The biosynthesis of spinosyn in Saccharopolyspora spinosa: Synthesis of the cross-bridging precursor and identification of the function of SpnJ

Article abstract of DOI:10.1021/ja076580i

Spinosyns are glycosylated polyketide-derived macrolides possessing a perhydro-as-indacene core that is presumably formed via a series of intramolecular cross-bridging reactions. The unusual structure of the spinosyn aglycone suggests an intriguing biosyn

Full text of DOI:10.1021/ja076580i

Published on Web 11/07/2007
The Biosynthesis of Spinosyn in Saccharopolyspora spinosa: Synthesis of
the Cross-Bridging Precursor and Identification of the Function of SpnJ
Hak Joong Kim, Rongson Pongdee, Qingquan Wu, Lin Hong, and Hung-wen Liu*
DiVision of Medicinal Chemistry, College of Pharmacy, and Department of Chemistry and Biochemistry,
UniVersity of Texas at Austin, Austin, Texas 78712
Received September 6, 2007; E-mail: h.w.liu@mail.utexas.edu
Scheme 1
Spinosyns A (1) and D (2) are polyketide-derived macrolides
produced by Saccharopolyspora spinosa.1 The combination of 1
and 2 serves as the active ingredient in a commercial insecticide,
Naturalyte. This mixture exhibits excellent insecticidal activity with
low mammalian toxicity and little detrimental environmental
effects.2 Structurally, the spinosyns consist of a 22-membered
macrolactone ring fused to a perhydro-as-indacene core scaffold.
In addition, they are glycosylated with tri-O-methylrhamnose and
forosamine at C-9 and C-17, respectively. The aglycone portion of
spinosyns is unusual among polyketide-derived secondary metabo-
lites due to the presence of three intramolecular carbon-carbon
bonds that constitute the as-indacene skeleton. The unusual nature
of the spinosyn aglycone suggests an intriguing biosynthetic
pathway which has generated much recent attention.3,4
The spinosyn (spn) biosynthetic gene cluster was cloned from
S. spinosa.3 Sequence analysis coupled with gene disruption
experiments suggested that macrolactone 4, which is derived from
the linear polyketide precursor 3, is a likely intermediate to form
the as-indacene aglycone (9),3,4 and the proteins encoded by spnF,
spnJ, spnL, and spnM could be involved in the cross-bridging
modifications. SpnJ exhibits sequence homology to a flavin-
dependent dehydrogenase (55% identity, 68% similarity) from
Streptomyces steffisburgensis,5 and the SpnF, SpnL, and SpnM
Scheme 2
proteins show resemblance to methyltransferase (SpnF and SpnL)
and lipase (SpnM),3 but lack part of the feature residues/motifs
conserved among members of each enzyme family. Hence, their
biological functions are not apparent and their roles in the cross-
bridging reactions cannot be established with certainty.
If 4 is indeed a precursor of the polycyclic aglycone and the
cross-bridging reactions are all post-cyclization events, oxidation
of the 15-hydroxyl group of 4 to a keto group (4 f 5) and
deprotonation at C-14 to form the C-3/C-14 linkage (5 f 6 f 7)
likely occur first (Scheme 1, route A). Subsequent deprotonation
at C-14 of 7 and elimination of the 11-hydroxyl group to give 8 is
Scheme 3
expected to trigger the second cross-coupling reaction (8 f 9) to
produce the as-indacene core. However, it is also possible that the
initial oxidation of the 15-hydroxyl group takes place on the linear
precursor 3 when it is still linked to the acyl carrier protein (ACP)
associated with the polyketide synthase SpnE (Scheme 1, route B).
The keto product 10 is next cleaved from ACP catalyzed by a
thioesterase (TE) to give 5, which then undergoes sequential
cyclizations in a similar manner as described above. In this pathway,
compound 4 is bypassed.
Although a [4 + 2] cyclization reaction has been speculated to
be a key step in the biosynthesis of many natural products, few
have been investigated at the molecular level.6 In the case of the
polyketide-derived lovastatin (14, Scheme 2), which contains an
octahydronaphthalene core, in vitro study with purified polyketide
synthase (PKS) showed that it is capable of converting a linear
N-acetylcysteamine mimic of the appropriate enzyme-bound
polyketide intermediate (12) to the correct isomer of the octahy-
dronaphthalene moiety (13).6b,f,g By analogy to 14, the cross-
bridging reactions in the spinosyn case could also take place when
the polyketide intermediate 10 is still tethered to ACP. In this
scenario, C-15 oxidation should occur on the polyketide intermedi-
9
14582
J. AM. CHEM. SOC. 2007, 129, 14582-14584
10.1021/ja076580i CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Scheme 4
Scheme 5
providing proof for a 1,3-syn relationship. Silyl ether protection
followed by a two-step Lemieux-Johnson oxidation of the terminal
olefin and reduction of the resulting aldehyde gave 26. It was crucial
to conduct the oxidation of the terminal alkene under neutral pH
as a nonbuffered reaction medium resulted in unwanted deprotection
of the triethylsilyl ether at C-15 which required reprotection.
Completion of the requisite sulfone 15 was achieved by conversion
of 26 to aryl sulfide 27 under Mitsunobu conditions followed by
oxidation employing hydrogen peroxide and ammonium molyb-
date.14
Our synthesis of aldehyde 16 is depicted in Scheme 5. Asym-
metric Brown allylation of dialdehyde 28, obtained from D-
mannitol,15 followed by benzoylation of the newly generated
secondary alcohols and removal of the isopropylidene protecting
group under acidic hydrolysis afforded diol 29 in excellent overall
yield. Oxidative cleavage of the vicinal diol in 29 facilitated by
lead tetraacetate followed by sodium borohydride reduction of the
ensuing aldehyde afforded primary alcohol 30. Next, protection of
the alcohol functionality in 30 as its PMB ether followed by
exchange of the benzoate protecting groups for silyl ethers provided
31. Following an oxidation of the terminal olefin in 31 using Jin’s
protocol,16 the resulting aldehyde was subjected to Takai olefina-
tion,17 affording the desired (E)-vinyliodide 32 in good overall yield
as a single stereoisomer. Oxidative removal of the PMB ether
employing DDQ in aqueous dichloromethane and oxidation of the
liberated hydroxy group with Dess-Martin periodinane then
provided aldehyde 16.
With fragments 15 and 16 in hand, we assembled the presumed
biosynthetic precursors 3 and 4 (Scheme 6). The use of KHMDS
in THF provided the best yield and stereoselectivity in forming
the C(12)-C(13) olefin by the Julia-Kocienski olefination18 pro-
tocol. Completion of the triene moiety was accomplished by a Heck
reaction involving 33 with ethyl dienoate 1719 under ligand-free
conditions,20 affording 34 as a mixture of isomers (E/Z ) 6/1 at
C(4)-C(5)) that were separated by flash column chromatography.
Chemoselective deprotection of the TES ether utilizing PPTS in
ethanol followed by saponification furnished seco-acid 35. Acid
35 was then subject to Yamaguchi macrolactonization21 followed
by global deprotection of the silicon-based protecting groups
employing HF•Pyr in cold ethanol for 4 days to give 4 in moderate
yields.22 The preparation of the linear substrate 3 was achieved in
a similar manner. Due to the difficulty of preparing the ACP-bound
linear polyketide chain, the intended linear substrate 3 was
synthesized as the N-acetylcysteamine (NAC) thioester 3′. Since
the corresponding N-acetylcysteamine (NAC)-linked thioesters have
been commonly used in the biosynthetic studies as surrogates for
ACP-bound acyl esters, 3′ is expected to be an effective substitute
for 3. As shown in Scheme 6, saponification of 34 afforded acid
36, which was then coupled with N-acetylcysteamine under
Yamaguchi esterification conditions. Final deprotection using
HF•Pyr complex led to the desired linear substrate 3′.22
ate 3 to give 10, which would then undergo cyclization to form
the tricyclic polyketide 11 (Scheme 1, route C). In this route, the
intermediacies of 4-8 are no longer necessary.
Intrigued by the complexity of the tetracyclic nucleus of the
spinosyns, especially the possible involvement of a Diels-Alder-
type [4 + 2] cyclization (7 f 8 or 10 f 11), we decided to study
the proposed cross-bridging reactions by first focusing on the initial
oxidation step to distinguish the three possible pathways. Herein,
we report the expression and purification of SpnJ, the chemical
synthesis of 3 and 4, and the results showing that 4, but not 3, is
a substrate for SpnJ. Our results clearly demonstrate that SpnJ is
the predicted oxidase and is specific for C-15 oxidation of mac-
rolactone 4. These findings provide compelling evidence establish-
ing the early sequence of events in the cross-bridging reactions.
Outlined in Scheme 3 is our retrosynthetic analysis for the
preparation of 3 and 4, in which two main couplings were
envisioned: a Julia-Kocienski olefination of sulfone 15 with
aldehyde 16 to establish the E-alkene geometry at C-12/C-13, and
a palladium-mediated Heck reaction to connect the C-5/C-6
carbon-carbon bond with 17. Once the linear polyketide 3 is
constructed, a Yamaguchi macrolactonization to form the ester
linkage at C-1 will afford 4.7
Our synthesis of the aryl sulfone 15, shown in Scheme 4, began
with the Weinreb amide 18 derived from δ-valerolactone.8 Asym-
metric addition9 of diethyl zinc catalyzed by (-)-(1S,2R)-N,N-
dibutylnorephedrine [(-)-DBNE] provided 19, whose enantiomeric
purity (92% ee) and absolute configuration were determined by
NMR analysis of the corresponding (R)- and (S)-Mosher esters.
Subsequent silylation followed by reduction furnished 20. Aldol
reaction10 between 20 and Crimmins’ oxazolidinethione auxiliary
21 gave the syn-aldol adduct 22. The configurations at C-16/C-17
were determined by application of the Zimmerman-Traxler transi-
tion state model as well as analysis of the J16,17 values, which are
in accord with literature precedent.11 Compound 23, derived from
22 via silylation and reduction, was subjected to Ley oxidation and
then Brown’s (+)-Ipc2B(allyl) reagent12 to yield 24 as a single
stereoisomer. The relative configurations at C-15/C-17 were
determined by conversion of 24 to the corresponding 1,3-acetonide
followed by application of Rychnovsky’s 13C NMR method,13
9
J. AM. CHEM. SOC. VOL. 129, NO. 47, 2007 14583

C O M M U N I C A T I O N S
Scheme 6
Figure 1. (1) TLC and (2) HPLC analysis of incubation mixture without
SpnJ (A) or with SpnJ (B)21 (TLC: silica gel, CH2Cl2/MeOH ) 93:7;
HPLC: C-18, 2% NH4OAc/CH3CN ) 70:30 to 20:80 over 120 min).
remaining enzymes (SpnF, SpnL, and SpnM) believed to be
involved in the cross-bridging event as well as their biological roles
leading to an understanding of the mechanism of formation for the
spinosyn aglycone.
Acknowledgment. This work was supported by grants from
the National Institutes of Health (GM35906 and GM54346).
Supporting Information Available: Experimental procedures
(PDF). This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Kirst, H. A.; Michel, K. H.; Martin, J. W.; Creemer, L. C.; Chio, E. H.;
Yao, R. C.; Nakatsukasa, W. M.; Boeck, L. D.; Occolowitz, J. L.; Paschal,
J. W.; Deeter, J. B.; Jones, N. D.; Thompson, G. D. Tetrahedron Lett.
1991, 32, 4839-4842.
With sufficient quantities of 3′ and 4 now available, we evaluated
whether the linear polyketide thioester 3′ or the macrolactone 4 is
a substrate for SpnJ, which is predicted to be an oxidase. Since the
functions of SpnF, SpnL, and SpnM are not obvious, it would be
interesting to examine whether SpnJ is a multifunctional enzyme
capable of catalyzing not only oxidation at C-15 but also depro-
tonation at C-14 to initiate the Michael addition to form a C-C
bond between C-3 and C-14, or the 1,4-elimination of the 11-OH
group setting the stage for a possible Diels-Alder-type [4 + 2]
cycloaddition.
To characterize SpnJ, the spnJ gene was cloned into a pET28b-
(+) vector and the resulting construct was expressed in E. coli
BL21(DE3) cells.22 After induction under 25 mM isopropyl â-D-
thiogalactopyranoside (IPTG), the N-terminal His6-tagged SpnJ
protein (60.6 kDa) was purified to near homogeneity using Ni-
NTA resin.22 The absorption spectrum of the purified protein is
typical for that of a flavoprotein. To determine the function of SpnJ,
SpnJ (20 µM) was incubated with 2 mM 3′ and 40 µM FAD in
Tris-HCl buffer (pH 8.0) at 29 °C. No new product was detected
after overnight incubation. Because 3′ cannot be processed by SpnJ,
it is reasonable to conclude that 3 is not a substrate for SpnJ.
In contrast, TLC and reverse-phase HPLC analysis of the
incubation mixture containing SpnJ (20 µM), 2 mM 4, and 40 µM
FAD in Tris-HCl buffer (pH 8.0) revealed the time-dependent
formation of a new product (nearly quantitative conversion after 4
h at 29 °C, see Figure 1), which was identified by 1H and 13C NMR
spectroscopy as well as HR-CIMS analysis as the C-15 ketone
derivative 5.22 These results clearly established that SpnJ is a flavin
oxidase and 4 is the substrate for SpnJ. The fact that no trace of
cross-bridging products was detected indicated that SpnJ is only a
monofunctional oxidase. These results provide strong support for
the conversion of 4 to 5 as the first step in the post-cyclization
reactions (vide supra) and set the stage for the study of the putative
Diels-Alderase activity.
(2) (a) Sparks, T. C.; Crouse, G. D.; Durst, G. Pest. Manag. Sci. 2001, 57,
896-905. (b) Salgado, V. L.; Sparks, T. C. In ComprehensiVe Molecular
Insect Science; Iatrou, K., Gill, S. S., Gilbert, L. I., Eds.; Pergamon:
Oxford, 2005; Vol. 6, pp 137-173.
(3) Waldron, C.; Matsushima, P.; Rosteck, P. R., Jr.; Broughton, M. C.; Turner,
J.; Madduri, K.; Crawford, K. P.; Merlo, D. J.; Baltz, R. H. Chem. Biol.
2001, 8, 487-499.
(4) Martin, C. J.; Timoney, M. C.; Sheridan, R. M.; Kendrew, S. G.;
Wilkinson, B.; Staunton, J. C.; Leadlay, P. F. Org. Biomol. Chem. 2003,
1, 4144-4147.
(5) Gullon, S.; Olano, C.; Abdelfattah, M. S.; Brana, A. F.; Rohr, J.; Mendez,
C.; Salas, J. A. Appl. EnViron. Microbiol. 2006, 72, 4172-4183.
(6) (a) Solanapyrone Synthase: Katayama, K.; Kobayashi, T.; Oikawa, H.;
Honma, M.; Ichihara, A. Biochim. Biophys. Acta 1998, 1384, 387-395.
(b) Lovastatin biosynthesis: Auclair, K.; Sutherland, A.; Kennedy, J.;
Witter, D. J.; Van den Heever, J. P.; Hutchinson, C. R.; Vederas, J. C. J.
Am. Chem. Soc. 2000, 122, 11519-11520. Macrophomate synthase: (c)
Ose, T.; Watanabe, K.; Mie, T.; Honma, M.; Watanabe, H.; Yao, M.;
Oikawa, H.; Tanaka, I. Nature 2003, 422, 185-189. (d) Guimaraes, C.
R.; Udier-Blagovic, M.; Jorgensen, W. L. J. Am. Chem. Soc. 2005, 127,
3577-3588. (e) Serafimov, J. M.; Westfeld, T.; Meier, B. H.; Hilvert, D.
J. Am. Chem. Soc. 2007, 129, 9580-9581. (f) Review: Oikawa, H. Bull.
Chem. Soc. Jpn. 2005, 78, 537-554. (g) Stocking, E. M.; Williams, R.
M. Angew. Chem., Int. Ed. 2003, 42, 3078-3115.
(7) The configurations at C-9, -11, -15,4 and -17 were assigned based on the
fact that the four ketoreductase (KR) domains in the spn PKS assembly
are all type A KRs (Caffrey, P. ChemBioChem 2003, 4, 654-657).22
(8) Molander, G. A.; McWilliams, J. C.; Noll, B. C. J. Am. Chem. Soc. 1997,
119, 1265-1276.
(9) Soai, K.; Yokoyama, S.; Hayasaka, T.; Ebihara, K. Chem. Lett. 1988,
843-846.
(10) Crimmins, M. T.; King, B. W.; Tabet, E. A.; Chaudhary, K. J. Org. Chem.
2001, 66, 894-902.
(11) Heathcock, C. H.; Pirrung, M. C.; Sohn, J. E. J. Org. Chem. 1979, 44,
4294-4299.
(12) Brown, H. C.; Randad, R. S.; Bhat, K. S.; Zaidlewicz, M.; Racherla, U.
S. J. Am. Chem. Soc. 1990, 112, 2389-2392.
(13) Rychnovsky, S. D.; Rogers, B. N.; Richardson, T. I. Acc. Chem. Res.
1998, 31, 9-17.
(14) Paquette, L. A.; Chang, S. K. Org. Lett. 2005, 7, 3111-3114.
(15) Saniere, M.; Lemerrer, Y.; Barbe, B.; Koscielniak, T.; Depezay, J. C.
Tetrahedron 1989, 45, 7317-28.
(16) Yu, W.; Mei, Y.; Kang, Y.; Hua, Z.; Jin, Z. Org. Lett. 2004, 6, 3217-
3219.
(17) Evans, D. A.; Black, W. C. J. Am. Chem. Soc. 1993, 115, 4497-4513.
(18) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley, A. Synlett 1998,
26-28.
In summary, the experiments described herein detail a convergent
synthesis of 4 postulated to be involved in spinosyn biosynthesis.
More importantly, we have validated the catalytic function of SpnJ
and established 4 as the precursor of the tricyclic nucleus of
spinosyns. Future work will encompass characterization of the
(19) Rodriguez, J.; Waegell, B. Synthesis 1988, 534-535.
(20) Jeffery, T. Tetrahedron Lett. 1985, 26, 2667-2670.
(21) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull. Chem.
Soc. Jpn. 1979, 52, 1989-1993.
(22) See Supporting Information for spectral data and experimental details.
JA076580I
9
14584 J. AM. CHEM. SOC. VOL. 129, NO. 47, 2007