In vitro biosynthesis of the antitumor agent azinomycin B

Article abstract of DOI:10.1021/ol052987l

Azinomycins have potential therapeutic value as antitumor agents; however, their biosynthesis is poorly understood. Here, we provide the first demonstration of a protein cell-free system capable of supporting complete in vitro biosynthesis of the antitumo

Full text of DOI:10.1021/ol052987l

ORGANIC
LETTERS
2006
Vol. 8, No. 6
1065-1068
In Vitro Biosynthesis of the Antitumor
Agent Azinomycin B
Chaomin Liu, Gilbert T. Kelly, and Coran M. H. Watanabe*
Department of Chemistry, Texas A&M UniVersity, College Station, Texas 77843
watanabe@mail.chem.tamu.edu
Received December 9, 2005
ABSTRACT
Azinomycins have potential therapeutic value as antitumor agents; however, their biosynthesis is poorly understood. Here, we provide the
first demonstration of a protein cell-free system capable of supporting complete in vitro biosynthesis of the antitumor agent azinomycin B.
The cell-free system is utilized to probe the cofactor dependence and substrate requirements of the pathway en route to azinomycin.
Azinomycins A (1a) and B (1b) (Figure 1) are antitumor
antibiotics isolated from two Streptomyces species, S. sah-
[3.1.0]hexane) ring system. Coupled with an epoxide moiety,
these structural functionalities impart the ability to form
interstrand cross-links with DNA via the electrophilic C10
and C21 carbons of azinomycin and the N7 positions of
suitably disposed purine bases.4
The novel architecture, intricate functionalization, and
compelling mode of action of the azinomycins have made
these agents attractive targets from both a synthetic and a
biosynthetic perspective. Synthetic routes to the azabicyclic
system have been reported including total synthesis of
azinomycin A.5 A variety of synthetic analogues have also
been generated.6 Considerably less is known about the
biosynthetic origin of these compounds.
Figure 1. Structures of azinomycins A and B.
achiroi1 and S. griseofuscus,2 respectively. Both compounds
exhibit in vitro cytotoxic activity at submicromolar levels
and demonstrate antitumor activities comparable to that of
mitomycin C in vivo.3 Unique to this class of natural products
is the presence of an aziridino[1,2-a]pyrrolidine (1-azabicyclo-
(3) Ishizeki, S.; Ohtsuka, M.; Irinoda, K.; Kukita, K.; Nagaoka, K.;
Nakashima, T. J. Antibiot. 1987, 40, 60.
(4) (a) For a review of the azinomycins, see: Hodgkinson, T. J.; Shipman,
M. Tetrahedron 2001, 57, 4467. (b) Lown, J. W.; Majumdar, K. C. Can. J.
Biochem. 1977, 55, 630. (c) Coleman, R. S.; Perez, R. J.; Burk, C. H.;
Navarro, A. J. Am. Chem. Soc. 2002, 124, 13008. (d) Armstrong, R. W.;
Salvati, M. E.; Nguyen, M. J. Am. Chem. Soc. 1992, 114, 3144. (e) Zang,
H.; Gates, K. S. Biochemistry 2000, 39, 14968. (f) Fujiwara, T.; Saito, I.;
Sugiyama, H. Tetrahedron Lett. 1999, 40, 315. (g) Coleman, R. S.; Burk,
C. H.; Navarro, A.; Brueggemeier, R. W.; Diaz-Cruz, E. S. Org. Lett. 2002,
4, 3545.
(1) Hata, T.; Koga, F.; Sano, Y.; Kanamori, K.; Matsumae, A.; Sugawara,
R.; Hoshi, T.; Shimi, T.; Ito, S.; Tomizawa, S J. Antibiot. 1954, A7, 107-
112.
(2) (a) Nagaoka, K.; Matsumoto, M.; Oono, J.; Yokoi, K.; Ishizeki, S.;
Nakashima, T. J. Antibiot. 1986, 39, 1527. (b) Yokoi, K.; Ishizeki, S.;
Nakashima, T. Chem. Pharm. Bull. 1986, 34, 4554.
(5) (a) Coleman, R. S.; Li, J.; Navarro, A. Angew. Chem., Int. Ed. 2001,
40, 1736. (b) Hodgkinson, T. J.; Shipman, M. Tetrahedron 2001, 57, 4467.
(c) Coleman, R. S. Strategies Tactics Org. Synth. 2004, 5, 51.
10.1021/ol052987l CCC: $33.50
© 2006 American Chemical Society
Published on Web 02/22/2006

Figure 2. Demonstration of in vitro biosynthesis. Controls: cpm
( 10. Exptl: cpm ( 50. (1) Conversion to the naphthoate. (2)
Conversion to azinomycin B. Reaction conditions: CFE + [1-14C]-
malonyl-CoA, acetyl-CoA, SAM, FAD, and NADPH.
Preliminary evaluation has been performed by Lowden and
co-workers establishing the polyketide origin of these
molecules.7 Here, we report details on the first cell-free
system capable of supporting in vitro biosynthesis of
azinomycin B. All steps from the initial construction of the
naphthoate fragment of the molecule to the formation of the
azabicyclo ring sytem have been achieved. The approach sets
the stage for protein purification and examination of discrete
Figure 4. Effect of protein inhibitors on azinomycin production.
Controls: cpm ( 10. Exptl: cpm ( 50. (A) Graphical representa-
tion of data. (B) Tabulated data.
enzyme activities or the identification of genes by reverse-
genetic techniques. In this report, we illustrate this method
of cell-free enzyme (CFE) preparation to investigate the
cofactor and substrate requirements of the pathway.
The fortified crude enzyme preparation was generated by
culturing S. sahachiroi and flash freezing the cells in liquid
nitrogen. The frozen material was transferred to a bead beater
containing cell-free extract buffer (pH 7.5) and glass beads.
The cells were pulverized (utilizing 10 intermittent cycles)
and centrifuged to generate the crude cell-free extract.
Initially, we examined the cell-free preparation for its
ability to support synthesis of the naphthoate core of
azinomycin and azinomycin B. The enzyme assay was
performed in duplicate by incubating the protein extract with
acetyl-CoA, cofactors (NADPH, SAM, and FAD), and
radiolabeled malonyl-CoA for 24 h. Following incubation,
the reactions were quenched by vortexing with dichlo-
(6) (a) LePla, R. C.; Landreau, C. A. S.; Shipman, M. S.; Hartley, J. A.;
Jones, G. D. D. Bioorg. Med. Lett. 2005, 15, 2861. (b) Hodgkinson, T. J.;
Kelland, L. R.; Shipman, M. Bioorg. Med. Lett. 2000, 10, 239. (c) Casely-
Hayford, M. A.; Pors, K.; Patterson, L. H.; Gerner, C.; Neidle, S.; Searcey,
M. Bioorg. Med. Lett. 2005, 15, 653. (d) Hartley, J. A.; Hazrati, A.; Kelland,
L. R.; Khanim, R.; Shipman, M.; Suzenet, F.; Walker, L. F. Angew. Chem.,
Int. Ed. 2000, 39, 3467. Casely-Hayford, M. A.; Pors, K.; James, C. H.;
Patterson, L. H.; Hartley, J. A.; Searcey, M. Org. Biomol. Chem. 2005, 3,
3585.
Figure 3. Effect of enzyme inhibitors on naphthoate production.
Reaction 1 exptl: cpm ( 50. Reactions 1-5: cpm ( 10. (A)
Graphical representation of data. (B) Tabulated data.
(7) (a) Corre, C.; Lowden, P. A. S. Chem. Commun. 2004, 990. (b) Corre,
C.; Landreau, C. A. S.; Shipman, M.; Lowden, P. A. S. Chem. Commun.
2004, 2600.
1066
Org. Lett., Vol. 8, No. 6, 2006

cofactors as detailed above. These results were also con-
firmed by HPLC co-injection.
As an additional control for these experiments, we
examined the effect of the FAS/PKS inhibitor cerulenin8 on
these reactions as well as that of P-450 inhibitors micona-
zole,9 metyrapone,10 and chloramphenicol.11 Reactions were
incubated with 10 and 100 µM of each inhibitor, respectively,
and assayed for both naphthoate and azinomycin B produc-
tion (Figures 3 and 4). As expected, cerulenin exhibited
inhibition against formation of the naphthoate (the putative
PKS product) and azinomycin B. Likewise, all P-450
inhibitors had a marked effect on both naphthoate and
azinomycin production. This suggests minimally the involve-
ment of a P-450 oxygenase in the first step of the biosynthesis
to generate what becomes the 3′-hydroxyl of the naphthoate
fragment (Figure 1), which is subsequently methylated, via
a SAM-dependent process, to give the final product. Comple-
tion of the biosynthesis of azinomycin B requires a number
of oxidative transformations and could also involve other
P-450-dependent processes.
Figure 5. Cofactor requirements of the azinomycin biosynthetic
pathway (conversion to azinomycin B). Controls: cpm ( 10.
Exptl: cpm ( 50. (1) Reaction conditions: CFE + [1-14C]malonyl-
CoA, acetyl-CoA, SAM, FAD, and NADPH. (2) Reaction condi-
tions: CFE + [1-14C]malonyl-CoA, acetyl-CoA, SAM and NAD-
PH. (3) Reaction conditions: CFE + [1-14C]malonyl-CoA, acetyl-
CoA, ATP, SAM, FAD, and NADPH. (4) Reaction conditions:
CFE + [1-14C]malonyl-CoA, acetyl-CoA, THF, SAM, FAD,
NADPH. (5) Reaction conditions: CFE + [1-14C]malonyl-CoA,
acetyl-CoA, SAM and FAD preparation.
The cell-free system was utilized to probe the cofactor
requirements of the azinomycin biosynthetic pathway. En-
zyme assays were performed as previously described varying
only the cofactor preparation. The results are shown in Figure
5 (reaction 1, SAM, FAD, and NADPH; reaction 2, SAM
and NADPH; reaction 3, ATP, SAM, FAD, and NADPH;
reaction 4, THF (tetrahydrofolate), SAM, FAD, and NADPH;
reaction 5, SAM and FAD). Not surprisingly, elimination
of NADPH from the reaction mixture abolished production
of azinomycin B. Neither removal of THF nor FAD, on the
other hand, appeared to have an effect. Azinomycin produc-
tion thus necessitates the involvement of two cofactors,
NADPH and SAM. The participation of SAM in the
biosynthesis has been demonstrated by Lowden and co-
workers through feeding experiments.7 Both NADPH and
romethane and centrifuged. The dichloromethane fraction
was transferred to new tubes and evaporated to dryness. The
organic residue was resolubilized in a minimal volume of
dichloromethane containing standards of azinomycin B or
naphthoate. Regions of the TLC plate corresponding to
products with appropriate Rf values (as observed by UV
analysis) were scraped from the plate and analyzed by
scintillation counting. The results of this initial experiment
are depicted in Figure 2 (reaction 1, conversion to the
naphthoate; reaction 2, conversion to azinomycin B). The
control experiment for these reactions consisted of boiled
cell-free extract (5 mL) incubated with tested substrates and
Figure 6. Schematic diagram depicting possible substrate requirements of the azinomycin B biosynthetic pathway.
Org. Lett., Vol. 8, No. 6, 2006
1067

biosynthetic pathway. Proposed intermediate building blocks
are diagrammatically represented in Figure 6. Enzyme assays
were performed as previously described with the following
modification. With the exception of the CFE enzyme control
reaction (reaction 1), to each protein extract was added
unlabeled malonyl-CoA, acetyl-CoA, and radiolabeled sub-
strate. The results are depicted in Figure 7. All suspected
intermediates (as shown in Figure 6) showed positive
incorporation, except for tyrosine (an amino acid with no
involvement in the pathway) and lysine which were included
as negative controls. Although lysine could be envisioned
to give rise to the azabicyclic ring system of azinomycin B
(Figure 8), lack of incorporation of S-adenosylmethionine
Figure 8. Schematic diagram depicting the formation of the
azabicyclic ring system from lysine.
(by way of 13C-methylmethionine)7 and background level
counts observed in vitro with radiolabeled lysine contradict
such a mechanism. Moreover, THF, a cofactor that can also
facilitate methyl transfer, did not exhibit an effect in vitro.
The higher levels of incorporation observed with malonyl-
CoA vs other amino acid substrates are a reflection of the
multiple malonyl units (five) incorporated into the naphthoate
fragment of the molecule. All other substrates are represented
only once.
Figure 7. Investigation of the principle building blocks of the
pathway. Controls: cpm ( 10. Exptl: cpm ( 50. (A) Graphical
representation. (B) Tabulated data.
Our results provide the first demonstration of a cell-free
enzyme system that supports biosynthesis of the interstrand
DNA cross-linker azinomycin B. Investigations with this
enzyme preparation impart important information regarding
the substrate and cofactor requirements of the pathway.
These studies pave the way for isotopic labeling studies
and mechanistic investigations that will ultimately provide
definitive proof for the intermediacy of proposed biosynthetic
precursors and the involvement of specific cofactors.
SAM can be envisioned to play critical roles in the
construction of the naphthoate fragment of the molecule.
Surprisingly, ATP (a necessary cofactor in NRPS biosyn-
thesis) did not appear to have much of an effect on
azinomycin B production. Presumably, ATP levels in the
cell are near saturation. Alternatively, although we suspect
that the azinomycins are constructed by a PKS/NRPS system,
without elucidation of the gene cluster, we cannot exclude
the possibility of another mechanism.
Acknowledgment. We thank the Welch Foundation, the
Elsa Pardee Foundation, Texas A&M University, and the
NIH CBI Training Program (T32 GM008523) for support
of this work.
The cell-free enzyme system was further exploited to
investigate the substrate requirements of the azinomycin
(8) (a) Omura, S. Bacteriol. ReV. 1976, 40, 681. (b) Omura, S. Methods
Enzymol. 1981, 520. (c) Hiltunen, M.; Soderhall, K. Appl. EnViron.
Microbiol. 1992, 58, 1043.
(9) (a) Rodrigues, A. D.; Gibson, G. G.; Ioannides, C.; Parke, D. V.
Biochem. Pharmacol. 1987, 36, 4277. (b) Rodrigues, A. D.; Lewis, D. F.;
Ionnides, C.; Parke, D. V. Xenobiotica 1987, 17, 1315.
(10) Williams, P. A.; Cosme, J.; Vinkovic, D. M.; Ward, A.; Angove,
H. C.; Day, P. J. Science 2004, 305, 683.
Supporting Information Available: Experimental details
on the preparation of the cell-free extract, enzyme assays,
and synthesis of the naphthoate fragment (modification of
Coleman et al.).5 This material is available free of charge
via the Internet at http://pubs.acs.org.
(11) Park, J. Y.; Kim, K. A.; Kim, S. L. Antimicrob. Agents Chemother.
2003, 47, 3464.
OL052987L
1068
Org. Lett., Vol. 8, No. 6, 2006