Synthesis and Incorporation of the First Polyketide Synthase Free Intermediate in Monocerin Biosynthesis

Article abstract of DOI:10.1002/anie.200352652

Enzyme essential: Feeding studies with Dreschlera ravenelii show that labeled dihydroisocoumarin 1 is metabolized into monocerin (2) in exceptionally high levels. These results provide evidence for the role of 1 as the first enzyme-free intermediate in the monocerin biosynthesis pathway and, thus, as the final product of the polyketide synthase (PKS) catalyzed part of the pathway. Compound 1 may therefore be of use in further studies to isolate and sequence the monocerin PKS gene.

Full text of DOI:10.1002/anie.200352652

Angewandte
Chemie
Monocerin Biosynthesis
Synthesis and Incorporation of the First
Polyketide Synthase Free Intermediate in
Monocerin Biosynthesis**
Lorraine C. Axford, Thomas J. Simpson,* and
Christine L. Willis
Monocerin (5) is a polyketide fungal metabolite that exhibits
antifungal, insecticidal, and plant pathogenic properties. It
has been isolated from several fungal species, including
Dreschlera monoceras,[1] D. ravenelii,[2] Exserohilum turcum,[3]
and, along with the related fusarentin ethers, for example, 4,
from Fusarium larvarum.[4] Polyketide metabolites are pro-
[*] L. C. Axford, Prof. T. J. Simpson, Prof. C. L. Willis
School of Chemistry
University of Bristol
Cantock's Close, Bristol, BS8 1TS (UK)
Fax: (+44)117-9298611
E-mail: tom.simpson@bristol.ac.uk
[**] We thank the Engineering and Physical Sciences Research Council
for support (GR/N21352) and for a research studentship (L.C.A.),
and Dr. M. P. Crump for a 600 MHz NMR spectrum.
Angew. Chem. Int. Ed. 2004, 43, 727 –730
DOI: 10.1002/anie.200352652
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
727

Communications
duced via linear, highly functionalized “poly-
ketide” intermediates that are assembled
from simple acyl coenzyme A (CoA) precur-
sors by multifunctional enzyme complexes—
polyketide synthases (PKSs)—the exact
nature of which varies according to the
source: bacteria, fungi, or higher plants.[5] In
fungi, all PKSs characterized to date belong
to the Type I class and consist of a single
multidomain protein encoded by a single
gene,[6] but the isolation and characterization
of fungal PKSs remains a significant chal-
lenge. To investigate this problem we have
developed
a polymerase chain reaction
(PCR) based method that uses oligonucleo-
tide primers designed on consensus sequen-
ces from the keto-synthase or methyl-trans-
Scheme 1. Proposed biosynthesis of monocerin (5) via PKS-bound and post-assembly
intermediates.
ferase domains of known PKS genes.[7]
A
complication with fungi, however, is that they frequently
contain several polyketide pathways and so isolation of the
desired PKS gene may still require a reverse genetics
approach. Ideally, this would involve an assay based on the
product of the PKS of interest, which is not usually the same
as the product isolated from fungal fermentations. The
isolated metabolite will almost always contain functionality
introduced during PKS-catalyzed assembly and cyclization
and also structural modifications effected by post-assembly
“tailoring” enzymes.[5] As these modifications are not neces-
sarily readily distinguishable, the exact structure of the PKS-
derived intermediate is often uncertain.
In the case of monocerin (5), stable-isotope labeling
studies[2] suggest that the immediate PKS-derived product
should be the dihydroisocoumarin 3. This would be formed as
shown in Scheme 1 by cyclization of the linear heptaketide 1
to give the enzyme-bound orsellinate analogue 2. Lactoniza-
tion by intramolecular displacement of the thioester linkage
to the PKS by the 9-hydroxy group would give 3 as the first
enzyme-free intermediate in monocerin biosynthesis. There is
much current interest in the engineering and heterologous
expression of PKS genes from different sources to permit the
rational production of novel structures.[6,8] The monocerin PKS
is likely to be of particular interest in this context as it produces
an intermediate, 1, with an initial high level of reductive
modification and ending with a more classical poly-b-ketide
moiety. Thus, a synthesis of 3 was required for two reasons: first,
in isotopically labeled form for incorporation studies to
establish its obligate intermediacy on the monocerin biosyn-
thetic pathway, and second as a standard for an assay (for
example, HPLC) to facilitate detection of monocerin PKS
activity in cell-free extracts of monocerin-producing cultures.
Subsequent isolation and purification of the PKS protein will
allow partial peptide sequencing and design of specific oligo-
nucleotide probes for isolation of the monocerin PKS gene.
Scheme 2. Synthesis of dihydroisocoumarin 3. Reagents and condi-
tions: a) NaN(SiMe3)2, THF, ꢀ788C; then MeCO2COCMe3, 5 h, 35%;
b) Sn(OTf)2, 1-ethylpiperidine, CH2Cl2, ꢀ408C, 4.5 h; then butanal,
ꢀ788C, 20 min, 88%; c) Et3N, MeNHOMe, CH2Cl2, 258C, 72 h, 92%;
d) TBDMSOTf, 2,6-lutidine, CH2Cl2, ꢀ258C, 1 h; then ꢀ258C!RT,
12 h, 98%; e) sec-BuLi, THF, ꢀ728C, 15 min; then 9, ꢀ728C, 1.25 h,
45%; f) HCl/EtOH (1%), 258C, 5 h, 87%; g) AcOH, Me4N·B(OAc)3H,
MeCN, 6.25 h, ꢀ108C, 92%; h) 3n HCl/dioxane (1:1), reflux, 48 h,
83%. Bn=benzyl, TBDMS=tert-butyldimethylsilyl Tf=triflate=tri-
fluoromethanesulfonyl, THF=tetrahydrofuran.
with pivaloyl chloride into the mixed anhydride,[10] which was
then used to acetylate the sodium salt of the thiazolidine-
thione 6. A stereoselective aldol reaction[11] of the acetylated
auxiliary 7 with butanal in the presence of tin triflate and 1-
ethylpiperidine gave alcohol 8 as a single diastereoisomer.
Displacement of the auxiliary with N,O-dimethylhydroxyl-
amine and triethylamine[11] followed by TBDMS protection of
the secondary alcohol gave the required Weinreb amide 9.
Dihydroisocoumarin
3 was prepared as shown in
Scheme 2by modification of our previously described con-
vergent route[9] with coupling of the Weinreb amide 9 and the
benzylic anion derived from benzamide 10 as the key carbon–
carbon bond-forming step. To obtain 3 with vicinal 13C labels
for incorporation studies, sodium [13C2]acetate was converted
728
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 727 –730

Angewandte
Chemie
The benzylic anion, formed by treatment of a rigorously
deoxygenated THF solution of benzamide 10[12] with sec-
butyllithium under argon, was treated with 9 to give ketone 11
in 45% optimized yield. Deprotection of the silyl ether and
stereoselective reduction[13] of the resultant b-hydroxyketone
gave the anti diol 12 as a single diastereoisomer. Lactone
formation and removal of the benzylic ethers were effected in
a single step by refluxing in 3n HCl.
must be so high that on release from the PKS it is immediately
further converted into monocerin, and so the steady-state
concentration of 3 will be negligible.
The results of this feeding study demonstrate that an
exceptionally high level of incorporation of dihydroisocou-
marin 3 into monocerin occurs, and this provides compelling
evidence for the role of 3 as an essential intermediate. The
level of incorporation, taken along with the previous stable-
isotope labeling experiments[2] is entirely consistent with the
proposed role of 3 as the first enzyme-free intermediate on
the pathway and, thus, the final product of the PKS-catalyzed
part of the pathway. This work has paved the way for the use
of 3 in further studies to isolate and sequence the monocerin
PKS gene. These are in progress.
Preliminary fermentation studies showed that monocerin
production in D. ravenelii commenced after about five days
and increased steadily to give a yield of approximately
200 mgLꢀ1 after a total of 14 days growth. [9,10-13C2]-Labeled
3 was pulse fed (30 mg in 0.6 mL of dimethyl sulfoxide
(DMSO), three aliquots at 2h intervals) to three flasks of
fresh culture after 96 h. After a further 144 h, the cultures
were worked up and monocerin (5) was isolated by flash
column chromatography and further purified by HPLC.
Initial analysis of the isolated metabolite by 1H NMR
spectroscopy showed that exceptionally high-level (60%)
incorporation of intact isotopic label from 3 had occurred.
The signal assigned to the proton 9-H shows the normal
(natural abundance) multiplet at d = 5.06 ppm flanked by
corresponding multiplets due to one-bond 13C–1H coupling
(1J = 159 Hz) from the incorporation of the 13C label from 3 at
Experimental Section
[9,10-13C2]-Labeled 3: This was prepared according to the route
shown in Scheme 2from sodium [1-,2
13C2]acetate (1.50 g,
17.86 mmol) to give dihydroisocoumarin 3 as an oil (169 mg);
[a]2D5 = + 15.6 (MeOH, c = 0.8); IR (film): n˜max = 3285, 1629, 1465,
1378, 1255, 1165 cmꢀ1; 1H NMR (400 MHz, CDCl3): d = 10.98 (s, 1H,
3-OH), 8.35 (brs, 1H, 5-OH), 6.28 (d, J = 2.2 Hz, 1H, ArH), 6.14 (brs,
1H, ArH), 4.78 (dm, J = 140 Hz, 1H, 9-H), 4.05 (brs, 1H, 11-OH),
4.05 (m, 1H, 11-H), 2.82 (m, 1H, 8-HH), 2.71 (dm, J = 13.9 Hz, 1H, 8-
HH), 1.94 (dm, J = 140 Hz, 1H, 10-HH), 1.69 (dm, J = 140 Hz, 1H,
10-HH), 1.30–1.55 (m, 4H, 12-H2 and 13-H2), 0.92(t, J = 6.8 Hz, 3H,
14-H3) ppm; 13C NMR (100 MHz, CDCl3), enriched signals only: d =
76.4 (d, J = 40 Hz, C-9), 41.8 (d, J = 40 Hz, C-10) ppm; MS (CI): m/z
(%): 269 [M+ + 1] (40%), 251 (100), 233 (16), 196 (22), 178 (15), 83
(29); HRMS (CI): calcd for C1213C2H19O5 [M+ + 1]: 269.1300; found:
269.1288.
1
C-9. Similar H–13C couplings are seen for the signals due to
the diastereotopic methylene protons centered at d = 2.52 and
2.09 ppm (1J = 130 and 135 Hz, respectively; Figure 1). In
addition to the one-bond coupling, the latter signal also shows
a clear two-bond 1H–13C coupling (2J = 6 Hz) to the vicinal 13
C
label incorporated at C-9. The high specific incorporation of 3
into monocerin was confirmed in the 13C NMR spectrum
(dC = 39.1 and 81.2ppm, 1J = 37 Hz) and the mass spectrum
([M+] 308 (73%), [M + 2] 310 (100%)).
We have analyzed the fermentation extracts for 3, but
within the limits of sensitivity of HPLC we were not able to
demonstrate its presence. Initially this may be surprising, but
given the remarkable efficiency with which 3 is incorporated
into monocerin, we would not expect it to be present in
observable quantities in the cultures. The rate of conversion
Isolation of monocerin (5): Dreschlera ravenelii (CBS 200.29) was
stored as a spore suspension in 20% glucose solution at ꢀ788C.
Potato dextrose agar plates were inoculated with spore suspension
(200 mL), spread with a glass spreader, and incubated upside down at
288C for seven days. A spore suspension prepared from this plate was
used to inoculate further potato dextrose agar plates and, after
growth, was stored in the dark at 48C. D. ravenelii was observed as
fluffy pale gray mycelia with a dark underside. A spore suspension
was made from two agar plates with sterile water (15 mL) and filtered
through glass wool. This was used to inoculate liquid culture medium
(100 mL) in five Erlenmeyer flasks (500 mL). The liquid medium was
made from d-fructose (50.0 g), mycological peptone (2.0 g), sodium
nitrate (2.0 g), potassium dihydrogen phosphate (1.0 g), potassium
chloride (0.5 g), magnesium sulfate heptahydrate (0.5 g), and iron(ii)
sulfate heptahydrate (0.01 g) dissolved in MilliQ purified water
(1000 mL). The fungus was grown as a static culture at 258C for
14 days. The mycelial mat was filtered through Merck filter paper and
washed with distilled water. The filtrate was extracted into ethyl
acetate (4 ” volume), the organic phases were combined and dried
over MgSO4, and the solvent was removed in vacuo to yield an orange
oil. This was purified by column chromatography on silica, with a
gradient eluent of 20!50% EtOAc in petrol, to give monocerin (5;
98 mg from 5 Erlenmeyer flasks, corresponding to 196 mgLꢀ1) as a
yellow oil; Rf = 0.31 (silica gel, EtOAc/petroleum ether 40–60 (1:1));
[a]2D1 = + 48.2(MeOH, c = 1.1), literature value[14] = + 53 (MeOH, c =
0.85); IR (film): n˜max = 2928, 2856, 2253, 1664, 1375, 1274, 1120, 787,
731 cmꢀ1; 1H NMR (400 MHz, CDCl3): d = 11.29 (s, 1H, ArOH), 6.60
(s, 1H, 6-H), 5.06 (ddd, J = 6.2, 3.3, 1.0 Hz, 1H, 9-H), 4.55 (d, J =
3.3 Hz, 1H, 8-H), 4.12(m, 1H, 11-H), 3.95 (s, 3H, OCH 3), 3.89 (s, 3H,
OCH3), 2.60 (ddd, J = 14.6, 8.6, 6.0 Hz, 1H, 10-HH), 2.16 (ddd, J =
14.6, 6.2, 1.0 Hz, 1H, 10-HH), 1.71 (m, 1H, 12-HH), 1.57 (m, 1H, 12-
HH), 1.40 (m, 2H, 13-H2), 0.92(t, J = 7.3 Hz, 3H, 14-H3) ppm;
13C NMR (100 MHz, CDCl3): d = 167.8 (C-1), 158.8 (C-5), 157.5 (C-
Figure 1. 600 MHz 1H NMR spectrum showing the 10-H2 signals of
monocerin (5) enriched from incorporation of [9,10-13C2]-labeled
dihydroisocoumarin 3.
Angew. Chem. Int. Ed. 2004, 43, 727 –730
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
729

Communications
3), 137.0 (C-4), 131.2(C-7), 104.5 (C-6), 101.9 (C-2), 81.3 (C-9), 78.8
(C-11), 74.5 (C-8), 60.7 (CH3O), 56.3 (CH3O), 39.1 (C-10), 38.1 (C-
12), 19.2 (C-13), 14.0 (C-14) ppm; MS (EI): m/z (%): 308 [M+]
(100%), 265 (52), 209 (30).
Feeding study: Culture medium (100 mL) in four Erlenmeyer
flasks (500 mL) was inoculated with spore suspension (1.2mL)
obtained from one agar plate (with 5 mL of sterile water) and
grown as a static culture at 258C. After 96 h, the cultures in three
flasks were fed with 13C-labeled compound 3 (30 mg) in DMSO
(0.6 mL, divided equally between the three flasks). The feeding was
pulsed into three batches, with each culture receiving 67 mL every two
hours. After a further 144 h, the mycelia were filtered and extracted
into ethyl acetate as usual. The organic extract was purified by column
chromatography, with a gradient eluent of 20!50% EtOAc in petrol,
to give a mixture of [13C2]-labeled monocerin and 3. This mixture was
then dissolved in ethyl acetate/hexane (ca. 1:1) and purified further by
preparative HPLC on a Luna 5m silica normal-phase column (250 ”
21.20 mm) at 254 nm with an ethyl acetate/hexane gradient. [9,10-
13C2]-Monocerin was isolated as
a
pale oil (4 mg); 1H NMR
(400 MHz, CDCl3): d = 11.28 (s, 1H, ArOH), 6.59 (s, 1H, 6-H), 5.06
(ddd, J = 6.1, 3.4, 1.2Hz, 0.40H, and ddd, J = 159, 5.6, 2.9 Hz, 0.60H,
9-H), 4.55 (m, 1H, 8-H), 4.12(m, 1H, 11-H), 3.95 (s, 3H, OCH 3), 3.90
(s, 3H, OCH3), 2.60 (ddd, J = 14.7, 8.7, 6.1 Hz, 0.40H, and dddd, J =
130, 14.7, 8.7, 6.1 Hz, 0.60H, 10-HH), 2.16 (ddd, J = 14.7, 6.6, 1.0 Hz,
0.40H, and dddd, J = 135, 14.7, 6.6, 6.0 Hz, 0.60H, 10-HH), 1.71 (m,
1H, 12-HH), 1.59 (m, 1H, 12-HH), 1.38 (m, 2H, 13-H2), 0.92(t, J =
7.3 Hz, 3H, 14-H3) ppm; 13C NMR (100 MHz, CDCl3), enriched
signals only: d = 81.2(d, J = 37 Hz, C-9), 39.1 (d, J = 37 Hz, C-
10) ppm; MS (EI): m/z (%): 310 [M+2] (100), 308 [M+] (73), 267
[M+2ꢀC3H7] (62), 265 [M+ꢀC3H7] (40), 209 (63), 83 (43); HRMS
(EI): calcd for C1413C2H20O6 [M+]: 310.1327; found: 310.1328.
Received: August 14, 2003 [Z52652]
Keywords: biosynthesis · fungal metabolites · monocerin ·
.
polyketide synthases · polyketides
[1] D. C. Aldridge, W. B. Turner, J. Chem. Soc. C 1970, 2598 – 2600.
[2] F. E. Scott, T. J. Simpson, L. A. Trimble, J. C. Vederas, J. Chem.
Soc. Chem. Commun. 1984, 756 – 758.
[3] D. J. Robeson, G. A. Strobel, Agric. Biol. Chem. 1982, 46, 2681 –
2683.
[4] J. F. Grove, M. Pople, J. Chem. Soc. Perkin Trans. 1 1979, 2048 –
2051.
[5] J. Staunton, K. J. Weissman, Nat. Prod. Rep. 2001, 18, 380 – 430.
[6] I. Fujii, A. Watanabe, U. Sankawa, Y. Ebizuka, Chem. Biol. 2001,
8, 189 – 197.
[7] L. E. H. Bingle, T. J. Simpson, C. M. Lazarus, Fungal Genet.
Biol. 1999, 26, 209 – 223; T. P. Nicholson, C. M. Lazarus, B. A. M.
Rudd, M. J. Dawson, T. J. Simpson, R. J. Cox, Chem. Biol. 2001,
8, 151 – 178.
[8] C. Khosla, J. Org. Chem. 2000, 65, 8127 – 8133.
[9] C. McNicholas, T. J. Simpson, N. J. Willett, Tetrahedron Lett.
1996, 37, 8053 – 8056.
[10] N. M. Kelly, R. G. Reid, C. L. Willis, P. L. Winton, Tetrahedron
Lett. 1995, 36, 8315 – 8318.
[11] N. Murakami, M. Sugimoto, M. Kobayashi, Bioorg. Med. Chem.
2001, 9, 57 – 67.
[12] P. Watts, PhD Thesis, University of Bristol, 1999; L. C. Axford,
T. J. Simpson, D. M. Smith, P. Watts, C. L. Willis, unpublished
results.
[13] R. F. Cunico, L. Bedell, J. Org. Chem. 1980, 45, 4797 – 4798;
D. A. Evans, K. T. Chapman, E. M. Carreira, J. Am. Chem. Soc.
1988, 110, 3560 – 3578.
[14] K. Mori, H. Takaishi, Tetrahedron 1989, 45, 1639 – 1646.
730
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 727 –730