermide6
(Figure 1). In addition, the substituted rigid and complex ring
system containing hydrogen-bond donors and acceptors would
also be favored in synthetic materials to generate a variety of
medicinal activities.7,8 Although numerous synthetic efforts
directed toward the development of pyrrolidin(on)e alkaloids
have been performed, an efficient and novel methodology for the
construction of the tetramic acid scaffolds would contribute to
the development of versatile chemical libraries with promising
biological activities.
The members of the chalcone synthase (CHS) superfamily of
type-III polyketide synthases (PKSs) are involved in the biogen-
esis of pharmaceutically and biologically important polyphenols,
including chalcone, stilbene, and benzalacetone.9,10 The type-III
PKSs are structurally and mechanistically simple enzymes that
accept free CoA thioesters as substrates without the involvement
of an acyl carrier protein and perform iterative condensation and
cyclization reactions to produce an array of chemically and
structurally divergent polyphenol scaffolds. Remarkably, they
exhibit unusually broad and promiscuous substrate specificities
Received: January 24, 2011
Published: March 10, 2011
r
2011 American Chemical Society
4746
dx.doi.org/10.1021/ja2006737 J. Am. Chem. Soc. 2011, 133, 4746–4749
|

Journal of the American Chemical Society
COMMUNICATION
Scheme 1. Proposed Mechanism for the Formation of (A)
Benzalacetone and (B) Tetramic Acid Derivatives by
R. palmatum BAS (the HMBC Spectrum Was Recorded
in DMSO-d6)
Figure 2. Reversed-phase HPLC and ESI-MS data for the products
derived from (A) L-phenylalanoyl-CoA and (B) L-tryptophanoyl-CoA.
The chromatograms were detected using (A) 250 and (B) 280 nm
UV light.
the R-carbon and enol carbon, respectively. The unnatural, novel
tetramic acid dimer 5 is likely to be produced by the non-
enzymatic, spontaneous dimerization of 3 between each tauto-
mer via an aldol-type condensation (Scheme 1B).
The structures of these enzyme reaction products were
confirmed by comparison with chemically synthesized authentic
compounds. Thus, according to the literature,25 condensation of
N-Boc-phenylalanine with Meldrum’s acid and subsequent cycli-
zation by refluxing in AcOEt gave the N-protected tetramic acid
derivative. During the final deprotection with trifluoroacetic acid,
5 was produced by spontaneous dimerization with the concomi-
tant removal of the Boc group. The chromatographic and
spectral data of synthetic 5 were identical to those of the
enzymatic product. It should be noted that the previous synthetic
work described by Courcambeck et al.7 did not lead to the direct
deprotection of the N-protected tetramic acid derivative. There-
fore, the spontaneous dimerization was not reported during their
synthetic process.
Interestingly, R. palmatum BAS efficiently accepted both L-
and D-phenylalanoyl-CoAs to produce the corresponding tetra-
mic acid derivatives in nearly equal yields. A steady-state kinetic
analysis gave KM = 11.7 μM and kcat = 27.8 minꢀ1 for L-
phenylalanoyl-CoA and KM = 3.3 μM and kcat = 8.1 minꢀ1 for
D-phenylalanoyl-CoA.
Next, we sought to extend this reaction to the bulkier aminoacyl-
CoA thioesters from L- and D-tryptophan. As in the case of
phenylalanoyl-CoA, R. palmatum BAS accepted both L- and D-
tryptophanoyl-CoAs as starter molecules with nearly equal efficien-
cies and carried out a single condensation with malonyl-CoA to
produce a 1:10 mixture of monomeric and dimeric tetramic acid
derivatives (4 and 6) with a combined yield of ∼10% (Figure 2B).
The structures of these products were also determined on the basis of
the spectroscopic data for synthetic materials. A steady-state kinetic
analysis gave KM = 9.8 μM and kcat = 14.7 minꢀ1 for L-tryptopha-
noyl-CoA and KM = 8.4 μM and kcat = 11.0 minꢀ1 for D-
tryptophanoyl-CoA, which are comparable to the corresponding
values for phenylalanoyl-CoA. These results indicate that the sub-
strate-binding pocket of R. palmatum BAS is large enough to
accommodate the bulky L- and D-aminoacyl-CoAs as starter mole-
cules, and they are supported by the results of docking studies based
on the crystal structure of R. palmatum BAS14 (Figure 3).
the tetramic acid scaffold by intramolecular lactamization
(Scheme 1B). To test this hypothesis, we investigated the
enzymatic synthesis of tetramic acid derivatives by utilizing R.
palmatum BAS and chiral aminoacyl-CoA thioesters that were
chemically synthesized from L- and D-amino acids.
According to the published method,24 the chiral aminoacyl-
CoAs derived from L- and D-phenylalanine or tryptophan were
chemically synthesized by the succinimide-based condensation
of the corresponding N-Boc amino acids with CoASH. Final
deprotection of the Boc group afforded the desired aminoacyl-
CoAs. When L-phenylalanoyl-CoA (1) and malonyl-CoA were
incubated with the purified recombinant R. palmatum BAS,12 the
enzyme reaction generated a 1:10 mixture of two products having
different parent ion peaks [M þ H]þ at m/z 190 (minor)
and 361 (major) in the liquid chromatographyꢀelectrospray
ionization mass spectrometry (LCꢀESI-MS) spectrum, with a
combined yield of 11% (Figure 2A). These products were not
detected in control experiments performed with the boiled
1
enzyme. The spectroscopic data (LCꢀESI-MS, UV, H and
13C NMR) of the minor compound were characteristic of the
tetramic acid 3 and identical to those of a chemically synthesized
authentic compound, as described below. The minor product 3 is
thus thought to be produced by the one-step decarboxylative
condensation of L-phenylalanoyl-CoA with malonyl-CoA fol-
lowed by lactam formation by the enzyme-bound diketide
intermediate (Scheme 1B).
On the other hand, the spectroscopic data (LCꢀESI-MS, UV,
1H and 13C NMR) of the major compound indicated that the
structure was the dehydrated dimer 5 of the minor product 3.
1
Thus, in the H NMR spectrum of 5 in DMSO-d6, two sets of
signals corresponding to phenylalanine were observed. Further-
more, a sharp singlet proton at δH 5.87, which was not apparently
related to phenylalanine, was exchangeable with CD3OD and
showed correlation to the olefinic carbon at δC 118.3 in the
HMQC spectrum. Finally, the HMBC correlations with the
carbons at δC 59.7 and δC 100.2 indicated the connectivity with
Tetramic acids (2,4-pyrrolidinediones) have been widely used
in the design of pseudopeptides as aspartyl, serine, and cysteine
protease inhibitors and as nonpeptidic neuraminidase inhibitors.
4747
dx.doi.org/10.1021/ja2006737 |J. Am. Chem. Soc. 2011, 133, 4746–4749

Journal of the American Chemical Society
COMMUNICATION
efficient and novel methodology for the construction of the
tetramic acid scaffold should thus contribute to the development
of versatile chemical libraries with promising biological activities.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental details and spec-
b
troscopic data. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
abei@mol.f.u-tokyo.ac.jp
Figure 3. Active-site structure of BAS. The diketide intermediate and
the tetramic acid product 6 from (A) L-tryptophanoyl-CoA and (B) D-
tryptophanoyl-CoA are shown as green stick models.
’ ACKNOWLEDGMENT
We thank Prof. K. Sekimizu and Dr. H. Hamamoto (Graduate
School of Pharmaceutical Sciences, The University of Tokyo) for
kindly providing bacterial strains. We also thank Dr. S. Fukuzawa
(Department of Chemistry, School of Science, The University of
Tokyo) for murine leukemia P388 cells. This work was finan-
cially supported by the Nagase Science and Technology Founda-
tion and by a Grant-in-Aid for Scientific Research from the
Ministry of Education, Culture, Sports, Science, and Technology
(MEXT), Japan.
Furthermore, the cytotoxic and antibacterial activities of tetramic
acid-containing natural products have also been reported. There-
fore, the above-obtained tetramic acid derivatives were tested for
their antibacterial activity against Staphylococcus aureus, Escher-
ichia coli, and Bacillus cereus and for their cytotoxic activity
against murine leukemia P388 cells. None of the products was
active against any of the bacteria, but the tetramic acid dimer D-5
derived from D-phenylalanine showed moderate cytotoxicity
(IC50 = 1 μg/mL) against murine leukemia P388 cells. It is quite
remarkable that the differences in the stereochemistry and
aromatic ring substituent significantly affected the cytotoxic
activity. Further investigations to elucidate the detailed mechan-
ism of the cytotoxicity are now in progress.
’ REFERENCES
(1) Royles, B. J. L. Chem. Rev. 1995, 95, 1981–2001.
(2) Schobert, R.; Schlenk, A. Bioorg. Med. Chem. 2008, 16, 4203–4221.
(3) Kanazawa, S.; Fusetani, N.; Matsunaga, S. Tetrahedron Lett.
1993, 34, 1065–1068.
(4) Matsunaga, S.; Fusetani, N.; Kato, Y.; Hirota, H. J. Am. Chem. Soc.
To the best of our knowledge, there have been no reports of
the generation of the tetramic acid (2,4-pyrrolidinedione) scaf-
fold by structurally simple type-III PKS. As mentioned above,
previous reports have shown that the tetramic acid scaffold is
produced by the modular and iterative type-I PKS/NRPS hybrid
megaenzyme systems.17ꢀ23 Thus, after polyketide chain elonga-
tion, NRPS appends an additional amino acid to the thioester
terminus, thereby generating a β-ketoamide aminoacetyl thio-
ester. It has been suggested that during the final cyclization step,
the NRPS reductive (R) domain catalyzes Dieckmann conden-
sation rather than reductive cleavage of the thioester to generate
tetramic acids.19ꢀ22 In contrast, the type-III PKS BAS initially
accepts the aminoacyl-CoA as a starter substrate and then recruits
malonyl-CoA for a Claisen condensation to generate the γ-
amino-β-ketothioester (Scheme 1B). The free γ-amino group
of the enzyme-bound intermediate could cleave the thioester
bond with concomitant intramolecular lactamization. Since the
cavity volume of the active site is apparently large enough for the
monomer (3 or 4), but not for the dimer (5 or 6) (Figure 3), it is
likely that the dimerization proceeds nonenzymatically.26 The
ability of the type-III PKS to accept the bulky aminoacyl-CoA as a
starter molecule suggests that further structural variants of tetra-
mic acids can be readily prepared from various amino acids.27 The
catalytic versatility of BAS is extremely useful for precursor-
directed biosynthesis.
1991, 113, 9690–9692.
(5) H€oltzel, A.; G€anzle, M. G.; Nicholson, G. J.; Hammes, W. P.;
Jung, G. Angew. Chem., Int. Ed. 2000, 39, 2766–2768.
(6) Gunasekera, S. P.; Gunasekera, P.; McCarthy, P. J. Org. Chem.
1991, 56, 4830–4833.
(7) Courcambeck, J.; Bihel, F.; Michelis, C. D.; Quelever, G.; Kraus,
J. L. J. Chem. Soc., Perkin Trans. 1 2001, 1421–1430.
(8) Anwar, M.; Cowley, A. R.; Moloney, M. G. Tetrahedron:
Asymmetry 2010, 21, 1758–1770.
(9) Austin, M. B.; Noel, J. P. Nat. Prod. Rep. 2003, 20, 79–110.
(10) Abe, I.; Morita, H. Nat. Prod. Rep. 2010, 27, 809–838.
(11) Abe, I. Top. Curr. Chem. 2010, 297, 45–66.
(12) Abe, I.; Takahashi, Y.; Morita, H.; Noguchi, H. Eur. J. Biochem.
2001, 268, 3354–3359.
(13) Abe, I.; Sano, Y.; Takahashi, Y.; Noguchi, H. J. Biol. Chem. 2003,
278, 25218–25226.
(14) Morita, H.; Shimokawa, Y.; Tanio, M.; Kato, R.; Noguchi, H.;
Sugio, S.; Kohno, T.; Abe, I. Proc. Natl. Acad. Sci. U.S.A. 2010,
107, 669–673.
(15) Abe, I.; Abe, T.; Wanibuchi, K.; Noguchi, H. Org. Lett. 2006,
8, 6063–6065.
(16) Abe, T.; Noma, H.; Noguchi, H.; Abe, I. Tetrahedron Lett. 2006,
47, 8727–8730.
(17) Wangun, H. V. K.; Hertweck, C. Org. Biomol. Chem. 2007,
5, 1702–1705.
(18) Sch€umann, J.; Hertweck, C. J. Am. Chem. Soc. 2007,
129, 9564–9565.
(19) Eley, K. L.; Halo, L. M.; Song, Z.; Powles, H.; Cox, R. J.; Bailey,
A. M.; Lazarus, C. M.; Simpson, T. J. ChemBioChem 2007, 8, 289–297.
(20) Halo, L. M.; Marshall, J. W.; Yakasai, A. A.; Song, Z.; Butts,
C. P.; Crump, M. P.; Heneghan, M.; Bailey, A. M.; Simpson, T. J.;
Lazarus, C. M.; Cox, R. J. ChemBioChem 2008, 9, 585–594.
In conclusion, we have demonstrated for the first time that
the structurally simple type-III PKS catalyzes the formation of a
series of unnatural, novel tetramic acid derivatives. Remarkably,
the novel tetramic acid dimer D-5 from D-phenylalanoyl-CoA
showed moderate cytotoxicity against a tumor cell line. This
4748
dx.doi.org/10.1021/ja2006737 |J. Am. Chem. Soc. 2011, 133, 4746–4749

Journal of the American Chemical Society
COMMUNICATION
(21) Sims, J. W.; Schmidt, E. W. J. Am. Chem. Soc. 2008,
130, 11149–11155.
(22) Liu, X.; Walsh, C. T. Biochemistry 2009, 48, 8746–8757.
(23) 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.
(24) Walker, K.; Fujisaki, S.; Long, R.; Croteau, R. Proc. Natl. Acad.
Sci. U.S.A. 2002, 99, 12715–12720.
(25) Poncet, J.; Jouin, P.; Castro, B.; Nicolas, L.; Boutar, M.;
Gaudemer, A. J. Chem. Soc., Perkin Trans. 1 1990, 611–616.
(26) Spontaneous dimerization of the chemically synthesized mono-
mers 3 and 4 also proceeded immediately.
(27) The lack of reports that type-III PKS produce tetramic acids in
nature would reflect the fact that aminoacyl-CoAs are very rare in nature.
4749
dx.doi.org/10.1021/ja2006737 |J. Am. Chem. Soc. 2011, 133, 4746–4749