Phorboxazole B synthetic studies: Construction of C(1-32) and C(33-46) subtargets

Article abstract of DOI:10.1039/b407240e

The convergent syntheses of the C(1-32) and C(33-46) domains of phorboxazole B are described. An iterative cyclocondensation strategy exploited the Jacobsen hetero-Diels-Alder (HDA) reaction as a platform for the synthesis of both the C(5-9) and C(11-15) tetrahydropyran rings. The use of 2-silyloxydiene coupling partners bearing an increasing resemblance to the phorboxazole skeleton was found to lead to a reduction in diastereoselectivity, however, in the case of the C(11-15) ring. The coupling of aldehyde 21 and 2-silyloxydiene 20 by this route provided a C(1-32) fragment which was elaborated to the macrolide core of phorboxazole B. The synthesis of the C(33-46) domain involved a Nozaki-Kishi coupling of aldehyde 31 and vinyl iodide 39. The syntheses of 31 and 39 were highly diastereoselective: an Evans [Cu(Ph-pybox)](SbF₆)₂-catalysed Mukaiyama aldol reaction formed the cornerstone of the synthesis of 31 whilst a Nagao-Fujita acetate aldol reaction provided a convenient means of installing the sole stereogenic centre of 39.

Full text of DOI:10.1039/b407240e

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Phorboxazole B synthetic studies: construction of C(1–32) and
C(33–46) subtargets†
Ian Paterson,* Alan Steven and Chris A. Luckhurst‡
University Chemical Laboratory, Lensfield Road, Cambridge, UK CB2 1EW.
E-mail: ip100@cam.ac.uk; Fax: +44 (0)1223 336362; Tel: +44 (0)1223 336407
Received 14th May 2004, Accepted 26th August 2004
First published as an Advance Article on the web 28th September 2004
The convergent syntheses of the C(1–32) and C(33–46) domains of phorboxazole B are described. An iterative
cyclocondensation strategy exploited the Jacobsen hetero-Diels–Alder (HDA) reaction as a platform for
the synthesis of both the C(5–9) and C(11–15) tetrahydropyran rings. The use of 2-silyloxydiene coupling
partners bearing an increasing resemblance to the phorboxazole skeleton was found to lead to a reduction in
diastereoselectivity, however, in the case of the C(11–15) ring. The coupling of aldehyde 21 and 2-silyloxydiene 20 by
this route provided a C(1–32) fragment which was elaborated to the macrolide core of phorboxazole B. The synthesis
of the C(33–46) domain involved a Nozaki–Kishi coupling of aldehyde 31 and vinyl iodide 39. The syntheses of 31
and 39 were highly diastereoselective: an Evans [Cu(Ph-pybox)](SbF6)2-catalysed Mukaiyama aldol reaction formed
the cornerstone of the synthesis of 31 whilst a Nagao–Fujita acetate aldol reaction provided a convenient means of
installing the sole stereogenic centre of 39.
Introduction
The isolation of phorboxazoles A and B (1 and 2, respectively,
Fig. 1) from the Indian Ocean sponge Phorbas sp. was
reported by Searle and Molinski in 1995.1 Their characteristic
21-membered oxazole-tris-tetrahydropyran macrolide ring
subtended by an oxazole-tetrahydropyran side-chain represents
a novel chemotype. Both 1 and 2 display essentially the same
profile in terms of biological activity, the antimitotic component
of which (1 displays a mean GI50 of 1.58 nM against the US NCI
panel of 60 human cancer cell lines2) has rendered them premier
medicinal targets. Whilst their precise mechanism of action is
subject to ongoing investigations, their ability to induce S-phase
cell cycle arrest in Burkitt lymphoma CA46 cells3 complements
the activity of antineoplastic agents which disrupt tubulin
polymerisation or microtubule disassembly. In addition to their
powerful biological activity, their scarcity and architectural
complexity have captured the attention of a number of
synthetic groups, culminating in five total syntheses to date,4–9
and numerous fragment syntheses.10,11
Our strategy was envisaged as offering a highly convergent
addition to the existing corpus, and targets the C(32–33) bond as
the site of the ultimate carbon–carbon bond formation (Fig. 1).
Herein, we report the synthesis of the C(1–32) and C(33–46)
domains which may enable this end to be realised.
Fig. 1 Structures of phorboxazoles A and B, and targeted C(32–33)
bond formation.
In common with the work of Smith,12 and Donaldson and
Results and discussion
Synthesis of a C(1–14) fragment
Greer,10a,13 2,3-dihydro-c-pyrone 3 was recognised as being a
suitable building block around which a synthesis of a C(1–14)
fragment could be fashioned (Scheme 1). Its enantioselective
synthesis via an HDA reaction between Danishefsky’s diene 4
and aldehyde 5 using Jacobsen’s HDA catalyst 6a14 proceeded in
89% yield with an acceptable 90% ee.
Attention then turned to the conjugate addition of a suitable
nucleophile, ideally corresponding to a propargylic synthon,15
to dihydropyrone 3.16 A rapid screen of the reaction of
dihydropyrone 7 (Scheme 2), a redundant compound from a
previous synthetic route,17 with allenylmagnesium bromide in
the presence of various additives, revealed a pronounced bias
towards carbonyl, rather than Michael, attack (Table 1, entries
1–3) with the result that alcohol 8 was the major product isolated.
This unanticipated mode of reactivity led to allylmagnesium
bromide being adopted as a surrogate nucleophile, with a view
to effecting a formal oxidation of the allyl group to the coveted
propargyl group at a later stage. Initial results with allylcopper
nucleophiles also showed a proclivity for carbonyl attack
Our previous studies towards the synthesis of the phorboxazoles
have led to the convergent synthesis of a pentacyclic C(4–32)
unit.11b Whilst this work provided a sophisticated example
of tetrahydropyranone assembly using Jacobsen’s HDA
methodology, there was no convenient provision for the
appendage of a C(1–3) unit which would allow the completion
of the macrolide core. Moreover, the installation of the native
functionality of the C(5–9) ring lacked step-efficiency. We now
report revised synthetic studies that tackle both of these issues.
† Electronic supplementary information (ESI) available: general
experimental information and procedures for the preparation of 6a,
6b, 17, 18, 21, 28, 36, 32, 40, 42. See http://www.rsc.org/suppdata/ob/
b4/b407240e/
‡ Current address: Department of Medicinal Chemistry, AstraZeneca
R&D Charnwood, Bakewell Road, Loughborough, UK LE11 5RH.
3 0 2 6
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Article Online
Table 1 Reactions of dihydropyrone 7 with allyl and allenyl metal reagents giving products 8–11 (see Scheme 2)
Entry
Nucleophile
Additivesa
Temperature/°C
Yield (%)
1
2
3
4
CuI, TMSCl, DMPU
CuBr·DMS (5 mol%), TMSCl, HMPA
CuCl (5 mol%), MnCl4Li2
TiCl4
−78 → −50
94
71
90
51
−78 → −45
0
0
5
6
7
CuI, DMPU, TMSCl
CuI, DMPU, TMSCl
TMSOTf
−78 → −45
−78
16b
31c
86d
−78 → −50
a Entries 1–3, 5 and 6 were performed in THF; entries 4 and 7 were performed in CH2Cl2. b The yield of alcohol 9 was 34% based on recovered
dihydropyrone 7. c Theyieldof alcohol 9was82%basedonrecovereddihydropyrone 7. Asmallamount(4%basedonrecovered 7)of tetrahydropyranone
10 was also isolated. d This yield referes to material isolated after treatment of the crude reaction mixture with TBAF (1 M THF)/AcOH.
Scheme 1 Reagents and conditions: a) 6a (5 mol%), EtOAc, BaO;
b) TFA, CH2Cl2.
(Table 1, entries 5 and 6) and furnished alcohol 9 as the only
Scheme 2 Reactions of dihydropyrone 7 with allyl and allenyl metal
product of any importance. The chemoselectivity displayed by
reagents (see Table 1).
these reactions was attributed to the ability of the ring oxygen
atom to deactivate the enone moiety towards the single electron
reliable methods were used for the elaboration of 12. Thus the
transfer (SET) processes which feature prominently in possible
dihydroxylation of the terminal olefin, Wittig methylenation of
mechanisms for copper(I)-initiated Michael additions to a,b-
the C-7 olefin and oxidative diol cleavage yielded aldehyde 13 in
unsaturated ketones.18 The use of reaction conditions which
74% yield over the three steps. The two-carbon homologation
failed to attenuate the excessive reactivity of allylic (and by
of aldehyde 13 was conveniently performed by subjecting the
possible analogy, allenyl/propargyl) copper(I)/cuprate reagents19
corresponding geminal dibromide 14 to the Grandjean et al.
may also have been at fault.
modification21 of the Corey–Fuchs reaction. In this way, the
With these failures fresh in our minds, attention turned to
dehydrobromination of 14 with NaHMDS gave alkynyl bromide
a method more akin to the classical Sakurai–Hosomi method
15, whose C-2-lithiate was subsequently quenched with methyl
of conjugate addition.20 Gratifyingly, the reaction of dihydro-
chloroformate so as to afford methyl ynoate 16 in 86% yield.
pyrone 7 with allyl tributylstannane and TMSOTf, followed by
The elaboration of the C-11 terminus of 16 was now required.
treatment with TBAF (1 M THF)–AcOH (10:1, v/v) exclusively
Thus the discharge of the BPS ether with HF·py was followed by
led to the formation of tetrahydropyranone 10 (entry 7). The
the oxidation of the resulting alcohol 17 to aldehyde 18 and its
trapping of the initially-formed adduct as the TMS enol ether
two-carbon homologation to enone 19 using a Ba(OH)2·xH2O-
was noteworthy as related experiments with a titanium enolate
promoted Horner–Wadsworth–Emmons (HWE) reaction.22 The
intermediate (albeit in the propargyl series) suffered from the
derivatisation of the methyl ketone of 19 to the corresponding
elimination of the tetrahydropyran oxygen atom leading to the
TES enol ether 20 proceeded in 95% yield.23
formation of the double Michael addition product, b-hydroxy
ketone 11, as the major product (entry 4).16
The transition to the series with tert-butyldiphenylsilyl (BPS)
Jacobsen HDA coupling studies
The aldehyde heterodienophile 21 which was to be coupled with
the C(1–14) domain was prepared by the DIBAl-H reduction
of the corresponding tert-butyl ester 2211a (Scheme 4). That
minimal overreduction was observed was explained by the
ability of the oxazole ester moiety to act as a Weinreb amide
surrogate.
protection at C-11 proceeded without incident, allowing the
gram-scale synthesis of the desired allyl adduct 12 (Scheme 3).
That the desired trans stereochemistry at C-5 and C-9 had been
installed was established via the extraction of the medium
vicinal coupling constants shown. With a sound route to
a basic C(5–9) tetrahydropyranone matrix now available,
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8
3 0 2 7

View Article Online
Scheme 3 Reagents and conditions: a) H2CCHCH2SnBu3, TMSOTf, CH2Cl2, −78 → −50 °C; b) TBAF (1 M THF)–AcOH (10:1, v/v), THF,
0 °C; c) OsO4 (4 mol%), NMO, tBuOH–THF–H2O (4:4:1); d) Ph3PCH2, THF, −40 → −20 °C; e) Pb(OAc)4, Na2CO3, CH2Cl2, 0 °C → RT; f) CBr4,
Ph3P, CH2Cl2, −10 °C; g) NaHMDS, THF, −98 °C; h) (i) nBuLi, THF, −78 °C, (ii) MeOC(O)Cl, HMPA; i) HF·py, MeCN, 0 °C; j) DMP, CH2Cl2; k)
MeC(O)CH2P(O)(OMe)2, Ba(OH)2·xH2O, THF; l) TESOTf, Et3N, Et2O, 0 °C.
Scheme 4 Reagents and conditions: a) (i) DIBAl-H, CH2Cl2, −78 °C, (ii) chromatographic removal of C(19–20) Z-diastereomer; b) 20, 6b (10 mol%),
4 Å molecular sieves, acetone; c) TBAF–AcOH (1:2, mol/mol), THF, 0 °C. # = Yield based on recovered 20.
In line with our previous studies, the construction of the
C(11–15) ring of the phorboxazole skeleton using the Jacobsen
HDA reaction between diene 20 and aldehyde 21 was targeted.
The lower reactivity of our diene coupling partner, when
compared to Danishefsky’s diene, necessitated the use of
the more reactive hexafluoroantimonate catalyst 6b,11b,14a as
3 0 2 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8

View Article Online
Table 2 Hetero-Diels–Alder reactions between 2-silyloxydienes and oxazole-containing aldehydes using catalyst 6b (see Scheme 5)
Heterodienophile
R2
Percentage yield (based on recovered diene)
46 (95)
dr [ee (%)]a
>97:3b [81]
42 (95)
3.2:1.0
41 (87)
44 (90)
36 (55)
1.5:1.0b
1.5:1.0
1.1:1.0
21
21
a The (11R,15R) stereoisomer predominated over the (11S,15S) stereoisomer in each case. b These diastereo- and enantio-selectivities refer to the
tetrahydropyranone products resulting from silyl enol ether hydrolysis.
opposed to the chloride catalyst 6a used previously in Scheme 1.
A brief optimisation of the reaction allowed two diastereomeric
silyloxydihydropyran products, 23 and 24, to be isolated in a
combined 36% yield and as a 1.1:1.0 mixture. Silyl enol ether
hydrolysis afforded tetrahydropyranones 25 and 26, whose
endo topology was ascertained from the detection of NOE
contacts at H-11/H-15. The comparison of their 1H NMR data
for H-10a/b with those of structurally-related compounds11b
allowed the tentative assignment of the absolute configurations
at C-11 and C-15 in 25 and 26. The use of the enantiomeric
catalyst failed to noticeably change the diastereomeric ratio
(25/26 = 1.0:1.3), indicating that a mismatched case of double
diastereodifferentiation was unlikely to be in operation. Instead,
Scheme 5 Reagents and conditions: 6b, 4 Å molecular sieves, acetone
(see Table 2).
it seemed likely that an indeterminate Lewis acid was promoting
the reaction through a manifold in which any chiral information
was not being effectively relayed to the transition state.
LiAl(OtBu)3H reduction of the newly-fashioned tetrahydropy-
ranone ring of 25 gave alcohol 27 with the C-13 configuration
of phorboxazole B. This stereochemical assignment rested
securely on the detection of NOE contacts at H-13/H-11 and
H-13/H-15, and the extraction of an axial–axial JH13–H12b value
and an axial–equatorial JH13–H12a value from the 1H NMR signal
for H-13, which implied axial hydride delivery so as to give an
equatorial alcohol.25,26
In order to permit the differential manipulation of the
oxygenated functionality at C-13 and C-24, the C-13 hydroxyl
group was protected as its BPS ether. This was treated in its crude
form with methanolic HCl so as to afford alcohol 28 (85% over
two steps from 27). The saponification of the methyl ester of
28 at C-1 to give carboxylic acid 29 was accomplished, with the
removal of a minimal amount of the C-13 BPS ether, using LiOH
in THF/water. The completion of the C(1–32) macrolactone ring
of 30 was achieved, as with the Evans synthesis,5 using a room
temperature Yamaguchi macrolactonisation reaction.27,28
These results, when taken together with our earlier studies,11b
indicate a pronounced fall off in the diastereoselectivity of
the Jacobsen HDA reaction as more and more of the native
phorboxazole functionality was built into the C-11 substituent
of the diene coupling partner, as in the general transformation
shown in Scheme 5. This is best exemplified by a comparison
of the stereoselectivities for a series of examples, as tabulated
in Table 2, which provide a sense of the terrain in which our
investigations have been conducted. The modest conversions
observed in these Jacobsen HDA reactions pointed to the
apparent inhibition of the catalytic cycle by either one of the
substrates and/or the silyloxydihydropyran product. A possible
source of this inhibition may have been competitive ligation
between the oxazole nitrogen atoms and the ligand heteroatoms
for the chromium(III) atom of the catalyst.24
Synthesis of a C(1–32) macrocyclic domain
Despite its limitations in terms of yield and diastereoselectivity,
the convergency of our stratagem allowed the manipulation of
25 to produce a macrocyclic fragment. In order to do this, the
issue of installing the C-13 stereogenic centre was addressed.
Given the epimeric nature of 1 and 2 at this position, a diastereo-
merically clean reduction at this stage was of more importance
than the absolute stereochemistry realised (Scheme 6). The
Synthesis of a C(33–38) fragment
Contemporaneous to these attempts to establish a convergent
entry into the macrolactone ring was our work directed towards
a synthesis of the corresponding C(33–46) fragment of the
phorboxazoles. Our approach was reliant on the preparation
and union of suitable C(33–38) and C(39–46) fragments
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8
3 0 2 9

View Article Online
Scheme 7 Retrosynthetic analysis of C(33–46) fragment.
Scheme 6 Reagents and conditions: a) LiAl(OtBu)3H, THF,
−78 → −10 °C; b) BPSCl, ImH, DMF; c) conc. HCl (1%), MeOH;
d) LiOH·H2O, THF, H2O; e) (i) 2,4,6-trichlorobenzoyl chloride, Et3N,
THF, (ii) DMAP, PhMe.
(Scheme 7). The former, as exemplified by aldehyde 31, was
envisaged as being accessible from ketolactone 32.
As with, though independent of, the work of the Evans
group,5,29 we were keen to exploit their copper(II)-pybox-
catalysed Mukaiyama aldol coupling studies for the
enantioselective installation of the C-37 stereogenic centre.
In our case (Scheme 8), this involved a Mukaiyama aldol
reaction between (benzyloxy)acetaldehyde 33 and dienolate
34 which was catalysed by [Cu((R,R)-Ph-pybox)](SbF6)2 (35).
This provided multigram quantities of alcohol 36 (95% ee
according to Kakisawa–Mosher analysis30). Its base-promoted
cyclisation, using the protocol of Sato et al.,31 formed
ketolactone 32 in almost quantitative yield. The installation of
the C-35 stereogenic centre was readily accomplished in a highly
diastereoselective fashion (R/S > 97:3 at C-35) by hydrogenating
37, the corresponding methyl enol ether of 32. This step also
served to unmask the C-38 hydroxyl group so as to give alcohol
38, which was subsequently oxidised to the fragile aldehyde 31,
in anticipation for our C(38–39) coupling studies.
With regard to a C(39–46) fragment with which aldehyde 31
could be coupled, we were drawn to the possibilities offered
by triene 39 (Scheme 7) with its native C-43 methoxy group
and terminal E-vinyl bromide in place. Triene 39 also featured
in the Evans synthesis where it was obtained from R-trityl
glycidol.5b Our phorboxazole retrosynthetic search, by contrast,
was coloured by a desire to avoid the chiral pool, and led us
to investigate the possibility of subjecting dienal 40 to an
asymmetric acetate-type aldol reaction.
Scheme 8 Reagents and conditions: a) [Cu((R,R)-Ph-pybox)]-
(SbF6)2 35 (8 mol%), CH2Cl2, −98 °C → −78 °C; b) PPTs, MeOH;
c) K2CO3, MeOH; d) (MeO)2SO2, K2CO3, acetone; e) H2, Pd–C, EtOAc;
f) Swern [O].
The synthesis of triene 39 was initiated by the MnO2-mediated
oxidation of the known alcohol 4132 and a Masamune–
Roush-type HWE homologation (Scheme 9).33 The DIBAl-H
reduction of 42, the Weinreb amide product of this sequence,
ineluctably gave the desired aldehyde 40 together with small but
variable amounts of the corresponding C(39–40) Z-diastereomer
43. An iodine-mediated isomerisation step was sufficient to
convert 43 to 40, increasing the efficiency of the transformation.
An auxiliary which is known to promote the diastereocontrolled
aldol reaction between itself and a,b-unsaturated aldehydes is
the 4(S)-isopropyl-1,3-thiazolidine-2-thione [4(S)-IPTT] of the
groups of Nagao and Fujita.34 The use of the tin(II) enolate of N-
acetyl-4(S)-IPTT 44 resulted in the virtually exclusive formation
of the desired diastereomer 45. The use of the titanium(IV)
3 0 3 0
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8

View Article Online
Scheme 9 Reagents and conditions: a) MnO2, CH2Cl2; b) MeON(Me)C(O)CH2P(O)(OEt)2, LiCl, DBU, MeCN–CH2Cl2 (3:1); c) DIBAl-H,
THF, −78 °C; d) I2, CH2Cl2; e) (i) 44, Sn(OTf)2, N-ethylpiperidine, CH2Cl2, −40 °C, (ii) 40, −98 → −78 °C; f) MeNHOMe·HCl, ImH, CH2Cl2; g) MeI,
Ag2O, Et2O, reflux; h) DIBAl-H, THF, −78 °C; i) nBu3SnCHI2, CrCl2, DMF; j) NBS, MeCN, 0 °C.
enolate, by contrast, as recommended by Urpí, Vilarrasa and
co-workers,35 proved less diastereoselective [45:(epi-C-43)-
45 = 5.5:1.0]. The manipulation of aldol adduct 45 involved
the initial conversion to Weinreb amide 46. A subsequent O-
methylation at C-43 so as to give methyl ether 47 was followed
by a DIBAl-H reduction. This sequence furnished aldehyde
48 in 64% yield over the three steps from 45. The conversion
of aldehyde 48 to the corresponding E-vinyl bromide 39 was
conveniently achieved via the corresponding vinyl stannane,
using the Hodgson variant36 of the Takai–Utimoto reaction.37
A comparison of the optical rotation of this material with that
of the Evans group verified that the correct C-43 stereochemistry
had been installed, as expected, during the Nagao–Fujita aldol
reaction.
Synthesis of the C(33–46) domain
With quantities of both vinyl iodide 39 and aldehyde 31 available,
the stage was set for examining their union. An analysis of the
functionality present in the coupling partners made the choice
of chemoselective coupling conditions a prerequisite. The vinyl
iodide of 39 was to be functionalised whilst leaving the vinyl
bromide moiety intact, and the resulting vinylmetal species
Scheme 10 Reagents and conditions: a) NiCl2–CrCl2 (10:1), tBuC5H4N,
THF; b) TPAP, 4 Å molecular sieves, CH2Cl2.
reacted with the C-38 aldehyde functionality of 31 in the
presence of the C-33 lactone. There was also a need to minimise
Outlook
any elimination of the base-sensitive methoxy groups at C-35 and
Whilst the viability of the implied retrosynthetic dissection
C-43. Attention quickly turned to the Nozaki–Kishi reaction.38
in Fig. 1 remains to be demonstrated if our synthetic studies
A rapid survey of possible reaction conditions, which included
of phorboxazole B are to reach a successful conclusion, the
varying the solvent composition, the relative stoichiometry of
synthesis of 30 and 50, our most advanced fragments to date,
have made the venture a rewarding one thus far. Compounds
30 and 50 have been synthesised in 19 linear steps (4.0% from
dihydropyrone 3) and 11 linear steps (7.1% from allylic alcohol
31 and 39, and the amounts and ratio of CrCl2/NiCl2 used was
briefly performed. 4-tert-Butylpyridine was used as an additive
on occasions, in line with the recommendations of Kishi and
co-workers.39 Whilst the yields for this process were modest
41) respectively. Their successful elaboration and union are pres-
(29%, Scheme 10) and dependent on the reaction conditions,
ently under study and may allow one of the phorboxazoles to be
the diastereoselectivity consistently showed a modest preference
procured via a longest linear sequence which is comparable to,
for the undesired C-38 epimer (R/S  1:3 at C-38).40 The
or better than, those reported to date.5,6
formation of a significant quantity (ca. 40%) of 49, the desiodo
analogue of vinyl iodide 39, would indicate that an oxidative
insertion into the C–I bond of 39 had taken place, despite its
steric encumbrance, but that the resulting vinylchromium(III)
nucleophile had been quenched prior to the coupling. Since the
allyl alcohol products of the reaction were inseparable at this
stage, and with a view to correcting the C-38 stereochemistry in
the future via a diastereoselective ketone reduction protocol, the
mixture was oxidised to the corresponding ketone 50 using the
Griffith–Ley protocol.41
Experimental
Dihydropyrone 3
A mixture of Jacobsen Schiff base chloride catalyst 6a14c
(7.6 mg, 16 lmol), BaO (63 mg, dried overnight at 150 °C and at
1 mm Hg prior to use) and dry EtOAc (100 lL) were pre-stirred
(60 min) at RT and in the absence of light. Aldehyde 5 (117 mg,
0.374 mmol) in dry EtOAc (200 lL, including washings) was
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8
3 0 3 1

View Article Online
subsequently added prior to cooling to 0 °C. Some of the extra
EtOAc was removed in vacuo over 10 min, so as to return the
solvent volume to ca. 0.2 mL, before Danishefsky’s diene 4
(60.7 lL, 0.312 mmol) was added. Once the reaction mixture
had been stirred at 0 °C and in the dark for 24 h, CH2Cl2
(1.0 mL) and then TFA (1 pipette drop) were added. A period
of warming to RT over 10 min was followed by the washing of
the reaction mixture through a short pad of silica gel atop a
short pad of Celite® with copious Et2O, before the combined
filtrates were concentrated in vacuo. The purification of the
residue by flash column chromatography on silica gel (light
petroleum–EtOAc–MeOH gradient, 80:10:1 → 40:10:1) gave
recovered aldehyde 5 (30.1 mg), dihydropyrone 3 (106 mg, 89%)
as a yellow oil and then a small amount of 4-methoxy-3-but-(E)-
en-2-one. The following data pertain solely to dihydropyrone 3
whose enantiomeric ratio was 95.1:4.9 according to chiral
HPLC analysis [light petroleum–iPrOH (249:1); Rt = 26 min (3:
95.1%), Rt = 28 min (ent-3: 4.9%); detection at k = 250 nm]. Rf
0.21 (light petroleum–EtOAc, 3:1); [a]2D0 −55 (c = 0.44, CHCl3);
mmax (thin film)/cm−1 2930 (s), 1681 (s, CO), 1596 (s, CC),
H-10b), 1.04 [9H, s, SiC(CH3)3]; 13C NMR (100.6 MHz, CDCl3)
dC 207.4, 135.5, 135.5, 133.6, 133.6, 133.4, 129.6, 129.6, 127.7,
127.6, 117.9, 71.4, 69.2, 59.9, 46.7, 46.3, 39.0, 36.8, 26.8, 19.1;
HRMS (+ESI) calc. for C26H34O3SiNa (MNa+) 445.2175, found
445.2190.
Aldehyde 13
To a solution of tetrahydropyranone 12 (7.79 g, 18.4 mmol)
and NMO (2.59 g, 22.1 mmol) in tBuOH–THF–water (225 mL,
t
4:4:1) at RT was added OsO4 in BuOH (3.75 g of 5%, w/w,
0.74 mmol). A period of stirring for 24 h without argon
protection was followed by the addition of Na2S2O3 solution
(100 mL) and stirring for 45 min, over which time the initial
yellow colour of the reaction mixture darkened. The partitioning
of the reaction mixture between CH2Cl2 (4 × 100 mL) and water
(100 mL) was followed by the drying (Na2SO4) of the combined
organic extracts and concentration in vacuo. This furnished a
1.0:1.0 mixture C-3 diol epimers (8.38 g) which was purified
by flash column chromatography on silica gel (CH2Cl2–MeOH
gradient, 100:1 → 20:1).
1
1273 (m), 1111 (s), 702 (m); H NMR (400 MHz, CDCl3) dH
To
a
chilled (0 °C), stirred suspension of dry
7.65 (4H, d, J = 7.8 Hz, ArH), 7.46–7.37 (6H, m, ArH), 7.29
(1H, d, J = 5.9, H-5), 5.40 (1H, d, J = 5.9, H-6), 4.68 (1H, dddd,
J = 12.5, 8.5, 4.4, 4.4, H-9), 3.86 (1H, ddd, J = 10.6, 4.7, 4.7,
H-11a), 3.78 (1H, ddd, J = 10.6, 5.4, 5.4, H-11b), 2.54 (1H,
dd, J = 16.8, 12.5, H-8a), 2.46 (1H, dd, J = 16.8, 4.4, H-8b),
2.06–1.98 (1H, m, H-10a), 1.92–1.84 (1H, m, H-10b) 1.05 [9H, s,
SiC(CH3)3]; 13C NMR (62.9 MHz, CDCl3) dC 192.5, 163.0, 135.5
(2C), 133.5 (2C), 129.7 (2C), 127.7 (2C), 107.1, 76.5, 59.2, 42.0,
37.2, 26.8, 19.2; HRMS (+ESI) calc. for C23H28O3SiNa (MNa+)
403.1705, found 403.1715.
triphenylphosphonium bromide (53.4 g, 149 mmol) in dry THF
(150 mL) was added PhLi in 70:30 cyclohexane–Et2O (77.7 mL,
1.80 M, 140 mmol) via syringe pump (2 mL min−1). Once 2 h had
elapsed from the start of the addition, the almost homogeneous
orange solution was cooled from 0 to −40 °C. A solution of the
diol mixture (8.38 g, 18.4 mmol) in dry THF (46 mL, including
washings) was subsequently added and the resulting mixture
allowed to warm slowly to −20 °C over 3 h. The reaction mixture
was subsequently quenched with pH 7 buffer solution (40 mL),
rapidly warmed to RT, diluted with water (200 mL) and extracted
with Et2O–CH2Cl2 (4 × 200 mL of 10:1). The combined organic
extracts were washed with brine (200 mL), dried (Na2SO4) and
concentrated in vacuo. Flash column chromatography on silica
gel, loading and eluting with Et2O–CH2Cl2–light petroleum
(3:1:1) gave a mixture of methylenated ketones (7.61 g) which
were epimeric at C-3. To a chilled (0 °C) mixture of this material
(7.61 g, 16.7 mmol) and Na2CO3 (3.91 g, 36.9 mmol) in dry
CH2Cl2 (80 mL) was added Pb(OAc)4 (8.99 g, 20.3 mmol) in a
portionwise fashion over 5 min. After 30 min stirring at 0 °C, the
flask was removed from the cooling bath. After a further 60 min,
the reaction mixture was washed through a pad of Celite® with
copious Et2O. The filtrate was washed with water (2 × 100 mL),
and then brine (100 mL), before being dried (Na2SO4) and
concentrated in vacuo. Flash column chromatography on silica
gel (light petroleum–EtOAc, 15:1 and then 7:1) gave aldehyde
13 (5.73 g, 74% from tetrahydropyranone 12) as a colourless
oil. Rf 0.32 (light petroleum–EtOAc, 5:1); [a]2D0 −25 (c = 0.21,
CHCl3); mmax (thin film)/cm−1 3072 (w), 2932 (m), 2857 (m), 1726
(s, CO), 1428 (m), 1111 (s), 1090 (s), 739 (m), 702 (s); 1H NMR
(400 MHz, CDCl3) dH 9.66 (1H, dd, J = 2.5, 1.8 Hz, H-3), 7.67–
7.64 (4H, m, ArH), 7.44–7.35 (6H, m, ArH), 4.79–4.78 (2H, m,
H-51a & H-51b), 4.21 (1H, dddd, J = 7.8, 7.5, 5.5, 4.0, H-5), 4.06
(1H, dddd, J = 8.5, 5.3, 5.0, 4.8, H-9), 3.75 (1H, ddd, J = 10.3,
7.8, 5.8, H-11a), 3.67 (1H, ddd, J = 10.3, 6.5, 5.8, H-11b), 2.63
(1H, ddd, J = 16.3, 7.8, 2.5, H-4a), 2.44 (1H, ddd, J = 16.3, 5.5,
1.8, H-4b), 2.38 (1H, dddd, J = 13.3, 4.8, 1.3, 1.3, H-8a), 2.36
(1H, dd, J = 13.3, 4.0, H-6a), 2.04–1.99 (2H, m, H-6b & H-8b),
1.88 (1H, dddd, J = 14.1, 8.5, 5.8, 5.8, H-10a), 1.65 (1H, dddd,
J = 14.1, 7.8, 6.5, 5.3, H-10b), 1.04 [9H, s, SiC(CH3)3]; 13C NMR
(100.6 MHz, CDCl3) dC 200.9, 141.1, 135.6 (2C), 133.9 (2C),
129.6 (2C), 127.6 (2C), 111.1, 69.7, 67.2, 60.4, 47.6, 39.6, 39.2,
35.6, 26.9, 19.2; HRMS (+ESI) calc. for C26H34O3SiNa (MNa+)
445.2175, found 445.2179.
Tetrahydropyranone 12
To a cold (−78 °C), stirred solution of dihydropyrone 3 (1.76 g,
4.61 mmol) in dry CH2Cl2 (20 mL) was added neat TMSOTf
(0.919 mL, 5.08 mmol) over 2 min. After a further 5 min, neat
allyl tributylstannane (3.15 mL, 10.2 mmol) was added over
5 min. After 2 h at −78 °C, the orange solution was allowed to
warm to −50 °C over a further 60 min. The quenching of the
reaction with MeOH–pH 7 buffer solution (8 mL of 3:1) and
rapid stirring for 5 min was followed by rapid warming to 0 °C,
its dilution with pH 7 buffer solution (20 mL) and subsequent
warming to RT. The partitioning of the reaction mixture between
Et2O (100 mL) and water (100 mL), and then brine (100 mL),
was followed by drying (MgSO4) and concentration in vacuo so
as to give the crude trimethylsilyl enol ether intermediate. This
was diluted with THF (30 mL) and cooled to 0 °C prior to the
addition of TBAF (1 M in THF)–AcOH (10.14 mL of 10:1,
v/v). A period of stirring without argon protection for 40 min
was followed by the partitioning of the reaction mixture between
Et2O (200 mL) and water (200 mL), and then brine (200 mL).
The organic phase was dried (Na2SO4) and concentrated in
vacuo before being diluted with Et2O (30 mL) and vigorously
stirred with KF on Celite® 42 (7.14 g of 1:1, w/w). After 60 min
the suspension was filtered, washing through with copious Et2O
and then EtOAc, before being concentrated in vacuo. Analysis
by 500 MHz NMR revealed S/R > 97:3 at C-5. Flash column
chromatography on silica gel (light petroleum–EtOAc, 20:1)
furnished tetrahydropyranone 12 (1.68 g, 86%) as a colourless
oil. Rf 0.59 (light petroleum–EtOAc, 10:1); [a]2D0 −14 (c = 5.3,
CHCl3); mmax (thin film)/cm−1 2931 (m), 2858 (m), 1720 (s, CO),
1428 (m), 1112 (s), 823 (m), 739 (m), 703 (s); 1H NMR (500 MHz,
CDCl3) dH 7.68–7.63 (4H, m, ArH), 7.45–7.35 (6H, m, ArH), 5.73
(1H, dddd, J = 17.1, 10.0, 7.0, 7.0 Hz, H-3), 5.10–5.03 (2H, m,
H-2a & H-2b), 4.40 (1H, dddd, J = 8.8, 6.3, 5.0, 5.0, H-9), 4.00
(1H, dddd, J = 7.3, 7.3, 6.3, 4.8, H-5), 3.79 (1H, ddd, J = 10.5,
8.0, 5.5, H-11a), 3.69 (1H, ddd, J = 10.5, 6.3, 5.5, H-11b), 2.57
(1H, dd, J = 14.3, 5.0, H-8a), 2.46 (1H, dd, J = 14.3, 4.8, H-6a),
2.38–2.18 (4H, m, H-6b, H-8b, H-4a & H-4b), 1.86 (1H, dddd,
J = 14.1, 8.8, 5.5, 5.5, H-10a), 1.68 (1H, dddd, 14.1, 8.0, 6.3, 5.0,
Geminal dibromide 14
To a chilled (−10 °C), stirred solution of CBr4 (16.2 g, 48.8 mmol)
in dry CH2Cl2 (20 mL) was added a pre-chilled (−10 °C) solution
of Ph3P (25.6 g, 97.4 mmol) in dry CH2Cl2 (35 mL, including
3 0 3 2
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8

View Article Online
washings) over 20 min. After a further 10 min, a pre-chilled
(−10 °C) solution of aldehyde 13 (5.15 g, 12.2 mmol) in dry
CH2Cl2 (25 mL, including washings) was added to the now
orange solution. The reaction was quenched after 30 min
with NaHCO3 solution (50 mL) before being rapidly warmed
to RT and washed through a Celite® pad with copious Et2O.
The phases of the filtrate were separated and the organic phase
washed with water (150 mL) and then brine (150 mL) before
being dried (Na2SO4). Flash column chromatography on silica
gel (light petroleum–EtOAc gradient, light petroleum → 10:1)
after dry-loading onto silica gel, gave geminal dibromide 14
(6.91 g, 98%) as a colourless oil. Rf 0.37 (light petroleum–
EtOAc, 10:1); [a]2D0 −35 (c = 2.1, CHCl3); mmax (thin film)/cm−1
2932 (m), 2857 (m), 1472 (w), 1428 (m), 1111 (s), 892 (w), 823
(m), 787 (w), 740 (m), 702 (s); 1H NMR (400 MHz, CDCl3) dH
7.68 (2H, dd, J = 5.4, 1.6 Hz, ArH), 7.66 (2H, dd, J = 5.4, 1.9,
ArH), 7.44–7.35 (6H, m, ArH), 6.39 (1H, dd, J = 7.5, 6.4, H-
3), 4.78 (1H, m, H-51a), 4.77 (1H, m, H-51b), 4.02 (1H, dddd,
J = 9.8, 5.4, 5.1, 4.9, H-9), 3.80–3.65 (3H, m, H-11a, H-11b &
H-5), 2.38–2.29 (3H, m, H-4a, H-6a & H-8a), 2.19 (1H, ddd,
J = 15.3, 7.5, 5.4, H-4b), 2.02–1.96 (2H, m, H-6b & H-8b),
1.86 (1H, dddd, J = 14.0, 8.4, 6.1, 5.4, H-10a), 1.62 (1H, dddd,
J = 14.0, 7.7, 6.8, 5.1, H-10b), 1.05 [9H, s, SiC(CH3)3]; 13C NMR
(100.6 MHz, CDCl3) dC 141.5, 135.6, 135.6, 135.1, 133.9, 133.8,
129.6 (2C), 127.6, 127.6, 110.8, 89.8, 70.1, 69.4, 60.5, 39.4, 39.3,
37.2, 35.9, 26.9, 19.2; HRMS (+ESI) calc. for C27H34Br2O2SiNa
(MNa+) 601.0597, found 601.0602.
light petroleum → 10:1) gave methyl ester 16 (4.82 g, 86%) as a
colourless oil. Rf 0.30 (light petroleum–EtOAc, 6:1); [a]2D0 −23
(c = 0.67, CHCl3); mmax (thin film)/cm−1 2947 (m), 1717 (s, CO),
1428 (m), 1256 (s), 1112 (s), 1082 (s), 740 (m), 703 (s); 1H NMR
(400 MHz, CDCl3) dH 7.68 (2H, dd, J = 5.4, 1.6 Hz, ArH), 7.66
(2H, dd, J = 5.4, 1.9, ArH), 7.44–7.35 (6H, m, ArH), 4.78 (1H,
m, H-51a), 4.77 (1H, m, H-51b), 4.02 (1H, dddd, J = 9.8, 5.4,
5.1, 4.9, H-9), 3.80–3.65 (3H, m, H-11a, H-11b & H-5), 3.72
(3H, s, CO2CH3), 2.38–2.29 (3H, m, H-4a, H-6a & H-8a), 2.19
(1H, ddd, J = 15.3, 7.5, 5.4, H-4b), 2.02–1.96 (2H, m, H-6b &
H-8b), 1.86 (1H, dddd, J = 14.0, 8.4, 6.1, 5.4, H-10a), 1.62 (1H,
dddd, J = 14.0, 7.7, 6.8, 5.1, H-10b), 1.05 [9H, s, SiC(CH3)3]; 13
C
NMR (100.6 MHz, CDCl3) dC 154.0, 140.8, 135.6, 135.6, 133.9,
133.8, 129.6, 129.6, 127.6 (2C), 111.3, 86.0, 74.5, 69.8, 69.7, 60.4,
52.5, 39.2, 38.6, 35.8, 26.9, 23.8, 19.2; HRMS (+ESI) calc. for
C29H36O4SiNa (MNa+) 499.2281, found 499.2271.
Enone 19
A sample of Ba(OH)2·8H2O (ca. 500 mg) was dried by heating
at 140 °C (oil bath temperature) for 3 h at <1 mm Hg. Dimethyl
(2-oxopropyl)phosphonate (416 mg, 2.51 mmol) in dry THF
(20 mL) was added to an accurately-weighed sample of the
dried Ba(OH)2·xH2O (215 mg, <1.25 mmol, glovebox), forming
an emulsion. After vigorous stirring for 30 min, aldehyde 18
(493 mg, 2.09 mmol) in THF–water (41 mL of 40:1, including
washings) was added. A period of stirring for 105 min was
followed by the pouring of the reaction mixture into chilled
(0 °C) NH4Cl solution (20 mL), the separation of the phases
and the back-extraction of the aqueous phase with Et2O
(2 × 30 mL). The combined organic fractions were washed with
brine (100 mL), dried (Na2SO4) and concentrated in vacuo. An
analysis of this material by 400 MHz 1H NMR revealed a single
C(11–12) alkene diastereomer (E/Z > 97:3). Flash column
chromatography on silica gel (light petroleum–EtOAc gradient,
8:1 → 2:1) gave enone 19 (543 mg, 94%) as a colourless oil. Rf
0.24 (light petroleum–EtOAc, 2:1); [a]2D0 −59 (c = 3.6, CHCl3);
mmax (thin film)/cm−1 2949 (w), 2239 (w, CC), 1714 (s, CO),
1674 (m), 1435 (w), 1255 (s), 1076 (m), 979 (w); 1H NMR
(500 MHz, CDCl3) dH 6.79 (1H, ddd, J = 16.1, 7.3, 7.0 Hz, H-
11), 6.11 (1H, d, J = 16.1, H-12), 4.86 (1H, m, H-51a), 4.85 (1H,
m, H-51b), 4.05 (1H, dddd, J = 6.8, 6.8, 6.0, 4.5, H-5), 3.92 (1H,
dddd, J = 7.0, 6.8, 6.0, 4.5, H-9), 3.75 (3H, s, CO2CH3), 2.62
(1H, dd, J = 17.2, 6.8, H-4a), 2.59–2.51 (1H, m, H-10a), 2.51
(1H, dd, J = 17.2, 6.8, H-4b), 2.45 (1H, dd, J = 13.6, 4.5, H-6a),
2.37 (1H, dd, J = 13.2, 4.5, H-8a), 2.33 (1H, ddd, J = 12.6, 7.3,
6.8, H-10b), 2.25 (3H, s, H-14), 2.13 (1H, dd, J = 13.6, 6.0, H-
6b), 2.04 (1H, dd, J = 13.2, 6.0, H-8b); 13C NMR (125.7 MHz,
CDCl3) dC 198.4, 153.9, 143.7, 139.9, 133.3, 112.0, 85.6, 74.4,
71.4, 70.1, 52.5, 39.0, 38.4, 36.5, 26.6, 23.4; HRMS (+ESI) calc.
for C16H20O4Na (MNa+) 299.1259, found 299.1270.
Alkynyl bromide 15
To a cold (−98 °C), stirred solution of geminal dibromide 14
(6.80 g, 11.8 mmol) in dry THF (117 mL) was added NaHMDS
in THF (13.0 mL of 1.00 M, 13.0 mmol) dropwise over 3 min
to leave a yellow solution. After stirring for a further 60 min,
the reaction was quenched with AcOH in THF (20 mL in
1:9, v/v), before being rapidly warmed to RT. Its partitioning
between Et2O (3 × 200 mL) and pH 7 buffer solution (100 mL)
was followed by the drying (MgSO4) of the organic phase and
concentration in vacuo. Flash column chromatography on silica
gel (light petroleum and then light petroleum–EtOAc, 20:1)
gave alkynyl bromide 15 as a colourless oil (5.79 g, 99%). Rf 0.47
(light petroleum–EtOAc, 7:1); [a]2D0 −16 (c = 14, CHCl3); mmax
(thin film)/cm−1 3071 (w, C–H), 2932 (w), 2857 (w), 1428 (m),
1106 (s), 1085 (s), 823 (m), 738 (m), 700 (s); 1H NMR (400 MHz,
CDCl3) dH 7.71–7.69 (4H, m, ArH), 7.44–7.38 (6H, m, ArH),
4.83 (1H, m, H-51a), 4.80 (1H, m, H-51b), 4.04 (1H, dddd,
J = 10.1, 5.8, 5.8, 5.4 Hz, H-9), 3.84–3.70 (3H, m, H-11a, H-
11b & H-5), 2.41–2.37 (4H, m, H-4a, H-4b, H-6a & H-8a), 2.12
(1H, dd, J = 13.2, 6.8, H-6b), 2.01 (1H, dd, J = 13.2, 5.8, H-8b),
1.91 (1H, dddd, J = 14.3, 8.7, 5.8, 5.8, H-10a), 1.66 (1H, dddd,
J = 14.3, 7.3, 7.3, 5.4, H-10b), 1.08 [9H, s, SiC(CH3)3]; 13C NMR
(100.6 MHz, CDCl3) dC 141.3, 135.6, 135.5, 133.9, 133.8, 129.5,
129.5, 127.6, 127.6, 110.9, 76.7, 70.0, 69.7, 60.5, 39.7, 39.2, 38.6,
35.8, 26.9, 24.8, 19.2; HRMS (+ESI) calc. for C27H33BrO2SiNa
(MNa+) 519.1331, found 519.1325.
Silyl enol ether 20
To a chilled (0 °C), stirred solution of enone 19 (430 mg, 1.56
mmol) and Et3N (0.651 mL, 4.67 mmol) in dry Et2O (20 mL) was
added TESOTf (0.528 mL, 2.33 mmol). After 60 min, the reac-
tion mixture was washed through a short plug of deactivated,
neutral alumina with copious Et2O. The concentration of the
filtrate in vacuo was followed by gravity column chromatography
on deactivated, neutral alumina (Et2O) so as to furnish silyl enol
ether 20 (576 mg, 95%) as a colourless oil. Rf 0.63 (light petro-
leum–EtOAc, 3:1); [a]2D0 −28 (c = 2.0, CHCl3); mmax (thin film)/
cm−1 2954 (w), 2240 (w, CC), 1717 (s, CO), 1251 (s), 1074 (s),
1017 (s), 965 (m), 749 (s); 1H NMR (400 MHz, CDCl3) dH 5.94–
5.92 (2H, m, H-12 & H-13), 4.86–4.84 (1H, m, H-51a), 4.82–4.81
(1H, m, H-51b), 4.26–4.25 (1H, m, H-14a), 4.23–4.22 (1H, m, H-
14b), 4.07–4.01 (1H, m, H-5), 3.79 (1H, dddd, J = 10.8, 6.6, 6.4,
6.4, H-9), 3.76 (3H, s, CO2CH3), 2.62–2.51 (2H, m, H-4a & H-
4b), 2.45 (1H, dd, J = 13.2, 4.2, H-6a), 2.42–2.32 (1H, m, H-8a),
2.36–2.27 (1H, m, H-10a), 2.29–2.20 (1H, m, H-8b), 2.17 (1H,
Methyl ester 16
To a cold (−78 °C), stirred solution of alkynyl bromide 15
n
(5.85 g, 10.1 mmol) in dry THF (100 mL) was added BuLi in
hexanes (7.87 mL of 1.54 M, 12.1 mmol) in a dropwise fashion.
After 30 min, methyl chloroformate (2.34 mL, 30.3 mL) and
then HMPA (5.27 mL, 30.3 mmol) were added to the pale yellow
solution of the lithium acetylide. A period of vigorous stirring
for a further 30 min was followed by careful quenching with
AcOH in THF (2 mL of 1:9, v/v). After rapid warming to RT,
the reaction mixture was partitioned between Et2O (3 × 200 mL)
and pH 7 buffer solution (100 mL). The combined organic
fractions were subsequently washed with brine (200 mL),
dried (MgSO4) and concentrated in vacuo. Flash column
chromatography on silica gel (light petroleum–EtOAc gradient,
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8
3 0 3 3

View Article Online
dd, J = 13.2, 6.3, H-6b), 2.03 (1H, dd, J = 13.4, 6.6, H-10b), 0.99
(9H, t, J = 7.8, SiCH2CH3), 0.71 (6H, q, J = 7.8, SiCH2CH3); 13C
NMR (100.6 MHz, CDCl3) dC 154.8, 154.0, 140.6, 130.3, 126.6,
111.6, 94.4, 85.9, 74.5, 72.4, 70.2, 52.6, 38.6, 38.3, 36.1, 23.6, 6.8,
4.9; HRMS (+ESI) calc. for C22H34O4SiNa (MNa+) 413.2124,
found 413.2134.
4.83–4.82 (1H, m, H-51a), 4.81–4.80 (1H, m, H-51b), 4.71 (1H,
dd, J = 11.7, 2.7, H-15), 4.02 (1H, dddd, J = 6.7, 6.3, 6.1, 4.0, H-
5), 3.99–3.93 (2H, m, H-9 & H-11), 3.75 (3H, s, CO2CH3), 3.54
(1H, ddd, J = 7.9, 5.0, 2.1, H-22), 3.44 (1H, d, J = 10.3, H-26),
3.43 (1H, dd, J = 9.9, 4.8, H-24), 2.79 (1H, dd, J = 14.8, 11.7,
H-14a), 2.61 (1H, dd, J = 14.8, 2.7, H-14b), 2.60–2.53 (3H, m,
H-4a, H-12a & H-21a), 2.45 (1H, dd, J = 17.1, 6.1, H-4b), 2.44
(3H, s, H-32), 2.42–2.36 (3H, m, H-6a, H-8a & H-12b), 2.31
(1H, ddd, J = 7.9, 6.3, 5.0, H-21b), 2.24 (1H, ddd, J = 14.3, 9.8,
5.1, H-10a), 2.17 (3H, s, H-48), 2.08 (1H, dd, J = 13.5, 6.7, H-
6b), 2.02 (1H, dd, J = 13.3, 6.1, H-8b), 1.79 (1H, qdd, J = 6.9,
4.8, 2.1, H-23), 1.73 (1H, ddq, J = 10.3, 9.9, 6.4, H-25), 1.62
(1H, ddd, J = 14.3, 7.7, 4.3, H-10b), 0.98 (3H, d, J = 6.9, H-50),
0.91 [9H, s, SiC(CH3)3], 0.75 (3H, d, J = 6.4, H-49), 0.06 [3H, s,
Si(CH3)aMe], 0.05 [3H, s, SiMe(CH3)b]; 13C NMR (125.7 MHz,
CDCl3) dC 205.7, 161.5, 160.6, 153.9, 140.8, 140.2, 138.1, 137.8,
137.1, 135.6, 134.5, 118.6, 118.0, 111.9, 88.8, 85.9, 74.4, 74.1,
71.6, 69.9, 68.8, 52.6, 46.9, 46.1, 39.3, 39.2, 39.2, 38.7, 36.5,
34.8, 25.8, 23.7, 18.1, 14.3, 13.9, 13.8, 5.9, −4.1, −4.8 (two
signals missing due to coincidence); HRMS (+ESI) calc. for
C43H60N2O9SiNa (MNa+) 799.3966, found 799.3976.
Silyloxydihydropyran 23
To a mixture of aldehyde 21 (78.0 mg, 0.156 mmol) and
activated, crushed 4 Å molecular sieves (102 mg) was added
Jacobsen Schiff base hexafluoroantimonate catalyst 6b14a in dry
acetone (100 lL of 0.156 M, 15.6 lmol). The vial was wrapped
in aluminium foil and the mixture left to pre-stir at RT for 3 h.
Silyl enol ether 20 (60.8 mg, 0.156 mmol) was then added in dry
acetone (0.3 mL, including washings) and the solvent volume
reduced to ca. 0.1 mL under a positive stream of argon. After
being stirred for 40 h in the absence of light, the reaction mixture
was washed through a cotton wool plug with copious Et2O and
concentrated in vacuo. Gravity column chromatography on
deactivated, neutral alumina (light petroleum–EtOAc gradient,
light petroleum → 2:1) gave residual silyl enol ether 20 (20.1 mg,
33%), a mixture of silyloxydihydropyrans 23–24 (50.1 mg, 36%,
23–24 = 1.1:1.0) and then residual aldehyde 21 (27.0 mg, 35%).
After purification by HPLC [light petroleum–EtOAc (2.8:1.0);
Rt = 20 min (24: 48.5%), Rt = 24 min (23: 51.5%); detection at
k = 254 nm], silyl enol ether 23 (26.2 mg, 19%) was isolated as a
colourless oil. Rf 0.18 (light petroleum–EtOAc, 1:3); [a]2D0 +36
(c = 0.34, CHCl3); mmax (thin film)/cm−1 2955 (s), 2929 (s), 2855
(m), 2241 (s, CC), 1717 (s, CO), 1253 (s), 1075 (s), 835 (s),
749 (m); 1H NMR (500 MHz, CDCl3) dH 7.49 (1H, s, H-30), 7.47
(1H, s, H-17), 6.66 (1H, ddd, J = 15.8, 8.5, 6.6, H-20), 6.32 (1H,
d, J = 15.8, H-19), 6.18 (1H, s, H-28), 4.88 (1H, dd, J = 1.7, 1.7,
H-12), 4.84–4.82 (1H, m, H-51a), 4.81–4.79 (1H, m, H-51b),
4.61 (1H, dd, J = 10.9, 3.2, H-15), 4.41 (1H, ddd, J = 7.3, 6.1,
1.7, H-11), 4.02 (1H, dddd, J = 6.8, 6.3, 6.1, 4.3, H-5), 4.01–3.97
(1H, m, H-9), 3.75 (3H, s, CO2CH3), 3.54 (1H, ddd, J = 7.9, 6.0,
1.9, H-22), 3.45 (1H, d, J = 10.1, H-26), 3.42 (1H, dd, J = 9.9,
5.7, H-24), 2.58–2.52 (1H, m, H-21a), 2.58–2.53 (1H, m, H-4a),
2.49–2.43 (4H, m, H-4b, H-6a, H-8a & H-14a), 2.44 (3H, s, H-
32), 2.34–2.28 (1H, m, H-21b), 2.28 (1H, dd, J = 15.9, 3.2, H-
14b), 2.13 (1H, dd, J = 12.9, 6.8, H-8b), 2.05 (1H, dd, J = 12.2,
6.1, H-6b), 2.01 (1H, ddd, J = 13.4, 7.3, 7.3, H-10a), 1.92 (3H,
s, H-48), 1.79 (1H, qdd, J = 5.7, 5.7, 1.9, H-23), 1.73 (1H, ddq,
J = 10.1, 9.9, 6.6, H-25), 1.58 (1H, ddd, J = 13.4, 7.3, 6.1, H-
10b), 0.98 (9H, t, J = 8.0, SiCH2CH3), 0.98 (3H, d, J = 5.7,
H-50), 0.91 [9H, s, SiC(CH3)3], 0.75 (3H, d, J = 6.6, H-49), 0.68
(6H, q, J = 8.0, SiCH2CH3), 0.06 [3H, s, Si(CH3)aMe], 0.05 [3H,
s, SiMe(CH3)b]; 13C NMR (125.7 MHz, CDCl3) dC 161.1, 160.6,
154.0, 148.4, 142.2, 140.6, 138.1, 137.8, 136.4, 135.5, 134.1,
118.6, 118.3, 111.6, 105.0, 88.8, 85.9, 77.4, 77.4, 74.5, 71.5, 69.9,
69.8, 69.7, 52.6, 39.4, 39.1, 39.1, 38.6, 36.4, 35.0, 34.8, 25.8, 23.8,
18.1, 14.3, 13.9, 13.8, 6.7, 5.9, 5.0, −4.1, −4.8; HRMS (+ESI)
calc. for C49H74N2O9Si2Na (MNa+) 913.4831, found 913.4815.
The data for tetrahydropyranone 26, which was prepared
using the analogous deprotection of silyl enol ether 24, were as
follows: Rf 0.51 (light petroleum–EtOAc, 2:3); [a]D20 +29 (c = 2.0,
CHCl3); mmax (thin film)/cm−1 2928 (m), 2238 (w, CC), 1716 (s,
CO), 1256 (s), 1107 (m), 1075 (s), 1029 (m), 887 (m), 835 (m),
1
775 (m); H NMR (700 MHz, CDCl3) dH 7.56 (1H, s, H-17),
7.49 (1H, s, H-30), 6.69 (1H, ddd, J = 16.2, 7.2, 6.9 Hz, H-20),
6.33 (1H, d, J = 16.2, H-19), 6.17 (1H, s, H-28), 4.81–4.80 (1H,
m, H-51a), 4.80–4.79 (1H, m, H-51b), 4.73 (1H, dd, J = 11.7,
2.9, H-15), 4.06–4.01 (3H, m, H-5, H-9 & H-11), 3.74 (3H, s,
CO2CH3), 3.54 (1H, ddd, J = 7.9, 7.9, 5.6, H-22), 3.44 (1H, d,
J = 10.3, H-26), 3.43 (1H, dd, J = 9.9, 4.8, H-24), 2.86 (1H, dd,
J = 14.9, 11.7, H-14a), 2.70–2.63 (1H, m, H-12a), 2.67 (1H,
dd, J = 17.2, 8.3, H-12b), 2.58–2.53 (1H, m, H-21a), 2.55 (1H,
dd, J = 14.9, 2.9, H-14b), 2.46–2.42 (1H, m, H-6a), 2.45–2.36
(2H, m, H-4a & H-4b), 2.44 (3H, s, H-32), 2.34–2.28 (1H, m, H-
21b), 2.30 (1H, dd, J = 13.5, 4.2, H-8a), 2.08 (1H, dd, J = 13.6,
5.1, H-6b), 1.97 (1H, dd, J = 13.5, 7.5, H-8b), 1.92 (3H, s, H-
47), 1.80–1.76 (1H, m, H-10a), 1.79 (1H, qdd, J = 6.9, 5.6, 4.8,
H-23), 1.74–1.71 (1H, m, H-10b), 1.73 (1H, ddq, J = 10.3, 9.9,
6.4, H-25), 0.98 (3H, d, J = 6.9, H-50), 0.91 [9H, s, SiC(CH3)3],
0.75 (3H, d, J = 6.4, H-49), 0.07 [3H, s, Si(CH3)aMe], 0.05 [3H,
s, SiMe(CH3)b]; 13C NMR (125.7 MHz, CDCl3) dC 206.0, 161.5,
160.6, 153.9, 140.8, 140.5, 138.1, 137.9, 136.9, 135.5, 134.7,
118.5, 118.1, 111.6, 88.9, 86.3, 77.5, 77.4, 74.1, 73.7, 71.4, 70.5,
68.2, 52.5, 48.0, 46.2, 40.5, 39.8, 39.3, 38.6, 36.4, 34.9, 25.8, 23.0,
18.1, 14.3, 13.9, 13.7, 5.9, −4.1, −4.8; HRMS (+ESI) calc. for
C43H60N2O9SiNa (MNa+) 799.3966, found 799.3951.
Alcohol 27
To a cold (−78 °C), stirred solution of tetrahydropyranone
25 (14.0 mg, 18.0 lmol) in dry THF (0.60 mL) was added
LiAl(OtBu)3H in THF (0.180 mL of 1.00 M, 0.180 mmol). The
warming of the reaction to −20 °C over 30 min was followed
by its stirring at between −20 and −10 °C for a further 90 min.
The careful addition of NH4Cl solution (5.0 mL) was followed
by rapid warming to RT and extraction with Et2O–EtOAc
(3 × 10 mL of 9:1). The combined extracts were washed with
brine (25 mL), dried (Na2SO4) and concentrated in vacuo.
Analysis by 500 MHz 1H NMR revealed an S/R ratio of >97:3
at C-13. Purification by silica gel PTLC (light petroleum–EtOAc,
4:1) gave alcohol 27 (13.9 mg, 99%) as a colourless oil. Rf 0.49
(EtOAc); [a]2D0 +17 (c = 0.85, CHCl3); mmax (thin film)/cm−1 3377
(br, O–H), 2926 (m), 2240 (w, CC), 1715 (m, CO), 1255
Tetrahydropyranone 25
To a chilled (0 °C) solution of silyloxydihydropyran 23
(26.2 mg, 29.4 lmol) in THF (2.5 mL) was added AcOH in
THF (58.8 lL of 1.00 M, 58.8 lmol) and then TBAF in THF
(29.4 lL of 1.00 M, 29.4 lmol). After stirring for 60 min, the
reaction mixture was poured into water (10 mL) and extracted
with CH2Cl2 (4 × 10 mL). The combined extracts were dried
(Na2SO4) and concentrated in vacuo. The purification of the
residue by silica gel PTLC (light petroleum–EtOAc, 2:3), gave
tetrahydropyranone 25 (21.0 mg, 92%) as a colourless oil. Rf
0.47 (light petroleum–EtOAc, 2:3); [a]2D0 +23 (c = 0.12, CHCl3);
mmax (thin film)/cm−1 2928 (s), 2240 (w, CC), 1717 (s, CO),
1257 (s), 1076 (s), 835 (m), 776 (m); 1H NMR (700 MHz, CDCl3)
dH 7.51 (1H, s, H-17), 7.49 (1H, s, H-30), 6.69 (1H, ddd, J = 16.2,
8.3, 6.3, H-20), 6.32 (1H, d, J = 16.2, H-19), 6.18 (1H, s, H-28),
1
(s), 1075 (s), 1030 (m), 835 (m), 775 (m), 753 (m); H NMR
(500 MHz, CDCl3) dH 7.48 (1H, s, H-30), 7.44 (1H, s, H-17),
6.65 (1H, ddd, J = 15.9, 8.0, 6.0, H-20), 6.31 (1H, d, J = 15.9,
H-19), 6.17 (1H, s, H-28), 4.81–4.78 (2H, m, H-51a & H-51b),
4.40 (1H, dd, J = 11.8, 2.4, H-15), 4.08 (1H, dddd, J = 8.4, 5.0,
3 0 3 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8

View Article Online
5.0, 5.0, H-5), 3.96–3.89 (1H, m, H-9), 3.92 (1H, dddd, J = 11.0,
11.0, 6.1, 4.9, H-13), 3.76 (3H, s, CO2CH3), 3.71–3.67 (1H, m, H-
11), 3.53 (1H, ddd, J = 8.8, 6.8, 2.2, H-22), 3.44 (1H, d, J = 10.4,
H-26), 3.42 (1H, dd, J = 10.1, 4.6, H-24), 2.69 (1H, dd, J = 17.3,
8.4, H-4a), 2.54 (1H, ddd, J = 14.1, 6.8, 6.0, H-21a), 2.46–2.39
(2H, m, H-6a & H-8a), 2.44 (3H, s, H-32), 2.43–2.37 (1H, m,
H-4b), 2.37–2.25 (1H, m, H-21b), 2.25 (1H, ddd, J = 12.2, 6.1,
2.4, H-14a), 2.23–2.18 (1H, m, H-12a), 2.13–2.09 (1H, m, H-
10a), 2.09–2.01 (1H, m, H-6b), 2.00 (1H, dd, J = 13.2, 5.5, H-
8b), 1.91 (3H, s, H-48), 1.78 (1H, qdd, J = 6.9, 4.6, 2.2, H-23),
1.76–1.69 (1H, m, H-25), 1.60 (1H, ddd, J = 12.2, 11.8, 11.0,
H-14b), 1.54 (1H, ddd, J = 13.9, 7.6, 5.3, H-10b), 1.28 (1H,
ddd, J = 12.2, 12.2, 11.0, H-12b), 0.97 (3H, d, J = 6.9, H-50),
0.90 [9H, s, SiC(CH3)3], 0.74 (3H, d, J = 6.5, H-49), 0.06 [3H, s,
syringe pump over 8 h to a solution of DMAP (4.8 mg, 39 lmol)
in dry PhMe (8.5 mL) which was being stirred at RT. At the
end of the addition, the transfer was quantified with more dry
PhMe (0.50 mL) before the cloudy solution was left to stir for a
further 12 h. The reaction mixture was subsequently diluted with
EtOAc (10 mL) and sequentially washed with citric acid solution
(10 mL), NaHCO3 solution (10 mL) and brine (10 mL). The
drying (Na2SO4) of the organic fraction and its concentration
in vacuo was followed by the purification of the residue by
silica gel PTLC (light petroleum–EtOAc, 1:1). This furnished
macrolactone 30 (2.4 mg, 70%) as a colourless oil. Rf 0.40 (light
petroleum–EtOAc, 1:1); [a]2D0 +17 (c = 0.13, CHCl3); mmax (thin
film)/cm−1 2924 (s), 2852 (s), 2229 (w), 1713 (s, CO), 1112 (s),
1094 (s), 704 (m); 1H NMR (500 MHz, CDCl3) dH 7.68–7.66 (4H,
m, ArH), 7.50 (1H, s, H-30), 7.45–7.35 (6H, m, ArH), 7.37 (1H,
s, H-17), 6.69 (1H, ddd, J = 15.9, 9.0, 6.3 Hz, H-20), 6.29 (1H,
d, J = 15.9, H-19), 6.24 (1H, s, H-28), 4.78–4.77 (1H, m, H-51a),
4.73–4.72 (1H, m, H-51b), 4.40 (1H, dd, J = 10.9, 4.6, H-24),
4.12–4.05 (1H, m, H-9), 4.09 (1H, dd, J = 10.3, 1.9, H-15),
3.95–3.86 (1H, m, H-11), 3.87 (1H, dddd, J = 9.8, 5.7, 4.9, 2.4,
H-5), 3.59 (1H, d, J = 10.3, H-26), 3.46–3.40 (1H, m, H-22), 3.46
(1H, app. dd, J = 11.0, 9.8, H-13), 2.96 (1H, dd, J = 17.3, 5.7, H-
4a), 2.66 (1H, ddq, J = 10.9, 10.3, 6.3, H-25), 2.53–2.48 (2H, m,
H-21a & H-21b), 2.45 (1H, dd, J = 13.4, 4.9, H-8a), 2.44 (3H, s,
H-32), 2.34 (1H, dd, J = 13.4, 9.8, H-6a), 2.23 (1H, dd, J = 17.3,
4.9, H-4b), 2.12 (1H, dd, J = 13.4, 2.4, H-6b), 2.08–2.00 (1H,
m, H-23), 2.01–1.94 (1H, m, H-14a), 2.00–1.96 (1H, m, H-10a),
1.96 (3H, s, H-48), 1.93 (1H, dd, J = 13.4, 4.9, H-8b), 1.76–1.69
(1H, m, H-10b), 1.75–1.66 (1H, m, H-12a), 1.75 (1H, ddd,
J = 12.2, 11.0, 10.3, H-14b), 1.49 (1H, ddd, J = 12.2, 12.2, 11.0,
H-12b), 1.06 [9H, s, SiC(CH3)3], 1.03 (3H, d, J = 6.8, H-50), 0.84
(3H, d, J = 6.3, H-49); 13C NMR (125.7 MHz, CDCl3) dC 160.9,
160.7, 153.6, 141.9, 141.5, 137.5, 136.6, 135.7 (2C), 135.7, 134.4,
134.2, 134.2, 129.8, 129.7, 127.6 (2C), 125.5, 119.7, 118.9, 110.8,
92.7, 89.4, 84.7, 78.3, 75.4, 72.9, 70.0, 69.2, 67.6, 65.9, 42.2, 39.1,
38.2, 38.0, 37.7, 34.2, 30.3, 29.7, 26.9, 23.8, 19.1, 14.1, 13.8, 13.5,
5.5; HRMS (+ESI) calc. for C52H62N2O8SiNa (MNa+) 893.4173,
found 893.4156.
Si(CH3)aMe], 0.05 [3H, s, SiMe(CH3)b] (OH signal missing); 13
C
NMR (125.7 MHz, CDCl3) dC 161.1, 160.6, 154.2, 142.2, 140.4,
138.1, 137.8, 136.3, 135.5, 134.0, 118.5, 118.3, 111.6, 88.8, 86.7,
77.4, 77.4, 74.1, 72.9, 71.2, 70.0, 68.6, 67.8, 52.8, 40.2, 39.8,
39.5, 39.4, 39.1, 38.9, 36.4, 34.8, 25.8, 23.2, 18.1, 14.3, 13.9,
13.8, 5.8, −4.1, −4.8; HRMS (+ESI) calc. for C43H62N2O9SiNa
(MNa+) 801.4122, found 801.4095.
Seco-acid 29
A solution of methyl ester 28 (4.6 mg, 5.1 lmol) and LiOH·H2O
(1.1 mg, 26 lmol) in THF (1.7 mL) and water (0.34 mL) was
vigorously stirred at RT, and without argon protection, for
2 h. It was subsequently partitioned between CH2Cl2–iPrOH
(5 × 10 mL of 9:1, v/v) and HCl solution (10 mL of 0.15 M)
before the combined organic extracts were dried (Na2SO4) and
concentrated in vacuo. The purification of the residue by silica
gel PTLC (EtOAc–iPrOH–AcOH, 90:10:1) gave seco-acid 29
(4.4 mg, 97%) as a white solid. Rf 0.11 (EtOAc–iPrOH–AcOH,
90:10:1); [a]2D0 +10.3 (c = 0.89, MeOH); mmax (thin film)/cm−1
3355 (br m, O–H), 2927 (m), 2857 (m), 2235 (w, CC), 1655 (m),
1582 (s), 1363 (s), 1105 (s), 1027 (s), 735 (s); 1H NMR (500 MHz,
CD3OD) dH 7.77 (1H, s, H-30), 7.69–7.65 (4H, m, ArH), 7.61
(1H, s, H-17), 7.47–7.36 (6H, m, ArH), 6.73 (1H ddd, J = 16.1,
7.9, 6.3, H-20), 6.32 (1H, d, J = 16.1, H-19), 6.17 (1H, s, H-28),
4.82–4.78 (1H, m, H-51a), 4.77–4.72 (1H, m, H-51b), 4.20 (1H,
dd, J = 11.7, 1.6, H-15), 3.96 (1H, dddd, J = 10.4, 10.4, 4.4, 4.4,
H-13), 3.85 (1H, dddd, J = 9.8, 5.0, 5.0, 5.0, H-5), 3.77–3.72
(1H, m, H-9), 3.68–3.58 (1H, m, H-22), 3.52 (1H, d, J = 10.4,
H-26), 3.46 (1H, dd, J = 10.1, 4.7, H-24), 3.39–3.34 (1H, m, H-
11), 2.54 (1H, ddd, J = 13.9, 6.6, 6.3, H-21a), 2.48–2.35 (2H, m,
H-4a & H-6a), 2.42 (3H, s, H-32), 2.41–2.30 (1H, m, H-21b),
2.31–2.22 (1H, m, H-8a), 2.26–2.14 (2H, m, H-4b & H-6b),
2.13–2.09 (1H, m, H-14a), 1.97–1.88 (1H, m, H-8b), 1.95–1.84
(1H, m, H-23), 1.90 (3H, s, H-48), 1.89–1.79 (2H, m, H-10a & H-
12a), 1.76–1.65 (1H, m, H-25), 1.71–1.58 (1H, m, H-14b), 1.43
(1H, ddd, J = 13.6, 6.3, 6.0, H-10b), 1.36–1.23 (1H, m, H-12b),
1.04 [9H, s, SiC(CH3)3], 0.99 (3H, d, J = 6.6, H-50), 0.83 (3H,
d, J = 6.6, H-49) (OH and COOH signals missing); 13C NMR
(125.7 MHz, CD3OD) dC 162.8, 162.7, 143.3, 142.7, 140.0, 138.9,
138.5, 137.6, 136.9, 136.9, 136.0, 135.4, 135.3, 131.0, 131.0,
128.8 (2C), 119.2, 118.6, 111.6, 90.2, 79.0, 77.2, 74.1, 72.1, 71.9,
70.7, 70.6, 40.3, 40.1, 40.0, 39.1, 37.2, 35.4, 30.8, 27.5, 24.3, 23.7,
19.9, 14.6, 13.8, 13.4, 6.1 (C-1, C-2 and C-3 signals missing due
to signal broadening); HRMS (+ESI) calc. for C52H64N2O9SiNa
(MNa+) 911.4279, found 911.4253.
Methyl enol ether 37
To a solution of ketolactone 32 (599 mg, 2.56 mmol) in dry
acetone (20 mL) was added (MeO)2SO2 (254 lL, 2.68 mmol).
Solid K2CO3 (371 mg, 2.68 mmol) was added portionwise over
6 h. A period of stirring for a further 48 h at RT led to the almost
complete consumption of 32. The reaction mixture was diluted
with Et2O (100 mL), washed with water (2 × 100 mL) and then
brine (100 mL), before being dried (MgSO4) and concentrated
in vacuo. The purification of the residue by flash column
chromatography on silica gel (light petroleum–EtOAc, 2:3) gave
methyl enol ether 37 (626 mg, 99%) as a colourless oil. Rf 0.22
(light petroleum–EtOAc, 2:3); [a]2D0 = +62 (c = 0.43, CHCl3);
mmax (thin film)/cm−1 2918 (s), 1704 (m, CO), 1621 (m, CC),
1
1223 (m), 1094 (m), 1051 (m), 819 (w); H NMR (500 MHz,
CDCl3) dH 7.37–7.30 (5H, m, ArH), 5.14 (1H, d, J = 1.4, H-34),
4.60 (2H, s, ArCHaHb & ArCHaHb), 4.58–4.54 (1H, m, H-37),
3.74 (3H, s, OCH3), 3.71–3.69 (2H, m, H-38a & H-38b), 2.72
(1H, ddd, J = 17.2, 11.8, 1.4, H-36a), 2.40 (1H, dd, J = 17.2, 4.1,
H-36b); 13C NMR (100.6 MHz, CDCl3) dC 172.7, 166.6, 137.7,
128.5, 127.9, 127.7, 90.2, 74.6, 73.7, 70.6, 56.0, 29.8; HRMS
(+ESI) calc. for C14H17O4 (MH+) 249.1127, found 249.1125.
Macrolactone 30
Alcohol 38
To a sample of seco-acid 29 (3.5 mg, 3.9 lmol) was added a
dry THF solution (738 lL) of 2,4,6-trichlorobenzoyl chloride
(64.0 × 10−3 M, 47.2 lmol) and Et3N (0.128 M, 94.4 lmol).
The resulting solution was stirred at RT for 45 min before the
volatiles were removed in vacuo using a needle inserted through
the septum. The vigour with which volatiles were removed at
this stage was controlled using occasional ice bath cooling. A
solution of the residue in dry PhMe (1.0 mL) was added via
To a solution of benzyl ether 37 (626 mg, 2.52 mmol) in undried
EtOAc (20 mL, winchester grade) was carefully added Pd–C
(980 mg of 10%, w/w). The solution was subsequently placed
under a hydrogen atmosphere (balloon pressure) and the
headspace purged. After 12 h of stirring at RT, the hydrogen
atmosphere was replaced with nitrogen. The washing of the
reaction mixture through a short Celite® pad, under a nitrogen
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8
3 0 3 5

View Article Online
stream and with copious EtOAc–MeOH (20:1), was followed
by its concentration in vacuo. Analysis by 400 MHz H NMR
chromatography on silica gel (light petroleum–EtOAc, 20:1 and
then 4:1) gave recovered dienal 40 (182 mg), and then aldol
adduct 45 [1.40 g, 96% based on N-acetyl-4(S)-IPTT 44] as a
yellow solid. Rf 0.41 (light petroleum–EtOAc, 1:1); mp 81–83 °C
(from Et2O); [a]D20 = +135 (c = 5.3, CHCl3); IR (CH2Cl2 solution)/
cm−1 3592 (w, O–H), 3016 (m), 2986 (s), 2968 (s), 1682 (s, CO),
1470 (m), 1364 (s), 1314 (s), 1168 (s), 1094 (s), 1043 (s), 965 (m);
1H NMR (400 MHz, CDCl3) dH 6.42 (1H, d, J = 15.6 Hz, H-41),
6.39 (1H, s, H-39), 5.81 (1H, dd, J = 15.6, 5.8, H-42), 5.16 (1H,
ddd, J = 7.9, 6.3, 1.0, CHiPr), 4.72 (1H, dddd, J = 8.7, 5.8, 4.1,
3.0, H-43), 3.70 (1H, dd, J = 17.5, 3.0, H-44a), 3.52 (1H, dd,
J = 11.5, 7.9, CHaHbS), 3.33 (1H, dd, J = 17.5, 8.7, H-44b), 3.04
(1H, dd, J = 11.5, 1.0, CHaHbS), 2.92 (1H, d, J = 4.1, OH), 2.41–
2.32 (1H, m, CHMe2), 1.97 (3H, s, H-47), 1.07 [3H, d, J = 6.8,
CH(CH3)aMe], 0.99 [3H, d, J = 6.8, CHMe(CH3)b]; 13C NMR
(100.6 MHz, CDCl3) dC 203.0, 172.3, 144.4, 131.6, 130.3, 84.5,
71.4, 68.4, 45.1, 30.8, 30.7, 20.1, 19.1, 17.8; HRMS (+ESI) calc.
for C14H20INO2S2Na (MNa+) 447.9878, found 447.9885.
1
revealed a single C-35 epimer (R/S > 97:3). Flash column
chromatography on silica gel (EtOAc–MeOH, 20:1) gave
recovered benzyl ether 37 (21 mg, 3%), and alcohol 38 (339 mg,
84%) as a colourless oil which solidified to a white solid on
standing. Rf 0.24 (EtOAc–MeOH, 20:1); mp 38–40 °C (from
CH2Cl2); [a]2D0 = −29 (c = 1.2, CHCl3); IR (thin film)/cm−1 3417
(br, O–H), 2922 (s), 1729 (s, CO), 1257 (s), 1090 (m); 1H NMR
(400 MHz, CDCl3) dH 4.34 (1H, dddd, J = 11.6, 5.5, 3.1, 3.1,
H-37), 3.83–3.77 (2H, m, H-38a & H-35), 3.73–3.67 (1H, m,
H-38b), 3.35 (3H, s, OCH3), 2.86 (1H, dd, J = 17.2, 5.6, H-34a),
2.63 (1H, dd, J = 6.7, 6.6, OH), 2.55 (1H, dd, J = 17.2, 7.2, H-
34b), 2.24 (1H, ddd, J = 13.7, 5.5, 4.2, H-36a), 1.72 (1H, ddd,
J = 13.7, 11.6, 8.5, H-36b); 13C NMR (100.6 MHz, CDCl3) dC
170.1, 77.7, 72.1, 64.5, 56.0, 36.2, 30.3; HRMS (+ESI) calc. for
C7H13O4 (MH+) 161.0814, found 161.0812.
Aldehyde 31
Weinreb amide 46
To a cold (−78 °C), stirred solution of (COCl)2 (0.149 mL,
1.71 mmol) in dry CH2Cl2 (4.0 mL) was added a solution of
DMSO in dry CH2Cl2 (1.21 mL of 2.82 M, 3.41 mmol) over
3 min. After 10 min stirring, alcohol 38 (91.0 mg, 0.568 mmol)
in dry CH2Cl2 (2.0 mL, including washings) was added dropwise
over 3 min. After a further 75 min at −78 °C, Et3N (0.792 mL,
5.68 mmol) was added dropwise so as to produce a slightly
cloudy solution. After 15 min, the solution was allowed to warm
to −20 °C over 30 min. After 10 min at −20 °C, light petroleum–
PhMe (10 mL of 3:1) was added and the reaction mixture was
rapidly warmed to RT. It was subsequently washed through a
short Celite® plug with copious light petroleum–PhMe (3:1)
before the filtrate was concentrated in vacuo. The purification
of the residue by flash column chromatography on silica gel
(EtOAc) gave aldehyde 31 (79.2 mg, 88%) as a colourless oil.
Rf 0.20 (EtOAc); [a]2D0 = −42 (c = 0.95, CHCl3); IR (CHCl3
solution)/cm−1 3012 (m), 2934 (m), 2835 (m), 1736 (s, CO),
A solution of aldol adduct 45 (290 mg, 0.682 mmol), N,O-
dimethylhydroxylamine hydrochloride (166 mg, 1.70 mmol)
and imidazole (232 mg, 3.41 mmol) in dry CH2Cl2 (3.6 mL) was
stirred at RT for 24 h before being poured into chilled (0 °C)
NH4Cl solution (10 mL). Extractions with Et2O (4 × 10 mL)
were followed by the washing of the combined organic fractions
with brine (40 mL), drying (Na2SO4) and concentration in vacuo
to a yellow oil. Flash column chromatography on silica gel
(light petroleum–EtOAc gradient, 3:1 → 2:1) gave 4(S)-IPTT
and then Weinreb amide 46 (160 mg, 72%) as a yellow oil. Rf
0.38 (EtOAc); [a]D20 = +13 (c = 2.8, CHCl3); mmax (thin film)/cm−1
3409 (w, O–H), 2918 (w), 1636 (s, CO), 1424 (m), 1386 (s),
1296 (m), 998 (m), 964 (s), 763 (w); 1H NMR (500 MHz, CDCl3)
dH 6.41 (1H, dq, J = 15.8, 0.6, H-41), 6.36 (1H, s, H-39), 5.79
(1H, dd, J = 15.8, 6.0, H-42), 4.59 (1H, dddd, J = 8.8, 6.0, 3.8,
2.9, H-43), 3.97 (1H, d, J = 3.8, OH), 3.68 (3H, s, OCH3), 3.19
(3H, s, NCH3), 2.72 (1H, dd, J = 17.3, 2.9, H-44a), 2.60 (1H,
dd, J = 17.3, 8.8, H-44b), 1.95 (3H, d, J = 0.6, H-47); 13C NMR
(125.7 MHz, CDCl3) dC 169.7, 142.2, 129.3, 129.0, 83.8, 68.6,
61.8, 39.5, 33.4, 21.9; HRMS (+ESI) calc. for C10H16INO3Na
(MNa+) 348.0073, found 348.0081.
1
1357 (m), 1236 (m), 1097 (s); H NMR (500 MHz, CDCl3) dH
9.71 (1H, s, H-38), 4.73 (1H, dd, J = 6.7, 2.8, H-37), 3.79–3.77
(1H, m, H-35), 3.21 (3H, s, OCH3), 2.84 (1H, dd, J = 17.9, 2.8,
H-34a), 2.65 (1H, dd, J = 17.9, 3.9, H-34b), 2.49 (1H, ddd,
J = 14.4, 2.8, 2.0, H-36a), 2.19 (1H, app. dd, J = 14.4, 6.7, H-
36b); 13C NMR (100.6 MHz, CDCl3) dC 199.4, 167.3, 80.1, 71.1,
56.1, 36.7, 29.0; HRMS (+EI) calc. for C6H9O3 ([M–CHO]+)
129.0552, found 129.0548.
Methyl ether 47
To a stirred mixture of alcohol 46 (26.9 mg, 82.7 lmol) and
Ag2O (192 mg, 0.829 mmol, glovebox) was added MeI in dry
Et2O (2.06 mL of 1:3). After being refluxed for 3 h, the reac-
tion mixture was cooled to RT and washed through a short
plug of silica gel with copious Et2O. Concentration in vacuo
gave methyl ether 47 (26.6 mg, 95%) as a colourless oil. Rf 0.33
(light petroleum–EtOAc, 1:1); [a]2D0 = +37 (c = 0.89, CHCl3);
mmax (thin film)/cm−1 2935 (w), 1658 (s, CO), 1385 (m), 1092 (s),
Aldol adduct 45
An excess of Sn(OTf)2 (ca. 3.0 g) was pre-washed with dry Et2O
(3 × 5 mL) and dried in vacuo (1 mm Hg) in order to remove
traces of CF3SO3H. To a cold (−55 °C), stirred suspension of an
accurately-weighed amount of this material (1.72 g, 4.12 mmol,
glovebox) in dry CH2Cl2 (6.0 mL), was added N-ethylpiperidine
(0.566 mL, 4.12 mmol) in a dropwise fashion. After 15 min, the
solution was slightly yellow in colour. A solution of N-acetyl-
4(S)-IPTT 44 (698 mg, 3.43 mmol, azeotropically dried 3 × with
PhMe) in dry CH2Cl2 (4 mL, including washings) was added
dropwise. The mixture was vigorously stirred at between −50
and −40 °C for 3 h and then between −40 and −38 °C for 90 min
before it was cooled to −98 °C. After the dropwise addition of
dienal 40 (915 mg, 4.12 mmol) in dry CH2Cl2 (6.6 mL, including
washings), the temperature was allowed to slowly rise to −78 °C.
The solution was vigorously stirred at this temperature until
150 min after the end of the addition of 40, at which point a pH 7
buffer–MeOH mixture (6 mL of 2:1) was added dropwise. The
mixture was rapidly warmed to RT and washed through a short
Celite® plug with copious Et2O. The partitioning of the filtrate
between Et2O (200 mL) and water (200 mL), and then brine
(200 mL), was followed by drying (MgSO4) and concentration
1
998 (m), 966 (m); H NMR (400 MHz, CDCl3) dH 6.39 (1H, s,
H-39), 6.37 (1H, dq, J = 15.8, 1.0, H-41), 5.67 (1H, dd, J = 15.8,
7.5, H-42), 4.17 (1H, ddd, J = 8.3, 7.5, 5.0, H-43), 3.68 (3H, s,
NOCH3), 3.28 [3H, s, C-43(OCH3)], 3.18 (3H, s, NCH3), 2.88
(1H, dd, J = 15.6, 8.3, H-44a), 2.48 (1H, dd, J = 15.6, 5.0, H-
44b), 1.97 (3H, d, J = 1.0, H-47); 13C NMR (100.6 MHz, CDCl3)
dC 171.3, 144.4, 133.2, 129.4, 84.5, 78.2, 61.3, 56.7, 38.2, 32.0,
20.1; HRMS (+ESI) calc. for C11H18INO3Na (MNa+) 362.0229,
found 362.0221.
Aldehyde 48
DIBAl-H in CH2Cl2 (1.67 mL of 1.00 M, 1.67 mmol) was
added in a dropwise fashion to a cold (−78 °C), stirred solu-
tion of Weinreb amide 47 (390 mg, 1.39 mmol) in dry THF
(10 mL). A period of stirring for 30 min was followed by the
dropwise addition of MeOH (0.50 mL). After 5 min, the reac-
tion mixture was cannulated into a vigorously stirred, biphasic
mixture of Et2O (100 mL) and Rochelle’s salt solution (100 mL)
1
in vacuo. An analysis of the residue by 500 MHz H NMR
showed no sign of (epi-C-43)-45. Its purification by flash column
3 0 3 6
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8

View Article Online
which was being held at −15 °C. The transfer was quantified
with more Et2O. The rapid warming of the reaction mixture
to RT and stirring for 30 min was followed by a phase separa-
tion and the back-extraction of the aqueous phase with Et2O
(2 × 25 mL). The combined organic fractions were washed with
brine (50 mL), dried (Na2SO4) and concentrated in vacuo. Flash
column chromatography on silica gel (light petroleum–EtOAc,
10:1 → 2:1) gave aldehyde 48 (303 mg, 94%) as a colourless
oil. Rf 0.57 (light petroleum–EtOAc, 1:1); [a]2D0 = +33 (c = 0.94,
CHCl3); IR (CH2Cl2 solution)/cm−1 3062 (m), 2981 (s), 2962 (s),
2930 (s), 2827 (m), 2732 (w), 1725 (s, CO), 1682 (s), 1298 (s),
1160 (m), 1116 (s), 1099 (s), 966 (s), 896 (s); 1H NMR (400 MHz,
CDCl3) dH 9.76 (1H, dd, J = 2.2, 1.9, H-45), 6.44 (1H, s, H-39),
6.37 (1H, d, J = 15.6, H-41), 5.63 (1H, dd, J = 15.6, 7.7, H-42),
4.15 (1H, ddd, J = 7.9, 7.7, 4.8, H-43), 3.29 (3H, s, OCH3), 2.73
(1H, ddd, J = 16.4, 7.9, 2.2, H-44a), 2.56 (1H, ddd, J = 16.4,
4.8, 1.9, H-44b), 1.97 (3H, s, H-47); 13C NMR (100.6 MHz,
CDCl3) dC 200.2, 144.1, 133.9, 128.4, 85.2 (2C), 56.5, 49.3, 20.0;
HRMS (+ESI) calc. for C9H13IO2Na (MNa+) 302.9858, found
302.9859.
6.1, H-43), 3.26 (3H, s, OCH3), 2.37–2.23 (2H, m, H-44a & H-
44b), 1.97 (3H, s, H-47).
Ketone 50
A stirred mixture of CrCl2 (55.1 mg, 0.448 mmol) and NiCl2
(5.8 mg, 45 lmol) was gently heated under vacuum with a heat
gun for 3 min. On cooling to RT, the flask was back-filled with
argon. A degassed solution (three freeze–pump–thaw cycles) of
triene 39 (20.0 mg, 56.0 lmol) in dry THF/4-tert-butylpyridine
(1.4 mL of 6:1, including washings) was added so as to produce
a homogeneous green solution. A solution of aldehyde 31 in dry
DMSO (0.830 mL of 0.135 M, 0.112 mmol) was subsequently
added dropwise before the flask was wrapped in aluminium foil.
A period of stirring for 9 h was followed by the addition of light
petroleum–EtOAc (10 mL of 1:1) and sodium serinate solution
(10 mL of 1 M, 10 mmol). The vigorous stirring of the biphasic
mixture for 40 min without argon protection was followed by
pouring into water (20 mL). The phases were separated and
the aqueous phase back-extracted with EtOAc (2 × 20 mL).
The combined organic phases were washed with water (50 mL)
and then brine (50 mL) before being dried (MgSO4) and
concentrated in vacuo. Flash column chromatography on a silica
gel-filled pipette column (light petroleum–EtOAc gradient,
15:1 → 3:1 → EtOAc) gave the desiodo triene 49 (5.6 mg) and
then an inseparable mixture of C-38 allylic alcohols (6.4 mg,
Triene 39
The DMF used in this reaction had been degassed prior to
use using three freeze–pump–thaw cycles. CrCl2 (930 mg,
7.57 mmol, glovebox), which was being stirred under vacuum,
was gently dried with a heat gun. The CrCl2 was placed
under argon on cooling to RT, and the flask was wrapped in
aluminium foil before being cooled to 0 °C. Dry DMF (13.0 mL)
was added and the flask was removed from the cooling bath.
After 15 min, a solution of aldehyde 48 (212 mg, 0.757 mmol)
1
29%, R/S = 1.0:3.4 by 400 MHz H NMR). To a solution of
this mixture (3.1 mg, 8.0 lmol) in dry CH2Cl2 (1.0 mL) and
activated, crushed 4 Å molecular sieves (30 mg) was added a dry
CH2Cl2 solution (80.0 lL) of TPAP (0.010 M, 0.80 lmol) and
NMO (0.149 M, 11.9 lmol). Complete conversion was achieved
after 2 h stirring at RT. The purification of the mixture using a
silica gel-filled pipette column (light petroleum–EtOAc gradient,
1:1 → 3:1) gave ketone 50 (1.8 mg, 58%) as a colourless oil. Rf
0.18 (light petroleum–EtOAc, 1:1); [a]2D0 = −4.0 (c = 0.13,
CHCl3); mmax (thin film)/cm−1 2918 (s), 1743 (s, OCO), 1676
(m, CO), 1582 (s, CC), 1438 (w), 1352 (w), 1216 (w), 1157
(w), 1094 (s), 971 (w); 1H NMR (500 MHz, CDCl3) dH 6.57 (1H,
s, H-39), 6.34 (1H, d, J = 15.7, H-41), 6.19 (1H, ddd, J = 13.6,
7.2, 7.2, H-45), 6.12 (1H, d, J = 13.6, H-46), 6.04 (1H, dd,
J = 15.7, 7.4, H-42), 4.70 (1H, dd, J = 7.2, 5.4, H-37), 3.78 (1H,
dddd, J = 9.2, 5.4, 5.0, 3.8, H-35), 3.74 (1H, ddd, J = 7.4, 6.7,
6.7, H-43), 3.29 (3H, s, OCH3), 3.27 (3H, s, OCH3), 2.79 (1H,
dd, J = 17.5, 5.0, H-34a), 2.68 (1H, dd, J = 17.5, 5.4, H-34b),
2.38–2.28 (3H, m, H-44a, H-44b & H-36a), 2.28 (3H, s, H-47),
2.24 (1H, ddd, J = 13.5, 9.2, 7.2, H-36b); 13C NMR (125.7 MHz,
CDCl3) dC 197.0, 168.3, 153.4, 137.4, 136.5, 133.2, 121.6, 106.8,
80.8, 80.6, 71.8, 56.8, 56.2, 38.8, 36.8, 30.9, 14.7; HRMS (+ESI)
calc. for C17H23BrO5Na (MNa+) 409.0627, found 409.0635.
n
and Bu3SnCHI2 43 (843 mg, 1.51 mmol) in dry DMF (4.4 mL,
including washings) was added to the green suspension. A
period of vigorous stirring for 60 min was followed by cooling
to 0 °C and quenching with pH 7 buffer solution (5.0 mL). The
mixture was poured into water (50 mL) and extracted with
EtOAc (3 × 50 mL). The combined extracts were washed with
water (3 × 100 mL) and then brine (100 mL) before being dried
(MgSO4) and concentrated in vacuo. An Et2O solution of the
residue was washed through a short plug of neutral, deactivated
alumina. The concentration of the filtrate in vacuo produced
the crude stannane (695 mg) as an ochre oil which consisted
of a chromatographically separable 8.2:1.0 mixture of E/Z
diastereomers at C(45–46). For convenience, the bulk sample
was not purified at this stage. The crude stannane mixture
was dissolved in dry MeCN (18 mL) and cooled to 0 °C. A
solution of NBS (240 mg, 1.35 mmol) in dry MeCN (4.0 mL)
was added dropwise and the resulting solution was stirred for
a further 30 min before Na2S2O3 solution–NaHCO3 solution
(10 mL of 1:1) was slowly added. The solution was removed
from the cooling bath, vigorously stirred for 10 min and poured
into water (50 mL). Extractions with Et2O (3 × 50 mL) were
followed by the washing of the combined extracts with water
(2 × 100 mL) and then brine (100 mL), drying (MgSO4) and
concentration in vacuo. An analysis of the residue by 400 MHz
1H NMR gave an E/Z alkene ratio at C(45–46) of 7.5:1.0.
Flash column chromatography on silica gel doped with 10%
w/w AgNO3 (light petroleum and then light petroleum–EtOAc,
45:1) gave triene 39 and its C(45–46) Z-diastereomer (268 mg,
99% over two steps from aldehyde 48) as a colourless oil which
was devoid of tributyltin residues. After the preparative HPLC
purification {light petroleum–Et3N (1%); Rt = 24.6 min [C(45–
46) Z-diastereomer: 11.8%], Rt = 26.7 min (39: 88.2%); detection
at k = 254 nm} of the sample, the spectral data of triene 39 were
found to be in agreement with those of Evans et al.,5b though
there was a discrepancy in the magnitude of the optical rotation.
Rf 0.14 (light petroleum–EtOAc, 50:1); [a]2D0 = +6.6 (c = 0.62,
Acknowledgements
We thank the EPSRC, Merck Research Laboratories, the Isaac
Newton Trust and Clare College, Cambridge, for financial
assistance and Dr David Wong for his contribution to some of
the experimental work.
References
1 P. A. Searle and T. F. Molinski, J. Am. Chem. Soc., 1995, 117,
8126–8131.
2 P. A. Searle, T. F. Molinski, L. J. Brzezinski and J. W. Leahy, J. Am.
Chem. Soc., 1996, 118, 9422–9423.
3 T. F. Molinski, Tetrahedron Lett., 1996, 37, 7879–7880.
4 C. J. Forsyth, F. Ahmed, R. D. Cink and C. S. Lee, J. Am. Chem.
Soc., 1998, 120, 5597–5598.
5 (a) D. A. Evans and D. M. Fitch, Angew. Chem., Int. Ed., 2000, 39,
2536–2540; (b) D. A. Evans, D. M. Fitch, T. E. Smith and V. J. Cee,
J. Am. Chem. Soc., 2000, 122, 10033–10046.
6 A. B. Smith III, P. R. Verhoest, K. P. Minbiole and M. Schelhaas,
J. Am. Chem. Soc., 2001, 123, 4834–4836; A. B. Smith III,
K. P. Minbiole, P. R. Verhoest and M. Schelhaas, J. Am. Chem. Soc.,
2001, 123, 10942–10953.
1
CH2Cl2) [lit.:5b +11.2 (c = 7.7, CH2Cl2)]; H NMR (400 MHz,
CDCl3) dH 6.41 (1H, s, H-39), 6.30 (1H, d, J = 15.6, H-41), 6.16
(1H, ddd, J = 13.7, 7.0, 7.0, H-45), 6.10 (1H, d, J = 13.7, H-46),
5.57 (1H, dd, J = 15.6, 7.8, H-42), 3.63 (1H, ddd, J = 7.8, 6.1,
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8
3 0 3 7

View Article Online
7 M. A. González and G. Pattenden, Angew. Chem., Int. Ed., 2003, 42,
1255–1258; G. Pattenden, M. A. González, P. B. Little, D. S. Millan,
A. T. Plowright, J. A. Tornos and T. Ye, Org. Biomol. Chem., 2003, 1,
4173–4208.
8 D. R. Williams, A. A. Kiryanov, U. Emde, M. P. Clark, M. A.
Berliner and J. T. Reeves, Angew. Chem., Int. Ed., 2003, 42, 1258–
1262.
reaction: R. T. Ruck and E. N. Jacobsen, Angew. Chem., Int. Ed.,
2003, 42, 4771–4774.
25 Kakisawa–Mosher analysis of the Mosher ester derivatives of the
equatorial alcohol resulting from the NaBH4-mediated reduction of
the undesired tetrahydropyranone 26 confirmed the configurational
assignment of this material and hence that of the desired
tetrahydropyranone 25. See ref. 30.
9 For a review of the total synthesis of the phorboxazoles, see:
L. O. Haustedt, I. V. Hartung and H. M. R. Hoffmann, Angew.
Chem., Int. Ed., 2003, 42, 2711–2716.
26 The global minimum conformer of tetrahydropyranone 25, as estab-
lished by molecular mechanics (MM2* forcefield, MacroModel ver-
sion 8.0), failed to explain the seeming preference that the sterically
bulky LiAl(OtBu)3H reducing agent showed for axial attack.
27 J. Inanaga, K. Hirata, H. Saeki, T. Katsuki and M. Yamaguchi, Bull.
Chem. Soc. Jpn., 1979, 52, 1989–1993.
28 After the completion of this manuscript, the Lindlar reduction of the
C(2–3) triple bond of macrolactone 30 also proved possible.
29 D. A. Evans, J. A. Murry and M. C. Kozlowski, J. Am. Chem.
Soc., 1996, 118, 5814–5815; D. A. Evans, V. J. Cee, T. E.
Smith, D. M. Fitch and P. S. Cho, Angew. Chem., Int. Ed.,
2000, 39, 2533–2536; D. A. Evans, M. C. Kozlowski, J. A. Murry,
C. S. Burgey, K. R. Campos, B. T. Connell and R. J. Staples, J. Am.
Chem. Soc., 1999, 121, 669–685.
10 (a) P. B. Greer and W. A. Donaldson, Tetrahedron, 2002, 58,
6009–6018, and refs. cited therein; (b) B. Liu and W.-S. Zhou, Tetra-
hedron Lett., 2003, 44, 4933–4935; (c) D. R. Li, Y. Q. Tu, G.-Q. Lin
and W.-S. Zhou, Tetrahedron Lett., 2003, 44, 8729–8732; (d) T. M.
Hansen, M. M. Engler and C. J. Forsyth, Bioorg. Med. Chem. Lett.,
2003, 13, 2127–2130; (e) J. S. Yadav and G. Rajaiah, Synlett, 2004,
1537–1539; (f) B. S. Lucas, L. M. Luther and S. D. Burke, Org. Lett.,
2004, 6, 2965–2968.
11 For our previous synthetic studies, see: (a) I. Paterson and
E. A. Arnott, Tetrahedron Lett., 1998, 39, 7185–7188; (b) I. Paterson
and C. A. Luckhurst, Tetrahedron Lett., 2003, 44, 3749–3754.
12 A. B. Smith III, P. R. Verhoest, K. P. Minbiole and J. J. Lim, Org.
Lett., 1999, 1, 909–912.
30 I. Ohtani, T. Kusumi, Y. Kashman and H. Kakisawa, J. Am. Chem.
Soc., 1991, 113, 4092–4096.
13 P. B. Greer and W. A. Donaldson, Tetrahedron Lett., 2000, 41,
3801–3803.
31 M. Sato, J. Sakaki, Y. Sugita, S. Yasuda, H. Sakoda and C. Kaneko,
Tetrahedron, 1991, 47, 5689–5708.
32 R. Baker and J. L. Castro, J. Chem. Soc., Perkin Trans. 1, 1990,
47–65.
33 M. A. Blanchette, W. Choy, J. T. Davis, A. P. Essenfeld,
S. Masamune, W. R. Roush and T. Sakai, Tetrahedron Lett., 1984,
25, 2183–2186.
34 Y. Nagao, Y. Hagiwara, T. Kumagai, M. Ochiai, T. Inoue,
K. Hashimoto and E. Fujita, J. Org. Chem., 1986, 51, 2391–2393.
35 A. González, J. Aiguadé, F. Urpí and J. Vilarrasa, Tetrahedron Lett.,
1996, 37, 8949–8952.
14 (a) A. G. Dossetter,T. F. JamisonandE. N. Jacobsen, Angew.Chem.,
Int. Ed., 1999, 38, 2398–2400; (b) G. D. Joly and E. N. Jacobsen,
Org. Lett., 2002, 4, 1795–1798; (c) K. Gademann, D. E. Chavez and
E. N. Jacobsen, Angew. Chem., Int. Ed., 2002, 41, 3059–3061.
15 For a rare example of conjugate propargyl delivery using an
organolithium, see: E. J. Corey and C. Rücker, Tetrahedron Lett.,
1982, 23, 719–722.
16 For examples of this behaviour with dihydropyrones, see: P. J.
Kocienski, P. Raubo, C. Smith and F. T. Boyle, Synthesis, 1999, 2087;
P. Kocienski, R. Narquizian, P. Raubo, C. Smith, L. J. Farrugia,
K. Muir and F. T. Boyle, J. Chem. Soc., Perkin Trans. 1, 2000, 2357.
17 Unpublished results from C. A. L.
36 D. M. Hodgson, A. M. Foley and P. J. Lovell, Tetrahedron Lett.,
1998, 39, 6419–6420.
37 K. Takai, K. Nitta and K. Utimoto, J. Am. Chem. Soc., 1986, 108,
7408–7410.
18 H. O. House, Acc. Chem. Res., 1976, 9, 59–67.
19 B. H. Lipshutz, E. L. Ellsworth, S. H. Dimock and R. A. J. Smith,
J. Org. Chem., 1989, 54, 4977–4979; B. H. Lipshutz, E. L. Ellsworth,
S. H. Dimock and R. A. J. Smith, J. Am. Chem. Soc., 1990, 112,
4404–4410.
38 For a comprehensive review, see: A. Fürstner, Chem. Rev., 1999, 99,
991–1045.
39 D. P. Stamos, X. C. Sheng, S. S. Chen and Y. Kishi, Tetrahedron
Lett., 1997, 38, 6355–6358.
40 The absolute configuration at C-38 was apparent by a comparison
of the 1H NMR spectrum of the allylic alcohol mixture with that of
compounds of known absolute configuration at this position.
41 S. V. Ley, J. Norman, W. P. Griffith and S. P. Marsden, Synthesis,
1994, 639–666.
42 T. Ando and J. Yamawaki, Chem. Lett., 1979, 45–46.
43 The nBu3SnCHI2 reagent was prepared in almost quantitative
yield from the analogous dibromo compound under Finkelstein
conditions (LiI, acetone). The dibromo compound was prepared
in 89% yield using the procedure of Hodgson and co-workers:
D. M. Hodgson, L. T. Boulton and G. N. Maw, Tetrahedron, 1995,
51, 3713–3724.
20 A. Hosomi and H. Sakurai, J. Am. Chem. Soc., 1977, 99,
1673–1675.
21 D. Grandjean, P. Pale and J. Chuche, Tetrahedron Lett., 1994, 35,
3529–3530.
22 I. Paterson, K.-S. Yeung and J. B. Smaill, Synlett., 1993, 774–776.
23 H. Emde, D. Domsch, H. Feger, U. Frick, A. Götz, H. H. Hergott,
K. Hofmann, W. Kober, K. Krägeloh, T. Oesterle, W. Steppan,
W. West and G. Simchen, Synthesis, 1982, 1–26.
24 However, Schiff base chromium(III) complexes, with similar
structures to 6a and 6b, have been used for hetero-ene reactions,
in conjunction with sub-stoichiometric amounts of the nitrogen-
containing Hünig’s base, with no apparent deleterious effect on the
3 0 3 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 0 2 6 – 3 0 3 8