Total synthesis of xerulinic acid

Article abstract of DOI:10.1002/anie.200453729

An inhibitor of the biosynthesis of cholesterol, xerulinic acid, has been synthesized for the first time. The convergent approach applied involves the palladium-catalyzed coupling of an enediynoic ester building block, a conjunctive C₆ bisstannane, and a bromine-containing methylenebutenolide (see scheme; R = Me₃SiCH₂CH ₂).

Full text of DOI:10.1002/anie.200453729

Angewandte
Chemie
Natural Products Synthesis
Total Synthesis of Xerulinic Acid**
Achim Sorg and Reinhard Brückner*
In 1990, Steglich, Anke, and co-workers identified a family of
three novel polyenes from the fungus Xerulina melanotricha
d˛rfelt: dihydroxerulin (1), xerulin (2), and xerulinic acid
[
1]
(
3). Dihydroxerulin (1) and xerulin (2) were characterized
Scheme 1. Retrosynthetic analysis of xerulinic acid.
would establish all C=C bonds and interconnect them by Cꢁ
[
6]
C-forming Stille couplings. This identified the enediyne
carboxylate 4, the hitherto unknown bisstannane 5, and the
[
7]
novel building block 6 —which ought to be generally useful
[
8]
for synthesizing g-alkylidenebutenolides —as our key pre-
cursors. Their syntheses are described in Schemes 2–4,
as 90:10–65:35 mixtures, whereas xerulinic acid (3) was
obtained pure. Each of these compounds inhibited the
biosynthesis of cholesterol in HeLa S3 cells by blocking
HMG-CoA synthase (EC 4.1.3.5); moreover, in the same cell
system, 3 suppressed the synthesis of RNA at IC ꢀ 100 mm.
5
0
The biological activity and structural uniqueness of the
xerulin family stimulated the interest of the synthetic
community. The first laboratory syntheses both of dihydro-
Scheme 2. a) LiNH (6 equiv), NH (l), ꢁ358C, 15 min; addition of 8,
2
3
3
.5 h; !RT, 17% (reference [9]: 21%); b) Dess–Martin periodinane
(
1.52 equiv), CH Cl , 08C!RT, 2.5 h, 79%; c) NaClO (2.1equiv),
2
2
2
[
2]
[3]
xerulin and xerulin were reported by our group. Rossi
KH PO (2.5 equiv), 2-methyl-2-butene (3.5 equiv), acetone/H O=3:2,
2
4
2
[
4]
et al. completed another synthesis of dihydroxerulin, and
08C, 89%; d) NBS (1.30 equiv), AgNO
3
(0.08 equiv), acetone, room
[
5]
temperature, 13 h, 79%; e) HOCH CH SiMe (1.2 equiv), DCC
Negishi, Alimardanov, and Xu prepared xerulin. Herein we
report the first total synthesis of xerulinic acid based on
unprecedented disconnections.
2
2
3
(
1.1 equiv), DMAP (0.05 equiv), ethyl acetate, 08C!RT, 2 h, 83%.
NBS=N-bromosuccinimide, DCC=dicyclohexylcarbodiimide,
DMAP=4-dimethylaminopyridine.
Scheme 1 outlines our retrosynthetic analysis. It differs
[
2]
[3]
from our routes to dihydroxerulin and xerulin in that we
intended to form CꢁC rather than C=C bonds in the final
respectively, and their coupling to give xerulinic acid (3) is
shown in Scheme 5.
steps. The latter approach—by Wittig reactions—had been
nonstereoselective irrespective of whether the ylides corre-
sponded to the right- or left-hand moiety of the target
The synthesis of enediyne carboxylate 4a began with the
synthesis of enediynol 9 from epichlorohydrin (8) and
lithiobutadiyne generated in situ from 1,4-dichlorobut-2-yne
[
2,3]
structure.
In contrast, the presently adopted approach
[
9]
(
7) and excess lithium amide (Scheme 2). Terminal bromi-
[
10]
[11]
[*] A. Sorg, Prof. Dr. R. Brückner
nation of 9 with NBS/AgNO3 led to bromoenediynol 10.
Institut für Organische Chemie und Biochemie
Universität Freiburg
Albertstrasse 21, 79104 Freiburg (Germany)
Fax: (+49)761-203-6100
The corresponding aldehyde 11 was generated by a Dess–
[12]
Martin oxidation of 10 (79% yield) and taken on to the
[13]
corresponding carboxylic acid 12 by a Lindgren oxidation
(89% yield). Esterification with 2-trimethylsilylethanol in the
E-mail: reinhard.brueckner@organik.chemie.uni-freiburg.de
presence of DCC and DMAP provided the unstable ester 4a
[
**] This work was financially supported by the Fonds der Chemischen
Industrie. We are grateful to Dr. Manfred Keller for valuable
spectroscopic support and to Melanie Waldrich for technical
assistance.
[
14]
in 83% yield.
The synthesis of bisstannane 5 proceeded highly stereo-
selectively (Scheme 3). It was assembled starting from the tin-
Angew. Chem. Int. Ed. 2004, 43, 4523 –4526
DOI: 10.1002/anie.200453729
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4523

Communications
Scheme 3. a) CBr (1.21 equiv), CH Cl , 08C, addition of PPh
3
4
2
2
(
1.10 equiv), 1 h, 82% (reference [16]: 81%); b) Na S (0.5 equiv),
2
+
ꢁ
Bu N HSO4 (0.7 mol%), H O/THF, room temperature, 7 h, 90%;
4
2
Scheme 4. a) Br (2.1equiv), CH Cl , 08C!RT, 2 h, 60% (refer-
2
2
2
c) KOH (30% on Al O ; 20 equiv), CBr F (4 equiv), THF, 08C, 30 min,
2
3
2 2
ence [25]: 40%); b) oleum/conc. H SO 2:1(v/v), 50–60 8C, 6 min;
2
4
7
1
3%; d) (NH ) Mo O (0.2 equiv), H O (10 equiv), EtOH, 08C!RT,
4
6
7
24
2
2
2
8%;[7] c) 1) P O (1.2 equiv), CH Cl , 08C!D, 1h, filtration; 2) NEt
4
10
2
2
3
h, 88%.
(1.03 equiv), CH Cl , 08C!D, 1h, 55%; d) HSnBu (1.10 equiv),
2
2
3
[
Pd(PPh ) ] (0.10 equiv), THF, 658C, 3 h, 51%.[7]
3 4
[
15]
containing alcohol 13, which was obtained by cuprostanny-
lation/protonolysis of propargyl alcohol. We transformed
compound 13 through nucleophilic attack first into the tin-
To couple bisstannane 5 and butenolide 6 in a 1:1 ratio
without too much competitive 1:2 coupling, we considered it
safest to perform one Sn!Li exchange per reagent molecule
[
16]
[28]
containing bromide 14 (82%) and subsequently into the
tin-containing sulfide 15 (90%). Oxidation with peroxo-
molybdate provided the corresponding sulfone 16 (88%).
Deprotonation of sulfone 16 in the presence of CBr F with
Al O -supported KOH
reaction to provide triene 5 in 73% yield. The resulting
sample of 5 represented a 96:4 mixture of the trans,trans,trans
and the trans,cis,trans isomers. This was inferred from the
ratio of the integrals for 2-H and 5-H (trans,trans,trans-5:
d2-H =d5-H =6.56 ppm, trans,cis,trans-5: d2-H =d5-H =7.08 ppm)
and from the magnitudes of the vicinal olefinic H,H coupling
first, a Li!Zn exchange next, and a Negishi coupling
[
17]
thereafter. The model compound 1,2-bis(tributylstannyl)-
ethylene (21) and butenolide 6 could indeed be combined
[
18]
[22]
in this manner, forming 1:1 coupling product 22 in 46% yield
(Scheme 5). To our delight, we could extend this procedure to
the 1,6-bis(stannyl)hexatriene 5 and butenolide 6 to obtain
the desired 1:1 product 23 in 63% yield. By comparison, the
combination of the same reagents through a Stille coupling in
THF in the presence of [Pd(dba) ] and AsPh led to 23 in only
2
2
[
19]
induced a Ramberg–Bäcklund
2
3
[
20]
2
3
44% yield. The all-trans configuration of 23 was proved by
1
H NMR spectroscopy (499.9 MHz, CDCl ; for the number-
3
constants in the major isomer (J = J5,6 = 18.6 Hz, J3,4
=
1
,2
[
21]
1
5.1 Hz).
1,6-Bis(tributylstannyl)hexatriene (5) ought to
be a valuable conjuctive C6 reagent in analogy to 1,2-
[
22]
bis(tributylstannyl)ethylene as a conjunctive C2 reagent
and 1,4-bis(methylstannyl)butadiene as a conjunctive C4
[
23,24]
reagent.
The first example of a conjunctive C reagent
6
is provided in the present study (see Scheme 5). Notably, we
were able to couple different electrophiles at the termini (C1
vs. C6) of 5. Such a differentiation is more difficult to realize
than the C1/C4 differentiation in the single unsymmetrical
biscoupling reported for the aforementioned C bisstanna-
4
[
23a]
ne).
We recently reported the two-step preparation of (bro-
momethylene)butenolide from dibromolevulinic acid
6
[
25a]
(
18;
Scheme 4) by oxidative cyclization (!19) and stereo-
[7]
selective reduction (!6). Since then we have been able to
shorten this synthesis by successively treating the same
[
26]
dibromolevulinic acid (18) in one pot with two reagents:
P O effects a dehydration, presumably giving rise to the enol
4
10
ester intermediate 20; thereafter triethylamine (instead of
triethylenediamine, which was used in ref. [26]) induces the b-
elimination of HBr. This gave four times as much bromobu-
Scheme 5. a) 21 (1.01 equiv), nBuLi (1.05 equiv), THF, ꢁ788C, 70 min;
then ZnCl (1.05 equiv), !ꢁ208C, 2 h; then 6, [Pd(PPh ) ] (5 mol%),
2
3 4
0
8C!RT, 45 min, 46%; b) 5 (1.3 equiv), nBuLi (1.3 equiv), THF,
ꢁ
788C, 20 min; then ZnCl (1.3 equiv), 30 min; then 6, [Pd(PPh ) ]
[
27]
2
3 4
tenolide 6 as our previous synthesis, with the same perfect
Z selectivity, both of which were prerequisites for 6 to become
a viable starting material for our study (we are unaware of
(
(
5 mol%), 08C, 1h, 63%; c) 6 (1.1 equiv), [Pd(dba) ] (5 mol%), AsPh3
2
20 mol%) THF, room temperature, 2 h, 44%; d) 4a (1.08 equiv),
[
Pd(dba) ] (6 mol%), AsPh (19 mol%), THF, room temperature, 5 h,
2
3
[
25b]
previous synthetic applications of this compound ).
73%; e) Bu NF (1.5 equiv), THF, 08C!RT, 2 h, 61%.
4
4
524
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 4523 –4526

Angewandte
Chemie
ing see Scheme 5): J6,7 = 14.9 Hz, J8,9 = 14.7 Hz, and J10,11
8.7 Hz.
The carbon skeleton of our target molecule 3 was
=
W. E. Childers, Jr., H. W. Pinnick, Tetrahedron 1981, 37, 2091 –
096 (2-methyl-2-butene used as hypochlorite scavenger).
14] Prepared by a procedure analogous to that of: a) N. Jeker, C.
Tamm, Helv. Chim. Acta 1988, 71, 1895 – 1903; b) G. T. Bourne,
D. C. Horwell, M. C. Pritchard, Tetrahedron 1991, 47, 4763 –
2
1
[
[
[
6]
[29]
[29]
completed by a Stille coupling ([Pd(dba) ],
AsPh3
)
2
between the brominated enediyne carboxylate 4a and the
stannylated heptatrienylidenebutenolide 23 (Scheme 5). Pro-
vided that exposure to atmospheric oxygen and daylight was
4774.
15] Prepared by a procedure analogous to that of: J.-F. Betzer, F.
Delaloge, B. Muller, A. Pancrazi, J. Prunet, J. Org. Chem. 1997,
62, 7768 – 7780.
[16] A. S.-Y. Lee, C.-W. Wu, Tetrahedron 1999, 55, 12531 – 12542.
17] Method: F. Reimnitz, Ph.D. thesis, Universität Freiburg, 2000,
[
30]
avoided, the reaction gave xerulinic acid ester 24 in 73%
yield. In the final step, this compound was deprotected by
[
treatment with anhydrous Bu NF in THF to give 3 in 61%
4
pp. 71–72, 147 – 148.
yield. The resulting specimen of synthetic xerulinic acid (3)
1
13
[18] The method is same as that described for the oxidation of
heterocycle-substituted sulfides en route to Julia–Lythgoe ole-
fins: P. A. Blakemore, P. J. Kocienski, S. Marzcak, J. Wicha,
Synthesis 1999, 1209 – 1215.
[19] T.-L. Chan, S. Fong, Y. Li, T.-O. Man, C.-D. Poon, J. Chem. Soc.
Chem. Commun. 1994, 1771 – 1772; pure THF was used instead
of tBuOH/THF (3:1).
was scrutinized by H, C, and 2D NMR spectroscopy at the
same field strengths (500 MHz/126 MHz) and in the same
solvent ([D ]DMSO) as the natural product. The juxtaposi-
6
[
1,31]
tion of our data and those from Steglich and co-workers
unambiguously showed that our compound and the natural
product are identical.
[
[
20] For a review, see: L. A. Paquette, Org. React. 1977, 25, 1 – 71.
In summary, we have completed the first, highly con-
vergent synthesis of xerulinic acid (3). Although xerulinic acid
is perhaps not “complex”, its preparation is by no means
simple and its tendency to “polymerize” cannot be over-
estimated. Our synthetic strategy towards 3 is distinct from
previous strategies aimed at the xerulin family of com-
1
21] 5: H NMR (499.9 MHz, CDCl , TMS): d = 0.83–0.97 (m; 6 ”
3
SnCH CH CH CH ), overlaps with 0.89 (t, J = 7.3 Hz; 6 ”
2
2
2
3
vic
SnCH CH CH CH ), 1.31 (tq, both Jvic = 7.3 Hz; 6 ”
2
2
2
3
SnCH CH CH CH ), 1.41–1.58 (m; 6 ” SnCH CH CH CH ),
2
2
2
3
2
2
2
3
6.15 (m , higher order; 3-H, 4-H), 6.29 (d, J = J6,5 = 18.6 Hz,
c
1,2
2
each peak flanked by Sn isotope satellites as 2d,
0.1 Hz, J117Sn,H = 67.2 Hz; 1-H, 6-H), 6.56 ppm (m , higher
order; 2-H, 5-H); C NMR (125.7 MHz, CDCl ): d = 9.6
flanked by Sn isotope satellites as 2d,
119Sn,H
J =
[
32]
2
7
pounds.
Bisstannane 5 should open a general route to
c
1
3
polyunsaturated molecules containing a trans,trans,trans-hex-
atriene moiety and lactone 6 should open a general route to g-
alkylidenebutenolides.
3
1
1
(
3
J
119Sn,C1’
119Sn,C1’’
= J =
1
1
44.8 Hz,
J
117Sn,C1’
= J
117Sn,C1’’
2 2 2 3
= 329.4 Hz; SnCH CH CH CH ),
1
as
3.7 (SnCH CH CH CH ), 27.3 (flanked by Sn isotope satellites
2
2
3
2
3
3
3
3
1d,
J
119Sn,C3’
= J
119Sn,C3’’
= J
117Sn,C3’
117Sn,C3’’
= J = 54.5 Hz;
Received: January 12, 2004 [Z53729]
SnCH CH CH CH ), 29.1 (flanked by Sn isotope satellites as
2
2
2
3
2
2
2
2
1
d,
J
119Sn,C2’
= J
119Sn,C2’’
= J
117Sn,C2’
= J
117Sn,C1’’
= 20.6 Hz;
Keywords: CꢁC coupling · inhibitors · lactones ·
SnCH CH CH CH ), 134.5 (flanked by Sn isotope satellites as
2 2 2 3
.
3
3
3
3
natural products · stereoselective synthesis
1d,
J
119Sn,C3 = J119Sn,C4 = J117Sn,C3 = J117Sn,C4 = 73.6 Hz; C3, C4),
1
1
J
35.6 (flanked by Sn isotope satellites as 2d,
119Sn,C6
J
119Sn,C1
=
1
1
1
= 380.0 Hz, J
117Sn,C1
= J
117Sn,C6
= 363.3 Hz; C1, C6), 146.9
2
2
(
J
flanked by Sn isotope satellites as 1d,
117Sn,C2
J
119Sn,C2
119Sn,C5
= J =
2
2
117Sn,C5
= J = 6.4 Hz; C2, C5); SELINCOR (S. Berger, J.
[
1] D. Kuhnt, T. Anke, H. Besl, M. Bross, R. Herrmann, U. Mocek,
B. Steffan, W. Steglich, J. Antibiot. 1990, 43, 1413 – 1420.
2] K. Siegel, R. Brückner, Synlett 1999, 1227 – 1230.
3] K. Siegel, R. Brückner, Chem. Eur. J. 1998, 4, 1116 – 1122.
4] R. Rossi, F. Bellina, A. Catanese, L. Mannina, D. Valensin,
Tetrahedron 2000, 56, 479 – 487.
Magn. Reson. 1989, 81, 561 – 564; 499.9 MHz/125.7 MHz,
CDCl ): J = 15.1 Hz, J3,2 = J4,5 = 10.1 Hz; elemental analysis
[
[
[
3
3,4
(
%): calcd for C H Sn (656.3): C 54.74, H 9.19; found: C 53.81,
30 60 2
H 8.97.
[
22] a) K. C. Nicolaou, T. K. Chakraborty, A. D. Piscopio, N.
Minowa, P. Bertinato, J. Am. Chem. Soc. 1993, 115, 4419 –
[
[
5] E.-i. Negishi, A. Alimardanov, C. Xu, Org. Lett. 2000, 2, 65 – 67.
6] For reviews, see: a) J. K. Stille, Angew. Chem. 1986, 98, 504 – 519;
Angew. Chem. Int. Ed. Engl. 1986, 25, 508 – 524; b) V. Farina,
G. P. Roth in Advances in Metal-Organic Chemistry, Vol. 5 (Ed.:
L. S. Liebeskind), JAI, Greenwich, Connecticut, 1996, pp. 1 – 53;
c) V. Farina, V. Krishnamurthy, W. J. Scott, Org. React. 1997, 50,
4
8
420; b) J. S. Panek, C. E. Masse J. Org. Chem. 1997, 62, 8290 –
291; c) C. E. Masse, M. Yang, J. Solomon, J. S. Panek, J. Am.
Chem. Soc. 1998, 120, 4123 – 4134; d) P. M. Pihko, A. M. P.
Koskinen, Synlett 1999, 1966 – 1968.
[
23] a) A. Kiehl, A. Eberhardt, M. Adam, V. Enkelmann, K. Müllen,
Angew. Chem. 1992, 104, 1623 – 1626; Angew. Chem. Int. Ed.
Engl. 1992, 31, 1588 – 1591; b) D. Nozawa, H. Takikawa, K. Mori,
J. Chem. Soc. Perkin Trans. 1 2000, 2043 – 2046.
[24] A fully conjugated 3,8-dimethyl-C10-bis(tributylstannane) has
also been synthesized and used as a building block: B. Vaz, R.
Alvarez, A. R. de Lera, J. Org. Chem. 2002, 67, 5040 – 5043.
[25] a) A. J. Manny, S. Kjelleberg, N. Kumar, R. de Nys, R. W. Read,
P. Steinberg, Tetrahedron 1997, 53, 15813 – 15826. b) The
aforementioned authors prepared lactone 6 for the first time
but the Experimental Section of their study details no more than
1
– 652.
[
[
7] A. Sorg, K. Siegel, R. Brückner, Synlett 2004, 321 – 325.
8] a) D. W. Knight, Contemp. Org. Synth. 1994, 1, 287 – 315; b) E.-i.
Negishi, M. Kotora, Tetrahedron 1997, 53, 6707 – 6738; c) R.
Brückner, Chem. Commun. 2001, 141 – 152; d) R. Brückner,
Curr. Org. Chem. 2001, 5, 679 – 718.
[
9] S. R. Landor, E. S. Pepper, J. Chem. Soc. C 1966, 2283 – 2285.
[
10] Prepared by a procedure analogous to that of: T. V. Bohner, R.
Beaudegnies, A. Vasella, Helv. Chim. Acta 1999, 82, 143 – 160.
1
13
[
11] All new compounds gave satisfactory H and C NMR spectra
and correct elemental analyses, except acid 12, unstable ester 4a,
and xerulinic acid (3), for all of which, however, correct high-
resolution mass spectra were obtained.
a
2% yield of 6 separated chromatographically from a
54%:8%:3%:2%:trace mixture of five compounds; accordingly,
6 becomes a synthetically useful reagent only on the grounds of
our two-step synthesis (Scheme 5).
[
[
12] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155 – 4156.
13] a) B. O. Lindgren, T. Nilsson, Acta Chem. Scand. 1973, 27, 888 –
[26] N. Kumar, R. W. Read (Unisearch Limited, Australia), PCT
Appl. 20020000639 2002 [Chem. Abstr. 2002, 136, 69697].
890 (resorcinol used as hypochlorite scavenger); b) B. S. Bal,
Angew. Chem. Int. Ed. 2004, 43, 4523 –4526
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4525

Communications
[
27] 6: P O (12.62 g, 44.45 mmol, 1.2 equiv) was added to a solution
4 10
of 18 (10.15 g, 37.04 mmol) in CH Cl (150 mL) at 08C. After
2
2
30 min, the solution was warmed to room temperature and
heated at reflux for 1.5 h. After cooling to room temperature, it
was filtered and concentrated in vacuo. A solution of the
intermediate (8.094 g) in CH Cl (75 mL) was cooled to 08C, and
2
2
NEt (5.29 mL, 3.84 g, 38.0 mmol, 1.02 equiv for 18) was added.
3
After 1 h, the mixture was first warmed to room temperature
and then heated at reflux for 1 h. Aqueous NH Cl (40 mL) was
4
added, and the mixture was extracted with CH Cl (6 ” 20 mL).
2
2
The combined organic extracts were dried with Na SO .
2
4
Evaporation at reduced pressure and flash chromatography
(
cyclohexane/EtOAc 20:1!2:1) provided 6 (3.54 g, 55%) as a
1
slightly yellow solid (m.p. 82–848C); H NMR (300.1 MHz,
CDCl , TMS): d = 6.11 (s; 1’-H), 6.32 (d, J = 5.6 Hz; 3-H),
7
3
3,4
1
3
.38 ppm (d, J4,3 = 5.6 Hz; 4-H); C NMR (125.7 MHz, CDCl ):
3
d = 92.4 (C1’), 120.8 (C3), 141.7 (C4), 152.4 (C5), 168.2 ppm
(
C2); elemental analysis (%): calcd for C H BrO (174.4): C
5 3 2
34.32, H 1.73; found: C 34.20, H 1.61.
[
28] For reviews, see: a) E. Erdik, Tetrahedron 1992, 48, 9577 – 9648;
b) E.-i. Negishi, F. Liu, in Metal-catalyzed Cross-coupling
Reactions (Eds.: F. Diederich, P. J. Stang), Wiley-VCH, Wein-
heim, 1998, pp. 1 – 42; c) P. Knochel in Metal-catalyzed Cross-
coupling Reactions (Eds.: F. Diederich, P. J. Stang), Wiley-VCH,
Weinheim, 1998, pp. 387 – 416.
[
29] a) V. Farina, Pure Appl. Chem. 1996, 68, 73 – 78; b) D. A.
Entwistle, S. I. Jordan, J. Montgomery, G. Pattenden, Synthesis
1
998, 603 – 612; c) C. Amatore, A. A. Bahsoun, A. Jutand, G.
Meyer, A. N. Ntepe, L. Ricard, J. Am. Chem. Soc. 2003, 125,
212 – 4222.
4
1
[
30] 24: H NMR (499.9 MHz, CDCl , CHCl internal standard): d =
3
3
0
.05 (s; SiMe ), 1.03 (m ; CH SiMe ), 4.26 (m ; OCH ), 5.78 (d,
3 c 2 3 c 2
J11,10 = 15.4 Hz; 11-H), 5.89 (d, J5,6 = 11.7 Hz; 5-H), 6.20 (d, J2,3
=
5
8
5
.4 Hz; 2-H), 6.32 (d, J17,16 = 16.0 Hz; 17-H), 6.41–6.56 (m; 7-H,
-H, 9-H), 6.80–6.88 (m; 6-H, 10-H, 16-H), 7.37 ppm (d, J3,2
.2 Hz; 3-H); an edited undecoupled HSQC NMR spectrum
=
1
(
“ H-coupled short-range H,C-COSY spectrum”; 499.9 MHz/
25.7 MHz, CHCl3 and CDCl3 internal standards in CDCl3,
1
respectively) revealed, amongst others: d = 6.44 ppm (ddd,
1
JH,C = 156.3 Hz, J9,8 = 12.8 Hz, J9,10 = 12.0 Hz; 9-H), d =
1
6
.51 ppm (ddd, J = 156.6 Hz, J8,9 = 14.6 Hz, J8,7 = 11.6 Hz; 8-
H,C
1
H), d = 6.53 (ddd, J = 154.3 Hz, J7,6 = 15.2 Hz, J7,8 = 11.3 Hz;
H,C
1
7
6
1
1
(
(
(
(
(
7
-H), 6.82 ppm (dd, J = 168.7 Hz, J16,17 = 16.1 Hz; 16-H), d =
H,C
1
.85 ppm (ddd, J = 157.6 Hz, J
= 15.1 Hz, J10,9 = 11.1 Hz;
H,C
10,11
1
0-H), d = 6.86 ppm (ddd, J = 159.2 Hz, J6,7 = 13.7 Hz, J6,5
=
H,C
1
3
2.3 Hz; 6-H);
C NMR (125.7 MHz, CDCl ): d = ꢁ1.5
3
Si(CH ) ), 17.3 (C2’), 63.3 (C1’), 78.3, 80.4, 82.5, and 85.3
3 3
C12, C13, C14, C15), 111.1 (C11), 114.4 (C5), 119.1 (C2), 123.6
C16), 128.4 (C6), 133.0 (C17), 134.6 (C9), 136.6 (C8), 137.5
C7), 142.5 (C3), 145.3 (C10), 149.7 (C4), 165.5 (C18), 169.2 ppm
C1); elemental analysis (%): calcd for C H O Si (392.1): C
2
3
24
4
0.38, H 6.16; found: C 70.15, H 6.13.
[
[
31] W. Steglich, personal communication. We are indebted to
Professor Steglich for sending copies of all the original NMR
spectra of compound 3.
32] A referee suggested that “the possibility of directly obtaining
xerulinic acid from xerulin should be commented on”. We are
unaware of such a possibility; furthermore, during alternative
attempts at the synthesis of xerulinic acid, we prepared both the
corresponding alcohol (“xerulinol”) and aldehyde (“xerulinal”)
but could not oxidize either to xerulinic acid: K. Siegel, Ph.D.
Thesis, Universität Freiburg, 2000.
4
526
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 4523 –4526