Synthesis of the anti-HIV agent (-)-hyperolactone C by using oxonium ylide formation-rearrangement

Article abstract of DOI:10.1002/chem.201101082

Starting from readily available (S)-styrene oxide an asymmetric synthesis is described of the naturally occurring anti-HIV spirolactone (-)-hyperolactone C, which possesses adjacent fully substituted stereocenters. The key step involves a stereocontrolled Rh^II-catalysed oxonium ylide formation-[2,3] sigmatropic rearrangement of an α-diazo-β-ketoester bearing allylic ether functionality. From the resulting furanone, an acid-catalysed lactonisation and dehydrogenation gives the natural product.

Full text of DOI:10.1002/chem.201101082

FULL PAPER
DOI: 10.1002/chem.201101082
Synthesis of the Anti-HIV Agent (ꢀ)-Hyperolactone C by Using Oxonium
Ylide Formation–Rearrangement
David M. Hodgson* and Stanislav Man[a]
Abstract: Starting from readily available (S)-styrene oxide an asymmetric synthesis
is described of the naturally occurring anti-HIV spirolactone (ꢀ)-hyperolactone C,
Keywords:
oxonium ylide · rhodium catalysis ·
sigmatropic
total synthesis
hyperolactone
·
which possesses adjacent fully substituted stereocenters. The key step involves a
stereocontrolled RhII-catalysed oxonium ylide formation–[2,3] sigmatropic rear-
rangement of an a-diazo-b-ketoester bearing allylic ether functionality. From the
resulting furanone, an acid-catalysed lactonisation and dehydrogenation gives the
natural product.
rearrangement
·
Introduction
Hyperolactone C (1) is a structurally interesting spirolactone
with adjacent quaternary stereocentres and forms part of a
small family of related lactones, hyperolactones A–D
(Scheme 1) originally isolated from the leaves and stems of
Hypericum chinese L.[1] First reported in 1995, hyperolac-
tone C has attracted increasing interest following the disclo-
sure in 2005 of a new anti-HIV agent, biyouyanagin A (3),
which was postulated to be biosynthetically derived by a
[2+2] cycloaddition from hyperolactone C and ent-zingiber-
ene (2).[2] While a subsequent biomimetic synthesis of
biyouyanagin A by Nicolaou and co-workers led to correc-
tion of the originally proposed stereochemistry of biyouya-
nagin A to that shown in Scheme 1, the synthesis did sup-
Scheme 1. Hyperolactones A–D and hyperolactone C (1) as precursor to
biyouyanagin A (3).
port the original biosynthetic hypothesis.[3] More significant
in the current context was the finding in follow-on struc-
ture–activity studies that the biological activity of biyouya-
nagin A resides in the hyperolactone C structural domain.[3b]
The structural complexity of hyperolactone C and potential
as a new lead in anti-HIV research combine to make it an
attractive target for synthetic studies.
Four of the five syntheses of hyperolactone C so far re-
ported are summarised in Scheme 2. The first synthesis by
Kinoshita in 2001, gave (ꢀ)-1 in 12 steps from malic acid
(4), and also unambiguously established the absolute config-
uration of the natural product.[4] However, introduction of
the second quaternary stereocentre by reaction of the eno-
late of lactone 6 with aldehyde 7 proceeded in favour of the
undesired diastereomer (40:60 d.r.). In 2004, Kraus and Wei
reported a five-step synthesis of rac-1 from methyl acetoace-
tate, which involved O-allylation of furanone 9 followed by
tandem Claisen rearrangement and lactonisation.[5] This is
an elegantly concise synthesis, in which the vicinal stereo-
centres are created with stereocontrol through a chair-like
transition state in the pericyclic step. But the transformation
of furanone 9 into rac-1 unfortunately proceeds in only 11%
yield, partly due to competing C-allylation. Nicolaouꢀs syn-
thesis of (ꢀ)-1, on the route to biyouyanagin A (3), used a
hydroxylactone intermediate 5 from Kinoshitaꢀs earlier syn-
thesis and required 11 steps from malic acid (4).[3] An im-
proved solution to introduction of the second quaternary
stereocentre was achieved by acetylide addition to the keto-
lactone obtained by oxidation of 5, which gave alcohol 8
(75:25 d.r.). Alcohol 8 was used in an impressive key-step
Pd-catalysed cascade sequence to give the full skeleton of
the natural product. Most recently, Xie and co-workers re-
ported in 2009 a six-step catalytic asymmetric synthesis of
(ꢀ)-1.[6] While involving the same disconnection and starting
materials as Kraus and Wei, in this case desired C-allylation
[a] Prof. D. M. Hodgson, Dr. S. Man
Department of Chemistry, Chemistry Research Laboratory
University of Oxford, Mansfield Road, Oxford, OX1 3TA (UK)
Fax : (+44)1865-285002
E-mail: david.hodgson@chem.ox.ac.uk
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101082.
Chem. Eur. J. 2011, 17, 9731 – 9737
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9731

of furanone 9 under Pd catalysis proceeded in high d.r.
(96:4) and excellent e.r. (99.5:0.5), although allylic regiose-
lectivity in the coupling was modest (70:30).
the aim of increasing the efficiency/d.r. (80:20, desired
isomer: minor isomers; the desired isomer is considered to
originate from endo-selective rearrangement on the face of
the ylide away from the phenyl group, as shown in 13). And
finally, to improve on the modest yield at the end of the syn-
thesis, principally believed to due to slow dehydrogenation.
This paper reports our promising results in these areas.
Results and Discussion
Racemic diazo ether 10 was originally prepared from a
Lewis acid-catalysed Mukaiyama-type aldol reaction be-
tween [bisACTHNUTRGENUG(N allyloxy)methyl]benzene and the silyl enol ether
of ethyl diazoacetoacetate,[7] and for which there was no
direct asymmetric version.[7] However, a potentially straight-
forward asymmetric synthesis of diazo ether 10 was consid-
ered that involved allylation of the precursor diazoalcohol
17 (Scheme 4), as this alcohol has previously been shown by
Kundu and Doyle to be available for R=Me in 80% yield
and 95.5:4.5 e.r. by a catalytic asymmetric aldol reaction
using benzaldehyde and the silyl enol ether of methyl a-di-
azoaceotacetate.[8] However, attempted allylation of rac-17
(R=Et)[9] using allyl halides (Cl, Br, I) gave no diazo ether
10 (using K2CO3 as base), was complicated by multiple
products (using NaH, or aq. NaOH/Bu4NI), or gave <50%
conversion with very low reproducibility when using allyl
bromide/Ag2O. Allyl trichloroacetimidate and catalytic
TfOH (or BF3·OEt2) also proved not to be viable, giving
<50% diazo ether 10 which was contaminated with signifi-
cant impurities including the corresponding a,b-unsaturated
ketone from elimination of water.
As alcohol 18 is available in various levels of enantioen-
richment by a variety of approaches,[10] we also considered
its allylation to give ether 19 using allyl trichloroacetimi-
date-catalytic TfOH followed by diazo transfer.[11] In our
hands, this allylation procedure gave an unworkable yield of
45% for ether 19 from rac-18,[12] which led us to consider al-
ternative strategies for the asymmetric synthesis of diazo
ether 10, where the ether was in place in a b-allyloxycarbon-
yl-type compound, prior to homologation to the b-ketoester
functionality.
Scheme 2. Overview of previous syntheses to hyperolactone C (1).
Reported in 2008, our seven-step approach to rac-hypero-
lactone C (1) from ethyl acetoacetate involved one-pot E-
selective cross-metathesis then oxonium ylide formation–
[2,3] sigmatropic rearrangement from diazo ether 10 and
methacrolein (11), followed by chemoselective aldehyde re-
duction, lactonisation and dehydrogenation (Scheme 3).[7] In
critically reviewing this first-generation synthesis, we identi-
fied several important areas for further investigation: First,
was to render the strategy an asymmetric one; secondly, to
develop an alternative to the difficult cross-metathesis using
methacrolein (100 equiv), which likely contributes to the
modest overall yield for the three-step sequence from diazo
ether 10 to lactol 15 by way of oxonium ylide 13. Thirdly, to
analyse and understand in more detail the origins of selec-
tivity in the key sigmatropic rearrangement (13!14), with
In considering the challenge of an asymmetric acetate-like
aldol reaction to directly generate b-alkoxycarbonyls, we
Scheme 3. Hodgsonꢀs 2008 oxonium ylide rearrangement-based synthesis of hyperolactone C (1).
9732
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9731 – 9737

Synthesis of (ꢀ)-Hyperolactone C
FULL PAPER
acetone cyanohydrin (68%),[17] although the corresponding
unsaturated nitrile was identified as a side-product (10–
15%). Using NaCN in EtOH/H2O,[18] gave low yields (<
60%) and purity of hydroxynitrile 23. Allylation of hydroxy-
nitrile 23 was initially complicated by very low yields and re-
producibility. As originally reported, reaction of hydroxyni-
trile 23 with allyl bromide in the presence of K2CO3 in ace-
tone (72 h) provided a moderate yield (60%) of ether nitrile
24,[19] but in our hands much lower yields (~35%) were ob-
tained. Using NaH as base led to a mixture of products con-
taining mainly the unsaturated nitrile from loss of water. Fi-
nally, in the presence of Ag2O and a large excess of allyl
bromide (10 equiv), ether nitrile 24 was obtained in excel-
lent yield (91%). Interestingly, this last transformation re-
quired the presence of trace water for the reaction to pro-
ceed, with no reaction being observed in dry solvents
(CH2Cl2, Et2O, acetone). Formation of ether (R)-19 from
ether nitrile 24 was achieved by a modified Blaise reaction
(81%).[20] The procedure benefited from simple in situ acti-
vation of zinc dust with MeSO3H and required only 2 equiv
of ethyl bromoacetate, which minimised and avoided the
formation of ethyl acetoacetate and diethyl butandioate by-
products, respectively. Diazo transfer[11] with ether (R)-19
gave diazo ether (R)-10 in 92% yield.
Scheme 4. Possible approaches to diazo ether 10.
were attracted to the work of Romea and Urpꢂ, who used
the Lewis acid-induced reaction of acetals with titanium
enolates of 4-substituted N-acetyl-1,3-thiazolidine-2-thio-
nes.[13]
ACHTUNGTRENNUNG
In
the
present
work
using
[bis-
(allyloxy)methyl]benzene[14] as the acetal, then of the 4-sub-
stituted thiazolidinethiones (4-substitutent=Ph, Bn, iPr, Me,
or tBu) and Lewis acids examined,[9] the (R)-phenylglycine-
derived N-acyl thione 20 with BF3·OMe2 proved optimal,
giving a 86:14 d.r. of thione ethers with diastereomerically
pure thione ether 21 being isolated in 74% yield after chro-
matography (Scheme 5). Interestingly, the tert-leucine-de-
rived N-acyl thiazolidinethione (4-substitutent=tBu) was
less diastereoselective (76:24 d.r.) with this acetal, and simi-
larly with benzaldehyde dimethyl acetal.[9] Masamune homo-
logation[15] of thione ether 21 by using Mg(O2CCH2CO2Et)–
imidazole gave ether R-19 in 80% yield with efficient
(87%) recovery of the chiral auxiliary.
Scheme 6. Synthesis of ether R-10 from (S)-styrene oxide (22).
With an efficient asymmetric synthesis of diazo ether (R)-
10 established, we re-evaluated and sought improvements to
the rest of our original racemic synthesis of hyperolactone C
(1) (Scheme 3). The (E)-enal 12 substrate for oxonium ylide
formation–rearrangement had previously been prepared in
21% yield from racemic diazo ether 10 by cross-metathesis
using a large excess of methacrolein.[7] In the present study,
ozonolysis followed by a Wittig olefination were found to
give improved access to (E)-enal (R)-12 (66% from diazo
ether (R)-10, Scheme 7). The intermediate diazoaldehyde 25
was best obtained (88%) after ozonolysis by using 1 equiv
of PPh3 (use of Me2S, or (MeO)3P were less satisfactory);
the tolerance of the diazo group to this chemistry is note-
worthy. High stereoselectivity (E/Z >99:1) was seen in the
Wittig olefination (75%), which was carried out using
1.3 equiv of the ylide 26 in CH2Cl2 (0.1m in diazoaldehyde
25) at RT for 50 h; a higher temperature (THF, reflux) led
to traces (~2%) of Z isomer being detected, a lower con-
Scheme 5. Chiral auxiliary-based synthesis of ether (R)-19.
Although the thiazolidinethione chemistry discussed
above provided access to ether (R)-19 on small-scale, the re-
quirement for careful chromatographic separation of thione
ether 21 from the minor diastereomer led to examination of
a further approach which avoided such a practically limiting
issue. This last approach evaluated cyanide ring-opening of
(S)-styrene oxide (22) followed by allylation and a Blaise re-
action to give ether (R)-19 (Scheme 6). After the optimisa-
tion described below and diazo transfer, this ultimately
became the preferred route to diazo ether (R)-10. Regiose-
lective ring-opening of (S)-styrene oxide[16] to give hydroxy-
nitrile 23 was best achieved via a modified reaction with
Chem. Eur. J. 2011, 17, 9731 – 9737
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9733

D. M. Hodgson and S. Man
centration (0.01m) led to longer reaction time or incomplete
conversion even in refluxing CH2Cl2, whereas higher con-
centration (0.2m) gave a decreased yield of (E)-enal 12
(65%).
and chromatographic purification was required. The 1,2-re-
duction was best effected using a 1:1 mixture of NaBH-
AHCTUNGRTEN(NGUN OAc)3/AcOH in EtOAc at 508C which gave an almost
quantitative yield and high purity of allylic alcohol 27 that
could subsequently be used without further purification,
giving dihydrohyperolactone 16a in 53% isolated yield over
the three steps from (E)-enal 12 (route B). Although the
isolated yield of 16a had improved, the change in 16a–d d.r.
from 79:13:6:2 to 65:3:28:4 in moving from (E)-enal 12 to
allylic alcohol 27 (Table 1, entries 2 and 5) indicated that the
latter displayed a significantly greater tendency to undergo
competing oxonium ylide rearrangement to the undesired
phenyl-bearing face (with high endo selectivity seen on both
faces). While the factors that influence stereoselectivity in
oxonium ylide rearrangements are not well understood, in
the present case temporary capping of the alcohol as its silyl
ether 28 for the ylide formation–rearrangement improved
the proportion of desired dihydrohyperolactone 16a at the
Scheme 7. (E)-Enal (R)-12 by ozonolysis–Wittig olefination of diazo
ether (R)-10.
Access to the pure (E)-enal 12 allowed a more
detailed study of the oxonium ylide formation–rear-
rangement chemistry (Scheme 8 and Table 1). Anal-
ysis was best carried out following reduction of the
intermediate aldehyde 14 and, in an improvement
to the original procedure, subsequent acid-cata-
lysed[6] conversion of the resulting crude lactol mix-
ture 15 to a mixture of dihydrohyperolactones 16a–
d, from which the major desired dihydrohyperolac-
tone isomer 16a could be isolated by chromatogra-
phy. However, this sequence (route A) only led to
16a in 33% yield over the 3 steps from (E)-enal 12,
with the 16a/16b/16c/16d d.r. for the process being
79:13:6:2 (Table 1, entry 2).[9] A closer analysis of
the crude reaction mixture after oxonium ylide re-
arrangement showed significant quantities of side-
products in addition to the diastereomeric mixture
of aldehyde 14; the proportion (~25%) of these
side-products was unaffected on using different cat-
(tfa)4]),[21]
alysts ([Rh2ACHTUNGTRENNUNG(OAc)4], [Rh2ACHTUNRGTEG(NUNN oct)4], or [Rh2ACTHNUGTRENNGUN
but a slight improvement (~3%) was observed if
the reaction temperature was lowered (ꢀ208C). Al-
though these side-products could not be character-
Scheme 8. Routes A–C to dihydrohyperolactones 16a–d from (E)-enal 12.
ised due their lability on silica gel, a major feature in the
crude H NMR indicated the preservation of allylaldehyde
functionality (tm, d=6.60, J=7.3 Hz; =CHCH2) which sug-
gests that [1,2] rearrangements[22] from oxonium ylide 13
might be competing.
expense of 16c (16a/16b/16c/16d 75:3:18:4), giving 16a in
63% yield from (E)-enal 12 (route C, 0.1 mmol scale;
Table 1, entry 8). This latter sequence was used to prepare
16a on 0.6 mmol scale (55% yield) to examine the last de-
hydrogenation step (see below).
1
On the basis that the aldehyde functionality in (E)-enal
12 might be facilitating competing undesired [1,2] shift path-
ways by assisting homolysis in the putative intermediate
oxonium ylide 13, we decided to examine if efficiency in the
oxonium ylide formation-rearrangement was improved by
1,2-reduction of (E)-enal 12 prior to ylide generation–rear-
rangement. Initially, reduction with NaBH3CN/AcOH in
THF at RT led to allylic alcohol 27 (82%) contaminated
with side products likely formed by conjugate reduction,
Variation in the facial selectivity of the rearrangement
due to whether allylic alcohol 27 or silyl ether 28 was em-
ployed led us to examine if such rearrangements also
showed changes depending on ester group size (Table 1). As
already noted above, facial selectivity was reduced with al-
lylic alcohol 27 (67:33) compared with (E)-enal 12 (91:9),
but interestingly the d.r. did vary for the corresponding Me
and tBu esters of 27 (71:3:22:4 and 50:2:44:4, respectively;
Table 1, entries 4 and 6), indicating that face selectivity de-
9734
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9731 – 9737

Synthesis of (ꢀ)-Hyperolactone C
FULL PAPER
Table 1. Results for routes A–C in Scheme 8, and for the corresponding
Experimental Section
Me and tBu ester series.[a]
Entry Route Ester 16a Yield
16a 16b 16c 16d Face selecti-
The synthesis of hyperolactone C (ꢀ)-1 from (S)-styrene oxide (22) is de-
scribed below. Full details of all other experiments discussed above are
given in the Supporting Information, which also includes 1H and 13C spec-
tra for all compounds.
[%][b]
vity[c]
1
2
3
4
5
6
7
8
9
A[d]
A[d]
A[d]
B[e]
B
Me
Et
tBu
Me
Et
tBu
Me
Et
(46)
(33)
(31)
(61)
(53)
(40)
(67)
(63)
(59)
74 19
79 13
78 12
4
6
5
3
2
5
4
4
4
3
4
3
93:7
92:8
(R)-3-Hydroxy-3-phenylpropanenitrile (23): (S)-(ꢀ)-Styrene oxide 97%
(22) (4.00 g, 33.3 mmol, 97% ee) and acetone cyanohydrin[26] (3.40 g,
40 mmol) were dissolved in THF (30 mL). Water (200 mL) and Et3N
(5.6 mL, 4.04 g, 40 mmol) were then added. The mixture was heated
under reflux for 30 h and then concentrated under reduced pressure. The
residue was purified by column chromatography (Et2O/petrol 1:1) to
give, as a yellow oil, hydroxynitrile 23 (3.32 g, 68%). [a]2D5 = +70 (c=0.5
90:10
74:26
68:32
52:48
85:15
78:22
74:26
71
65
50
81
75
72
3
3
2
4
3
2
22
28
44
12
18
23
B
C[e]
C
C
tBu
in CHCl3); lit.[27] [a]D28 =+58 (c=1.1); Rf (Et2O/petrol 1:1)
= 0.21;
[a] Ratio of 16a–d determined by crude 1H NMR analysis.[9] [b] Isolated
yield, and based on (E)-enal 12 (or corresponding E-enal of Me/tBu
esters). [c] Selectivity for rearrangement occurring on opposite vs same
face of ylide (cf. 13) as Ph group. [d] Oxonium ylide formation–rear-
rangement carried out at ꢀ208C (24 h). [e] Same d.r. also observed at
ꢀ208C.
1H NMR (400 MHz, CDCl3): d=7.46–7.30 (m, 1H; Ar), 4.96 (t, J=
6.2 Hz, 1H; O-CH), 3.21 (brs, 1H; OH), 2.70 ppm (d, J=6.2 Hz, 2H;
CH2); 13C NMR (101 MHz, CDCl3): d=140.9 (C, quat), 128.7 (2ꢃAr),
ꢁ
128.5 (Ar), 125.4 (2ꢃAr), 117.5 (C N), 69.6 (OCH), 27.7 ppm (CH2); IR
(neat): n˜ = 702, 757, 1028, 1057, 1087, 1220, 1309, 1412, 1455, 1495, 2255
(C N), 2900, 2930, 2969, 3033, 3065, 3089, 3443 cmꢀ1; LRMS (ESI+):
ꢁ
m/z: calcd for C9H9NNaO: 170.06; found: 170.07 [M+Na+].
(R)-3-Allyloxy-3-phenylpropanenitrile (24): Hydroxynitrile 23 (3.31 g,
22.5 mmol) was dissolved in CH2Cl2 (56 mL). Silver(I) oxide (5.86 g,
25.3 mmol) and allyl bromide (19.5 mL, 27.2 g, 22.50 mmol) were added.
The reaction mixture was stirred for 7 d at room temperature in the dark,
then filtered and concentrated under reduced pressure. The residue was
purified by column chromatography (CH2Cl2/petrol 2:3) to give, as a
light yellow oil, ether nitrile 24[19] (3.83 g, 91%) and (CH2Cl2/Et2O 9:1)
recovered starting alcohol 23 (0.21 g, 6%). Rf (EtOAc/petrol 1:9) = 0.37;
Rf (CH2Cl2) = 0.51; [a]D25 =+94 (c=0.5 in CHCl3); 1H NMR (400 MHz,
CDCl3): d=7.50–7.31 (m, 5H; Ar), 5.99–5.82 (m, 1H; CH=CH2), 5.29
(dd, J=17.3, 0.9 Hz, 1H; O-CHaHb), 5.22 (d, J=10.4 Hz, 1H; O-CHaHb),
4.64 (dd, J=7.2, 5.7 Hz, 1H; Ar-CH), 4.01 (dd, J=12.8, 4.9 Hz, 1H; =
CHaHb), 3.85 (dd, J=12.8, 6.2 Hz, 1H; =CHaHb), 2.80 (dd, J=16.7,
7.3 Hz, 1H; NC-CHaHb), 2.71 ppm (dd, J=16.7, 5.6 Hz, 1H; NC-
CHaHb); 13C NMR (101 MHz, CDCl3): d=138.8 (C, quat.), 133.8 (=CH),
creased with increasing ester group size (74:26 (Me ester),
68:32 (Et ester), 52:48 (tBu ester). The same trend, but less
pronounced, was observed in the Me/Et/tBu ester seriesꢀ
corresponding to (E)-enal 12, and silyl ether 28.[23]
Dehydrogenation of dihydrohyperolactone 16 to hypero-
lactone C (1) had originally been achieved in poor conver-
sion with DDQ under extended (6 d) reflux in benzene
(Scheme 3),[7] and no improvement was seen in toluene. Ac-
tivation of dihydrohyperolactone 16a as the silyl enol ether
29 followed by addition of DDQ at RT,[24] proved a milder
and more effective last step in the synthesis, giving hypero-
lactone C (ꢀ)-1 in 81% yield (Scheme 9), with data consis-
tent with the literature.[25]
ꢁ
128.9 (2ꢃAr), 128.8 (Ar), 126.3 (2ꢃAr), 117.6 (=CH2), 117.1 (C N), 76.2
(OCH), 69.8 (OCH2), 27.0 ppm (NC-CH2); IR (neat): n˜ = 702, 853, 930,
ꢁ
994, 1024, 1074, 1091, 1135, 1422, 1455, 1494, 1647, 2252 (C N), 2867,
2984, 3032, 3066 cmꢀ1; LRMS (ESI+): m/z: calcd for: C12H13NNaO:
210.09; found: 210.09 [M+Na+].
Ethyl (R)-5-(allyloxy)-3-oxo-5-phenylpentanoate (19): Zn powder (1.96 g,
30 mmol, dust <10 mm, 98+ %) was added to a solution of MeSO3H
(43 mg, 0.45 mmol) in THF (30 mL) at room temperature. After heating
under reflux for 10 min, a solution of ether nitrile 24 (18.7 g, 10 mmol) in
THF (10 mL) was added in one portion. Then ethyl bromoacetate
(2.20 mL, 3.32 g, 20 mmol) was added dropwise over 60 min. After heat-
ing under reflux for one additional hour, an aq. solution of HCl (50 mL,
2m) was added at room temperature and the mixture was stirred for
90 min. The mixture was extracted with CH2Cl2 (5ꢃ50 mL). The com-
bined extracts were washed with saturated aq. NaHCO3 (30 mL) and
brine (30 mL), dried (Na2SO4) and concentrated. The residue was puri-
fied by column chromatography (EtOAc/petrol 1:9) to give, as a light
yellow oil, ether 17 (2.23 g, 81%). [a]2D5 =+76 (c=0.6 in CHCl3). Rf
Scheme 9. Completion of synthesis of hyperolactone C (ꢀ)-1.
Conclusion
In summary, we have completed an asymmetric synthesis of
natural hyperolactone C (ꢀ)-1 in 10 steps starting from (S)-
styrene oxide. Key features include the use of the Blaise re-
action to construct a diazo ether; subsequent chemoselective
ozonolysis, in the presence of diazo functionality, is followed
by Wittig olefination to give an E-enal. The E-enal is used
in an improved oxonium ylide rearrangement, where factors
influencing reaction efficiency and stereoselectivity have
been identified. Conversion of the E-enal through a stream-
lined process involving 1,2-reduction, RhII-catalysed oxoni-
um ylide formation–rearrangement and acid-catalysed lacto-
nisation gives dihydrohyperolactone C in 53% yield involv-
ing no intermediate purifications. Finally, a mild and effi-
cient DDQ oxidation proctol gives the natural product.
1
(EtOAc/petrol 1:9) = 0.22; H NMR (400 MHz, CDCl3): major tautomer
d=7.28–7.41 (m, 5H; Ar), 5.93–5.80 (m, 1H; CH=CHaCHb), 5.21 (dq,
J=17.2, 1.6 Hz, 1H; CH=CHaCHb), 5.15 (dq, J=10.4, 1.4 Hz, 1H; CH=
CHaCHb), 4.83 (dd, J=9.0, 4.1 Hz, 1H; CHO), 4.18 (q, J=7.2 Hz, 2H;
CH3CH2), 3.89 (ddt, J=12.6, 5.3, 1.4 Hz, 1H; CHaCHbO), 3.78 (ddt, J=
12.6, 6.1, 1.4 Hz, 1H; CHaCHbO), 3.49 (s, 2H; CH2COOEt), 3.11 (dd, J=
15.9, 9.0 Hz, 1H; CHaCHbC=O), 2.73 (dd, J=15.9, 4.1 Hz, 1H;
CHaCHbC=O), 1.27 ppm (t, J=7.2 Hz, 3H; CH3), minor tautomer (~
8%) d=4.99 (s, 1H), 4.73 (dd, J=8.5, 5.1 Hz, 1 H; O-CH), 2.50 ppm (dd,
J=14.3, 5.1 Hz, 1H); 13C NMR (101 MHz, CDCl3): major tautomer d=
200.7 (C=O), 166.9 (O-C=O), 140.8 (Ar, quat), 134.4 (=CH), 128.6 (2ꢃ
Ar), 128.0 (Ar), 126.5 (2ꢃAr), 117.0 (=CH2), 77.1 (OCH), 69.6 (=C-CH2-
O), 61.3 (CH2CH3), 51.1 (CH2), 50.3 (CH2), 14.0 ppm (CH3); IR (neat): n˜
=702, 759, 927, 1030, 1135, 1193, 1238, 1319, 1368, 1454, 1648, 1719,
Chem. Eur. J. 2011, 17, 9731 – 9737
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9735

D. M. Hodgson and S. Man
1746, 2872, 2906, 2935, 2983, 3031, 3065 cmꢀ1; HRMS (ESI+): m/z: calcd
for C16H20NaO4: 299.1254; found: 299.1253 [M+Na+].
solution of (E)-enal 12 (228 mg, 0.66 mmol) and glacial AcOH (0.19 mL,
200 mg, 3.5 mmol) in EtOAc (18 mL). After heating for 5.5 h at ~508C,
water (18 mL) was added and the organic layer separated. The water
layer was extracted with EtOAc (3ꢃ18 mL). The combined organic
layers were washed with sat. aq. NaHCO3 (3ꢃ14 mL) and brine (1ꢃ
14 mL). The solution was dried (Na2SO4) and concentrated under re-
duced pressure to give, as a pale yellow oil, allylic alcohol 27 (219 mg,
Ethyl (R)-5-(allyloxy)-2-diazo-3-oxo-5-phenylpentanoate (10): Et3N
(1.17 mL, 0.846 g, 8.36 mmol) was added to a solution of ether 17 (2.20 g,
7.96 mmol) and p-acetamidobenzensulfonylazide (2.0 g, 8.33 mmol) in
dry CH2Cl2 (32 mL). After stirring for 24 h at room temperature, the mix-
ture was concentrated under reduce pressure. The residue was triturated
with a mixture of Et2O/petrol 1:1 (4ꢃ30 mL). The combined solutions
were concentrated under reduced pressure and the residue was purified
by column chromatography (Et2O/petrol 1:9) to give, as a light yellow
oil, diazo ether 10[7] (2.21 g, 92%). [a]d25 =+10 (c=0.5 in CHCl3). Rf
(Et2O/petrol 1:9) = 0.18; 1H NMR (400 MHz, CDCl3): d=7.41–7.28 (m,
5H; Ar), 5.93–5.80 (m, 1H; CH=CHaHb), 5.21 (dq, J=17.2, 1.7 Hz, 1H;
=CHaHb), 5.14 (dq, J=10.4, 1.4 Hz, 1H; =CHaHb), 4.95 (dd, J=9.0,
4.2 Hz, 1H; OCH), 4.29 (q, J=7.1 Hz, 2H; CH2CH3), 3.91 (ddt, J=12.7,
5.1, 1.5 Hz, 1H; CHaCHbO), 3.80 (ddt, J=12.7, 6.0, 1.4 Hz, 1H;
CHaCHbO), 3.58 (dd, J=16.3, 9.0 Hz, 1H; CHaCHbC=O), 3.01 (dd, J=
16.3, 4.2 Hz, 1H; CHaCHbC=O), 1.32 ppm (t, J=7.1 Hz, 3H; CH3);
13C NMR (101 MHz, CDCl3): d=189.9 (C=O), 161.1 (O=C-O), 141.2
(Ar, quat.), 134.7 (=CH), 128.4 (2ꢃAr), 127.8 (Ar), 126.7 (2ꢃAr), 116.7
(=CH2), 76.7 (OCH), 69.6 (OCH2), 61.4 (CH2CH3), 48.1 (O=CCH2),
14.3 ppm (CH3); IR (neat): n˜ = 702, 925, 1022, 1089, 1131, 1206, 1307,
1373, 1454, 1657, 1718, 2135, 2861, 2910, 2983, 3031, 3065 cmꢀ1; LRMS
(ESI+): m/z: calcd for C16H18N2NaO4: 325.12; found: 325.10 [M+Na+].
96%). Rf (Et2O)
=
0.50; [a]2D5 =+20 (c=0.5 in CHCl3); 1H NMR
(400 MHz, CDCl3): d=7.42–7.27 (m, 5H; Ar), 5.62–5.54 (m, 1H; =CH),
4.91 (dd, J=9.0, 4.1 Hz, 1H; O-CH), 4.28 (q, J=7.2 Hz, 2H; CH2CH3),
4.00 (s, 2H; CH2OH), 3.97–3.86 (m, 2H; =CHCH2), 3.56 (dd, J=16.6,
9.0 Hz, 1H; OCHCHaHb), 2.99 (dd, J=16.6, 4.1 Hz, 1H; OCHCHaHb),
1.57 (s, 3H; =CCH3), 1.31 ppm (t, J=7.2 Hz, 3H; CH2CH3); 13C NMR
(101 MHz, CDCl3): d=189.8 (C=O), 161.0 (O=C-O), 141.2 (C, quat.),
139.2 (C, quat.), 128.4 (2ꢃAr), 127.7 (Ar), 126.7 (2ꢃAr), 121.3 (=CH),
76.6 (OCH), 67.8 (CH2OH), 64.7 (=CHCH2), 61.3 (OCH2CH3), 48.1 (O=
CCH2), 14.2 (OCH2CH3), 13.7 ppm (=CCH3); IR (neat): n˜ = 1017 (s),
1073 (s), 1133 (m), 1206 (s), 1307 (s), 1374 (s), 1454 (w), 1653 (s), 1716
(s), 2136 (s), 2771 (w), 2921 (m), 3031 (w), 3062 (w), 3086 (w), 3432 cmꢀ1
(br); HRMS (ESI+): m/z: calcd for C18H22N2NaO5: 369.1421; found:
369.1422 [M+Na+].
Ethyl (R)-2-diazo-5-[(2E)-3-methyl-4-(trimethylsilyloxy)but-2-enyloxy]-3-
oxo-5-phenylpentanoate (28): TMSCN (125 mL, 99 mg, 1 mmol) was
added to a solution of allylic alcohol 27 (219 mg, ~0.63 mmol) in CH2Cl2
(1 mL) at room temperature. After stirring for 60 min, the reaction mix-
ture was concentrated under reduced pressure (CAUTION: toxic gas li-
Ethyl (R)-2-diazo-3-oxo-5-(2-oxoethoxy)-5-phenylpentanoate (25):
A
stream of ozone was bubbled through a solution of diazo ether 10
(302 mg, 1 mmol) in CH2Cl2 (33 mL) at ꢀ788C. When the mixture turned
blue, excess ozone was removed by a stream of oxygen then nitrogen.
Ph3P (262 mg, 1 mmol) was added at ꢀ788C, and after stirring for ~20 h
at room temperature, the solution was concentrated under reduced pres-
sure. The residue was purified by column chromatography; gradient elu-
tion Et2O/petrol 2:3 ! Et2O. Diazoaldehyde 25 was obtained as a light
yellow oil (267 mg, 88%). Rf (Et2O/petrol 2:3) = 0.07; Rf (Et2O) = 0.49;
[a]2D5 =+27 (c=0.5 in CHCl3); 1H NMR (400 MHz, CDCl3): d=9.67 (t,
J=1.0 Hz, 1H; O=CH), 7.46–7.28 (m, 5H; Ar), 4.99 (dd, J=9.1, 3.8 Hz,
1H; O-CH), 4.31 (q, J=7.1 Hz, 2H; CH2CH3), 3.95 (t, J=1.0 Hz, 2H;
O=CHCH2), 3.68 (dd, J=16.6, 9.1 Hz, 1H; O-CHCHaHb), 3.07 (dd, J=
16.6, 3.8 Hz, 1H; O-CHCHaHb), 1.33 ppm (t, J=7.1 Hz, 3H; CH2CH3);
13C NMR (101 MHz, CDCl3): d=200.8 (O=CH), 189.5 (C=O), 161.1 (O=
C-O), 139.6 (Ar, quat.), 128.7 (2ꢃAr), 128.4 (Ar), 126.8 (2ꢃAr), 78.4 (O-
CH), 74.2 (O=CHCH2), 61.4 (CH2CH3), 47.8 (O-CHCH2), 14.2 ppm
(CH2CH3); IR (neat): n˜ = 702 (m), 729 (s), 1022 (s), 1133 (s), 1207 (s),
1309 (s), 1374 (s), 1653 (s), 1715 (s), 2137 (s), 2826 (w), 2909 (w), 2935
(w), 2984 (w), 3032 (w), 3063 cmꢀ1 (w); HRMS (EI/EF+): m/z: calcd for
C15H17N2O5: 305.1132; found: 305.1179 [M+H+].
berated), to give, as a pale yellow oil silyl ether 28 (256 mg, 97%). [a]D25
=
+20 (c=0.5 in CHCl3); 1H NMR (400 MHz, CDCl3): d=7.41–7.25 (m,
5H; Ar), 5.58 (tm, J=6.6 Hz, 1H; =CH), 4.91 (dd, J=9.1, 4.3 Hz, 1H;
O-CH), 4.28 (q, J=7.1 Hz, 2H; CH2CH3), 3.98 (s, 2H; CH2OSi), 3.95–
3.85 (m, 2H; =CHCH2), 3.53 (dd, J=16.3, 9.0 Hz, 1H; OCHCHaHb),
3.02 (dd, J=16.4, 4.3 Hz, 1H; OCHCHaHb), 1.51 (s, 3H; =CHCH3), 1.31
(t, J=7.1 Hz, 3H; CH2CH3), 0.12 ppm (s, 9H; SiCH3); 13C NMR
(101 MHz, CDCl3): d=189.9 (C=O), 161.2, (O=C-O), 141.4 (C, quat.),
138.3 (C, quat.), 128.4 (2ꢃAr), 127.7 (Ar), 126.8 (2ꢃAr), 121.2 (=CH),
76.9 (OCH), 67.7 (CH2OSi), 65.0 (=CHCH2), 61.4 (OCH2CH3), 48.2 (O=
CCH2), 14.3 (OCH2CH3), 13.6 (=CCH3), ꢀ0.5 ppm (SiCH3); IR (neat): n˜
= 843 (m), 877 (m), 1019 (w), 1064 (m), 1206 (w), 1251 (w), 1307 (s),
1373 (m), 1660 (m), 1719 (s), 2135 (s), 2856 (w), 2927 (w), 2958 cmꢀ1 (w);
HRMS (ESI+): m/z: calcd for C21H30N2NaO5Si: 441.1816; found:
441.1815 [M+Na+].
A
ACHTUNGTREN[NUGN 4.4]nonane-4,6-
A
CHTUNGTRENNUNG
tion to a solution of silyl ether 28 (256 mg, 0.61 mmol) in dry CH2Cl2
(66 mL) at room temperature. After stirring for 2 h, MeSO3H (44 mL,
65 mg, 0.68 mmol) was added. After stirring overnight at room tempera-
ture, the reaction mixture was filtered through pad of celite and concen-
trated under reduced pressure. The residue was purified by column chro-
matography (EtOAc/petrol 1:9). First eluted a yellow oil, an 86:14 mix-
ture of 16c and 16d (27 mg, 15%);[9] Rf (Et2O/petrol 1:4) = 0.25. Second
eluted a yellow oil, dihydrohyperolactone C 16a (99 mg, 60%, 55% from
(E)-enal 12). Rf (EtOAc/petrol 1:9) = 0.16; Rf (Et2O/petrol 1:4) = 0.19;
[a]D25 =ꢀ59 (c=0.5 in CHCl3); lit.[3b] [a]D25 =ꢀ59 (c=0.52 in CHCl3);[25]
1H NMR (400 MHz, CDCl3): d=7.58–7.51 (m, 2H; Ar), 7.46–7.32 (m,
3H; Ar), 6.00 (dd, J=17.4, 10.9 Hz, 1H; CH=CH2), 5.43–5.38 (m, 2H; =
CH2), 5.35 (dd, J=10.0, 6.6 Hz, 1H; O-CH), 4.67 (d, J=8.6 Hz, 1H; O-
CHaHb), 4.03 (d, J=8.6 Hz, 1H; O-CHaHb), 2.95 (dd, J=18.7, 6.6 Hz,
1H; O=C-CHaHb), 2.68 (dd, J=18.7, 10.0 Hz, 1H; O=C-CHaHb),
1.39 ppm (s, 3H; CH3); 13C NMR (101 MHz, CDCl3): d=208.4 (C=O),
170.6 (O-C=O), 139.8 (Ar, quat.), 135.7 (CH=CH2), 128.9 (2ꢃAr), 128.7
(Ar), 126.2 (2ꢃAr), 119.4 (=CH2), 90.6 (C(O)CC(O2), quat.), 78.6
(ArCH), 73.7 (OCH2), 48.6 (CCH3), 45.0 (O=CCH2), 18.7 ppm (CH3);
Ethyl (R)-2-diazo-5-[(2E)-3-methyl-4-oxobut-2-enyloxy]-3-oxo-5-phenyl-
pentanoate (12): A solution of diazoaldehyde 25 (267 mg, 0.88 mmol)
and ylide 26 (363 mg, 1.14 mmol) in CH2Cl2 (8.8 mL) was stirred for 50 h
at room temperature. The reaction mixture was concentrated under re-
duced pressure and the residue purified by column chromatography
(Et2O/petrol 3:7) to give, as a light yellow oil, (E)-enal 12[7] (228 mg,
75%). Rf (Et2O/petrol 4:6)
=
0.39; [a]2D5 =17 (c=0.5 in CHCl3);
1H NMR (400 MHz, CDCl3): d=9.41 (s, 1H; CH=O), 7.42–7.28 (m, 5H;
Ar), 6.57 (tq, J=5.6, 1.4 Hz, 1H; =CH), 4.96 (dd, J=9.4, 3.8 Hz, 1H; O-
CH), 4.30 (q, J=7.2 Hz, 2H; O-CH2CH3), 4.22–4.17 (m, 2H; OCH2CH=
), 3.64 (dd, J=16.8, 9.4 Hz, 1H; O-CHCHaCHb), 3.00 (dd, J=16.8,
3.8 Hz, 1H; O-CHCHaCHb), 1.62 (m, J=1.2 Hz, 3H; =CCH3), 1.33 ppm
(t, J=7.2 Hz, 3H; O-CH2CH3); 13C NMR (101 MHz, CDCl3): d=194.5
(O=CH), 189.7 (C=O), 161.1 (O-C=O), 149.8 (=CH), 140.4 (C, quat.),
139.1 (C, quat.), 128.7 (2ꢃAr), 128.2 (Ar), 126.8 (2ꢃAr), 77.7 (O-CH),
65.5 (O-CH2CH=), 61.5 (O-CH2CH3), 48.0 (O-CHCH2), 14.3 (O-
CH2CH3), 9.4 ppm (=CCH3); IR (neat): n˜ = 702 (m), 746 (w), 1020 (m),
1095 (m), 1205 (m), 1308 (s), 1374 (m), 1453 (w), 1655 (s), 1691 (s), 1716
(s), 2137 (s), 2714 (w), 2834 (w), 2910 (w), 2984 (w), 3031 (w), 3063 cmꢀ1
(w); LRMS (ESI+): m/z: calcd for C18H20N2NaO5: 367.13; found: 367.11
[M+Na+].
IRACTHNUTRGNE(NUG neat): n˜ =700 (m), 732 (w), 767 (w), 936 (w), 1007 (m), 1056 (w),
1106 (s), 1171 (m), 1272 (w), 1368 (w), 1455 (w), 1751 (s), 1961 (s), 2918
(w), 2974 cmꢀ1 (w); LRMS (ESI+): m/z: calcd for C16H16NaO4: 295.09;
found: 295.10 [M+Na+]. Last eluted a yellow oil, an 73:27 mixture of
16b and 16a (6 mg, 4%);[9] Rf (Et2O/petrol 1:4) = 0.13.
Ethyl
(R)-2-diazo-5-[(2E)-3-methyl-4-hydroxybut-2-enyloxy]-3-oxo-5-
(OAc)3 (0.70 g, 3.3 mmol) was added to a
phenylpentanoate (27): NaBHAHCNUTGTRENNUNG
9736
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9731 – 9737

Synthesis of (ꢀ)-Hyperolactone C
FULL PAPER
A
E
[7] D. M. Hodgson, D. Angrish, S. P. Erickson, J. Kloesges, C. H. Lee,
Org. Lett. 2008, 10, 5553–5556.
(1) (hyperolactone C): Et3N (21 mL, 15 mg, 0.15 mmol, freshly distilled
from CaH2) and TMSOTf (23 mL, 28 mg, 0.125 mmol) were successively
added to a solution of dihydrohyperolactone C 16a (27 mg, 0.1 mmol) in
CH2Cl2 (4.5 mL) at room temperature. After stirring for 1.5 h, DDQ
(45 mg, 0.2 mmol) was added in one portion. The reaction mixture was
stirred overnight, then concentrated under reduced pressure. The residue
was purified by column chromatography (Et2O/petrol 3:7) to give, as a
pale yellow solid, hyperolactone C (1) (22 mg, 81%). M.p. 106–1108C
(lit.[1a] m.p. 1048C), rac m.p. 114–1208C (lit.[7] 101–1028C); Rf (Et2O/
petrol 3:7) = 0.14; Rf (Et2O/petrol 4:6) = 0.25; [a]2D5 =ꢀ257 (c=0.02 in
CHCl3); lit.[3b] [a]2D5 =ꢀ273 (c=0.69 in CHCl3); lit.[6] [a]2D0 =ꢀ237 (c=0.10
in CHCl3); lit.[1b] [a]D22 =ꢀ271, (c=0.11 in CHCl3); [a]2D5 =ꢀ264 (c=0.01
in EtOH); lit.[3b] [a]D =ꢀ356 (c=0.02 in EtOH); 1H NMR (500 MHz,
CDCl3): d=7.89–7.80 (m, 2H; Ar), 7.65–7.57 (m, 1H; Ar), 7.56–7.46 (m,
2H; Ar), 5.99 (dd, J=17.6, 10.9 Hz, 1H; CH=CH2), 5.99 (s, 1H; Ar-C=
CH), 5.27 (d, J=10.9 Hz, 1H; CH=CHaHb), 5.26 (d, J=17.6 Hz, 1H;
CH=CHaHb), 4.96 (d, J=8.6 Hz, 1H; O-CHaHb), 4.12 (d, J=8.6 Hz, 1H;
O-CHaHb), 1.53 ppm (s, 3H; CH3); 13C NMR (101 MHz, CDCl3): d=
196.5 (C=O), 187.2 (Ar-C=CH), 168.1 (CO2), 134.2 (CH=CH2), 133.6
(Ar), 129.0 (2ꢃAr), 127.7 (Ar, quat.), 127.4 (2ꢃAr), 119.1 (CH=CH2),
100.2 (=CH-C=O), 93.1 (C(O)CC(O2), quat.), 74.1 (O-CH2), 48.8 ppm
(CCH3, quat.), 19.5 (CH3); IR (KBr): n˜ = 775 (m), 870 (m), 1101 (s),
1345 (m), 1607 (m), 1706 (s), 1781 (s), 1919 (w), 2986 (w), 3004 (w), 3025
(w), 3116 cmꢀ1 (w); LRMS (ESI+): m/z: calcd for C16H14NaO4: 293.08;
found: 293.06 [M+Na+].
[8] K. Kundu, M. P. Doyle, Tetrahedron: Asymmetry 2006, 17, 574–577.
[9] See the Supporting Information for details.
[10] a) S. Kiyooka, M. A. Hena, J. Org. Chem. 1999, 64, 5511–5523;
b) R. Villano, M. De Rosa, C. Salerno, A. Soriente, A. Scettri, Tetra-
hedron: Asymmetry 2002, 13, 1949–1952; c) M. Hayashi, T. Inoue,
Y. Miyamoto, N. Oguni, Tetrahedron 1994, 50, 4385–4398; d) M.
Hayashi, K. Yoshimoto, N. Hirata, K. Tanaka, N. Oguni, K. Harada,
A. Matsushita, Y. Kawachi, H. Sasaki, Isr. J. Chem. 2001, 41, 241–
246; e) K. Narkunan, B.-J. Uang, Synthesis 1998, 1713–1714.
[11] T. Yakura, K. Matsui, K. Matsuzaka, M. Yamashita, Heterocycles
2009, 79, 353–358. In this paper, both allylation, using allyl trichlor-
oacetimidate, and diazo transfer have been reported on the corre-
sponding Me ester of 18; yields of ~50% and >90%, respectively
were indicated in correspondence from Professor Yakura.
[12] C. Xu, C. Yuan, Tetrahedron 2005, 61, 2169–2186.
[13] a) A. Cosp, P. Romea, F. Urpꢂ, J. Vilarrasa, Tetrahedron Lett. 2001,
42, 4629–4631; b) A. Cosp, I. Larrosa, I. Vilasꢂs, P. Romea, F. Urpꢂ,
J. Vilarrasa, Synlett 2003, 1109–1112; c) J. Baiget, A. Cosp, E.
Gꢅlvez, L. Gꢆmez-Pinal, P. Romea, F. Urpꢂ, Tetrahedron 2008, 64,
5637–5644; d) E. Gꢅlvez, P. Romea, F. Urpꢂ, Org. Synth. 2009, 86,
81–91.
[14] J. B. Brogan, J. E. Richard, C. K. Zercher, Synth. Commun. 1995, 25,
587–593.
[15] D. W. Brooks, L. D.-L. Lu, S. Masamune, Angew. Chem. 1979, 91,
76–77; Angew. Chem. Int. Ed. Engl. 1979, 18, 72–74.
[16] Styrene oxide is readily (commercially) available in high e.r. by Ja-
cobsen hydrolytic kinetic resolution: S. E. Schaus, B. D. Brandes,
J. F. Larrow, M. Tokunaga, K. B. Hansen, A. E. Gould, M. E.
Furrow, E. N. Jacobsen, J. Am. Chem. Soc. 2002, 124, 1307–1315.
[17] a) D. Mitchell, T. M. Koenig, Tetrahedron Lett. 1992, 33, 3281–3284;
b) D. Mitchell, T. M. Koenig, Synth. Commun. 1995, 25, 1231–1238.
[18] A. Kamal, G. B. R. Khanna, R. Ramu, Tetrahedron: Asymmetry
2002, 13, 2039–2051.
Acknowledgements
We thank the European Union for a Marie Curie Fellowship (PIEF-GA-
2009-237270 to S.M.), and Professor T. Yakura (Toyama University) for
some details on the preparation of the methyl esters of 10 and 19.
[19] D.-Y. Ma, D.-X. Wang, J. Pan, Z.-T. Huang, M.-X. Wang, Tetrahe-
dron: Asymmetry 2008, 19, 322–329.
[20] H. Shin, B. S. Choi, K. K. Lee, H.-W. Choi, J. H. Chang, K. W. Lee,
D. H. Nam, N.-S. Kim, Synthesis 2004, 2629–2632.
[21] Catalyst dependent d.r. has been observed in the oxonium ylide for-
mation-[2,3] sigmatropic rearrangement of the corresponding Me
ester of diazo ether 10, see ref. [11].
[22] Nitrogen, Oxygen and Sulfur Ylide Chemistry (Ed.: J. S. Clark),
Oxford University Press, Oxford, 2002, pp. 1–113.
[1] a) Y. Aramaki, K. Chiba, M. Tada, Phytochemistry 1995, 38, 1419–
1421; b) S. L. Crockett, W. Schꢄhly, F. Belaj, I. A. Khan, Acta Crys-
tallogr. Sect. E 2004, 60, o2174–o2176; c) W. Schꢄhly, S. L. Crockett,
W. M. F. Fabian, Chirality 2005, 17, 250–256.
[2] N. Tanaka, M. Okasaka, Y. Ishimaru, Y. Takaishi, M. Sato, M. Oka-
moto, T. Oshikawa, S. U. Ahmed, L. M. Consentino, K.-H. Lee, Org.
Lett. 2005, 7, 2997–2999.
[3] a) K. C. Nicolaou, D. Sarlah, D. M. Shaw, Angew. Chem. 2007, 119,
4792–4795; Angew. Chem. Int. Ed. 2007, 46, 4708–4711; b) K. C.
Nicolaou, T. R. Wu, D. Sarlah, D. M. Shaw, E. Rowcliffe, D. R.
Burton, J. Am. Chem. Soc. 2008, 130, 11114–11121; see also:
c) K. C. Nicolaou, S. Sanchini, T. R. Wu, D. Sarlah, Chem. Eur. J.
2010, 16, 7678–7682; d) K. C. Nicolaou, S. Sanchini, D. Sarlah, G.
Lu, T. R. Wu, D. K. Nomura, B. F. Cravatt, B. Cubit, J. C de La
Torre, D. R. Burton, Proc. Natl. Acad. Sci. USA 2011, 108, 6715–
6720.
[23] Little variation in d.r. values are also observed for oxonium ylide
formation–rearrangement of Me/Et/tBu esters corresponding to
diazo ether 10 (91:9,[11] 90:10,[7] and 87:13, respectively).
[24] a) I. Ryu, S. Murai, Y. Hatayama, N. Sonoda, Tetrahedron Lett.
1978, 19, 3455–3458; b) C.-L. Chin, D. D.-P. Tran, K.-S. Shia, H.-J.
Liu, Synlett 2005, 417–420.
[25] Conversion of 16a to (ꢀ)-1 and comparison of the data (in particu-
1
lar H NMR: d CH2O) for 16a–d with the dihydrohyperolactone re-
ported in ref. [3b], indicates that the configuration drawn at the Ph-
bearing centre in the latter should be corrected by inversion.
[4] T. Ueki, M. Doe, R. Tanaka, Y. Morimoto, K. Yoshihara, T, Ki-
noshita, J. Heterocycl. Chem. 2001, 38, 165–172.
[26] R. F. B. Cox, R. T. Stormont, Org. Synth. 1935, 15, 1–2.
[27] P. N. Liu, P. M. Gu, F. Wang, Y. Q. Tu, Org. Lett. 2004, 6, 169–172.
[5] G. A. Kraus, J. Wei, J. Nat. Prod. 2004, 67, 1039–1040.
[6] a) C. Du, L. Li, Y. Li, Z. Xie, Angew. Chem. 2009, 121, 7993–7996;
Angew. Chem. Int. Ed. 2009, 48, 7853–7856 (corrigendum: C. Du, L.
Li, Y. Li, Z. Xie, Angew. Chem. 2009, 121, 9375; Angew. Chem. Int.
Ed. 2009, 48, 9211); b) Y. Wu, C. Du, C. Hu, Y. Li, Z. Xie, J. Org.
Chem. 2011, 76, 4075–4081.
Received: April 8, 2011
Published online: July 15, 2011
Chem. Eur. J. 2011, 17, 9731 – 9737
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9737