Total synthesis of cyclic diterpene tonantzitlolone based on a highly stereoselective substrate-controlled aldol reaction and ring-closing metathesis

Article abstract of DOI:10.1002/chem.200600082

The total synthesis of the cyclic diterpene ent-tonantzitlolone (ent-1) is presented. Key steps for assembling the macrocyclic core structure of 1 are a highly selective aldol reaction and an E selective ring-closing metathesis reaction. A detailed invest

Full text of DOI:10.1002/chem.200600082

FULL PAPER
DOI: 10.1002/chem.200600082
Total Synthesis of Cyclic Diterpene Tonantzitlolone Basedon a
Highly Stereoselective Substrate-ControlledAldol Reaction and
Ring-Closing Metathesis
Christian Jasper,[a] Alexander Adibekian,[b] Torsten Busch,[c] Monika Quitschalle,[c]
[c]
Rüdiger Wittenberg,[d] andAndreas Kirschning*
In memory of Jasmin Jakupovic
Abstract: The total synthesis of the cyclic diterpene ent-tonantzitlolone (ent-1) is
presented. Key steps for assembling the macrocyclic core structure of 1 are a
highly selective aldol reaction and an E selective ring-closing metathesis reaction.
A detailed investigation of these two steps and the final transformations towards
the completion of the synthesis is disclosed.
Keywords: aldol reactions · natural
products · ring-closing metathesis ·
terpenes · total synthesis
Introduction
Recently, we communicated the total synthesis of the antitu-
mor active diterpene tonantzitlolone (1) as its enantiomer
(ent-1), thereby elucidating the absolute configuration of the
natural product (Figure 1).[1] The new natural compound
was first isolated from the endemic Mexican plant Stillingia
sanguinolenta in 1990 by X. A. Dominguez et al. (Departa-
Figure 1. Structures of natural tonantzitlolone (1) and its congener 2.
mento de Quimica, ITESM, Monterrey, Mexico). In a de-
tailed reinvestigation of the same plant, the structures of 1
and a congener 2 (OAc at C-4’ of side chain) were elucidat-
ed in collaboration with J. Jakupovic in 1997.[2] Preliminary
biological evaluation indicated high activity and selectivity
against human breast and kidney cancer cell lines.[3] Interest-
ingly, when PtK2 potoroo kidney cells were incubated with
10 mgmLꢀ1 of 1 for one day a high number of mitotic cells
(around 20%) showed a monoastral half spindle instead of
a normal bipolar spindle apparatus.[4]
Our retrosynthetic analysis is depicted in Scheme 1. The
final steps of the synthesis are a dihydroxylation of an ole-
finic double bond at C-4 and C-5 in 3 which is followed by
an esterification of the (E)-3-methyl-pent-2-enoic acid there-
by introducing the side chain at C-8. The macrocyclisation,
one key transformation of the synthesis, is achieved by ring-
closing metathesis, therefore an acyclic precursor 4 is re-
quired which can be obtained by a substrate-controlled
aldol reaction (with anti-Felkin–Anh selectivity) of ketone 5
with aldehyde 6. Tetrahydrofuran 7 is prepared from acyclic
diester 8 which originates from methyl geranate 9, the start-
ing material.
Our synthesis is highlighted by a set of substrate-control-
led reactions for introducing the stereogenic centres in the
western and the northern part of the molecule. In this full
account, we particularly focus on the factors that govern the
selectivities of the aldol reaction and the ring-closing meta-
[a]Dr. C. Jasper
Universität Basel, Departement Chemie
Spitalstrasse 51, 4065 Basel (Switzerland)
[b]A. Adibekian
present address: ETH Zürich
Laboratory of Organic Chemistry
Wolfgang-Paulistrasse 10, 8093 Zürich (Switzerland)
[c]T. Busch, M. Quitschalle, Prof. Dr. A. Kirschning
Institut für Organische Chemie, Leibniz Universität Hannover
Schneiderberg 1b, 30167 Hannover (Germany)
Fax : (+49)511-726-3011
E-mail: andreas.kirschning@oci.uni-hannover.de
[d]R. Wittenberg
DS-Chemie GmbH, Straubinger Strasse 12
28219 Bremen (Germany)
Chem. Eur. J. 2006, 12, 8719 – 8734
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8719

eocontrol affording the anti-
Felkin–Anh product 14, basi-
cally as a single diastereoisom-
er (Scheme 2)[10] and 2) the
base KHMDS was the best
choice despite the fact that po-
tassium is not a common coun-
terion in chelation-controlled
aldol reactions.[11] The facial
differentiation around the al-
dehyde group cannot be the
governing factor to explain the
anti-Felkin–Anh selectivity at
C-7/C-8 as this should lead to
the reversed stereochemical
result.[12] The reaction of
Z
enolates with a-methyl chiral
aldehydes requires additional
steric considerations for the
transition state, and an excel-
lent model was first introduced
by Roush et al. (Scheme 3).[13]
For a deeper understanding of
the possible transition state of
Scheme 1. Retrosynthetic analysis of ent-1 including the key aldol reaction and ring-closing metathesis (RCM).
PG=protecting group.
thesis and discuss the late-stage transformations towards the
synthetic completion of tonantzitlolone (ent-1).
this remarkably substrate-controlled aldol reaction, we stud-
ied the roles of the counterion as well as the aldehyde, and
additionally altered the relative configuration around the
tetrahydrofuran ring.
Results andDiscussion
Thus, treatment of a-siloxy ketone 19 (simplified analogue
of ketone 13), prepared by established functional group in-
Preparation of the RCM pre-
cursor: In our synthetic ap-
proach,[1] the macrocyclic core
of ent-1 is divided in a more
complex south fragment 10
and two smaller northern frag-
ments 11 and 14. After Julia–
Kocienski olefination between
10 and 11,[5] the resulting diene
12 could be easily transformed
to the advanced ketone 13 in
three steps (Scheme 2). The Z
enolate of ketone 13 was then
coupled with the aldehyde 6[6]
to furnish the aldol product 14
with a 7,8-anti-8,9-syn relation-
ship[7] in high yield.[8] Ketone
14 was reduced to the syn diol
15 with DIBAL-H and further
elaborated to the 1,3-acetonide
Scheme 2. Assembly of the fragments to generate RCM-precursor 16 (numbering of positions according to
Figure 1): a) LDA, THF, ꢀ788C, 20 min, 10, ꢀ788C to RT, 16 h, D, 4 h; b) i) MeOH/H2O, pTsOH, 508C, 3 h,
95+3.5% 12; ii) TBSCl (1.6 equiv), CH2Cl2, imidazole, DMAP (cat.), RT, 30 min, 99%, iii) Dess–Martin peri-
odinane (1.5 equiv), CH2Cl2, 08C to RT, 2 h, 95%; c) KHMDS (1.0 equiv), THF, ꢀ788C, 0.5 h, 6 (ca.
5.0 equiv), 2 h, dr>98:2; d) DIBAL-H (8.0 equiv), CH2Cl2, ꢀ78 to ꢀ308C, 4 h; e) i) pTsOH (0.4 equiv), 2,2-
DMP, RT, 4 h; ii) TBAF (4.4 equiv), THF, RT, 16 h. LDA=lithium diisopropylamide, TBS=tert-butyldimethyl-
silyl, DMAP=4-dimethylaminopyridine, KHMDS=potassium hexamethyldisilazide, DIBAL-H=diisobutyla-
luminium hydride, pTs=p-toluenesulfonyl, DMP=dimethoxypropane, TBAF=tetra-n-butylammonium fluo-
ride.
16.[9]
Detailed study of the aldol re-
action: Two aspects of the
aldol reaction were particularly
noteworthy: 1) the reaction
proceeded with excellent ster-
8720
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8719 – 8734

Total Synthesis of Tonantzitlolone
FULL PAPER
Scheme 3. Stereochemical aspects of the aldol reaction.
terconversions from tetrahydrofuran 17, with bases that
differ in the countercation and coupling of the enolate
formed with aldehyde (S)-6, afforded diastereoisomeric
aldol products 20 and 21. The results that are listed in
Scheme 4 indicate a decrease in selectivity when the Lewis
acidity of the counterion is increased, pointing against a
tight and closed transition state. One possible rationale for
this unexpected trend can be linked to the size of the cation
which becomes part of a chelated Zimmerman–Traxler tran-
sition state which includes the oxygen atom of the tetrahy-
drofuran ring.[14,15]
Scheme 4. The role of the counter ion on the stereoselectivity of the
aldol reaction: a) NaH (2.5 equiv), THF, D, 0.5 h, BnBr (1.2 equiv), RT,
over night; b) MeOH/H2O 1:1, pTsOH (0.25 equiv), 508C, 3 h; c) TBSCl
(1.9 equiv), CH2Cl2, imidazole, DMAP (cat.), RT, 3 h; d) Dess–Martin
periodinane (1.7 equiv), CH2Cl2, 08C to RT, 6.5 h, 87% over 4 steps;
e) HMDS base (1.1 equiv), THF, ꢀ788C, 0.5 h, (S)-6 (ca. 5.0 equiv), 2 h.
HMDS=hexamethyldisilazide.
Secondly, we studied the role of the aldehyde in the aldol
reaction by coupling the (S)-aldehyde 6 with the enantio-
meric tetrahydrofuran ent-13. Tetrahydrofuran ent-13 was
prepared according to the sequence described for 13.[1] In
fact, it has the correct absolute configuration of the natural
tonantzitlolone (1). Thus, under our established reaction
conditions, the aldol reaction yielded three aldol products
23–25 of which isomer 23 was separated from the mixture of
the other aldol byproducts 24 and 25 (Scheme 5) by column
chromatography. As the configuration of the newly formed
stereogenic centres at C-8 and C-9 could not unequivocally
be determined, we reduced the keto group of the separated
fractions and transformed the 1,3-diol moiety into the cyclic
isopropylidenes 26–28. At this stage, the two byproducts 27
and 28 could be separated by repeated chromatography so
that all three isomers could be analytically inspected in
detail. The 13C NMR spectroscopic chemical shifts (d=29.7
and 19.1 ppm) of the acetonide methyl groups in 26 revealed
the 1,3-syn relative configuration.[15,16] NOE contacts and
Scheme 5. Aldol reaction between ent-13 and (S)-6: a) KHMDS (1.1 equiv), THF, ꢀ788C, 30 min, (S)-6 (ca. 5.0 equiv), 2 h; b) DIBAL-H (1m in hexane,
5–10 equiv), CH2Cl2, ꢀ78 to ꢀ308C; c) 2,2-DMP, pTsOH (0.4 equiv), RT, 4 h.
Chem. Eur. J. 2006, 12, 8719 – 8734
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8721

A. Kirschning et al.
the values of the coupling con-
stants (J) strongly indicate an
all-syn relationship between
the four stereogenic centres
(Figure 2). Although not selec-
tive, the aldol addition be-
tween aldehyde 6 and the Z
enolate derived from ketone
ent-13 favourably proceeds by
a Felkin transition state.
In a similar fashion byprod-
uct 27 was analysed. The
13C NMR spectroscopic chemi-
cal shifts (d=25.9, 24.8 ppm)
of the acetonide methyl groups
revealed an anti configuration
at C-8 and C-10. The values of
the coupling constants (J) fur-
ther support an 8,9-anti-9,10-
syn configuration (Figure 2).
Thus, it seems that 27 is the
result of a Felkin selective ad-
Figure 2. NOE contacts and coupling constants in 26 and 27.
dition of the
E enolate of
ketone ent-13 to the aldehyde
(S)-6 followed by an anti-se-
lective 1,3-reduction. Equili-
bration towards the E-enolate
is not surprising due to a pro-
longed reaction time necessary
for this mismatched transfor-
mation while the anti-induc-
tion can possibly be attributed
to solvent influences.[17]
Finally, the second byprod-
uct 28 also results from an anti
1,3-reduction as determined
from the chemical shifts of the
Scheme 6. Synthesis of model ketone 35: a) i) N-Ts-l-valine (1.0 equiv), BH3·THF (1.0 equiv), CH2Cl2, RT,
20 min, ꢀ788C, 30 (1.5 equiv), 29, ꢀ78 to ꢀ308C, buffer pH 7, 73%; ii) LiAlH4 (5.0 equiv), Et2O, 0 8C to RT,
overnight, 90%; b) i) pTsOH (cat.), 2,2-DMP (5.0 equiv), DMF, RT, 30 min, H2O, 15 min, K2CO3, 95%; ii) 5
acetonide
methyl
carbon
mol% Ti
ACHTREUNG
atoms (d=26.9, 24.8 ppm).
ꢀ258C, allyl alcohol, 10 h, 51%, dr=4:1; c) 32 (mixture of diastereoisomers), HO
ACHTREUNG
The
1H NMR
(J(8,9)=3.0 Hz,
coupling
25 mol% pTsOH, CH2Cl2/THF 18:1, RT, 2 h, then 2,2-DMP, 10 h, 79%, dr=3:1; d) i) MeOH, 20 mol%
pTsOH, 508C, 18 h, 92%; ii) imidazole (3.5 equiv), DMAP (0.2 equiv), TBSCl (2.4 equiv), CH2Cl2, RT, 10 h,
85%; iii) Dess–Martin periodinane (2 equiv), CH2Cl2, 08C to RT, 6 h, 92%.
constants
ACHTREUNG
J
ACHTREUNG
a
quence, the relative configuration between C-7 and C-8 has
to be anti.[17,18] Thus, acetonide 28 is formed by anti-Felkin
selective addition of the Z enolate of ketone ent-13 to the
aldehyde (S)-6 followed by 1,3-anti-selective reduction with
DIBAL-H.
In addition, further insight into the possible transition
states of the aldol reaction were collected by altering the
relative configuration around the tetrahydrofuran ring. As
outlined in Scheme 6, we synthesized the model ketone 35
by a similar route to the one previously developed.[1] Direct
reduction of the Kiyooka aldol product[9] to triol 31 im-
proved the yield from 73 to 90% as this improved protocol
minimizes the retroaldol reaction.[19] Protection of the 1,3-
diol followed by Sharpless asymmetric epoxidation with
(+)-diethyl l-tartrate delivered epoxy alcohol 32, this time
with reduced selectivity (drꢁ4:1).[20] The separation was
achieved for the cyclisation products 33 and 34. Final trans-
formations included protections, deprotection and oxidation
which led to the 11,14-anti as well as syn-configured tetrahy-
drofurans 35 and 36 which reflects the mediocre selectivity
for the Sharpless epoxidation.
When the simplified 11,14-cis configured tetrahydrofuran
36 was coupled with aldehyde (S)-6, the anti-Felkin–Anh
product 37 was formed, as expected, as a single diaster-
eoisomer (structurally proven as cyclic disiloxane 38). In
contrast, the 11,14-anti-configured tetrahydrofuran 35 af-
8722
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8719 – 8734

Total Synthesis of Tonantzitlolone
FULL PAPER
forded a diastereoisomeric mixture of aldol products 39a
and 39b (3:1) with aldehyde (S)-6 (Scheme 7). For structural
elucidation both diastereoisomers were reduced and con-
verted into the dioxasilinanes 40 and 41. Dioxasilinanes[21]
were chosen because acetonide formation proceeded ineffi-
ciently. At each stage of the two-step sequence, we checked
the diastereoisomeric ratio (after reduction: ~3:1; after pro-
tection: ~4:1). Determination of relevant coupling con-
stants[13,22] and NOE in 40 and 41 (contacts between H-7, H-
8 and H-9) allowed us to elucidate the configuration of all
stereogenic centres. Thus, inversion of configuration at C-11
again led to the preferred Felkin–Anh aldol product 39a.
Obviously, the enolate moiety and not the aldehyde plays
the major role for stereocontrol, because inversion of one
stereogenic centre around the tetrahydrofuran ring thereby
switching from 11,14-syn- to a 11,14-trans configuration
leads to reversal of the face selectivity during the aldol proc-
ess. Rationalization of these results is hampered by the fact
that very few examples of aldol reactions comprising a Z
enolate derived from an a-alkoxy ketone and an a-chiral al-
dehyde can be found in the literature.[12] Additionally, potas-
sium is a very uncommon counterion for these processes. It
should be taken into account that it is only a weak Lewis
acid and does not necessarily promote closed transition
states under standard conditions which is required for the
Zimmerman–Traxler one, particularly as THF is a polar sol-
vent with coordinative power. Still, the proposed transition-
state model remains feasible and is assured by the high den-
sity of oxygen functionalities around the enolate moiety.
Hence, by combining the Felkin–Anh rule with the Zimmer-
man–Traxler model the following transition states for the
various aldol reactions can be proposed, which are based on
the considerations first descri-
bed by Roush et al.[13]
When employing the 11,14-
syn tetrahydrofuran 13 with al-
dehyde (S)-6 (also applies to
tetrahydrofuran 36 and alde-
hyde (S)-6), the two Felkin–
Anh transition states 42a and
42b which are rotamers suffer
from 1,3-syn-pentane interac-
tion or an unfavourable
gauche relationship, respec-
tively (Scheme 8).[23] Both of
these interactions are reduced
to a minimum in the anti-
Felkin–Anh transition state
42c in which potassium addi-
tionally chelates the oxygen
atom of the tetrahydrofuran
ring. Here, the unfavourable
interactions present in rotam-
ers 42a and 42b are reduced
to a minimum for which some
evidence has been collected
before,[13] albeit in the case
presented here the difference
in size between R1 (allyl and
methyl) in 42a and 42c seems
not to be as pronounced as to
explain the excellent stereose-
lectivity of this transformation.
A closer view onto the transi-
tion state 42c reveals the
methyl substituent at C11 to
be
a crucial factor, which
would point to the axial H-
substituent of the aldehyde in
similar transition-state confor-
mations of 42a and 42b (not
depicted in Scheme 8).
Scheme 7. Aldol reaction of diastereoisomeric ketones 35 and 36 with aldehyde (S)-6: a) KHMDS (1.1 equiv),
THF, ꢀ788C, 0.5 h, (S)-6 (ꢁ5.0 equiv), 15 min for 35 affording 39a/39b: 64% and dr=73:27; and 40 min for
36 affording 37: 88% and dr>98:2; b) i) DIBAL-H, (1m in hexane, 5 equiv), CH2Cl2, ꢀ78 to ꢀ308C, 5 h, 81%
(after reisolation and reuse of the starting materials); ii) (tBu)2Si(OTf)2, (1.5 equiv), 2,6-lutidine, CH2Cl2, RT,
N
24 h, 96%; c) i) mixture of 39a/39b, Et2BOMe (1.1 equiv), THF/MeOH 3:1, ꢀ788C, 15 min, NaBH4
(3.3 equiv), 8 h, 67%, dr=80:20; ii) (tBu)2Si
A
Chem. Eur. J. 2006, 12, 8719 – 8734
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8723

A. Kirschning et al.
smoothly converted into the
corresponding ketone 47 by
using the Dess–Martin reagent.
Our first experimental ef-
forts towards ring closure were
conducted with dienes 16, 27
and 51 (obtained from silyl
ether 26)[26] which all contained
an 8,10-acetonide protection
and
a
11,14-cis-disubstituted
tetrahydrofuran but differed in
the relative configuration be-
tween
C-7
and
C-11
(Scheme 11). The matched-
case aldol product 16 afforded
macrocycles 49 with pro-
Scheme 8. Possible transition states for the aldol reaction of potassium enolate derived from tetrahydrofuran
13 with aldehyde (S)-6 (matched case).
nounced
Z selectivity (E/Z
By switching from 13 to ent-13 and subsequent reaction
with aldehyde (S)-6 under the same conditions, a complex
mixture of diastereoisomers was obtained with the Felkin–
Anh product 38 (corresponding to a reaction through transi-
tion state 43a) as the main diastereoisomer. This result
clearly indicates that the differentiation between Me and R1
(allyl) in transition states 43c and 43a cannot be the govern-
ing factor for stereocontrol (vide supra), as this should lead
again to a product with anti-Felkin–Anh control (corre-
sponding to a reaction through transition state 43c). Conse-
quently, the C11 methyl/aldehyde-H interaction seems to
dominate the stereochemical outcome of this reaction. Con-
sidering the low yield of the transformation between ent-13
and (S)-6, some contribution of a stronger gauche interac-
tion in 43a in comparison to 42c making both transition
states (43a and 43c) unlikely cannot be completely neglect-
ed.
In the case of the 11,14-trans tetrahydrofuran 35 two
Scheme 9. Possible transition states for the aldol reaction of potassium
enolate derived from tetrahydrofuran ent-13 with aldehyde (S)-6 (mis-
matched case).
closed transition states similar to those presented in
Schemes 8 and 9 can be drawn, from which again it becomes
evident, that the stereochemistry at C-14 has almost no
bearing on the selectivity of the aldol reaction.[24]
ꢁ1:6) when subjected to the RCM with Grubbs II cata-
lyst.[27] This selectivity decreased, when the three positions
at C-7, C-9 and C-10 were simultaneously inverted as in
diene 27.[28] Remarkably, when only the position at C-7 is
relatively inverted with respect to the rest of the molecule
as in the mismatched aldol product 51, the RCM to macro-
cycles 52 proceeded with little E selectivity.
We planned to use macrocycle 49 for finishing the total
synthesis of tonantzitlolone, but we found that dihydroxyla-
tion of the olefinic double bond at C-4/C-5 with OsO4
proved to be unsuccessful, as only the minor E isomer could
be converted. We expected that removal of the cyclic acetal
should enhance the flexibility of the western chain. Thus,
diene 47 afforded the RCM product 53 in excellent yield,
but displayed no stereoselectivity (E/Zꢁ1:1, Scheme 12).
Due to the bulky TES groups, neither the E nor the Z com-
ponent of this inseparable mixture could be dihydroxylated
Studies on the stereocontrol of the RCM: To advance the
synthesis, oxidation at C-9 of alcohol 16 was necessary
which ought to set the stage for the RCM.[25] However, all
oxidation procedures tested failed. Additionally, we were
unable to develop an orthogonal protecting-group strategy
for diol 15. Under basic conditions, the TBS-group rapidly
migrates to the neighbouring hydroxy groups at C-8 and C-
10, furnishing regioisomers 44 and 45 (Scheme 10). This ob-
servation was exploited for a differentiation strategy. Conse-
quently, desilylation afforded the corresponding triol which
in the following was selectively reprotected at C-8 and C-10
with TES-Cl leaving the central hydroxyl group at C-9 in 46
free. In contrast, when TMS-Cl was employed as silylating
agent, the discrimination between the three hydroxyl groups
was less selective, for which different steric requirements of
the two reagents can be made responsible. Alcohol 46 was
8724
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8719 – 8734

Total Synthesis of Tonantzitlolone
FULL PAPER
Scheme 10. Differentiation strategy and oxidation of the C-9 position:
a) NaH (1.9 equiv), THF, 08C, 15 min; b) TBAF (2.3 equiv), THF, RT,
4 h; c) imidazole (6.1 equiv), TES-Cl (5.2 equiv), DMF, RT to 308C, 26 h;
d) Dess–Martin periodinane (4.2 equiv), CH2Cl2, RT, 1.5 h.
while epoxidation with the smaller reagent mCPBA also led
to oxidation of the olefinic double bond at C-1/C-2. An im-
portant message from this experiment is the observation
that formation of an E-configured macrocyclisation product
has to be envisaged to terminate the synthesis and that the
functional and protecting-
Scheme 11. Ring-closing metathesis of acetonides 16, 27 and 51:
a) Grubbs II (10 mol%), CH2Cl2, D, 2 h; b) TBAF (5.6 equiv), THF, RT,
19.5 h; c) Grubbs II (10 mol%), CH2Cl2, D, 5.5 h.
group pattern in the western
half can exert a strong influ-
ence on this selectivity. There-
fore, we removed the silyl pro-
tection and conducted a RCM
macrocyclisation with dihy-
droxy ketone 54. Indeed, this
substrate could be cyclised
yielding the macrocycle 55a
with good E selectivity in 85%
yield (E/Zꢁ5:1). When basic
instead of neutral Al2O3 was
used during workup, partial
tautomerisation of dihydroxy
ketone 54 was observed af-
fording one isomeric byprod-
Scheme 12. Ring-closing metathesis of ketones 47 and 54: a) Grubbs II (13 mol%), CH2Cl2, 08C, 2 h; b) TBAF
(2.7 equiv), THF, 08C, 0.5 h; c) Grubbs II (9.5 mol%), CH2Cl2, 408C, 1.5 h.
uct 55b of unknown configura-
tion at C-9.[29] This RCM ex-
periment clearly demonstrates
that the degree of flexibility in the western half is of vital
importance for the stereochemical outcome of the ring-clos-
ing metathesis reaction. This example adds to the increasing
number of reports[30] in which subtle changes in the molecu-
lar structure of a given precursor or the reaction conditions
can have a dramatic effect on the stereochemical outcome
of a RCM reaction. Despite the fact that the macrocyclisa-
tion is commonly very efficient in terms of yields, the E/Z
selectivity of alkene formation is often low and hard to pre-
dict. In particular, examples for the ring-closing metathesis
of macrocyclic natural products shed a light on this problem
as it became evident that various factors are responsible for
the stereochemical outcome of the macrocyclisation proc-
ess.[31] In the present case, protecting groups and the config-
uration of several stereogenic centres in the acyclic precur-
sor can dramatically reverse the stereoselectivity of the
RCM, even if these governing groups are not closely located
to the reaction centre.[32] In the present case, basically
almost complete reversal of E/Z selectivity was achieved by
simple functional-group manipulations in related dienes 16
and 54.[33]
Chem. Eur. J. 2006, 12, 8719 – 8734
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8725

A. Kirschning et al.
Finalization of the synthesis:
With macrocycle 55a in hand,
we initiated studies on the
direct introduction of the ester
side chain at position 8 to
avoid additional protection
strategies in the western half.
We found that only under Ya-
maguchi conditions could (E)-
3-methylpent-2-enoic
acid[34]
be coupled yielding a complex
mixture of products. We there-
fore had to carry out the che-
moselective dihydroxylation of
the olefinic double bond at C-
4 and C-5 first, before intro-
ducing the side chain. The di-
hydroxylation cleanly proceed-
ed at the desired position with
slight preference for the unde-
sired (4R,5R)-diol 56a under
conditions of substrate control
(56a/56b 1.3:1) (Scheme 13).
When the dihydroxylation
was carried out under Sharp-
less conditions[20a] with the b-
DHQ-CLB-ligand the (4R,5R)-
diol 56a (56a/56b 9:1) was
preferentially formed while
the a-DHQ-CLB-ligand re-
sembles the mismatched sce-
nario and yielded a mixture of
diols 56a and 56b (1:1.3).[35]
Exploiting again the observa-
tion that the choice of func-
tionality in the western part
has a strong impact on the
stereochemical outcome of re-
actions around C-4 and C-5
we first formed the protected
dihydroxy ketone 57 from the
8,10-diol 55, which then was
treated with Sharpless asym-
metric dihydroxylation con-
ditions again by using a-
DHQ-CLB as the ligand
(Scheme 14).[36] After desilyla-
tion of the diastereoisomeric
mixture the desired hemiacetal
56b with improved selectivity
(56a/56b 1:3) was isolated. It
is noteworthy that deprotec-
tion was required for lactol
formation to occur, so that the
need for triol differentiation
became necessary in the fol-
lowing.[37] As differentiation
Scheme 13. Dihydroxylation of macrocyclic alkene 55a: a) NMO (2.5 equiv), MSA (3.5 equiv), OsO4
(6.3 mol%), tBuOH/H2O 2:1, 08C to RT, 3.5 h, 86%; b) K2CO3 (3.4 equiv), K3Fe(CN)6 (2.7 equiv), MSA
(2.3 equiv), OsO4 (10 mol%), b-DHQD-CLB (22 mol%), tBuOH/H2O 1:1, 08C, 2 h, 78%, drꢁ9:1; c) K2CO3
(3.0 equiv), K3Fe(CN)6 (3.8 equiv), MSA (3.0 equiv), OsO4 (4 mol%), a-DHQD-CLB (23 mol%), tBuOH/
H2O 1:1, 08C to RT, 5 h, 84%; NMO=N-methylmorpholin-N-oxide, MSA=methanesulfonamide, DHQD-
CLB=dihydroquinidine p-chlorobenzoate.
Scheme 14. Completion of the synthesis: a) NEt3 (6.9 equiv), TMSCl (4.7 equiv), DMF, 08C, 1 h, b) i) K2CO3
(3.1 equiv), K3Fe(CN)6 (3.0 equiv), MSA (3.0 equiv), OsO4 (10 mol%), a-DHQD-CLB (36 mol%), tBuOH/
H2O 1:1, 08C, 2.5 h, ii) TBAF (1.2 equiv), THF, 08C, 10 min, 83% (2 steps); c) XOH (14.9 equiv), DIC
(14.4 equiv), CH2Cl2, RT, 1 h, DMAP (1.2 equiv), 28 h; d) TPAP (0.4 equiv), NMO (excess), CH2Cl2, RT, 1 h;
TMS=trimethylsilyl, DIC=diisopropylcarbodiimide, TPAP=tetraisopropyl perruthenate.
8726
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8719 – 8734

Total Synthesis of Tonantzitlolone
FULL PAPER
Darmstadt) and spots were detected either by UV-absorption or by char-
ring with H2SO4/4-methoxybenzaldehyde in methanol. Preparative
column chromatography was performed on silica gel 60 (E. Merck,
Darmstadt). For experimental and spectroscopic data of ent-1, 11, 12, 13,
14, 15, 16, 17, 29–31, 46, 47, 49, 54, 55a, 56a,b, 57, 58a–c and 59 please
refer to the Supporting Information of our previous publication.[1]
by various protecting-group strategies could not be ach-
ieved, direct esterification was tested. Esterification with dii-
sopropyl carbodiimide was the method of choice and yielded
a mixture of all possible position isomers (58a–c). Gratify-
ingly, the C-10 ester 58a quantitatively rearranged to the de-
sired C-8 ester 58b because of the 8,10-syn relationship of
the two adjacent hydroxy groups.
Finally, the two resulting esters were oxidized with TPAP/
NMO furnishing ent-tonantzitlolone (ent-1),[38] and ketone
59 (derived from 58c).
Aldehyde 10: A flame-dried 10 mL flask was charged with alcohol 17
(205 mg, 0.79 mmol) and dry CH2Cl2 (4 mL). Then molecular sieves (4 Š,
120 mg), N-methylmorpholine-N-oxide (138.5 mg, 1.18 mmol, 1.5 equiv)
and tetra-n-propylperruthenate (15.3 mg, 0.04 mmol, 5.1 mol%) were
added portionwise and the solution was stirred for 2.5 h. After this time,
the dark black solution was filtered (5 g of silica gel) and eluted with hex-
anes/ethyl acetate 8:1. Removal of the solvent under reduced pressure af-
forded the target aldehyde 10 (184 mg, 0.72 mmol, 91%) as a colourless
1
oil. H NMR (400 MHz, CDCl3): d=0.97, 1.01 (2s, 6H; C
ACHTREUNG
Conclusion
3H; OCCH3), 1.27, 1.36 (2s, 6H; O2C
ACHTREUNG
H-13), 1.81–1.90 (m, 1H; H-13), 1.93–2.01 (m, 1H; H-12), 3.71–3.76 (m,
1H; H-9), 3.89–3.97 (m, 3H; H-9, H-10, H-14), 9.56 ppm (s, 1H; H-1);
We described the first total synthesis of the diterpene to-
nantzitlolone ent-1 and importantly elucidated its absolute
configuration. Important focus was devoted to the aldol re-
action of the potassium enolate derived from an a-siloxy
ketone with an a-methyl chiral aldehyde in a highly stereo-
controlled manner yielding the anti-Felkin–Anh product.[39]
Factors that may govern this unusual aldol reaction were
studied. Additionally, we studied the influence of neighbour-
ing oxygen functionalities and relative configuration of ster-
eogenic centres on the E/Z selectivity of the key ring-closing
metathesis process. These principal studies enabled us to de-
velop a highly stereoselective total synthesis of this compact
and highly functionalised macrocyclic diterpene which
makes effective use of substrate-controlled reactions. Details
on the biological properties and the biological target of to-
nantzitlolone, its (ꢀ)-enantiomer and various derivatives
will be reported in a separate account.
13C NMR (100 MHz, CDCl3): d=17.1, 19.6, 21.6 (3q; C
ACHTREUNG
25.0, 26.3 (2q; O2C
ACHTREUNG
A
1458 (m), 1370 (m), 1211 (m), 1155 (m), 1068 (s), 854 cmꢀ1 (s); HRMS-
ESI: m/z: calcd for C14H25O4: 257.1747 [M+H]+; found: 257.1753.
Ketofuran 19: A 10 mL round-bottomed flask was charged with alcohol
17 (157.3 mg, 0.61 mmol) and dry THF (4 mL). NaH (60 mg, 60% sus-
pension, 1.5 mmol, 2.5 equiv) was added and the mixture was heated
under refluxing conditions for 30 min. Then, benzyl bromide (90 mL,
0.76 mmol, 1.2 equiv) was added and the solution was stirred overnight
at RT. Saturated aqueous NH4Cl solution was added, and the aqueous
layer was separated and extracted with CH2Cl2. The combined organic
solutions were dried (MgSO4) and concentrated under reduced pressure.
A 100 mL round-bottomed flask was charged with the resulting benzyl
ether 18 and with 30 mL of MeOH/H2O 1:1. After addition of pTsOH
(25 mg, 0.15 mmol, 0.25 equiv) the solution was warmed to 508C for 3 h
and then cooled to RT. The reaction was stopped by addition of saturated
aqueous NaHCO3 solution and the mixture was extracted with ethyl ace-
tate. The combined organic phases were dried (MgSO4) and concentrated
under reduced pressure.
A flame-dried 25 mL round-bottomed flask was charged with the crude
diol (~0.61 mmol) and dry CH2Cl2 (10 mL). Imidazole (100 mg,
1.47 mmol, 2.4 equiv) was added, followed by TBSCl (170 mg, 1.13 mmol,
1.9 equiv) and 4-DMAP (10 mg, 0.08 mmol, 0.1 equiv). After 3 h, the re-
action was quenched by addition of saturated aqueous NH4Cl solution.
The aqueous layer was separated and extracted with CH2Cl2, and the
combined organic solutions were dried (MgSO4) and concentrated under
reduced pressure.
Experimental Section
General remarks andstarting materials
:
1H NMR, 13C NMR, 1H/13C-
COSY and NOESY spectra were measured on Avance 200/DPX
(Bruker) with 200 MHz (50 MHz), Avance 400/DPX (Bruker) 400 MHz
(100 MHz) and Avance 500/DRX (Bruker), respectively, by using tetra-
methylsilane as the internal standard. If not otherwise noted, CDCl3 and
[D4]MeOH are the solvents for all NMR experiments (reference:
CHCl3 =d=7.26 ppm, [D4]MeOH=d=3.31 ppm). Multiplicities are de-
scribed by using the following abbreviations: s=singlet, d=doublet, t=
A 50 mL round-bottomed flask was charged with the crude alcohol and
CH2Cl2 (15 mL). The solution was cooled to 08C, then Dess–Martin peri-
odinane (340 mg, 0.8 mmol, 1.3 equiv) was added and the reaction mix-
ture was warmed to RT and stirred for 6.5 h. During that time two addi-
tional portions (50 mg) of Dess–Martin periodinane were added. Then
the reaction was terminated by addition of saturated aqueous NaHCO3/
Na2S2O3 solution. The aqueous layer was separated and extracted with
CH2Cl2. The combined organic solutions were dried (MgSO4) and con-
centrated under reduced pressure. Flash-column chromatography (hex-
anes/ethyl acetate 40:1) afforded the desired ketone 19 (221.7 mg,
0.53 mmol, 87% over 4 steps) as an oil. [a]2D3 =ꢀ35.9 (c=1.07 in CHCl3);
1H NMR (400 MHz, CDCl3, CH2Cl2 =5.3 ppm): d=0.13 (2s, 6H; Si-
triplet, q=quartet, m=multiplet, sext.
= sextet, br=broad and ps=
pseudo. Chemical shift values of 13C NMR spectra are reported as values
in ppm relative to residual CDCl3 (ca. 77.0 ppm) or [D4]MeOH (ca.
49.0 ppm) as internal standards. The multiplicities refer to the resonances
in the off-resonance spectra and were elucidated by using the distortion-
less enhancement by the polarisation transfer (DEPT) spectral editing
technique, with secondary pulses at 90 and 1358. Multiplicities are report-
ed by using the following abbreviations: s=singlet (due to quaternary
carbon), d=doublet (methine), q=quartet (methyl) and t=triplet (meth-
ylene). Numbering of protons and carbon atoms are commonly related to
numbering in tonantzitlolone (1). Mass spectra were recorded on a type
LCT-spectrometer (Micromass) and on a type VG autospec (Micromass).
Ion mass (m/z) signals are reported as values in atomic mass units fol-
lowed, in parentheses, by the peak intensities relative to the base peak
A
N
1.61–1.70 (m, 1H; H-12), 1.79–1.88 (m, 2H; H-13), 2.16–2.24 (m, 1H; H-
12), 3.17, 3.25 (2d, J=8.8 Hz, 2H; H-1), 3.98–4.03 (m, 1H; H-14), 4.43,
4.47 (2d, J=12.4 Hz, 2H; PhCH2), 4.58, 4.63 (2d, J=19.5 Hz, 2H; H-9),
7.28–7.40 ppm (m, 5H; Ph); 13C NMR (50 MHz, CDCl3): d=ꢀ5.4, ꢀ5.3
(100%). Optical rotations [a]were collected on
a Polarimeter 341
(2q; Si
N
N
(Perkin–Elmer) at
a wavelength of 589 nm and are given in
10ꢀ1 degcm2 gꢀ1. All solvents used were of reagent grade and were further
dried. Reactions were monitored by TLC on silica gel 60 F254 (E. Merck,
(q; tBu), 26.0 (t; C-13), 36.2 (t; C-12), 37.6 (s; C-15), 66.5 (t; C-9), 73.3,
77.4 (t, C-1; PhCH2), 83.5 (d; C-14), 87.6 (s; C-11), 127.4, 127.4, 128.3, (d;
Chem. Eur. J. 2006, 12, 8719 – 8734
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8727

A. Kirschning et al.
Ph), 138.8, (s; Ph), 211.6 ppm (s; C-10); HRMS-ESI: m/z: calcd for
C24H40O4SiNa: 443.2594 [M+Na]+; found: 443.2596.
chromatography (hexanes/ethyl acetate 40:1!10:1) delivered the corre-
sponding diol (5.1 mg, 11 mmol, 48%) as an oil, along with recovered
starting material 23 (4.9 mg, 10 mmol, 43%). [a]2D3 =ꢀ13.7 (c=0.54 in
Standard procedure for the aldol reaction 19!20/21: The preparation of
aldehyde (S)-6 started from the corresponding alcohol which was dis-
solved in CH2Cl2 (c=0.15m). The solution was cooled to 08C and Dess–
Martin periodinane (1.2 equiv) was added. The mixture was warmed to
RT and stirred for 1 h. After this time the reaction was terminated by ad-
dition of aqueous Na2S2O3/NaHCO3 solution and stirred for an additional
hour. The aqueous layer was separated and extracted with CH2Cl2. The
combined organic solutions were dried (MgSO4) and cautiously concen-
trated (p>600 mbar).
CHCl3); 1H NMR (400 MHz, CDCl3): d=0.14, 0.16 (2s, 6H; Si
ACHTREUNG
0.91 (s, 9H; tBu), 0.96, 1.03 (2s, 6H; C
ACHTREUNG
CHCH3), 1.08 (d, J=6.9 Hz, 3H; CHCH3), 1.21 (s, 3H; OCCH3), 1.57–
1.70 (m, 2H; H-12, H-13), 1.74–1.89 (m, 3H; H-6, H-7, H-13), 2.03–2.10
(m, 1H; H-12), 2.18–2.25 (m, 1H; H-6), 2.64 (d, J=7.5 Hz, 1H; 8-OH),
2.78 (d, J=5.1 Hz, 1H; 10-OH), 2.83 (brsext., J=6.5 Hz, 1H; H-3), 3.30
(dt, J=2.4, 7.1 Hz, 1H; H-8), 3.47 (t, J=5.1 Hz, 1H; H-10), 3.71 (dd, J=
8.5, 6.4 Hz, 1H; H-14), 3.83 (dd, J=5.6, 2.6 Hz, 1H; H-9), 4.94 (ddd, J=
10.0, 1.7, 1.4 Hz, 1H; H-5), 4.97 (ddd, J=17.1, 1.7, 1.4 Hz, 1H; H-5),
5.00–5.06 (m, 2H; H-4’ ”2), 5.36 (dd, J=15.8, 6.3 Hz, 1H; H-2), 5.44 (dd,
J=15.8, 0.7 Hz, 1H; H-1), 5.77 (dddd, J=16.9, 10.3, 7.4, 6.5 Hz, 1H; H-
5’), 5.78 ppm (ddd, J=17.1, 10.4, 6.6 Hz, 1H; H-4); 13C NMR (100 MHz,
A 25 mL round-bottomed flask was charged with ketone 19 and with dry
THF (c=0.04m). The solution was cooled to ꢀ788C, before the HMDS-
base (as a solution, 1.1 equiv) was introduced dropwise. After 30 min (in
the case of LiHMDS and NaHMDS the solution was warmed to ꢀ208C
for that time and then recooled to ꢀ788C), the crude aldehyde 6 in dry
THF was added and the solution was stirred for 2 h. The reaction was
then quenched carefully by addition of MeOH, followed by saturated
aqueous NH4Cl solution. The flask was warmed up to RT in a cold bath,
then the aqueous layer was separated and extracted with MTBE. The
combined organic solutions were dried (MgSO4), concentrated and fil-
tered over a short column of silica (hexanes/ethyl acetate 40:1). The
product ratio 20:21 was determined by HPLC-ESI (for Li: 4:1, Na: 8:1
and K: 18:1).
CDCl3): d=ꢀ4.4, ꢀ3.3 (2q; Si
ACHTREUNG
20.2 (q; CHCH3), 21.9 (q; OCCH3), 24.7, 24.7 (2q; C
AHCTREUNG
tBu), 27.2 (t; C-13), 35.4 (d; C-7), 35.5 (t; C-12), 38.5 (t; C-6), 39.0 (s; C-
15), 40.6 (d; C-3), 73.9 (d; C-8), 73.9 (d; C-10), 77.0 (d; C-9), 84.1 (s; C-
11), 85.6 (d; C-14), 112.9 (t; C-5), 116.5 (t; C-4’), 132.8 (d; C-2), 134.7 (d;
C-1), 136.9 (d; C-5’), 143.3 ppm (d; C-4); HRMS-ESI: m/z: calcd for
C28H52O4SiNa: 503.3533 [M+Na]+; found: 503.3533.
pTsOH (6.3 mg, 37 mmol, 0.4 eq) was added to the diol (41.4 mg,
86 mmol), obtained as described above, dissolved in 2,2-dimethoxypro-
pane (2.2 mL). After 4 h, the reaction was stopped by addition of saturat-
ed aqueous NaHCO3 solution. The solution was transferred to a separat-
ing funnel and extracted twice with CH2Cl2. The organic phase was dried
(MgSO4) and concentrated under reduced pressure. Gel filtration (hex-
anes/ethyl acetate 80:1) afforded the desired product 26 (37.6 mg,
Aldol products 23: A 25 mL round-bottomed flask was charged with
ketone ent-13 (60.6 mg, 0.16 mmol) and dry THF (3.6 mL). The solution
was cooled to ꢀ788C, before KHMDS (0.25 mL, 0.66m in toluene, 1.1
equiv) was introduced dropwise. After 30 min, crude aldehyde 6 (about
1.2 mmol) in dry THF (0.4 mL) was added and the solution was stirred
for 2 h. The reaction mixture was carefully hydrolysed by addition of
MeOH followed by saturated aqueous NH4Cl solution. The flask was
slowly warmed to RT in the cold bath, then the aqueous layer was sepa-
rated and extracted with MTBE. The combined organic solutions were
dried (MgSO4) and concentrated. Flash chromatography (hexanes/ethyl
acetate 40:1) furnished the desired aldol product 23 (30 mg, 0.063 mmol,
39%) as a colourless oil and as an inseparable mixture of the two diaster-
eoisomers 24 and 25 which were directly employed in the reduction/pro-
tection sequence that followed. Compound 23: [a]2D3 =+42.6 (c=0.99 in
1
72 mmol; 84%) as a colourless oil. [a]2D3 =ꢀ2.1 (c=1 in CHCl3); H NMR
(400 MHz, CDCl3): d=0.13 (s, 6H; Si(CH3)2), 0.88 (d, 3H, J=6.5 Hz;
T
CHCH3), 0.91 (s, 9H; tBu), 0.95, 0.99 (2s, 6H; C
6.8 Hz, 3H; CHCH3), 1.22 (s, 3H; OCCH3), 1.39, 1.46 (2s, 6H; O2C-
(CH3)2), 1.53–1.77 (m, 4H; H-6, H-12 ”2), H-13), 1.86–1.97 (m, 2H; H-7,
A
AHCTREUNG
H-13), 2.21–2.27 (m, 1H; H-6), 2.81 (sext.q, J=6.8, 1.1 Hz, 1H; H-3),
3.19 (d, J=10.0 Hz, 1H; H-8), 3.51 (s, 1H; H-10), 3.77 (dd, J=8.2,
4.9 Hz, 1H; H-14), 3.83 (s, 1H; H-9), 4.94 (ddd, J=10.4, 1.7, 1.4 Hz, 1H;
H-5), 4.97 (ddd, J=17.2, 1.6, 1.4 Hz, 1H; H-5), 5.00–5.05 (m, 2H; H-4’
”2), 5.30 (dd, J=15.9, 6.5 Hz, 1H; H-2), 5.42 (dd, J=15.9, 1.0 Hz, 1H;
H-1), 5.76 (dddd, J=16.6, 10.7, 8.5, 5.9 Hz, 1H; H-5’), 5.78 ppm (ddd, J=
17.2, 10.4, 6.7 Hz, 1H; H-4); 13C NMR (100 MHz, CDCl3): d=ꢀ2.0, ꢀ1.2
CHCl3); 1H NMR (400 MHz, CDCl3): d=ꢀ0.02, 0.09 (2s, 6H; Si
N
N
CHCH3), 1.07 (d, J=6.9 Hz, 3H; CHCH3), 1.35 (s, 3H; OCCH3), 1.46–
1.55 (m, 1H; H-13), 1.74–1.92 (m, 5H; H-12 ”2, H-13, H-6, H-7), 2.04
(brd, J=10.7 Hz, 1H; 8-OH), 2.33 (dddd, J=13.2, 6.3, 3.3, 1.6 Hz, 1H;
H-6), 2.81 (sext.td, J=6.7, 1.4, 0.8 Hz, 1H; H-3), 3.80 (dd, J=8.0, 6.7 Hz,
1H; H-14), 3.90 (brt, J=9.4 Hz, 1H; H-8), 4.93 (ddd, J=10.3, 1.7,
1.2 Hz, 1H; H-5), 4.96 (ddd, J=17.2, 1.7, 1.6 Hz, 1H; H-5), 5.00–5.07 (m,
2H; H-4’ ”2), 5.10 (d, J=1.3 Hz, 1H; H-9), 5.36 (dd, J=15.9, 6.6 Hz,
1H; H-2), 5.45 (dd, J=15.9, 0.8 Hz, 1H; H-1), 5.76 (ddd, J=17.2, 10.3,
6.4 Hz, 1H; H-4), 5.77 ppm (dddd, J=16.5, 10.6, 7.6, 6.3 Hz, 1H; H-5’);
(2q; Si
(q; CHCH3), 21.5 (q; OCCH3), 24.0, 24.9 (2q; C
26.6 (t; C-13), 29.7 (q; O2C(CH3)2), 31.5 (d; C-7), 34.00 (t; C-12), 37.1 (t;
C-6), 39.7 (s; C-15), 40.7 (d; C-3), 66.2 (d; C-9), 79.2 (d; C-8), 80.8 (d; C-
10), 84.5 (s; C-11), 85.1 (d; C-14), 99.4 (s; O2C(CH3)2), 112.7 (t; C-5),
A
ACHTREUNG
AHCTREUNG
AHCTREUNG
AHCTREUNG
116.8 (t; C-4’), 131.6 (d; C-2), 135.7 (d; C-1), 136.2 (d; C-5’), 143.5 ppm
(d; C-4); IR (ATR): n˜ =2976 (s), 2932 (m), 2859 (s), 1639 (w), 1463 (m),
1381 (m), 1257 (m), 1199 (m), 1121 (s), 1091 (s), 1027 (w), 936 (w), 912
(m), 861 (w), 835 (m), 804 (w), 773 (m), 696 cmꢀ1 (w); HRMS-ESI: m/z:
calcd for C31H56O4SiNa: 543.3846 [M+Na]+; found: 543.3846.
13C NMR (100 MHz, CDCl3): d=ꢀ5.1, ꢀ4.1 (2q; Si
C
Protectedtriols 27 and28 : Acetonides 27 (14.6 mg, 28 mmol, 41%) and
28 (6.6 mg, 13 mmol, 19%) were obtained from a mixture of hydroxyke-
tones 24 and 25 by performing the reduction/protection sequence descri-
bed for hydroxyketone 23 after subsequent separation by flash-column
chromatography (hexanes/ethyl actetate 150:1).
(CH3)2), 25.8 (t; C-13), 26.0 (q; tBu), 36.7 (d; C-7), 37.4 (t; C-12), 38.2 (t;
C-6), 39.0 (s; C-15), 40.7 (d; C-3), 75.0 (d; C-8), 76.3 (d; C-9), 86.5 (d; C-
14), 88.7 (s; C-11), 112.9 (t; C-5), 116.4 (t; C-4’), 132.4 (d; C-2), 134.9 (d;
C-1), 136.9 (d; C-5’), 143.2 (d; C-4), 212.8 ppm (s; C-10); IR (ATR): n˜ =
3558 (w), 3078 (w), 2960 (m), 2929 (m), 2857 (m), 1727 (s), 1639 (m),
1462 (m), 1388 (m), 1363 (m), 1253 (m), 1160 (m), 1101 (m), 1043 (s),
1000 (m), 976 (s), 910 (s), 836 (s), 777 (s), 676 (m), 613 cmꢀ1 (w); HRMS-
ESI: m/z: calcd for C30H53NO4SiNa: 542.3642 [M+Na+MeCN]+; found:
542.3642.
Compound27 : Rf =0.68 (petroleum ether/ethyl acetate 10:1); [a]D23
+3.3 (c=1.07 in CHCl3); 1H NMR (400 MHz, CDCl3): d=0.08, 0.09 (2s,
6H; Si(CH3)2), 0.86 (d, 3H, J=6.8 Hz; CHCH3), 0.90 (s, 9H; tBu), 0.93,
0.96 (2s, 6H; C(CH3)2), 1.06 (d, J=6.9 Hz, 3H; CHCH3), 1.22 (s, 3H;
OCCH3), 1.32 (d, J=0.5 Hz, 3H; O2C(CH3)2), 1.46 (d, J=0.5 Hz, 3H;
O2C(CH3)2), 1.48–1.81 (m, 4H; H-12 ”2, H-13 ”2), 2.04 (dddt, J=13.5,
=
AHCTREUNG
AHCTREUNG
AHCTREUNG
Protectedtriol 26 : A 5 mL round-bottomed flask was charged with aldol
product 23 (10.8 mg, 23 mmol) and anhydrous CH2Cl2 (1.1 mL) and
cooled to ꢀ788C, before DIBAL-H (0.2 mL, 1m in hexane, 8.7 equiv)
was added. After 22 h, the reaction mixture was warmed to ꢀ308C and
was then quenched with H2O (0.2 mL). After reaching RT, saturated
aqueous Na/K tartrate solution was added and stirring was continued for
1 h. The aqueous layer was separated and extracted with CH2Cl2. The
combined organic solutions were dried (MgSO4) and concentrated. Flash
AHCTREUNG
8.8, 7.1, 1.3 Hz, 1H; H-6), 2.18 (ddddd, J=13.5, 7.3, 6.6, 1.3, 1.2 Hz, 1H;
H-6), 2.47 (ddqdd, J=7.9, 7.0, 6.8, 0.9, 0.7 Hz, 1H; H-7), 2.80 (sext.q, J=
6.7, 1.2 Hz, 1H; H-3), 3.72 (t, J=6.8 Hz, 1H; H-14), 3.76 (d, J=5.1 Hz,
1H; H-10), 4.18 (dd, J=9.5, 0.8 Hz, 1H; H-8), 4.21 (dd, J=9.5, 5.1 Hz,
1H; H-9), 4.93 (ddd, J=10.4, 1.7, 1.3 Hz, 1H; H-5), 4.94–4.99 (m, 1H;
H-5), 4.95–5.04 (m, 2H; H-4’ ”2), 5.29 (dd, J=15.9, 6.7 Hz, 1H; H-2),
8728
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8719 – 8734

Total Synthesis of Tonantzitlolone
FULL PAPER
5.40 (dd, J=15.9, 0.9 Hz, 1H; H-1), 5.77 (ddd, J=17.2, 10.4, 6.5 Hz, 1H;
H-4), 5.80 ppm (dddd, J=17.1, 10.3, 7.1, 6.6 Hz, 1H; H-5’); 13C NMR
0.89 mmol). After 2 h, complete conversion of the starting material was
observed, before 2,2-dimethoxypropane (7 mL) was introduced. The solu-
tion was stirred for 10 h and then terminated by addition of saturated
aqueous NaHCO3 solution. The aqueous layer was separated and extract-
ed with methyl tert-butylmethyl ether (5”20 mL). The combined organic
layers were dried (MgSO4) and concentrated under reduced pressure.
Flash chromatography (hexanes/ethyl acetate 50:1) enabled us to sepa-
rate both alcohols and gave 33 (535 mg, 2.07 mmol) and 34 (170 mg,
0.66 mmol) in an overall yield of 79%.
(100 MHz, CDCl3): d=ꢀ4.2, ꢀ3.1 (2q; Si
(s; tBu), 20.3 (q; CHCH3), 21.5 (q; OCCH3), 24.2, 25.0 (2q; C
25.0, 27.1 (2q; O2C(CH3)2), 25.4 (t; C-13), 26.5 (q; tBu), 33.6 (d; C-7),
ACHTREUNG
AHCTREUNG
ACHTREUNG
39.0 (t; C-12), 39.3 (s; C-15), 39.7 (t; C-6), 40.7 (d; C-3), 72.5 (d; C-8),
80.5 (d; C-9), 82.8 (d; C-10), 84.0 (s; C-11), 86.8 (d; C-14), 106.3 (s; O2C-
A
138.2 (d; C-5’), 143.9 ppm (d; C-4); HRMS-ESI: m/z: calcd for
C31H56O4SiNa: 543.3846 [M+Na]+; found: 543.3799.
Compound33
:
[a]2D3 =ꢀ2.2 (c=1 in CHCl3); 1H NMR (400 MHz,
Compound28 : Rf =0.58 (petroleum ether/ethyl acetate 10:1); 1H NMR
CDCl3): d=0.82 (s, 3H; C
N
N
OC(CH3)), 1.32, 140 (2brs, 6H; CH3 acetonide), 1.66 (m, 1H; 12-H),
A
(400 MHz, CDCl3): d=0.08, 0.11 (2s, 6H; Si
0.95 (d, J=6.9 Hz, 3H; CHCH3), 0.97, 1.00 (2s, 6H; C
J=6.9 Hz, 3H; CHCH3), 1.17 (s, 3H; OCCH3), 1.33, 1.46 (2s, 6H; O2C-
(CH3)2), 1.52–1.58 (m, 1H; H-13), 1.66–1.91 (m, 4H; H-6, H-12 ”2, H-
ACHTREUNG
1.80 (m, 2H; 13-H), 1.94 (ddd, 1H, J=9.8, 7.7, 3.8 Hz; 12-H’), 3.17 (m,
2H; 1-H), 3.47 (dd, 1H, J=12.6, 10.9 Hz; 14-H), 3.70 (dd, 1H, J=8.2,
6.6 Hz; 10-H), 3.85 (dd, 1H, J=9.6, 5.8 Hz; 11-H), 3.97 ppm (dd, 1H, J=
8.2, 6.8 Hz; 10-H’); 13C NMR (100 MHz, CDCl3): d=18.4 (s; OCCH3),
AHCTREUNG
ACHTREUNG
13), 1.95–2.06 (m, 1H; H-7), 2.62–2.68 (m, 1H; H-6), 2.81 (sext.q, J=6.8,
1.1 Hz, 1H; H-3), 3.72 (t, J=7.0 Hz, 1H; H-14), 3.97 (d, J=6.5 Hz, 1H;
H-10), 4.22 (dd, J=6.5, 3.0 Hz, 1H; H-9), 4.25 (dd, J=3.0, 1.5 Hz, 1H;
H-8), 4.93 (ddd, J=10.4, 1.7, 1.3 Hz, 1H; H-5), 4.96 (ddd, J=17.1, 1.7,
1.6 Hz, 1H; H-5), 4.91–5.01 (m, 2H; H-4’ ”2), 5.31 (dd, J=15.9, 6.9 Hz,
1H; H-2), 5.49 (dd, J=15.9, 1.1 Hz, 1H; H-1), 5.78 (ddd, J=17.1, 10.4,
6.5 Hz, 1H; H-4), 5.82 ppm (dddd, J=17.1, 10.1, 8.2, 6.1 Hz, 1H; H-5’);
22.6, 22.9 (2q; C
33.9 (t; C-12), 37.3 (s; C-15), 66.0 (t; C-9), 73.0 (t; C-1), 81.2 (d; C-14),
83.5 (s; C-11), 88.8 (d; C-10), 109.5 ppm (s; OC(CH3)2); IR (ATR): n˜ =
(CH3)2), 25.2, 26.4 (2q; CH3 acetonide), 27.0 (t; C-13),
AHCTREUNG
3445 (br w), 2968 (m), 2873 (m), 1456 (w), 1369 (m), 1260 (m), 1210 (m),
1155 (m), 1041 (s), 898 (w), 854 cmꢀ1 (m); HRMS-ESI: m/z: calcd for
C14H26O4Na: 281.1729 [M+Na]+; found: 281.1722.
13C NMR (100 MHz, CDCl3): d=ꢀ4.3, ꢀ3.5 (2q; Si
C
Physical and analytical details for compound 34 are described in our pre-
vious communication.[1]
(2q; C
A
G
Furanyl ketone 35: A 100 mL round-bottomed flask was charged with
acetonide 33 (434 mg, 1.68 mmol) and MeOH (20 mL). After addition of
pTsOH (63 mg, 0.34 mmol; 0.2 equiv), the solution was warmed to 508C
for 18 h and then cooled to RT. The reaction was terminated by addition
of saturated aqueous NaHCO3 solution and the organic phase was re-
moved under reduced pressure. The aqueous phase was extracted with
ethyl acetate. The combined organic phases were dried (Na2SO4) and the
crude product was purified by flash-column chromatography (ethyl ace-
tate/MeOH 100:1) to provide the desired triol (338 mg, 1.55 mmol; 92%)
36.0 (d; C-7), 36.4 (t; C-6), 38.3 (t; C-12), 39.3 (s; C-15), 40.6 (d; C-3),
73.7 (d; C-8), 83.3 (d; C-10), 83.6 (s; C-11), 84.0 (d; C-9), 86.0 (d; C-14),
106.8 (s; O2C(CH3)2), 112.6 (t; C-5), 114.9 (t; C-4’), 131.6 (d; C-2), 135.6
(d; C-1), 139.4 (d; C-5’), 143.6 ppm (d; C-4); HRMS-ESI: m/z: calcd for
C31H56O4SiNa: 543.3846 [M+Na]+; found: 543.3846.
U
Epoxyalcohols 32a and32b : A flame-dried three-neck 250 mL round-
bottomed flask equipped with a pressure equalised dropping funnel was
charged with powdered molecular sieves (4 Š, 910 mg) and dry CH2Cl2
(45 mL). The solution was cooled to ꢀ158C, and then Ti
1
U
as an oil. [a]2D3 =+1.3 (c=1 in CHCl3); H NMR (400 MHz, [D4]MeOH):
0.56 mmol) was added, followed by (ꢀ)-diethyl l-tartrate (131 mg,
0.67 mmol) in dry CH2Cl2 (5 mL) and tert-butyl hydroperoxide (3.91 mL,
22.28 mmol, 5.5m in CH2Cl2). After 30 min, the suspension was cooled to
ꢀ258C before the allylic alcohol (2.7 g, 11.14 mmol) dissolved in dry
CH2Cl2 (70 mL) was introduced dropwise by means of a dropping funnel.
The reaction was stirred for 10 h at ꢀ258C and then filtered through a
pad of Celite. The filtrate was mixed with H2O (25 mL) and warmed to
RT. Then a mixture of aqueous NaOH solution (30%, 30 mL) saturated
with NaCl was added and the mixture was vigorously stirred for 1 h. The
aqueous layer was separated and extracted with CH2Cl2. The combined
organic layers were dried over anhydrous MgSO4 and concentrated
under reduced pressure. Flash chromatography (hexanes/ethyl acetate
3:1) afforded a 4:1 mixture of the two diastereoisomeric epoxides 32a
and 32b in 51% yield as a colourless oil. The mixture was directly em-
ployed for the next step. Selected analytical data were collected and sig-
nals labelled with * refer to the minor diastereoisomer 32b. Numbering
of protons and carbon atoms are according to IUPAC nomenclature;
1H NMR (400 MHz, [D4]MeOH, TMS=0 ppm): d=0.70*, 0.98* (2s, 6H;
d=0.85, 0.86 (2s, 6H; C(CH3)2), 1.11 (s, 3H; OCCH3), 1.61 (m, 1H; 12-
G
H), 1.80 (m, 2H; 13-H), 2.09 (m, 1H; 12-H’), 3.32 (m, 2H; 1-H), 3.46 (m,
1H; 14-H), 3.57 (d, 1H, J=6.8 Hz; 8-H), 3.78 (d, 1H, J=8.2 Hz; 9-H’),
3.97 ppm (dd, 14H, J=8.2, 6.8 Hz; 10-H); 13C NMR (100 MHz,
[D4]MeOH): d=19.2 (q; C
N
G
27.2 (t; C-13), 35.0 (t; C-12), 38.7 (s; C
AHCTREUNG
1), 78.7 (d; C-14), 85.1 (s; OCCH3), 85.8 ppm (d; C-10); HRMS-ESI:
m/z: calcd for C11H22O4Na: 241.1416 [M+Na]+; found: 241.1409.
Imidazole (140 mg, 2.09 mmol; 3.5 equiv), followed by tert-butyldimethyl-
silyl chloride (216 mg, 1.43 mmol; 2.4 equiv) and 4-DMAP (7 mg,
0.12 mmol; 0.2 equiv) was added to a 20 mL round-bottomed flask charg-
ed with the deprotected triol (130 mg, 0.59 mmol) and dry CH2Cl2
(5 mL). After 10 h, the reaction was stopped by addition of saturated
aqueous NH4Cl solution. The aqueous layer was extracted once with
CH2Cl2, and then the combined organic solutions were dried (Na2SO4)
and concentrated under reduced pressure. Gel filtration over silica gel
(hexanes/ethyl acetate 10:1) afforded the TBS-protected alcohol (223 mg,
0.5 mmol, 85%) as
a
colourless oil. [a]2D3 =ꢀ4.7 (c=1 in CHCl3);
1H NMR (400 MHz, CDCl3): d=0.01, 0.07 (2s, 12H; Si
(CH3)2 ”2), 0.79,
A
C
A
H
(s, 3H; CH3 at C-3), 1.36 (2 s, 6H; CH3 acetonide), 1.35, ꢀ1.41 (m, 2H;
4-H), 1.50, ꢀ1.59 (m, 1H; 5-H), 1.80, ꢀ1.89 (m, 1H; 5-H’), 2.96 (dd, 1H,
J=6.7, 4.2 Hz; 2-H), 2.99* (dd, 1H, J=6.6, 4.2 Hz; 2-H), 3.26 (d, 1H, J=
11.4 Hz; 8-H), 3.44 (dd, 1H, J=9.7, 1.8 Hz; 6-H), 3.57 (d, 1H, J=
11.4 Hz; 8-H’), 3.67 (dd, 1H, J=12.1, 6.7 Hz; 1-H’), 3.82 ppm (dd, 1H,
J=12.1, 4.2 Hz; 1-H); 13C NMR (100 MHz, CDCl3): d=16.8, 18.3 (2q; C-
0.82 (2s, 6H; C(CH3)2, 2-Me), 0.88, 0.89 (2s, 18H; tBu ”2), 1.10 (brs,
G
3H; OCCH3), 1.58 (m, 1H; 12-H), 1.74 (m, 2H; 13-H), 2.04 (ddd, 1H,
J=18.5, 11.8, 9.3 Hz; 12-H’), 3.34, 3.28 (2d, 2H, J=9.4 Hz; 1-H), 3.56
(m, 1H; 9-H), 3.58 (dd, 1H, J=16.7, 7.2 Hz; 14-H), 3.73 (dd, 1H, J=9.4,
3.1 Hz; 9-H’), 3.78 ppm (t, 1H, J=9.4, 6.6 Hz; 10-H); 13C NMR
(100 MHz, CDCl3): d=ꢀ5.4, ꢀ5.2 (4q; Si
19.5, 21.0 (2q; C(CH3)2), 23.2 (q; OCCH3), 26.1 (2s; tBu), 26.4 (t; C-13),
34.3 (t; C-12), 38.3 (s; C(CH3)2), 64.1 (t; C-9), 70.0 (t; C-1), 77.1 (d; C-
(CH3)2 ”2), 18.4 (2s; tBu ”2),
A
AHCTREUNG
21.9*, 29.7* (2q; methyl acetonide), 24.5* (t; C-5), 24.8 (t; C-5), 33.0* (t;
C-4), 33.04 (t; C-4), 34.7* (s; C-7), 35.6 (s; C-7), 60.5* (s; C-3), 61.1* (t;
C-1), 61.5 (s; C-3), 61.6 (t; C-1), 62.4* (d; C-2), 63.3 (d; C-2), 72.2 (t; C-
8), 77.0* (d; C-6), 77.5 (d; C-6), 98.8*, 98.8 ppm (s; C-9).
AHCTREUNG
14), 83.2 (s; OCCH3), 84.4 ppm (d; C-10); HRMS-ESI: m/z: calcd for
C23H50O4Si2Na: 469.3145 [M+Na]+; found: 469.3146.
A 100 mL round-bottomed flask was charged with the TBS-protected al-
cohol (36 mg, 81 mmol) described above which was dissolved in dry
CH2Cl2 (20 mL). The solution was cooled to 08C, then Dess–Martin peri-
odinane (60 mg, 162 mmol, 2.0 equiv) was added. After addition, the reac-
tion mixture was warmed to RT and stirred for 6 h, before saturated
Tetrahydrofurans 33 and 34: A 100 mL round-bottomed flask was charg-
ed with epoxides 32a and 32b (904 mg, 3.45 mmol) and dry CH2Cl2
(37 mL).
A solution of ethylene glycol (256 mg, 4.14 mmol) in dry
CH2Cl2/THF (48 mL, 10:1) was added, followed by dry pTsOH (119 mg,
Chem. Eur. J. 2006, 12, 8719 – 8734
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8729

A. Kirschning et al.
aqueous NaHCO3 and Na2S2O3 solutions were added. The aqueous layer
was separated and extracted once with CH2Cl2. The combined organic
solutions were dried (Na2SO4) and concentrated under reduced pressure.
Flash-column chromatography (hexanes/ethyl acetate 100:1) afforded the
desired ketone 35 (35 mg, 75 mmol, 92%). [a]2D3 =+1.2 (c=1 in CHCl3);
d=0.02, 0.09 (2s, 12H; Si
(d, 3H, J=6.1 Hz; CH3CH), 0.89, 0.92 (2s, 18H; tBu ”2), 1.27 (s, 3H;
OCCH3, 6-Me), 1.72–1.97 (m, 6H; CH3 CH, CH2CH=, 12-H, 13-H), 2.54
A
N
ꢀ
(m, 1H; CH2CH=), 3.29, 3.41 (2d, 2H, J=9.4 Hz; 1-H), 3.78–3.93 (m,
3H; 14-H, 8-H), 4.88* (d, 1H, J=1.0 Hz; 9-H), 4.9 (d, 1H, J=1.3 Hz; 9-
1H NMR (400 MHz, CDCl3): d=0.02, 0.07 (2s, 12H; Si
0.83 (2s, 6H; C(CH3)2), 0.88, 0.90 (2s, 18H; tBu ”2), 1.28 (brs, 3H;
(CH3)2 ”2), 0.79,
H), 5.05 (m, 2H; =CH2), 5.78 ppm (m, 1H; CH=); 13C NMR (100 MHz,
ꢀ
ACHTREUNG
[D6]benzene=128.1 ppm): d=ꢀ5.3, ꢀ5.1 (4q, Si
G
OCCH3), 1.72 (m, 3H; 12-H, 13-H), 2.13 (m, 1H; 12-H’), 3.26, 3.42 (2d,
2H, J=9.3 Hz; 1-H), 3.78 (t, 1H, J=7.4 Hz; 14-H), 4.72, 4.66 ppm (2d,
2H, J=19.4 Hz; 9-H); 13C NMR (100 MHz, CDCl3): d=ꢀ5.0, ꢀ4.7 (4q;
CH3CH), 15.8* (q; CH3CH), 18.5, 18.8 (2s; tBu ”2), 18.5, 18.9 (2s, tBu
”2), 20.3, 20.8 (q; C(CH3)2), 25.9 (t; C-13), 26.1, 26.2 (2q; tBu ”2), 26.8
(q; OCCH3), 35.7 (t; C-12), 36.8* (d; CH3CH), 36.8 (d; CH3CH), 37.7*
(t; CH2CH=), 37.8 (t, CH2CH=), 38.8 (s; C(CH3)2), 70.4 (t; C-1), 75.2 (d;
AHCTREUNG
Si
N
N
AHCTREUNG
OCCH3), 26.4 (t; C-13), 26.4, 26.5 (2q; tBu), 36.0 (t; C-12), 38.7 (s; C-
(CH3)2), 66.6 (t; C-9), 70.2 (t; C-1), 85.4 (d; C-14), 88.0 (s; OCCH3),
C-8), 75.5* (d; C-8), 75.9* (d; C-9), 77.0 (d; C-9), 86.2* (d; C-14), 86.3
(d; C-14), 88.7 (s; OCCH3), 88.9* (s; OCCH3), 116.5 (t; =CH2), 116.7* (t;
ACHTREUNG
214.3 ppm (s; C-10); HRMS-ESI: m/z: calcd for C23H48O4Si2Na: 467.2989
ꢀ
ꢀ
=CH2), 137.3* (d; CH=), 137.5 (d; CH=), 212.3 (s; C-10), 213.3* ppm
(s; C-10); HRMS-ESI: m/z: calcd for C29H58O5Si2Na: 565.3721 [M+Na]+;
found: 565.3723.
[M+Na]+; found: 467.2981.
Furanyl ketone 36: Acetonide 34 (418 mg, 1.62 mmol) was treated by the
same method as that for the of deprotection acetonides described for the
sequence 33!35 to yield the corresponding triol (292 mg, 1.34 mmol;
Aldol product 37: A 50 mL round-bottomed flask was charged with
ketone 36 (92 mg, 241 mmol) and dry THF (10 mL). The solution was
cooled to ꢀ788C, before KHMDS (530 mL, 265 mmol, 1.1 equiv, 0.5m in
toluene) was introduced dropwise. After 30 min, aldehyde 6 in dry THF
(5 mL) was added and the solution was stirred for another 40 min. The
reaction was terminated by addition of saturated aqueous NH4Cl solu-
tion. The aqueous layer was separated and extracted with MTBE (3”
10 mL). The combined organic layers were dried (Na2SO4) and concen-
trated under reduced pressure. Flash-column chromatography (hexanes/
ethyl acetate 20:1) afforded the desired product 37 (115 mg, 212 mmol;
83%) as
a
colourless oil. [a]2D3 =+26.2 (c=1 in CHCl3); 1H NMR
(400 MHz, [D4]MeOH): d=0.88, 0.89 (s, 6H; C(CH3)2), 1.15 (s, 3H;
A
OCCH3), 1.61 (m, 1H; 12-H), 1.76 (m, 2H; 13-H), 2.08 (m, 1H; 12-H’),
3.33 (m, 2H; 1-H), 3.48 (t, 1H, J=8.4 Hz; 14-H), 3.57 (d, 1H, J=8.8 Hz;
9-H), 3.73 (d, 1H, J=6.0 Hz; 9-H’), 3.87 ppm (dd, 1H, J=8.8, 6.0 Hz; 10-
H); 13C NMR (100 MHz, [D4]MeOH): d=20.0 (p; OCCH3), 21.7, 21.8
(2q; C
N
N
9), 70.6 (t; C-1), 78.3 (d; C-14), 84.8 (d; C-10), 85.0 ppm (s; OCCH3);
HRMS-ESI: m/z: calcd for C11H22O4Na: 241.1416 [M+Na]+; found:
241.1411.
88%) as
a
colourless oil. [a]2D3 =ꢀ21.8 (c=1 in CHCl3); 1H NMR
(400 MHz, [D6]benzene, C6H6 =7.16 ppm): d=0.05, 0.09 (2s, 12H; Si-
A
U
This triol (200 mg, 0.92 mmol) was subjected to the procedure for the si-
lylation described for the sequence 33!35 to yield the corresponding al-
cohol (339 mg, 0.76 mmol; 83%) as a colourless oil.
CH3CH), 0.98, 1.01 (2s, 18H; tBu ”2), 1.41 (s, 3H; OCCH3), 1.57–2.20
(m, 6H; CH3CH, CH2CH=, 12-H, 13-H), 2.72 (m, 1H; CH2CH=), 3.33,
3.48 (2d, 2H, J=9.2 Hz; 1-H), 3.72 (t, 1H, J=6.1 Hz; 14-H), 3.81 (d, 1H,
J=8.9, 6.3 Hz; 8-H), 4.93 (m, 1H; 9-H), 5.23 (m, 2H; CH2=), 5.90 ppm
[a]2D3 =ꢀ2.5 (CHCl3 in c=1); 1H NMR (400 MHz, CDCl3): d=0.01, 0.06
(2s, 12H; Si
N
N
(m, 1H; CH=); 13C NMR (100 MHz, [D6]benzene=128.1 ppm): d=
ꢀ
tBu ”2), 1.13 (brs, 3H; OCCH3), 1.55 (m, 1H; 12-H), 1.72 (m, 2H; 13-
H), 2.03 (m, 1H; 12-H’), 3.37, 3.33 (2d, 2H, J=9.3 Hz; 1-H), 3.49 (m,
1H; 14-H), 3.60 (dd, 1H, J=10.2, 7.2 Hz; 9-H), 3.75 (dd, 1H, J=10.2,
3.6 Hz; 9-H’), 3.80 ppm (dd, 1H, J=7.2, 3.6 Hz; 10-H); 13C NMR
ꢀ5.3, ꢀ5.1 (4q; Si
ACHTREUNG
19.5, 21.2 (q; C
ACHTREUNG
AHCTREUNG
(100 MHz, CDCl3): d=ꢀ5.4, ꢀ5.2 (4q; Si
19.7, 21.1 (2q; C(CH3)2), 21.9 (q; OCCH3), 26.1 (2q; tBu), 26.4 (t; C-13),
34.7 (t; C-12), 38.3 (s; C(CH3)2), 63.9 (t; C-9), 70.0 (t; C-1), 76.9 (d; C-
(CH3)2 ”2), 18.5 (2s; tBu ”2),
OCCH3), 116.6 (t; CH2=), 137.4 (d; CH=), 212.0 ppm (s, C-10); HRMS-
ACHTREUNG
ESI: m/z: calcd for C29H58O5Si2Na: 565.3721 [M+Na]+; found: 565.3713.
AHCTREUNG
14), 82.0 (d; C-10), 83.2 ppm (s; OCCH3); HRMS-ESI: m/z: calcd for
General procedure for silylation of 1,3-diols: A round-bottomed flask
was charged with the diol (1 equiv) and dry CH2Cl2 (0.1 mmolmLꢀ1). 2,6-
Lutidine (3 equiv) was added, followed by di-tert-butylsilyl ditriflate
(1.5 equiv). After stirring for 24 h at RT the reaction was quenched by
addition of saturated aqueous NaHCO3 solution. The aqueous layer was
separated and extracted with CH2Cl2. The combined organic solutions
were washed with brine, dried (Na2SO4) and concentrated under reduced
pressure. Flash-column chromatography afforded the desired silinanes as
colourless oils.
C23H50O4Si2Na: 469.3145 [M+Na]+; found: 469.3139.
This alcohol (126 mg, 0.28 mmol) was treated by the same method as that
for the oxidation described for isomer 33!35 to yield ketone 36 (110 mg,
0.25 mmol; 89%) as a colourless oil. [a]2D3 =ꢀ17.4 (c=1 in CHCl3);
1H NMR (400 MHz, CDCl3): d=0.02, 0.08 (2s, 12H; Si
(CH3)2 ”2), 0.83,
G
0.85 (2s, 6H; C(CH3)2), 0.88, 0.91 (2s, 18H; tBu ”2), 1.29 (brs, 3H;
U
OCCH3), 1.72 (m, 3H; 12-H, 13-H), 2.12 (m, 1H; 12-H’), 3.30, 3.41 (2d,
2H, J=9.2 Hz; 1-H), 3.79 (t, 1H, J=7.4 Hz; 14-H), 4.73, 4.63 ppm (2d,
2H, J=19.6 Hz; 9-H); 13C NMR (400 MHz, CDCl3): d=ꢀ5.4, ꢀ5.2 (4q;
Dioxasilinane 38: A 50 mL round-bottomed flask was charged with aldol
product 37 (27 mg, 49 mmol)) and anhydrous CH2Cl2 (10 mL). The mix-
ture was then cooled to ꢀ788C, before DIBAL-H (150 mL, 147 mmol, 3
equiv) was added. Additional DIBAL-H (100 mL, 2 equiv) was added
after 2 h. After 5 h, the reaction mixture was allowed to warm to ꢀ308C
and was then hydrolysed with water (0.25 mL) and aqueous NaOH (4m,
0.25 mL). Na2SO4 was added and the mixture was filtered. The residue
was washed and the combined organic solutions were concentrated under
reduced pressure. Flash-column chromatography delivered the diol
(111 mg, 21 mmol, 42%), as an oil, along with recovered starting material
(13 mg, 23 mmol, 49%). [a]2D3 =ꢀ9.8 (c=1 in CHCl3); 1H NMR
Si
N
N
OCCH3), 25.9, 26.0 (4q; tBu), 26.0 (t; C-13), 36.4 (t; C-12), 38.4 (s; C-
(CH3)2), 66.6 (t; C-9), 70.0 (t; C-1), 83.3 (d; C-14), 87.6 (s; OCCH3),
ACHTREUNG
211.8 ppm (s; C-10); HRMS-ESI: m/z: calcd for C23H48O4Si2Na: 467.2989
[M+Na]+; found: 467.2979.
Aldol products 39a and 39b: A 50 mL round-bottomed flask was charged
with ketone 35 (79 mg, 178 mmol) and dry THF (8 mL). The solution was
cooled to ꢀ788C, before KHMDS (392 mL, 1.1 equiv, 0.5m in toluene)
was introduced dropwise. After 30 min, the aldehyde 6 in dry THF
(5 mL) was added and the solution was stirred for another 15 min. The
reaction was then stopped by dropwise addition of pH 7 phosphate
buffer under vigorous stirring. The flask was warmed to RT in the cold
bath, then the aqueous layer was separated and extracted with MTBE
(3”10 mL). The combined organic solutions were dried over Na2SO4 and
concentrated under reduced pressure. Flash chromatography (hexanes/
ethyl acetate 50:1) furnished an inseparable mixture (3:1) of two diaster-
eoisomeric aldol products 39a and 39b* (62 mg, 114 mmol, 64%) as a col-
ourless oil. [a]2D3 =+1.2 (c=1 in CHCl3); 1H NMR (400 MHz, CDCl3):
(400 MHz, CDCl3): d=0.03, 0.13, 0.17 (3s, 12H; Si
A
6H; C(CH3)2), 0.84 (d, J=6.8 Hz, 3H; CH3CH), 0.89, 0.91 (2s, 18H; tBu
AHCTREUNG
”2), 1.19 (s, 3H; OCCH3), 1.60 (dt, J=12.4, 8.3 Hz; H-12), 1.70–1.85 (m,
3H; H-13, H-13, H-7), 1.96 (dtt, J=13.7, 8.3, 1.0 Hz, 1H; H-6), 2.15
(ddd, J=12.4, 8.8, 4.0 Hz, 1H; H-12), 2.50 (dddt, J=13.7, 6.2, 3.2, 1.6 Hz,
1H; H-6), 2.72 (d, J=8.9 Hz, 1H; 8-OH), 2.85 (d, J=3.8 Hz, 1H; 10-
OH), 3.17 (td, J=9.2, 0.5 Hz, 1H; H-8), 3.32 (d, J=9.4 Hz, 1H; H-1),
3.38 (d, J=9.4 Hz, 1H; H-1), 3.48 (dd, J=6.8, 3.8 Hz, 1H; H-10), 3.80
8730
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8719 – 8734

Total Synthesis of Tonantzitlolone
FULL PAPER
(dd, J=6.8, 0.5 Hz, 1H; H-9), 3.83 (dd, J=9.4, 6.3 Hz, 1H; H-14), 4.99–
5.07 (m, 2H; H-4 ”2), 5.79 ppm (dddd, J=16.9, 10.4, 8.3, 6.2 Hz, 1H; H-
CH3CH), 18.4, 18.5 (2s; tBu ”2), 19.7, 19.9 (2s; tBu ”2), 20.3, 20.5 (q; C-
A
5); 13C NMR (400 MHz, CDCl3): d=ꢀ5.4, ꢀ5.4, ꢀ4.6, ꢀ3.3 (4q; Si
(CH3)2
(CH3)2),
(2q; tBu ”2), 29.9 (t; C-12), 34.4 (t; C CH3CH), 38.4 (t; CH2 CH=),
38.5 (s; C(CH3)2), 70.1 (t; C-1), 75.4 (d; C-8), 77.4 (d; C-9), 83.7 (d;
ꢀ
ꢀ
”2), 15.9 (q; CH3CH), 18.5, 18.8 (2s; tBu ”2), 20.0, 21.2 (2q; C
G
ACHTREUNG
21.6 (q; OCCH3), 26.1, 26.3 (2q; tBu ”2), 27.1 (t; C-13), 35.6 (t; C-12),
35.6 (d; C-7), 37.3 (t; C-6), 38.3 (s; C-15), 70.0 (t; C-1), 73.8 (d; C-9), 74.1
(d; C-8), 77.5 (d; C-10), 82.0 (d; C-14), 83.5 (s; C-11), 116.3 (t; C-4),
137.4 ppm (d; C-5); HRMS-ESI: m/z: calcd for C29H60O5Si2Na: 567.3877
[M+Na]+; found: 567.3878.
C-10). 84.8 (d; C-14), 86.2 (s; OCCH3), 115.9 (t; CH2=), 130.0 ppm (d;
CH=); HRMS-EI: m/z: calcd for C37H76O5Si3: 684.5001 [M]+; found:
684.5002.
Compound41
:
[a]2D0 =ꢀ7.7 (c=1 in CHCl3); 1H NMR (500 MHz,
CDCl3): d=0.01–0.15 (4s, 12H; Si(CH3)2 ”2), 0.78, 0.87 (2s, 6H; C-
A
AHCTREUNG
This diol (11 mg, 21 mmol) was treated by the same method as that de-
scribed in the general procedure for the silylation of 1,3-diols to yield sili-
nane 38 (14 mg, 20 mmol, 96%) as a colourless oil. [a]2D3 =ꢀ6.2 (c=1 in
ꢀ
ꢀ
H, 12-H, CH3CH, CH2 CH=), 2.55 (m, 1H; CH2 CH=), 3.27, 3.37 (m,
2H; 1-H), 3.57 (dd, 1H, J=10.2, 1.1 Hz; 8-H), 3.66 (dd, 1H, J=10.6,
3.3 Hz; 14-H), 3.83 (dd, 1H, J=1.1, 1.1 Hz; 9-H), 4.02 (d, 1H, J=1.1 Hz;
10-H), 4.99 (m, 2H; CH2=), 5.78 ppm (m, 1H; CH=); 13C NMR
CHCl3); 1H NMR (400 MHz, CDCl3): d=0.03, 0.03 (2s, 6H; Si
ACHTREUNG
0.12 (s, 3H; Si
A
ACHTREUNG
ACHTREUNG
(100 MHz, CDCl3): d=ꢀ5.4, ꢀ5.1 (4q; Si
ACHTREUNG
1.01, 1.06 (2, 18H; tBu ”2), 1.27 (s, 3H; OCCH3), 1.49–1.64 (m, 2H; H-
12, H-13), 1.71–1.81 (m, 2H; H-13, CH2CH=), 1.85–1.98 (m, 2H; H-12,
CH3CH), 2.69–2.76 (m, 1H; CH2CH=), 3.32 (d, 1H, J=9.5 Hz; H-1),
3.50 (d, 1H, J=9.5 Hz; H-1), 3.94 (d, 1H, J=8.9 Hz; H-8), 3.97–4.07 (m,
2H; H-9, H-14), 4.35 (d, 1H, J=4.6 Hz; H-10), 4.94–5.04 (m, 2H; CH2=),
18.4, 18.5 (2s; tBu ”2), 20.1, 20.4 (2s, tBu ”2), 20.3, 20.5 (2q; C
ACHTREUNG
26.0 (t; C-13), 26.0, 26.1 (2q; tBu ”2), 26.2 (q; OCCH3), 27.4, 27.6 (2q;
tBu ”2), 29.8 (t; C-12), 34.3 (d; CH3CH), 38.3 (t; CH2 CH=), 38.5 (s; C-
ꢀ
A
5.79 ppm (dddd, 1H, J=17.1, 10.1, 8.5, 5.6 Hz, CH=); 13C NMR
ꢀ
C-14), 86.0 (s; OCCH3), 116.9 (t; CH2=), 125.7 ppm (d; CH=); HRMS-
EI: m/z: calcd for C37H76O5Si3: 684.5001 [M]+; found: 684.5004.
(100 MHz, CDCl3): d=ꢀ5.4, ꢀ5.3, ꢀ4.5, ꢀ2.0 (4q; Si
G
CH3CH), 18.4, 18.5 (2s; tBu ”2), 19.9, 21.2 (2q; C
(CH3)2), 20.9, 21.4 (2s;
9,10-Dihydroxy-8-O-(tert-butyldimethylsilyl)ether 44 and 8,9-dihydroxy-
10-O-(tert-butyldimethylsilyl)ether 45: TBS-ether 15 (19 mg, 39 mmol) in
THF (1 mL) was added to a suspension of washed NaH (60% in paraffin
oil, 3 mg, 75 mmol, 1.9 equiv) in THF (1 mL) at 08C. The reaction mix-
ture was stirred at this temperature for 15 min until aqueous NH4Cl
(0.5 mL) was added. Direct purification by column chromatography over
silica gel (petroleum ether/ethyl acetate 40:1) afforded the two regioiso-
meric TBS ethers 44 (10 mg, 20.8 mmol, 53%) and 45 (7.5 mg, 15.6 mmol,
39%) as colourless oils.
tBu ”2), 22.5 (q; OCCH3), 26.0 (t; C-13), 26.0, 26.3 (2q; tBu ”2), 27.9,
28.4 (2q; tBu ”2), 34.3 (t; C-12), 36.4 (d; CH3CH), 38.4 (t; CH2CH=),
38.5 (s; C(CH3)2), 70.1 (t; C-1), 69.7 (d; C-10), 70.8 (d; C-8), 80.5 (d; C-
R
ꢀ
9), 81.5 (d; C-14), 85.0 (s; OCCH3), 116.2 (t; CH2=), 137.2 ppm (d, CH=);
HRMS-EI: m/z: calcd for C37H76O5Si3: 684.5001 [M]+; found: 684.5004.
Dioxasilinanes 40 and41 : A mixture of both diastereoisomeric ketones
39a and 39b (21 mg, 38 mmol) in a mixture of THF/MeOH (7 mL, 3:1)
was cooled to ꢀ788C. Et2BOMe (6 mL, 42 mmol, 1.1 equiv) in THF
(1 mL) was then added and the reaction mixture was stirred for 15 min,
before adding NaBH4 (4.7 mg, 125 mmol, 3.3 equiv) suspended in 5 mL
THF. After stirring for 8 h, the reaction mixture was warmed to 08C and
hydrolysed by dropwise addition of an aqueous solution of AcOH (1n,
10 mL). The mixture was extracted with ethyl acetate (3”10 mL). The
combined organic layers were dried (Na2SO4) and concentrated under re-
duced pressure. Purification by column chromatography (hexanes/ethyl
acetate 50:1) yielded both diols (14 mg, 26 mmol, 67%) as a colourless
Compound44
[a]2D3 =+12.0 (c=1.0 in CHCl3); 1H NMR (400 MHz, CDCl3): d=0.09,
0.12 (2s, 6H; Si(CH3)2), 0.92 (s, 9H; tBu), 0.96, 1.03 (2s, 6H; C(CH3)2),
: TLC: Rf =0.44 (petroleum ether/ethyl acetate 10:1);
T
ACHTREUNG
ꢀ
0.98 (d, J=6.9 Hz, 3H; CH3CH), 1.09 (d, J=6.9 Hz, 3H; CH3 CH), 1.21
(s, 3H; OCCH3), 1.46–1.60 (m, 1H; CH3CH at C-7), 1.62–1.84 (m, 3H;
ꢀ
H-7, H-13 ”2), 1.86–1.97 (m, 1H; CH2 CH= at C-6), 2.08–2.22 (m, 2H;
H-6, H-12), 2.69 (d, J=9.4 Hz, 1H; 10-OH), 2.83 (pshex., J=6.7 Hz, 1H;
H-3), 3.20 (d, J=2.5 Hz, 1H; 9-OH), 3.31 (d, J=9.4 Hz, 1H; H-10), 3.64
(dd, J=6.6, 2.5 Hz, 1H; H-8), 3.71 (dd, J=8.5, 6.5 Hz, 1H; H-14), 3.88
(dd, J=6.6, 2.5 Hz, 1H; H-9), 4.90–5.04 (m, 4H; H-5, H-4’), 5.36 (dd, J=
16.0, 6.3 Hz, 1H; H-2), 5.45 (d, J=16.0 Hz, 1H; H-1), 5.79 (ddd, J=17.1,
10.4, 6.6 Hz, 1H; H-4), 5.70–5.80 ppm (m, 1H; H-5’); 13C NMR
1
oil. The data labelled with * refer to the minor diastereoisomer. H NMR
(400 MHz, CDCl3): d=0.01–0.17 (m, 12H; Si
A
6H; C(CH3)2), 0.82 (d, 3H, J=6.4 Hz; CH3CH), 0.88–0.92 (4s, 18H; tBu
ACHTREUNG
ꢀ
”2), 1.26 (s, 3H; OCCH3), 1.50–2.00 (m, 6H; 12-H, 13-H, CH3CH, CH2
ꢀ
CH=), 2.62 (m, 1H; CH2 CH=), 3.26, 3.36 (m, 2H; 1-H), 3.59 (m, 1H; 9-
(100 MHz, CDCl3): d=ꢀ4.5, ꢀ3.9 (2q; Si
24.7 (5q; OCCH3, CHCH3 ”2, C(CH3)2), 18.5 (s; tBu), 26.2 (q; SiC-
(CH3)3), 27.0 (t; C-13), 34.4 (t; C-12), 35.5 (t; C-6), 35.9 (d; C-7), 38.9 (s;
(CH3)2), 17.3, 20.0, 23.5, 24.3,
H), 3.65 (t, 1H, J=4.3 Hz; 14-H), 3.79 (dd, 1H, J=11.1, 3.2 Hz; 8-H), 4.0
AHCTREUNG
ꢀ
(d, 1H, J=3.5 Hz; 10-H), 4.99 (m, 2H; =CH2), 5.78 ppm (m, 1H; CH=);
AHCTREUNG
13C NMR (100 MHz, CDCl3): d=ꢀ5.4, ꢀ5.2 (4q; Si
(CH3)2 ”2), 15.5* (q;
C-15), 40.5 (d; C-3), 71.2 (d; C-9), 74.4 (d; C-10), 78.0 (d; C-8), 86.1 (s;
C-11), 86.2 (d; C-14), 112.7 (t; C-5), 115.5 (t; C-4’), 132.5 (d; C-2), 134.7
(d; C-1), 138.0 (d; C-5’), 143.2 ppm (d; C-4); IR (ATR): n˜ =3479 (brs),
3077 (w), 2960 (s), 2930 (s), 2857 (m), 1731 (w), 1639 (w), <1462 (m),
1414 (w), 1387 (m), 1362 (w), 1251 (m), 1088 (s), 1060 (m), 993 (w), 912
(m), 888 (w), 836 (m), 777 (m), 673 cmꢀ1 (w); HRMS-ESI: m/z: calcd for
C28H52O4SiNa: 503.3533 [M+Na]+; found: 503.3540.
CH3CH), 15.8 (q; CH3CH), 18.4, 18.5 (2s; tBu ”2), 18.5*, 18.6* (2s; tBu
”2), 19.6, 19.7 (q; C
A
(q; OCCH3), 26.4* (q; OCCH3), 30.5 (t; C-12), 30.6* (t; C-12), 34.3* (d;
ꢀ
ꢀ
CH3CH), 34.4 (d; CH3CH), 36.8* (t; CH2 CH=), 36.9 (d; CH2 CH=),
38.9 (s; C(CH3)2), 70.0 (t; C-1), 75.4 (d; C-8), 75.4* (d; C-8), 77.1* (d; C-
ACHTREUNG
9), 77.2 (d; C-9), 83.2* (d; C-10), 83.3 (d; C-10). 84.4 (t; C-14), 84.5* (d;
C-14), 88.7 (s; OCCH3), 88.9* (s; OCCH3), 116.9* (t; CH2=), 115.9 (t;
CH2=), 125.7* (d; CH=), 130.0 ppm (d; CH=); HRMS-ESI: m/z: calcd
for C29H60O5Si2Na: 567.3877 [M+Na]+; found: 567.3879.
Compound45
1H NMR (200 MHz, CDCl3, CHCl3): d=0.15, 0.20 (2s, 6H; Si
0.92 (s, 9H; tBu), 0.96, 0.98 (2s, 6H; C(CH3)2), 0.99 (d, J=6.8 Hz, 3H;
:
TLC: Rf =0.24 (petroleum ether/ethyl acetate 10:1);
ACHTREUNG
AHCTREUNG
A mixture of both diastereoisomeric diols (14 mg, 26 mmol) was treated
by the same method as that for the silylation to yield a mixture of sili-
nanes 40 and 41 (4:1, 13 mg, 19 mmol, 73%) as a colourless oil.
CH3CH), 1.07 (d, J=6.9 Hz, 3H; CH3CH), 1.16 (s, 3H; OCCH3), 1.36–
1.76, 1.80–2.07 (m, 6H; H-6, H-7, H-12, H-13), 2.23–2.39 (m, 1H; H-6),
2.71–2.90 (m, 1H; H-3), 2.87 (d, J=4.3 Hz, 1H; 8-OH), 3.09 (d, J=
7.7 Hz, 1H; 9-OH), 3.31 (dd, J=9.5, 5.3 Hz, 1H; H-8), 3.51 (ddd, J=7.7,
5.3, 2.1 Hz, 1H; H-9), 3.62 (d, J=2.1 Hz, 1H; H-10), 3.75 (dd, J=8.1,
5.8 Hz, 1H; H-14), 4.88–5.10 (m, 4H; H-5, H-4’), 5.32 (dd, J=15.9,
5.9 Hz, 1H; H-2), 5.45 (d, J=15.9, 1H; H-1), 5.67–5.90 ppm (m, 2H; H-
4, H-5’); IR (ATR): n˜ =3482 (br, m), 3077 (w), 2959 (s), 2930 (s), 2859
(m), 1639 (w), 1463 (m), 1385 (m), 1251 (m), 1110 (m), 1082 (w), 1027
(m), 993 (m), 911 (m), 836 (m), 780 (m), 683 cmꢀ1 (w).
Compound40
:
[a]2D0 =ꢀ2.0 (c=1 in CHCl3); 1H NMR (500 MHz,
CDCl3): d=0.01, (2s, 6H; Si
(2s, 6H; C
18H; tBu ”2), 1.00–1.07 (2s, 18H; tBu ”2), 1.18 (s, 3H; OCCH3), 1.56–
G
G
A
ꢀ
ꢀ
2.20 (m, 6H; 13-H, 12-H, CH3CH, CH2 CH=), 2.67 (m, 1H; CH2 CH=),
3.29, 3.38 (m, 2H; 1-H), 3.56 (dd, 1H, J=2.3, 4.5 Hz; 8-H), 3.67 (dd, 1H,
J=10.5, 3.7 Hz; 14-H), 3.83 (dd, 1H, J=2.3, 2.2 Hz; 9-H), 3.96 (d, 1H,
J=2.2 Hz; 10-H), 4.99 (m, 2H; CH2=), 5.77 ppm (m, 1H; CH=);
Macrocyclic silyl ether 50: A 10 mL round-bottomed flask was charged
with TBS alcohol 27 (2.9 mg, 5.57 mmol) and dry CH2Cl2 (2 mL). Grubbs
13C NMR (100 MHz, CDCl3): d=ꢀ5.4, ꢀ5.1 (4q; Si
(CH3)2 ”2), 14.3 (q;
Chem. Eur. J. 2006, 12, 8719 – 8734
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8731

A. Kirschning et al.
II catalyst (0.5 mg, 0.6 mmol) was added in dry CH2Cl2 (1 mL) and the
solution was refluxed under nitrogen for 4.5 h and then concentrated.
Flash chromatography over silica gel (hexanes/ethyl acetate 20:1) afford-
ed the desired product 50 (1.7 mg, 3.44 mmol, 62%), as a mixture of E/Z
isomers (E/Z 1:2.2). 1H NMR (400 MHz, CDCl3): d=0.83 (s, 3H; H-17),
0.87 (d, J=6.8 Hz, 3H; H-19), 0.87 (s, 3H; H-16), 0.89 (s, 9H; Si-
(s), 923 (w), 885 (m), 854 (m), 805 (m), 718 cmꢀ1 (m); HRMS-ESI: m/z:
calcd for C25H41NO4Na: 442.2933 [M+Na]+; found: 442.2920.
Macrocyclic 8,10-O-(triethylsilyl) ether 53: Grubbs II catalyst (10 mg,
11.8 mmol, 12.5 mol%, c=0.0038m) was added to a solution of acyclic
8,10-bis-silylated ketone 49 (56 mg, 94 mmol) in absolute CH2Cl2 (25 mL)
at 08C. After 2 h, the solution was saturated with air, filtered over neu-
tral Al2O3 and concentrated. Column chromatography (petroleum ether/
ethyl acetate 60:1) afforded the two inseparable (E/Z 1:1) macrocycles 53
(51.5 mg, 91 mmol; 97%) as colourless oils. TLC: Rf =0.65 (petroleum
ether/ethyl acetate 40–60:1); the interpretation of the 1H NMR spectra
was hampered due to line broadening for which we make a slow confor-
mational equilibrium of the macrocycle responsible; 13C NMR (100 MHz,
CDCl3): d=5.05, 5.11, 6.79, 6.90, 15.58, 21.74, 24.58, 25.88, 26.22, 28.08,
29.68, 30.63, 32.24, 33.81, 37.78, 41.74, 78.69, 85.65, 85.69, 86.33, 87.25,
128.82, 132.30, 132.84, 133.70, 135.25, 136.37 ppm (all observed signals of
the mixture are listed as detected and are not comprehensive. Character-
istically, no CH2 signals were detected with the DEPT-135 method re-
flecting the absence of terminal olefins); HRMS-ESI: m/z: calcd for
C32H60O4Si2Na: 587.3928 [M+Na]+; found: 587.3957. Further analytical
evidence for the macrocyclic product 53 was collected from the derivative
60 collected after mCPBA oxidation (unknown configuration around the
oxirane ring).
ACHTREUNG
A
ACHTREUNG
1H; H-3), 3.71 (d, J=5.7 Hz, 1H; H-10), 4.10 (dd, J=9.9, 1.4 Hz, 1H; H-
8), 4.19 (dd, J=9.9, 5.7 Hz, 1H; H-9), 5.38 (d, J=16.1 Hz, 1H; H-1), 5.43
(dd, J=16.1, 5.5 Hz, 1H; H-2), 5.45 (dt, J=11.2, 8.2 Hz, 1H; H-5),
5.55 ppm (dd, J=11.2, 8.4 Hz, 1H; H-4); HRMS-EI calcd for C29H52O4Si:
492.3635 [M]+; found: 492.3636.
Trienol 51: A 5 mL round-bottomed flask was charged with TBS alcohol
26 (16.5 mg, 32 mmol) and THF (1 mL). TBAF·3H2O (56 mg, 0.18 mmol,
5.6 equiv) was added in THF (0.5 mL) and the solution was stirred for
19.5 h. Silica gel was added and the mixture was concentrated. A gel fil-
tration over silica gel (hexanes/ethyl acetate 40:1) afforded the desired
product 51 (11.6 mg, 29 mmol, 91%) as a colourless oil. [a]2D3 =ꢀ14.6 (c=
0.9 in CHCl3); 1H NMR (400 MHz, CDCl3): 0.94 (d, J=6.5 Hz, 3H; H-
19), 0.99 (s, 3H; H-17), 1.00 (s, 3H; H-16), 1.06 (d, J=6.8 Hz, 3H; H-20),
1.26 (s, 3H; H-18), 1.37 (s, 3H; O2C
N
N
1.52 (ddd, J=11.9, 8.4, 7.9 Hz, 1H; H-12), 1.68–1.77 (m, 2H; H-13 ”2),
1.88 (dtt, J=13.6, 8.5, 0.9 Hz, 1H; H-6), 2.03 (dddq, J=9.6, 8.5, 6.5,
3.8 Hz, 1H; H-7), 2.09 (ddd, J=11.9, 8.0, 6.0 Hz, 1H; H-12), 2.35 (dddt,
J=13.6, 5.8, 3.8, 1.9 Hz, 1H; H-6), 2.81 (sext.q, J=6.9, 1.2 Hz, 1H; H-3),
3.25 (d, J=9.6 Hz, 1H; H-8), 3.47 (d, J=1.0 Hz, 1H; H-10), 3.71 (dd, J=
8.2, 6.8 Hz, 1H; H-14), 3.83 (dt, J=2.7, 1.4 Hz, 1H; H-9), 4.08 (d, J=
2.7 Hz, 1H; 9-OH), 4.91 (dt, J=10.4, 1.6 Hz, 1H; H-5), 4.95 (dt, J=17.1,
1.6 Hz, 1H; H-5), 4.98–5.08 (m, 2H; H-4’ ”2), 5.28 (dd, J=15.7, 7.2 Hz,
1H; H-2), 5.50 (dd, J=15.7, 1.0 Hz, 1H; H-1), 5.78 (ddd, J=17.1, 10.4,
6.5 Hz, 1H; H-4), 5.80 ppm (dddd, J=17.0, 10.2, 8.4, 5.8 Hz, 1H; H-5’);
The inseparable metathesis products 53 (39 mg, 69 mmol) were dissolved
in CH2Cl2 (1.2 mL) and added to a suspension of mCPBA (43 mg,
174 mmol; 2.5 equiv) in CH2Cl2 (3 mL) at ꢀ208C. After 2 h, the mixture
was warmed to 108C and a saturated aqueous NaHCO3 solution (3 mL),
a saturated aqueous Na2S2O3 solution (3 mL) and CH2Cl2 (10 mL) were
added. The layers were separated and the aqueous phase was twice ex-
tracted with CH2Cl2. The combined organic extracts were dried (Na2SO4)
and concentrated under reduced pressure. GC analysis of the crude prod-
uct (39 mg) revealed three main products, of which only macrocycle 60
(4 mg, 6.9 mmol, 10%) could be isolated by flash-column chromatography
(hexanes/ethyl acetate 40:1) as a colourless oil. TLC: Rf =0.23 (petro-
13C NMR (100 MHz, CDCl3): 15.8 (q; C-19), 19.1 (q; O2C
(q; C-20), 23.4 (q; C-17), 23.8 (q; C-16), 24.6 (q; C-18), 27.1 (t; C-13),
29.9 (q; O2C(CH3)2), 32.7 (d; C-7), 35.4 (t; C-12), 36.4 (t; C-6), 39.2 (s;
C-15), 40.7 (d; C-3), 63.6 (d; C-9), 75.6 (d; C-10), 77.4 (d; C-8), 84.9 (s;
C-11), 87.0 (d; C-14), 99.0 (s; O2C(CH3)2), 112.5 (t; C-5), 116.2 (t; C-4’),
(CH3)2), 20.4
ACHTREUNG
AHCTREUNG
131.4 (d; C-2), 136.2 (d; C-1), 136.9 (d; C-5’), 143.6 ppm (d; C-4); IR: n˜ =
3473 (m), 2964 (m), 2930 (w), 2871 (m), 1639 (m), 1456 (m), 1377 (m),
1259 (m), 1201 (m), 1169 (m), 1091 (m), 1059 (w), 1024 (m), 983 (s), 909
(s), 861 (m), 813 (m), 721 (w), 667 (w), 642 cmꢀ1 (w); HRMS-ESI: m/z:
calcd for C25H42O4Na: 429.2981 [M+Na]+; found: 429.2523.
Macrocyclic alcohol 52: A 25 mL round-bottomed flask was charged with
alcohol 51 (6.4 mg, 15.7 mmol) and dry CH2Cl2 (4 mL). Grubbs II catalyst
(1.4 mg, 1.6 mmol) was added in dry CH2Cl2 (3 mL) and the solution was
heated under refluxing conditions under nitrogen for 5.5 h and then con-
centrated under reduced pressure. Flash chromatography over silica gel
(hexanes/ethyl acetate 40:1) afforded the desired product 52 (4.7 mg,
12.4 mmol, 79%) as a mixture of E/Z isomers (E/Z 1.3:1). The E isomer
was isolated by flash chromatography over silica gel (hexanes/ethyl ace-
tate 100:1). [a]2D3 =ꢀ14.5 (c=0.94 in CHCl3); 1H NMR (400 MHz,
leum ether/ethyl acetate 40:1); [a]2D3 =ꢀ16.1 (c=0.4 in CHCl3); 1H NMR
(400 MHz, CDCl3): d=0.55–0.63 (m, 9H; Si(CH2CH3)3, H-19), 0.64–0.72
A
(m, 6H; Si
N
N
(d, J=7.0 Hz, 3H; H-20), 1.17, 1.25 (2s, 6H; H-16, H-17), 1.26–1.40,
1.60–1.80 (m, 7H; H-6, H-7, H-12, H-13), 2.49 (d, J=2.5 Hz, 1H; H-1),
2.79 (d, J=2.5 Hz, 1H; H-2), 2.91–3.01 (m, 1H; H-3), 3.58 (dd, J=10.7,
4.5 Hz, 1H; H-14), 4.06 (brs, 1H; H-10), 5.04 (brs, 1H; H-8), 5.20 (t, J=
11.1 Hz, 1H; H-4), 5.33 ppm (dt, J=11.1, 3.3 Hz, 1H; H-5); 13C NMR
CDCl3): 0.91, 1.10 (2s, 6H; C
1.00 (d, J=6.8 Hz, 3H; CHCH3), 1.19 (s, 3H; OCCH3), 1.31 (ddd, J=
12.5, 11.6, 8.2 Hz, 1H; H-12), 1.39 (s, 3H; O2C(CH3)2), 1.40 (s, 3H; O2C-
(CH3)2), 1.57–1.70 (m, 2H; H-13 ”2), 1.82–1.95 (m, 3H; H-6, H-7, 9-
(CH3)2), 0.92 (d, J=6.1 Hz, 3H; CHCH3),
(100 MHz, CDCl3): d=5.0, 5.1 (t; Si
A
AHCTREUNG
AHCTREUNG
ACHTREUNG
OH), 2.22 (dt, J=13.3, 2.7 Hz, 1H; H-6), 2.52 (ddd, J=12.6, 8.4, 1.5 Hz,
1H; H-12), 2.78–2.87 (m, 1H; H-3), 3.27 (d, J=9.2 Hz, 1H; H-8), 3.50
(d, J=1.0 Hz, 1H; H-10), 3.53 (dd, J=10.6, 5.1 Hz, 1H; H-14), 4.07 (d,
J=10.9 Hz, 1H; H-9), 5.30 (ddd, J=15.2, 9.5, 2.5 Hz, 1H; H-4), 5.37 (dd,
J=16.0, 5.8 Hz, 1H; H-2), 5.49 (ddd, J=15.2, 10.1, 2.7 Hz, 1H; H-5),
5.57 ppm (dd, J=16.0, 1.4 Hz, 1H; H-1); 13C NMR (100 MHz, CDCl3):
C-16, C-17), 29.7, 30.6, 30.9, 32.5, 36.1, 41.1, 58.3 (d; C-1), 59.6 (d; C-2),
77.2 (d; C-8), 78.3 (d; C-10), 85.7 (d; C-14), 86.6 (s; C-11), 129.8 (d; C-5),
130.7 ppm (d; C-4). C-3, 6, 7, 12, 13 and 15 could not be assigned precise-
ly, C-9 was not detected; IR (ATR): n˜ =2958 (s), 2912 (w), 2877 (s), 1724
(m), 1458 (m), 1415 (w), 1384 (m), 1325 (w), 1297 (w), 1239 (m), 1124
(s), 1101 (s), 1054 (w), 1005 (m), 975 (w), 921 (m), 896 (m), 872 (w), 846
(m), 808 (m), 795 (m), 744 (s), 687 cmꢀ1 (w); HRMS-ESI: m/z: calcd for
C32H60O5Si2Na: 603.3877 [M+Na]+; found: 603.3865.
19.2 (q; O2C
(q; C-16), 28.1 (t; C-13), 28.6 (q; C-18), 30.1 (q; O2C
ACHTREUNG
AHCTREUNG
12), 33.0 (d; C-7), 38.3 (s; C-15), 38.7 (t; C-6), 40.9 (d; C-3), 63.2 (d; C-
9), 77.6 (d; C-10), 80.8 (d; C-8), 82.3 (s; C-11), 86.6 (d; C-14), 99.4 (s;
Macrocyclic 8-keto-9,10-diol (55b): Experimental details for the synthesis
of the target olefin 55a are described in our previous communication.[1]
When the workup was conducted according to the procedure described
above for triethylsilylated macrocycles 53, with the exception that basic
O2C(CH3)2), 130.1 (d; C-5), 132.9 (d; C-1), 133.4 (d; C-2), 133.9 ppm (d;
G
C-4); IR: n˜ =3528 (w), 2962 (m), 2923 (m), 2866 (m), 1453 (m), 1380 (m),
1297 (w), 1260 (m), 1201 (m), 1156 (m), 1060 (s), 1027 (m), 1003 (m), 966
8732
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8719 – 8734

Total Synthesis of Tonantzitlolone
FULL PAPER
instead of neutral Al2O3 was employed, partial tautomerisation towards
55b was observed.
der Chemischen Industrie. Additional support was kindly made possible
by a graduate fellowship provided by the state of Lower Saxony for R.W.
We are grateful to AnalytiCon Discovery (Potsdam) for an authentic
sample of tonantzitlolone 1 and to Solvay Pharmaceutical Research Lab-
oratories for the donation of chemicals. We are indebted to G.J. for help-
ful discussions.
Compound55b : TLC: Rf =0.13 (petroleum ether/ethyl acetate 10:1);
[a]2D3 =+48.2 (c=1.0, CHCl3); 1H NMR (400 MHz, CDCl3): d=0.88, 1.05
(2 s, 6H; H-16, H-17), 1.03 (d, J=6.7 Hz, 3H; H-20), 1.09 (s, 3H; H-18),
1.17 (d, J=7.0 Hz, 3H; H-19), 1.74–1.95 (m, 4H; H-6, H-12 ”2, H-13),
1.97–2.09 (m, 2H; H-6, H-13), 2.47–2.57 (m, 1H; H-7), 2.67–2.80 (m, 2H;
H-3, 8-OH), 3.54 (brs, 1H; 10-OH), 3.81 (dd, J=8.0, 3.0 Hz, 1H; H-14),
4.35 (brs, 1H; H-8), 4.57 (brs, 1H; H-10), 5.16–5.24 (m, 1H; H-5), 5.27
(dd, J=15.4 Hz, 7.9 Hz, 1H; H-2), 5.33 (d, J=15.4 Hz, 1H; H-1),
5.40 ppm (dd, J=15.5, 8.8 Hz, 1H; H-4); 13C NMR (100 MHz, CDCl3):
d=20.4 (q; C-19), 21.0 (q; C-18), 21.3 (q; C-20), 24.2, 25.7 (q; C-16, C-
17), 26.7 (t; C-13), 32.3 (t; C-12), 34.6 (t; C-6), 39.6 (s; C-15), 40.3 (d; C-
7), 42.0 (d; C-3), 77.9 (d; C-8), 81.8 (d; C-10), 85.4 (s; C-11), 88.2 (d; C-
14), 127.8 (d; C-2), 132.7 (d; C-5), 133.8 (d; C-1), 137.2 (d; C-4),
212.4 ppm (s; C-9); IR (ATR): n˜ =3477 (br), 2961 (m), 2930 (w), 2871
(m), 1707 (m), 1454 (m), 1374 (m), 1327 (m), 1239 (m), 1156 (w), 1121
(m), 1074 (m), 1059 (s), 1024 (s), 982 (s), 910 (w), 883 (w), 829 (w), 769
(w), 726 (w), 666 cmꢀ1 (w); HRMS-ESI: m/z: calcd for C20H32O4Na:
359.2198 [M+Na]+; found: 359.2202.
[1]C. Jasper, R. Wittenberg, M. Quitschalle, J. Jakupovic, A. Kirschn-
ing, Org. Lett. 2005, 7, 479–482.
[2]a) X. A. Dominguez, H. V. Sanchez, E. G. Gomez Lopez, G. Dräger,
E. Kunst, A. Kirschning, F. Tsichritzis, F. Jeske, J. Jakupovic, unpub-
lished results; b) F. Jeske, PhD Thesis, TU Berlin (Germany), 1997.
[3]J. Jakupovic, F. Sasse, unpublished results.
[4]We are indebted to Dr. Florenz Sasse, Gesellschaft für Biotechnolo-
gische Forschung, Braunschweig (Germany), for performing biologi-
cal tests with tonantzitlolone and derivatives.
[5]P. R. Blakemore, J. Chem. Soc. Perkin Trans. 1 2002, 2563–2585 and
references therein.
Macrocyclic tetrols 56a/56b: Procedure A: NMO (15 mg, 128 mmol,
2.5 equiv) and MSA (17 mg, 179 mmol, 3.5 equiv) were added to a solu-
tion of olefin 55a (17 mg, 50.5 mmol) in tBuOH (2 mL) and H2O (1 mL).
Then the solution was cooled to 08C and a solution of OsO4 in tBuOH
(2.5 wt%, 40 mL, 32.4 mg, 3.2 mmol, 6.3 mol%) was added dropwise. The
solution was slowly warmed to RT and after 3.5 h, terminated by addition
of Na2S2O3·5H2O (46 mg, 185 mmol, 3.7 equiv). After 10 min, the aqueous
layer was separated and extracted twice with CH2Cl2. The combined or-
ganic solutions were dried (Na2SO4) and concentrated under reduced
pressure. Purification by flash-column chromatography (hexanes/EtOAc
1:1) afforded the inseparable mixture (1.3:1) of tetraols 56a and 56b
(16 mg, 43.2 mmol, 86%) as a colourless oil. TLC: Rf =0.16 (petroleum
ether/ethyl acetate 1:1); [a]D20 =+38.8 (c=1.0 in CHCl3); IR (ATR): n˜ =
3335 (s, br), 2964 (m), 2929 (m), 2871 (m), 1716 (m), 1455 (m), 1386 (m),
1326 (s), 1154 (s), 1098 (w), 1052 (m), 1025 (s), 991 (s), 914 (w), 878 (w),
827 (w), 764 (w), 667 cmꢀ1 (w); HRMS-ESI: m/z: calcd for C20H34O6Na:
393.2253 [M+Na]+; found: 393.2258.
[6]a) D. A. Evans, M. D. Ennis, D. J. Mathre, J. Am. Chem. Soc. 1982,
104, 1737–1739; b) D. A. Evans, S. L. Bender, J. Morris, J. Am.
Chem. Soc. 1988, 110, 2506–2526. Because of its volatility, aldehyde
6 was used in excess to ensure complete conversion of the enolate
moiety.
[7]Throughout the text and the graphics the numbering of carbon
atoms in substructures correspond to the numbering used for to-
nantzilolone.
[8]Elucidation of the absolute configuration was not possible before
ring-closing metathesis reaction and formation of a macrocycle of
type 3 had been conducted (see also reference [1]).
[9]S.-i. Kiyooka, H. Kuroda, Y. Shimasaki, Tetrahedron Lett. 1986, 27,
3009–3012. Elucidation of the 8,10-syn relationship in 15 was ach-
ieved by NMR spectroscopic analysis of the acetonide 16 (see refer-
ences [1, 15 and 16]and discussion of the following section).
[10]Schinzer and co-workers disclosed a related aldol reaction as part of
the total syntheses of epothilones A and B with identical preference
for the anti-Felkin–Anh selectivity: a) D. Schinzer, A. Bauer, O. M.
Bçhm, A. Limberg, M. Cordes, Chem. Eur. J. 1999, 5, 2483–2491;
b) D. Schinzer, A. Bauer, J. Schieber, Chem. Eur. J. 1999, 5, 2483–
2491.
1
Compound56a : H NMR (500 MHz, [D4]MeOH): d=0.91, 1.17 (2s, 6H;
H-16, H-17), 1.04 (d, J=7.0 Hz, 3H; H-20), 1.03–1.10 (m, 1H; H-6), 1.07
(d, J=7.0 Hz, 3H; H-19), 1.29 (s, 3H; H-18), 1.57 (ddd, J=12.5, 8.7,
6.7 Hz, 1H; H-12), 1.80–1.93 (m, 2H, H-6; H-13), 1.96–2.08 (m, 1H; H-
13), 2.13–2.30 (m, 3H; H-3, H-7, H-12), 3.15 (dd, J=7.6, 2.4 Hz, 1H; H-
4), 3.37 (ddd, J=10.9, 7.6, 1.7 Hz, 1H; H-5), 3.75 (dd, J=8.7, 6.8 Hz, 1H;
H-14), 4.46 (brs, 2H; H-8, H-10), 5.47 (dd, J=15.9, 8.6 Hz, 1H; H-2),
5.53 ppm (d, J=15.9 Hz, 1H; H-1); 13C NMR (125 MHz, [D4]MeOH):
d=18.4 (q; C-19), 19.5 (q; C-20), 25.6 (q; C-18), 27.2–27.6 (q, q, t; C-16,
C-17, C-13), 33.2 (t; C-12), 34.1 (t; C-6), 36.3 (d; C-7), 39.5 (s; C-15), 41.9
(d; C-3), 73.3 (d; C-5), 76.2 (d; C-8), 78.7 (d; C-10), 79.5 (d; C-4), 85.4 (s;
C-11), 88.3 (d; C-14), 130.6 (d; C-2), 136.9 (d; C-1), 212.1 ppm (s; C-9).
[11]In our first synthetic endeavour, we tested other enolisation proto-
cols (LDA, Bu2BOTf/Et3N, Cy2BCl/Et3N, TiCl4/(iPr2)EtN and
R
Me2AlCl/Et3N), which all proved to be insufficient: R. Wittenberg,
H. Monenschein, G. Dräger, C. Beier, G. Jas, C. Jasper, A. Kirschn-
ing, Tetrahedron Lett. 2004, 45, 4457–4460.
[12]a) A. Mengel, O. Reiser, Chem. Rev. 1999, 99, 1191–1223.
[13]a) W. R. Roush, J. Org. Chem. 1991, 56, 4151–4157; b) C. Gennari,
S. Vieth, A. Comotti, A. Vulpetti, J. M. Goodmann, I. Paterson, Tet-
rahedron 1992, 48, 4439–4458.
[14]Addition of 18-crown-6 to a potassium enolate led to reduced dia-
stereoselectivity (10:1). Crown ethers have cavities with different
space which are able to size-select cations. To some extent chelate
22 resembles a cavity of a crown ether.
[15]D. A. Evans, D. L. Rieger, J. R. Gage, Tetrahedron Lett. 1990, 31,
7099–7100.
[16]S. D. Rychnovsky, B. Rogers, G. Yang, J. Org. Chem. 1993, 58, 3511–
3515.
[17]In their original communication Kiyooka et al. reported on the non-
selectivity of the 1,3-syn reduction with DIBAL-H in CH2Cl2 while
the use of THF assures a high level of selectivity (see reference [9]).
When we first applied this protocol, CH2Cl2 proved to be the best
choice. However, reduction of hydroxyketones 23–25 most likely did
not display the same solvent dependence.
1
Compound56b : H NMR (500 MHz, [D4]MeOH): d=0.90, 1.12 (2s, 6H;
H-16, H-17), 1.04 (d, J=6.6 Hz, 3H; H-20), 1.08 (d, J=6.9 Hz, 3H; H-
19), 1.22 (s, 3H; H-18), 1.45–1.68 (m, 2H; H-6, H-12), 1.83–1.95 (m, 1H;
H-13), 1.92–2.06 (m, 1H; H-13), 2.08–2.19 (m, 3H; H-6, H-7, H-12),
2.28–2.37 (m, 1H; H-3), 3.26 (dd, J=10.2, 1.4 Hz, 1H; H-4), 3.57 (pst,
J=7.0 Hz, 1H; H-5), 3.71 (dd, J=8.4, 6.1 Hz, 1H; H-14), 4.28 (brs, 1H;
H-10), 4.55 (d, J=2.2 Hz, 1H; H-8), 5.23 (dd, J=15.8, 9.6 Hz, 1H; H-2),
5.56–5.50 ppm (m, 1H; H-1); 13C NMR (125 MHz, [D4]MeOH): d=18.1
(q; C-19), 19.2 (q; C-20), 24.8 (q; C-18), 26.0, 27.2–27.6 (q, q, t; C-16, C-
17, C-13), 33.9 (t; C-12), 34.1 (t; C-6), 35.9 (d; C-7), 39.5 (s; C-15), 41.7
(d; C-3), 72.0 (d; C-5), 75.8 (d; C-4), 77.1 (d; C-10), 78.7 (d; C-8), 85.1 (s;
C-11), 88.1 (d; C-14), 133.6 (d; C-2), 136.9 (d; C-1), 210.4 ppm (s; C-9).
[18]Note: The 7,8- syn-8,9-syn aldol product 23 is reduced syn selectively
by DIBAL-H. This means that 28 cannot have a 7,8-syn-8,9-syn-
9,10-anti configuration.
Acknowledgements
This work was supported by a KekulØ scholarship for C.J. and by a gradu-
ate scholarship provided by the Stiftung Stipendien-Fonds des Verbandes
[19]All aldol products described in this report tended to undergo retro-
aldol reaction.
Chem. Eur. J. 2006, 12, 8719 – 8734
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8733

A. Kirschning et al.
[20]a) H. C. Kolb, M. S. VanNieuwenhze, K. B. Sharpless, Chem. Rev.
1994, 94, 2483–2547; b) Y. Gao, R. M. Hanson, J. M. Klunder, S. Y.
Ko, H. Masamune, K. B. Sharpless, J. Am. Chem. Soc. 1987, 109,
5765–5780.
[21]In related contexts the di- tert-butylsiloxy group has been utilised
before: a) K.-M. Chen, G. E. Hardtmann, K. Prasad, O. Repic, M. J.
Shapiro, Tetrahedron Lett. 1987, 28, 155–158; b) I. Paterson, J. G.
Cumming, R. A. Ward, S. Lamboley, Tetrahedron 1995, 51, 9393–
9412; c) D. A. Evans, S. W. Kaldor, T. K. Jones, J. Clardy, T. J. Stout,
J. Am. Chem. Soc. 1990, 112, 7001–7031.
olaou, P. G. Bulger, D. Sarlah, Angew. Chem. 2005, 117, 4564–4601;
Angew. Chem. Int. Ed. 2005, 44, 4490–4527.
[26]Syntheses of macrocycles 48 and 50 were also used to further sup-
port elucidation of the relative stereochemistry of 25 and 26.
[27]Hoveyda-type catalysts turned out not to be suited for this RCM.
[28]Dienes 16 and 27 are enantiomeric in the furan part and at C-3. Be-
cause focus is directed to the stereotetrade between C-7 and C-10
relative changes between 16 and 27 are highlighted for this part rela-
tive to the furan moiety.
[29]The NMR spectroscopic data for 55b clearly show an E configura-
[22]I. Paterson, M. V. Perkins, Tetrahedron 1996, 52, 1811–1834.
[23]R. W. Hoffmann, Angew. Chem. 2000, 112, 2134–2150; Angew.
Chem. Int. Ed. 2000, 39, 2054–2070.
[24]The dominant stereochemical role of the enolate compared to the
aldehyde is also ideally supported by two related model reactions in
which we used 1) simplified achiral aldehyde 61 and 2) simplified
ketone 64. Silyl migration after aldol reaction led to isomeric ke-
tones 62a and 62b which both gave dihyroxy ketone 63 upon desily-
tion at C-4/C-5 (J(4,5)>15 Hz).
A
[30]A short overview can be found in J. Prunet, Angew. Chem. 2003,
115, 2932–2936; Angew. Chem. Int. Ed. 2003, 42, 2826–2830.
[31]a) K. Nakashima, R. Ito, M. Sono, M. Tori, Heterocycles 2000, 53,
301–314; b) A. Fürstner, K. Radkowski, Chem. Commun. 2001,
671–672; c) A. Fürstner, T. Dierkes, O. R. Thiel, G. Blanda, Chem.
Eur. J. 2001, 7, 5286–5298; d) A. Fürstner, K. Radkowski, C. Wirtz,
R. Goddard, C. Lehmann, R. Mynott, J. Am. Chem. Soc. 2002, 124,
7061–7069; e) E. A. Couladouros, A. P. Mihou, E. A. Bouzas, Org.
Lett. 2004, 6, 977–980.
[32]Switches in stereoselectivity for the RCM macrocyclisation exerted
by a remote substituent were described for the syntheses of epothi-
lone and salicylhalamides A and B: a) D. Meng, D.-S. Su, A. Balog,
P. Bertinato, A. J. Sorensen, S. J. Danishefsky, Y.-H. Zheng, T.-C.
Chou, L. He, S. B. Horwitz, J. Am. Chem. Soc. 1997, 119, 2733–
2734; b) A. Fürstner, O. R. Thiel, G. Blanda, Org. Lett. 2000, 2,
3731–3734. However, for both natural product projects a reliable
prediction of the stereochemical outcome of the RCM could not be
given.
[33]Besides the increased flexibility of the western fragment in 16 com-
pared to 54, also the presence of functional groups with coordinative
properties for metal complexes (such as the Grubbs II catalyst) in
dihydroxy ketone 55 can be made responsible for the observed se-
lectivity.
[34]a) H. E. Zimmerman, J. D. Robbins, R. D. McKelvey, C. J. Samuel,
L. R. Sousa, J. Am. Chem. Soc. 1974, 96, 4630–4643; b) W. Oppolz-
er, J.-P. Barras, Helv. Chim. Acta 1987, 70, 1666–1675.
[35]It needs to be noted that the stereochemical outcome of the asym-
metric dihydroxylation is reversed to the prediction given by the
Sharpless mnemonic. Asymmetric induction with the commercially
available dihydroxylation kit (AD-mix) did not result in any conver-
sion. Dihydroxylation product 56a was only present in the open
form, while diastereoisomer 56b was also present in the lactol
form.[36]
[36]The high chemoselectivity of the dihydroxylation manifests itself
both in the complete differentiation of the two olefinic double
bonds and by the reisolation of pure Z-configured 4,5-alkene after
the oxidation.
[37]Not surprisingly, lactol formation only occurs with the desired dihy-
droxylation product 56b, and not with 56a, which does not display
the required stereochemical topolgy. Interestingly, this lactol forma-
tion of 56b is only complete when 56a is removed, although no in-
termolecular phenomenona/interactions of the mixture could be de-
termined.
[38]Optical rotations for ent-tonantzitlolone (ent-1) was determined to
be [a]2D0 =ꢀ119 (c=0.06 in CHCl3) (tonantzitlolone 1: [a]2D0 =+134
(c=0.25 in CHCl3)). Apart from these facts all spectroscopic data
(NMR, IR, MS) were in full accordance with those of the authentic
material.
lation. Basically no other diastereoisomer was detected in the crude
reaction mixtures while the second aldol reaction between 64 and
(S)-6 yielded the two diastereomeric products 65 in a 3:2 ratio, as
1
was judged from the H NMR spectra (refer also to reference [13a]).
[25]Reviews on alkene metathesis including ring-closing metathesis
(RCM): a) M. Schuster, S. Blechert, Angew. Chem. 1997, 109, 2124–
2145; Angew. Chem. Int. Ed. Engl. 1997, 36, 2036–2056; b) A. Fürst-
ner, Top. Catal. 1997, 4, 285–299; c) S. K. Armstrong, J. Chem. Soc.
Perkin Trans. 1 1998, 371–388; d) R. H. Grubbs, S. Chang, Tetrahe-
dron 1998, 54, 4413–4450; e) A. Fürstner, Angew. Chem. 2000, 112,
3140–3172; Angew. Chem. Int. Ed. 2000, 39, 3012–3043; f) T. M.
Trnka, R. H. Grubbs, Acc. Chem. Res. 2001, 34, 18–29; g) K. C. Nic-
[39]Lithium enolates tend to form dimers which may also prevent for-
mation of transition states that involve coordination by the furan
oxygen atom.
Received: January 19, 2006
Revised: July 24, 2006
Published online: September 6, 2006
8734
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8719 – 8734