A new method for the preparation of non-terminal alkynes: Application to the total syntheses of tulearin A and C

Article abstract of DOI:10.1002/chem.201404873

Lactones are known to react with the reagent generated in situ from CCl₄ and PPh₃ in a Wittig-type fashion to give gem-dichloro-olefin derivatives. Such compounds are now shown to undergo reductive alkylation on treatment with organolithium reagents RLi to furnish acetylene derivatives bearing the substituent R at their termini (R=Me, n-, sec-, tert-alkyl, silyl); the reaction can be catalyzed with either Cu(acac)₂ or Fe(acac)₃/1,2-diaminobenzene. Two alkynol derivatives prepared in this way from readily accessible lactone precursors served as the key building blocks for the total syntheses of the cytotoxic marine macrolides tulearin A (1) and C (2). The assembly of these fragile targets hinged upon ring closing alkyne metathesis (RCAM) followed by a formal trans-reduction of the resulting cycloalkynes via trans-hydrosilylation/protodesilylation.

Full text of DOI:10.1002/chem.201404873

DOI: 10.1002/chem.201404873
Full Paper
&
Natural Products |Hot Paper|
A New Method for the Preparation of Non-Terminal Alkynes:
Application to the Total Syntheses of Tulearin A and C
Konrad Lehr, Saskia Schulthoff, Yoshihiro Ueda, Ronaldo Mariz, Lucie Leseurre,
Barbara Gabor, and Alois Fꢀrstner*[a]
Abstract: Lactones are known to react with the reagent
generated in situ from CCl4 and PPh3 in a Wittig-type fashion
to give gem-dichloro-olefin derivatives. Such compounds are
now shown to undergo reductive alkylation on treatment
with organolithium reagents RLi to furnish acetylene deriva-
tives bearing the substituent R at their termini (R=Me, n-,
sec-, tert-alkyl, silyl); the reaction can be catalyzed with
either Cu(acac)2 or Fe(acac)3/1,2-diaminobenzene. Two alky-
nol derivatives prepared in this way from readily accessible
lactone precursors served as the key building blocks for the
total syntheses of the cytotoxic marine macrolides tulearin A
(1) and C (2). The assembly of these fragile targets hinged
upon ring closing alkyne metathesis (RCAM) followed by
a formal trans-reduction of the resulting cycloalkynes via
trans-hydrosilylation/protodesilylation.
Introduction
the C12ꢀC13 bond by ring-closing olefin metathesis
(RCM).[10,11] As in many related cases reported in the literature,
this transformation was stereo-unselective, furnishing an E/Z
ratio of 1.9:1.[8] Attempts at improving this result by variation
of the catalyst and changing of the peripheral protecting
groups in the cyclization precursor were to no avail. This out-
come reiterates the notion that inherently E-selective olefin
metathesis catalysts are on high demand.[12]
Alkyne metathesis at large is greatly benefitting from a new
generation of catalysts, which combine excellent reactivity
with an outstanding selectivity and an unprecedented ease of
handling.[1] To fully harness the potential of this transformation,
however, it is also necessary to develop new and effective
gateways to the required starting materials. Although the
recent past has seen the first successful metathesis reactions
of terminal acetylene derivatives,[2–5] the use of non-terminal al-
kynes still dominates the field. Therefore improved access
routes to alkynes in general and methyl-capped alkynes in par-
ticular are on high demand.
This aspect became apparent during a program directed to-
wards the total synthesis of the tulearins, a small family of
macrolides isolated from a Madagascan sponge of the Fasca-
plysinopsis genus (Scheme 1).[6,7] Although the limited supply
from the natural source made a detailed biological evaluation
impossible, preliminary data for tulearin A (1) indicated a differ-
ential cytotoxicity against two human leukemia cell lines.[6] This
may well be the reason for the immediate interest that these
targets arose in the preparative community,[8,9] although their
stereostructure had originally not been established[6] and could
be assigned only after a crystalline derivative of 1 had be ob-
tained.[7] Therefore, a prompt foray by Cossy and co-workers
targeted what ultimately turned out to be an unnatural
isomer;[8] their synthesis was based on a macrocyclization at
[a] Dr. K. Lehr, S. Schulthoff, Dr. Y. Ueda, Dr. R. Mariz, Dr. L. Leseurre, B. Gabor,
Prof. A. Fꢀrstner
Max-Planck-Institut fꢀr Kohlenforschung
45470 Mꢀlheim/Ruhr (Germany)
E-mail: fuerstner@kofo.mpg.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201404873.
Scheme 1. Retrosynthetic analysis of the tulearins.
Chem. Eur. J. 2015, 21, 219 – 227
219
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
We saw an opportunity to solve the problem by recourse to
ring-closing alkyne metathesis (RCAM).[1,13] When combined
with a proper semi-reduction method, this transformation
opens a reliable entry into macrocyclic alkenes of either config-
uration. In addition, the tulearin case might provide a rigorous
test for the available alkyne metathesis catalysts, all of which
are high-valent and hence inherently Lewis acidic Schrock alky-
lidynes of molybdenum or tungsten.[14] It needs to be seen if
this property can be sufficiently tempered by the ancillary li-
gands that a motif as acid-sensitive as the pentadienyl ester
substructure in 1 and 2 will subsist.
Lithium reagents other than MeLi can also be used. Specifi-
cally, nBuLi, sBuLi, tBuLi and Me3SiCH2Li furnished the corre-
sponding non-terminal alkynes 5b–e in good to excellent
yields, whereas Me2PhSiLi[24] opened an unorthodox entry into
silylated alkynes (entry 11). In contrast, PhLi was found unsuita-
ble, as it got rapidly oxidized even at ꢀ788C with formation of
biphenyl as the only detectable organic product (entry 9);
MeMgBr or Me2Zn gave no conversion.
Table 1. Optimization of the alkylative elimination with formation of
non-terminal alkynes.[a]
If RCAM is to be used as the key transform, the retrosynthet-
ic disconnection of tulearin A (1) as the parent compound un-
ravels alkynes of type A and B as the key building blocks
(Scheme 1). Fragment A might be obtained from an alkynol
derivative of type C, which in turn could derive from lactone F.
Pattern recognition suggests that the very same lactone F
should also give access to the acid sector D (see the color-
coded carbon atoms); such a double use of a the same source
of chirality consolidates the substrate basis and reduces the
overall preparative burden. The third key fragment E can be
traced back to a lactone of type G.
From the conceptual viewpoint, it is the conversion of two
different lactones into the corresponding non-terminal alkynes
(F ! C, G ! E) which forms the very basis of the blueprint
shown in Scheme 1. Although such a transformation is certain-
ly possible by established means, its central role in the project-
ed total synthesis advocated a search for alternative solutions.
One possibility, together with its implementation into the
actual total synthesis of tulearin A and C, is outlined below.[15]
Entry RLi
Solvent T [8C]
t
[h]
Additives (mol%)
Yield
[%][b]
1
2
3
MeLi[c]
MeLi
MeLi
Et2O
Et2O
THF
RT
RT
RT
96
48
2
50[c]
90
80
(GC)
89
4
5
MeLi
MeLi
Et2O
Et2O
RT
RT
4
2
Cu(acac)2 (10)
Fe(acac)3 (10) +1,2-di-
aminobenzene (50)
70
6
7
8
9
nBuLi
sBuLi
tBuLi
PhLi
Et2O
Et2O
Et2O
Et2O
RT
4
1
1
!1
4
76
84
82
[d]
ꢀ78
ꢀ78
ꢀ78
RT
10
11
Me3SiCH2Li Et2O
Me2PhSiLi Et2O
Cu(acac)2 (10)
86
83
ꢀ78
to RT
6
[a] Unless stated otherwise, all reactions were carried out using 5 equiva-
lents of the respective organolithium reagent. [b] Yield of isolated materi-
al, unless stated otherwise. [c] Using 2.1 equivalents of MeLi. [d] Biphenyl
was the only distinct product; RT=room temperature.
Results and Discussion
Formation of non-terminal alkynes by reductive alkylation
Because of their activated carbonyl character, many lactones
are amenable to Wittig-type olefinations with phosphorous
ylides, which ordinary esters would not undergo.[16] This in-
cludes the reagent generated in situ from CCl4 and PPh3 which
furnishes the corresponding dichloro-olefin derivatives.[17–20] In
this way, multigram amounts of compound 4 were prepared
from lactone 3 (Table 1). A host of other lactones with ring
sizes of 5–8 reacted similarly well, whereas larger lactones in-
creasingly resemble regular esters and therefore participated
less willingly; for representative results, see Table 3.
These experimental results allow conclusions to be drawn
with regard to the likely mechanism of this transformation
(Scheme 2). Metal–halogen exchange upon reaction of 4 with
RLi is thought to generate a metal vinylidene of type 6 in the
first place. One might envisage that this intermediate collapses
in the presence of excess RLi with formation of the lithium ace-
tylide 7, which then gets alkylated by the R-Cl concomitantly
formed (path A). However, the observation that sBuLi and even
tBuLi are particularly effective reagents makes this scenario
highly unlikely, because the efficient alkylation of a lithium ace-
tylide by sBuCl or tBuCl is improbable, if not impossible.[25]
Rather, metal vinylidenes of type 6 are known to be amphi-
philic; it is their electrophilic rendition that is believed to be
the key enabling feature,[26–28] in that it allows such species to
intercept a second equivalent of RLi to give 8 prior to reduc-
tive elimination (path B).[29] Under this premise, it is plausible
that the initial Li/Cl exchange is rate-determining; otherwise, it
would be difficult to explain why tBuLi reacts much faster than
MeLi (cf. Table 1, entries 2 and 8). The handicap of the un-
branched alkyllithium reagents in this triggering event might
be overcome with the help of an adequate catalyst that ren-
Treatment of the resulting dichloro-olefins with excess lithi-
um sand in THF is known to engender a reductive elimination
with formation of the corresponding terminal alkynes.[21] We
reasoned that replacement of metallic lithium by organolithi-
um reagents RLi might allow non-terminal alkynes to be pre-
pared in a similar fashion.[22] In fact, treatment of compound 4
as the model substrate with MeLi in Et2O resulted in a very
clean formation of the corresponding methyl-capped acetylene
derivative 5a (R=Me), but the reaction was slow and excess
MeLi had to be used to ensure full conversion (Table 1, en-
tries 1 and 2). The use of THF engendered a much faster rate
but led to a slight drop in yield (entry 3).[23]
Chem. Eur. J. 2015, 21, 219 – 227
www.chemeurj.org
220
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. Two conceivable mechanistic scenarios; the available evidence
suggests that pathway B sketches the essence of the alkylative elimination
reaction, cf. Text.
ders the metal-halogen exchange more facile. In fact, addition
of Cu(acac)2 (10 mol%) to a mixture of 4 and MeLi in Et2O led
to a significant rate acceleration without compromising the
yield or purity of the resulting product (entry 4). Even faster is
the reaction in the presence of Fe(acac)3 (5–10 mol%), al-
though it was mandatory to use 1,2-diaminobenzene as an ad-
ditive (entry 5);[30–32] in its absence, the reaction was unselec-
tive and product isolation therefore troublesome.
Scheme 3. Optimization of the alkylative elimination reaction in the series of
g-butyrolactone derived gem-dichloro-olefins; for the results, see Table 2.
The relevance of the catalyzed variants became fully appar-
ent when we started to explore the scope of this new method
for the preparation of non-terminal alkynes. The case of the
five-membered lactone derivative
9
is representative
Table 2. Optimization of the alkylative elimination shown in Scheme 3.[a]
(Scheme 3 and Table 2): although treatment of the derived di-
chloro-olefin 10 with MeLi in Et2O furnished the expected
methyl capped acetylene derivative 12 in reasonable yield,
substantial amounts of the corresponding allene 15 were de-
tected (entry 1). Actually, 15 became the major product when
the reaction was carried out in THF (entry 2), which is thought
to derive from an allylic deprotonation of the substrate[33] fol-
lowed by reductive elimination of the resulting organolithium
intermediate and formal cross coupling of the gem-dichloro-
olefin subunit. Although this alkylative allene formation is also
potentially valuable, it can be suppressed upon addition of cat-
alytic amounts of either Cu(acac)2 or Fe(acac)3/1,2-diaminoben-
zene (entries 3 and 5); in this case, the iron catalyst afforded
significantly better yields of the desired alkyne 12. This exam-
ple also illustrates the importance of the 1,2-diaminobenzene
additive, although the exact role of this protic modifier is cur-
rently not clear. In any case, an iron catalyzed cross coupling
of 10 with formation of product 16 prevailed in its absence
(entry 4).[34]
Entry Solvent t [h] Additives [mol%]
Yield [%][b]
12 15
16
1
2
3
4
5
Et2O
THF
Et2O
Et2O
Et2O
72
3
20
4
–
–
60 10
25 58
58 5 (GC)
32 <5 (GC) 46
<5 (GC)
<5 (GC)
<5 (GC)
Cu(acac)2 (10)
FeCl2 (10)[c]
Fe(acac)3 (5) + 1,2-
diaminobenzene (25)
2
85 <5 (GC) <5 (GC)
[a] All reactions were carried out with MeLi (5 equiv) at ambient tempera-
ture. [b] Yields of isolated materials, unless indicated. [c] FeCl2 of 99.99%
purity was used.
was of immediate interest for a total synthesis of kendomycin
recently completed by our laboratory.[36] For all products de-
rived from five-membered lactone precursors (entries 14–21),
the iron catalyzed variant was quintessential for success as it
allows competing allene formation to be suppressed. This strik-
ing dichotomy of alkyne versus allene formation is apparent
from entries 14/15 and 19/20.
Table 3 shows a selection of non-terminal alkynes made by
the uncatalyzed or catalyzed variants of the new protocol.
Overall, the alkylative elimination worked well for a host of
gem-dichloro-olefins of different ring size. Entry 13 shows that
benzoannulated lactone derivatives can also be used, provided
the copper catalyzed protocol is chosen. In the present con-
text, the results of entries 3–5 and 16 are arguably most rele-
vant, because the resulting alkyne served as key building block
en route to tulearin A and C (see below).[35] Moreover, entry 1
shows that a lactone with a chiral center a to the carbonyl
group can be transformed into the corresponding non-termi-
nal alkyne without loss in optical purity; this particular product
Figure 1. Substrates not amenable to the alkylative alkyne synthesis.
Chem. Eur. J. 2015, 21, 219 – 227
www.chemeurj.org
221
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
bohydrate literature confirms that such derivatives are
prone to different types of fragmentation reactions and/
or decomposition on exposure to an organometallic
base.[33]
Table 3. Formation of methyl-capped alkynes or allenes by alkylative elimina-
tion.[a]
Entry Substrate
Dichloro-
olefination
[%]
Method Product
Yield [%][b]
Total Synthesis of Tulearin C
1
2
81
C
B
92[c]
56[d]
As outlined in the introduction, a lactone of type F (=
compound 19) was meant to serve as the key building
block for the synthesis of the tulearins. This compound
is best prepared on scale by desymmetrization of cheap
dimethyl 3-methylglutarate (17) with the aid of pig liver
esterase, followed by recrystallization of the cinchoni-
dine salt of the resulting crude acid (Scheme 4).[37,38]
Monoester 18 (ꢁ98% ee, 20 g scale) was easily trans-
formed into 19 by selective reduction followed by lacto-
nization.
[d]
3
4
5
A
B
C
90
77
85
92
95
6
7
A
B
89
88
8
92
92
B
88
To access the common northern sector of all tulearins,
19 was converted into the corresponding alkynol by the
novel reductive alkylation method outlined above. The
derived aldehyde 21 was subjected to asymmetric
alkyne addition under the aegis of (ꢀ)-N-methylephe-
drine (28) to give the propargyl alcohol derivative 22 in
excellent diastereomeric purity (dr 98:2).[39] This product
was subjected to hydroalumination/iodination in Et2O
as the solvent;[40] attempts at trapping the organoalumi-
num intermediate with MeX (X=I, OTf) were unsuccess-
ful. While 23 turned out to be surprisingly labile in un-
protected form, the corresponding TBS-ether 24 was
more manageable. An iron-catalyzed cross-coupling
with MeMgBr[41] proceeded quantitatively but led to an
unexpected loss of stereochemical integrity of the react-
ing double bond (E/Z 2:1) even when performed at
ꢀ408C. Similar problems were initially encountered with
more traditional palladium catalysis; careful optimiza-
tion, however, showed that the use of meticulously han-
dled Me2Zn allowed this surprisingly facile isomerization
to be suppressed.[42,43] Under these conditions, the re-
quired alcohol 25 was obtained in high yield, basically
as a single isomer after deprotection under standard
conditions.
9
10
B
C
84
95
11
30
88
B
80
12
13
A
C
<20
60
14
91
A
58
85
15
16
D
D
87
83
(R=MOM)
(R=MOM)
17
18
62
(R=PMB)
D
D
69
(R=PMB)
88 (R=Bn)
70
(R=Bn)
19
20
95
A
D
84
80
As discussed in the Introduction, it was planned to
use lactone 19 twice en route to tulearin C. To this end,
this substrate was subjected to a Claisen condensation
with ethyl acetate followed by a ruthenium catalyzed
asymmetric hydrogenation of the resulting ketoester
29, which is in an inconsequential equilibrium with the
corresponding hemiacetal form (Scheme 5). Although
the use of (R)-BINAP as ligand gave a respectable result
(76%, dr 96:4),[44] (R)-SYNPHOS (34) led to a truly impec-
cable outcome (89%, dr 99:1).[45] The primary alcohol in
the resulting product 30 had to be protected prior to
a Frꢂter–Seebach alkylation which set the missing
methyl branch with a workable selectivity of 85:15 in favor of
the required anti-isomer.[46] Product 32 thus formed after MOM
protection of the aldol subunit was elaborated into iodide 33
in readiness for a Negishi-type fragment coupling.[47]
[d]
21
D
55[d]
[a] Method A: MeLi (5 equiv), Et2O, RT; method B: MeLi (5 equiv), THF, RT; method
C: MeLi (5 equiv), Et2O, Cu(acac)2 (10 mol%), RT; method D: MeLi (5 equiv), Et2O,
Fe(acac)3 (5–10 mol%), 1,2-diaminobenzene (25–50 mol%), RT. [b] Yield of isolated
material. [c] The starting lactone and the alkyne product had the same optical
purity (ee 94%). [d] The yield over both steps is given; Bn=benzyl; MOM=me-
thoxymethyl; PMB=p-methoxybenzyl; TBS=tert-butyldimethlysilyl.
Limitations were encountered with substrates carrying
oxygen substituents a and/or b to the former lactone carbonyl
group, which were decomposed under the standard conditions
(Figure 1). This outcome is not totally surprising since the car-
Chem. Eur. J. 2015, 21, 219 – 227
www.chemeurj.org
222
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
almost quantitatively within no more than 15 min in
the presence of the alkylidyne ate-complex 49 as the
catalyst.[51] This gratifying result stands in marked
contrast to the difficulties encountered in the final
stages of the projected synthesis. It was not totally
unexpected that the ruthenium catalyzed trans-hy-
drosilylaton of cycloalkyne 43 with BnMe2SiH might
be challenging.[52] The [Cp*Ru]+ catalyst exhibits high
affinity to 1,3-dienes, a signature subunit of com-
pound 43.[53] Yet, the solvent-free conditions previ-
ously developed by our group allowed 43 to be con-
verted into an inconsequential mixture of regioiso-
meric alkenylsilanes 44 in respectable yield.[54,55] At
this stage, only the global deprotection was missing
to complete the total synthesis of tulearin C (2). How-
ever, the cleavage of the CꢀSiBnMe2 group with
TBAF mandated elevated temperatures, which result-
ed in concomitant elimination of the MOM-protected
aldol subunit. Although the resulting product 45 was
a dead end anyway, attempts at removing the isopro-
pylidene acetal under mildly acidic conditions point-
ed to yet another Achilles heel: thus, acetal cleavage
instigated rearrangement by ionization of the fragile
dienyl-ester. Trapping of the mesomerically stabilized
pentadienyl cation intermediate by the proximal
C12ꢀC13 double bond relays the positive charge and
Scheme 4. a) i) Pig liver esterase, NaOH (1m), MeOH, pH 7 buffer, 78%; ii) (ꢀ)-cinchoni-
dine, aq. acetone, then aq. HCl (ꢁ98% ee); b) i) aq. LiOH; (ii) LiBH4, THF; iii) aq. HCl, 92%;
c) CCl4, PPh3, THF, reflux, 92%; d) MeLi, Et2O, RT, 90%; e) Dess–Martin periodinane,
CH2Cl2; f) 27, Zn(OTf)2, (ꢀ)-N-methyl-ephedrine (28), (iPr)2NEt, toluene, MS 3 ꢃ, 57% (over
both steps), dr 98:2; g) i) Red-Al, Et2O, MS 3 ꢃ, 08C ! RT; ii) I2, ꢀ208C, 99%; h) TBSOTf,
2,6-lutidine, CH2Cl2, 94%; i) Me2Zn, (dppf)PdCl2·CH2Cl2 (5 mol%), THF, Et3N, 658C, 90%; j)
TBAF, THF, 96%; k) HCꢂCMgBr, Pd(PPh3)4 (3 mol%), THF, 72%; dppf=1,1’-bis(diphenyl-
phosphino)ferrocene; Red-Al=sodium bis-(2-methoxyethoxy)-aluminium-hydride;
TBAF=tetra-n-butylammonium fluoride; Tf=trifluoromethanesulfonyl.
Treatment of iodide 33 with Zn/Cu[48] gave the correspond-
ing functionalized organozinc reagent that was coupled with
alkenyl iodide 37; the latter was made from 3-butyn-1-ol (35)
via zirconocene catalyzed hydroalumination/iodination[49] fol-
lowed by alkylation of the quite unstable primary triflate 36
with excess propynyllithium (Scheme 6). A subsequent asym-
metric dihydroxylation of the alkene site in the resulting prod-
uct 38 furnished diol 39,[50] which was swiftly elaborated into
the required acid sector 40 as deemed appropriate for the
total synthesis.
allows the C9-OH group to terminate the cascade with forma-
tion of a tetrahydropyranyl ring.[56] Since product 48 was ob-
Esterification of 25 and 40 set the stage for the crucial ring-
closure of diyne 42 via RCAM (Scheme 7), which proceeded
Scheme 6. a) i) Zn/Cu, toluene/DMA (19:1), 708C; ii) Pd(PPh3)4 (5 mol%), 37,
608C, 72%; b) AD-mix a, MeSO2NH2, tBuOH, H2O, 08C, 83%; c) 2,2-
dimethoxypropane, pTsOH, 91%; d) aq. LiOH, MeOH/THF (1:1), quant.; e)
HCl, MeOH, 99%; f) TBSOTf, 2,6-lutidine, CH2Cl2, 90%; g) aq. LiOH, MeOH/
THF (1:1), 99%; h) Dibal-H, Cp2ZrCl2, THF, RT, then I2, ꢀ788C, 80%; i) Tf2O,
CH2Cl2, pyridine, ꢀ208C ! RT; j) propynyllithium, THF, ꢀ208C ! 08C, 76%
(over both steps); DMA=dimethylacetamide.
Scheme 5. a) EtOAc, LDA, THF, ꢀ788C, 90%; b) H2 (10 atm), RuCl3 (1 mol%),
(R)-SYNPHOS (34) (1 mol%), EtOH, 808C, 89% (dr 99:1) c) TBDPSCl, Et3N,
DMAP cat., CH2Cl2, 95%; d) LDA, THF/DMPU, ꢀ788C ! ꢀ408C, then MeI,
ꢀ788C!08C, 79% (dr 85:15); e) MOMCl, (iPr)2NEt, CH2Cl2, reflux, 97%; f)
TBAF, THF, 83%; g) I2, PPh3, imidazole, CH2Cl2, 90%; MOM=methoxymethyl;
TBDPS=tert-butyldiphenylsilyl.
Chem. Eur. J. 2015, 21, 219 – 227
www.chemeurj.org
223
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
To meet the high chemical de-
mands of the final deprotection,
(EtO)3SiH was used in lieu of
BnMe2SiH for the subsequent
trans-hydrosilylation
of
51,
which had again to be per-
formed under solvent-free condi-
tions. Protodesilylation of the re-
sulting alkenylsiloxanes 52 was
now achieved with the aid of
AgF in protic medium at ambi-
ent temperature.[54] Likewise, the
final desilylation was possible
under sufficiently mild condi-
tions that left the labile aldol
subunit intact. This sequence al-
lowed tulearin C (2) to be ob-
tained in good, albeit somewhat
variable, yield over the final
three steps. Careful NMR inspec-
tion of the crude material
showed that no other stereoiso-
mer was present in more than
3%. After purification, the spec-
tra of the synthetic samples
compared very favorably to
those of the authentic natural
product.[6,7]
Scheme 7. a) 25, EDC·HCl, DMAP, CH2Cl2, 08C ! RT, 84%; b) 49 (5 mol%), MS 5 ꢃ, toluene, 508C, 98%; c)
Me2BnSiH, [Cp*Ru(MeCN)3]PF6 (10 mol%), neat, 67%; d) TBAF, THF, 508C, 78%; e) TFA cat., CH2Cl2, 46%; Cp*=pen-
tamethylcyclopentadienyl; DMAP=4-dimethylaminopyridine; EDC=1-ethyl-3-(3-dimethylaminopropyl)carbodi-
imide; TBAF=tetra-n-butylammonium fluoride; TFA=trifluoroacetic acid.
Total Synthesis of Tulearin A
Tulearin A (1) differs from its
sister compound tulearin C (2) only by a single carbamate. For
the lack of any electronic or steric bias however, it seemed un-
reasonable to assume that this group could be attached selec-
tained as a single diastereomer with all-equatorial substituents
on the trans-decaline-type scaffold, it is reasonable to assume
a concerted process via a charge-delocalized intermediate of
type 47, which is a characteristic trait of many cation–p cycliza-
tion reactions.[57,58]
Although this initial foray had not been successful, the gath-
ered intelligence allowed us to refine our tactics. Since tulear-
in C and its immediate precurors are obviously highly sensitive
to acid as well as base, it seemed prudent to resort solely to
silyl protecting groups which should be removable under suffi-
ciently mild conditions. To this end, the acid segment was
modified in that 39 was transformed into the all-TBS-protected
building block 41 (Scheme 6), which was then esterified with
25 to give diyne 50 (Scheme 8). As expected, this change in
the protecting groups had no effect on the macrocyclization
by RCAM, which was again rapid and productive. In view of
the rather sensitive character of the target, this outcome at-
tests to an excellent application profile of the latest generation
of alkyne metathesis catalysts, which draw their activity and se-
lectivity from the synergy between the molybdenum center
and the triarylsilanolate ligands.[51,59] Although 49 and conge-
ners are high-valent complexes with a formal d0 electron
count, the actual Lewis acidity is tempered to the extent that
even exceedingly acid sensitive motifs remain intact.
Scheme 8. a) 25, EDC·HCl, DMAP, CH2Cl2, 98%; b) 49 (4 mol%), toluene
(2 mm), MS 5 ꢃ, 508C, 96%; c) (EtO)3SiH, [Cp*Ru(MeCN)3]PF6 (10 mol%), 08C;
d) AgF, THF/MeOH/H2O; e) TBAF, THF, 43–60% (over three steps).
Chem. Eur. J. 2015, 21, 219 – 227
www.chemeurj.org
224
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tively to the C8-OH group of 2
or any of its immediate synthetic
precursors. Therefore, it was nec-
essary to discriminate the secon-
dary hydroxyl groups from the
very beginning (Scheme 1).
To this end, the commercial
lactone 53 was transformed into
the methyl-capped alkynol deriv-
ative 55 by dichloro-olefination,
iron catalyzed reductive alkyla-
tion of 54 and standard protect-
ing
group
management
(Scheme 9).[60] Product 55 was
then converted into the some-
what unstable aldehyde 56
which was reacted without delay
with iodide 33b in the presence
of CrCl2 and catalytic amounts of
cobalt phthalocyanine in DMF to
give alcohol 57 in an acceptable
yield;[61,62] a slight excess of the
aldehyde partner was used to
drive the conversion. Since 57
was formed as a 3:1 mixture of
diastereomers, an oxidation/re-
duction sequence had to
remedy the stereochemical out-
come.[63,64] Subsequent treat-
ment with TESOTf/lutidine fol- Scheme 9. a) MOMCl, (iPr)2NEt, CH2Cl2, reflux, 94%; b) CCl4, PPh3, THF, reflux, 87%; c) MeLi, Fe(acac)3 (5 mol%), 1,2-
diaminobenzene (25 mol%), Et2O, 83%; d) TBSOTf, 2,6-lutidine, CH2Cl2, 90%; e) Me2BBr, CH2Cl2, ꢀ788C, 90%; f)
lowed by careful saponification
SO3·pyridine, DMSO, CH2Cl2, 08C, 93%; g) 33b, CrCl2, cobalt phthalocyanine (5 mol%), DMF, 54%; h) Dess–Martin
of the methyl ester furnished the
periodinane, CH2Cl2, 90%; i) L-Selectride, THF, ꢀ788C, 79%; j) TESOTf, 2,6-lutidine, 08C, 93%; k) aq. LiOH, THF/
required acid segment 58 that
MeOH, 64%; l) 25, EDC·HCl, DMAP, CH2Cl2, 86%; m) 60 (5 mol%), toluene (2 mm), MS 5 ꢃ, 508C, 96%; n) (EtO)3SiH,
features a proper pattern of or- [Cp*Ru(MeCN)3]PF6 (10 mol%), 08C; o) AgF, THF/MeOH/H2O, 56% (over both steps); p) Cl3CC(O)NCO, CH2Cl2,
ꢀ788C, then MeOH, NaHCO3, 50%; q) 3HF·Et3N, Et3N, MeCN, 408C, 68%; r) upon storage at 48C, cf. Text; TES=tri-
thogonal silyl groups on the sec-
ethylsilyl.
ondary alcohol functions.
Next, 58 was esterified with
25 and the resulting diyne 59 subjected to a high yielding
RCAM, for which the neutral alkylidyne complex 60 rather than
the ate-complex 49 was used, which had been developed in
a parallel program in our laboratory.[59,65] The ease with which
61 was formed by this catalyst is remarkable if one considers
that this particular compound is unstable even when stored in
the refrigerator; it eventually undergoes an irreversible strain-
driven ring expansion via two consecutive [3,3]-rearrange-
ments of the pentadienyl ester subunit with formation of lac-
tone 62, the structure of which has been unambiguously as-
signed. The inherent chemical instability of 61 expressed in
this behavior is also thought to be the major reason why the
final sequence of ruthenium catalyzed trans-hydrosilylation fol-
lowed by protodesilylation and selective cleavage of the TES
ether off the C8-OH group was challenging. Yet product 63
could be obtained in 56% yield over the three operations that
enabled the subsequent selective installation of the carba-
mate.[66] Finally, global desilylation was achieved with the help
of Et3N·HF in MeCN in the presence of excess Et3N, whereas
more conventional conditions were totally inadequate. The
spectral data of synthetic tulearin A (1) thus formed compare
favorably with those of the natural product isolated from the
Madagascan Fascaplysinopsis sponge (see the Supporting Infor-
mation).[6,7]
Conclusion
The total syntheses of tulearin A and C reported herein spon-
sored the development of a new method for the conversion of
lactones into non-terminal alkynes. First, a Wittig-type reaction
with CCl4/PPh3 converts the substrate into a gem-dichloro-
olefin, which is then treated with an organolithium reagent RLi
to give the corresponding acetylene derivative incorporating
the substituent R at its terminus (R=methyl, n-, sec- and tert-
alkyl, silyl). Such reductive alkylation reactions are best per-
formed in the presence of catalytic amounts of either Cu(acac)2
or Fe(acac)3/1,2-diaminobenzene and are thought to proceed
via electrophilic metal vinylidene species as the key reactive in-
termediates.
Chem. Eur. J. 2015, 21, 219 – 227
www.chemeurj.org
225
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[8] A. L. Mandel, V. Bellosta, D. P. Curran, J. Cossy, Org. Lett. 2009, 11, 3282–
3285.
[9] J. S. Yadav, M. Venkatesh, N. Swapnil, A. R. Prasad, Tetrahedron Lett.
2013, 54, 2336–2339.
[10] a) T. M. Trnka, R. H. Grubbs, Acc. Chem. Res. 2001, 34, 18–29; b) A. Fꢀrst-
ner, Angew. Chem. Int. Ed. 2000, 39, 3012–3043; Angew. Chem. 2000,
112, 3140–3172; c) K. C. Nicolaou, P. G. Bulger, D. Sarlah, Angew. Chem.
Int. Ed. 2005, 44, 4490–4527; Angew. Chem. 2005, 117, 4564–4601;
d) A. H. Hoveyda, A. R. Zhugralin, Nature 2007, 450, 243–251; e) A.
Fꢀrstner, Chem. Commun. 2011, 47, 6505–6511.
[11] Metathesis in Natural Product Synthesis: Strategies, Substrates and Cat-
alysts (Eds.: J. Cossy, S. Arseniyadis, C. Meyer), Wiley-VCH, Weinheim,
2010.
[12] A. Fꢀrstner, Science 2013, 341, 1357 (UNSP 1229713).
[13] a) A. Fꢀrstner, G. Seidel, Angew. Chem. Int. Ed. 1998, 37, 1734–1736;
Angew. Chem. 1998, 110, 1758–1760; b) A. Fꢀrstner, O. Guth, A. Rumbo,
G. Seidel, J. Am. Chem. Soc. 1999, 121, 11108–11113; c) A. Fꢀrstner, C.
Mathes, C. W. Lehmann, Chem. Eur. J. 2001, 7, 5299–5317.
[14] R. R. Schrock, Chem. Rev. 2002, 102, 145–179.
[15] For a preliminary communication, see: K. Lehr, R. Mariz, L. Leseurre, B.
Gabor, A. Fꢀrstner, Angew. Chem. Int. Ed. 2011, 50, 11373–11377;
Angew. Chem. 2011, 123, 11575–11579.
[16] M. Lakhrissi, Y. Chapleur, Angew. Chem. Int. Ed. Engl. 1996, 35, 750–752;
Angew. Chem. 1996, 108, 833–834.
[17] M. Suda, A. Fukushima, Tetrahedron Lett. 1981, 22, 759–762.
[18] a) M. Lakhrissi, Y. Chapleur, J. Org. Chem. 1994, 59, 5752–5757; b) M.
Lakhrissi, Y. Chapleur, Synlett 1991, 583–585; c) C. Taillefumier, Y. Chap-
leur, Chem. Rev. 2004, 104, 263–292.
This methodology allowed the chiral building blocks re-
quired for the synthesis of the marine macrolides tulearin A (1)
and C (2) to be obtained in excellent yield and optical purity.
Their elaboration into the target structures involved a sequence
of ring-closing alkyne metathesis (RCAM) to forge the macrocy-
clic frames, followed by semi-reduction of the resulting cycloal-
kynes to the corresponding E-alkenes by trans-hydrosilylation/
protodesilylation. The RCAM step worked very nicely despite
the exceptional sensitivity of the target compounds towards
acid as well as base. This favorable outcome attests to the re-
markable selectivity profile of the latest generation of molyb-
denum alkylidyne catalysts endowed with silanolate
groups.[51,59] In contrast, the trans-hydrosilylation/protodesilyla-
tion chemistry was more difficult than anticipated. Specifically,
the presence of a 1,3-diene subunit in the substrate dims the
activity of the required cationic Cp*Ru catalyst, and the condi-
tions necessary for the protodesilylation of the resulting alke-
nylsilanes endangered the very fragile products. Although
these challenges were ultimately met, it became clear that im-
proved and/or complementary methodology for the chemo-
and stereoselective reduction of alkynes to E-alkenes is desir-
able to enable applications to even more demanding target
compounds. First forays in this direction have recently been re-
ported by our laboratory, which are currently subject to in-
tense scrutiny.[67–70]
[19] For a review on gem-dihaloalkenes, see: G. Chelucci, Chem. Rev. 2012,
112, 1344–1462.
[20] Such lactone-derived dichloro-olefins tend to decompose upon stor-
age; however, most of the compounds described herein can be kept in
a refrigerator under Ar for several days.
[21] a) J. S. Yadav, V. Prahlad, M. C. Chander, J. Chem. Soc. Chem. Commun.
1993, 137–138; b) J. S. Yadav, C. Venugopal, Synlett 2007, 2262–2266.
[22] a) For a singular example in which a gem-diiodo-alkenyl ether had been
transformed into a non-terminal alkyne on treatment with a large
excess of MeMgBr. see: R. M. Black, G. B. Gill, J. Chem. Soc. Perkin Trans.
1 1980, 410–418; b) in contrast to the results reported herein, treat-
ment of acyclic allyl 1,1-dichlorovinyl ethers with 2.2 equivalents of
nBuLi in THF did not result in incorporation of the nBu-groups into the
products; rather, terminal alkynyl allyl ether derivatives were formed
which undergo a spontaneous Claisen rearrangement, see: A. Christo-
pher, D. Brandes, S. Kelly, T. Minehan, Org. Lett. 2006, 8, 451–454.
[23] Other (co)solvents investigated led to dramatic losses; hexane retards
the reaction, whereas the use of CH2Cl2, NMP or DMPU as additives fur-
nished intractable mixtures.
Experimental Section
All experimental details can be found in the Supporting Informa-
tion. The material contains compound characterization and copies
of spectra of new compounds.
Acknowledgements
Generous financial support by the MPG, the Fonds der Chemi-
schen Industrie, and the Kyoto University Education Program
(scholarship to Y. U.) is gratefully acknowledged. We thank the
analytical departments of our Institute for excellent support,
and Dr. P. Persich and Dr. M. Fuchs for contributing two exam-
ples of the new alkyne synthesis.
[24] I. Fleming, R. S. Roberts, S. C. Smith, J. Chem. Soc. Perkin Trans. 1 1998,
1209–1214.
[25] This conclusion is supported by a control experiment using 1-chloro-1-
heptyne: as expected, treatment with MeLi merely furnished 1-heptyne
after aqueous work-up, which shows that a lithium acetylide was
formed but not trapped by the methyl chloride generated in situ.
[26] For a review on a-heteroatom substituted alkenyllithium reagents, see:
M. Braun, Angew. Chem. Int. Ed. 1998, 37, 430–451; Angew. Chem. 1998,
110, 444–465.
[27] For pioneering studies on the reaction of alkali metal vinylidene species
with nucleophiles, see: a) G. Kçbrich, F. Ansari, Chem. Ber. 1967, 100,
2011–2020; b) M. Duraisamy, H. M. Walborsky, J. Am. Chem. Soc. 1984,
106, 5035–5037; c) J. Rachon, V. Goedken, H. M. Walborsky, J. Am.
Chem. Soc. 1986, 108, 7435–7436; d) M. Topolski, M. Duraisamy, J.
Rachon, J. Gawronski, K. Gawronska, V. Goedken, H. M. Walborsky, J.
Org. Chem. 1993, 58, 546–555; e) J. Barluenga, M. A. Rodriguez, P. J.
Campos, G. Asensio, J. Am. Chem. Soc. 1988, 110, 5567–5568; f) H. Yana-
gisawa, K. Miura, M. Kitamura, K. Narasaka, K. Ando, Bull. Chem. Soc. Jpn.
2003, 76, 2009–2026.
Keywords: alkynes
·
hydrosilylation
·
iron catalysis
·
metathesis · molybdenum · natural products
[1] A. Fꢀrstner, Angew. Chem. Int. Ed. 2013, 52, 2794–2819; Angew. Chem.
2013, 125, 2860–2887.
[2] B. Haberlag, M. Freytag, C. G. Daniliuc, P. G. Jones, M. Tamm, Angew.
Chem. Int. Ed. 2012, 51, 13019–13022; Angew. Chem. 2012, 124, 13195–
13199.
[3] P. Persich, J. Llaveria, R. Lhermet, T. de Haro, R. Stade, A. Kondoh, A.
Fꢀrstner, Chem. Eur. J. 2013, 19, 13047–13058.
[4] R. Lhermet, A. Fꢀrstner, Chem. Eur. J. 2014, 20, 13188–13193.
[5] For the first implementation into the total synthesis of a complex natu-
ral product, see: J. Willwacher, A. Fꢀrstner, Angew. Chem. Int. Ed. 2014,
53, 4217–4221; Angew. Chem. 2014, 126, 4301–4305.
[6] A. Bishara, A. Rudi, M. Aknin, D. Neumann, N. Ben-Califa, Y. Kashman,
Org. Lett. 2008, 10, 153–156.
[28] For DFT-studies indicating a low-lying LUMO, see: a) P. v. R. Schleyer, T.
Clark, A. J. Kos, G. W. Spitznagel, C. Rohde, D. Arad, K. N. Houk, N. G.
Rondan, J. Am. Chem. Soc. 1984, 106, 6467–6475; b) T. Kimura, T. Satoh,
Tetrahedron Lett. 2013, 54, 5072–5074.
[7] A. Bishara, A. Rudi, I. Goldberg, M. Aknin, Y. Kashman, Tetrahedron Lett.
2009, 50, 3820–3822.
Chem. Eur. J. 2015, 21, 219 – 227
www.chemeurj.org
226
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[29] X-ray structures of 1-chloro-1-lithio (magnesio)ethene derivatives indi-
cate considerable rehydridization of the vinylidene C atom (higher s-
character) and a very significant elongation of the CꢀCl bond, cf: a) G.
Boche, M. Marsch, A. Mꢀller, K. Harms, Angew. Chem. Int. Ed. Engl. 1993,
32, 1032–1033; Angew. Chem. 1993, 105, 1081–1082; b) G. Boche, K.
Harms, M. Marsch, A. Mꢀller, J. Chem. Soc. Chem. Commun. 1994, 1393–
1394; for additional spectroscopic evidence, see: c) D. Seebach, H.
Siegel, J. Gabriel, R. Hꢄssig, Helv. Chim. Acta 1980, 63, 2046–2053.
[30] B. D. Sherry, A. Fꢀrstner, Acc. Chem. Res. 2008, 41, 1500–1511.
[31] For an indication that iron salts might catalyze metal/halogen ex-
change, see: L. Melzig, C. R. Diꢅne, C. J. Rohbogner, P. Knochel, Org. Lett.
2011, 13, 3174–3177.
[32] A variety of other additives was tested, including tmeda, 2,2’-bipyridine
and ethylene-1,2-diamine, which had little effect on the outcome.
[33] A. Bandzouzi, Y. Chapleur, Carbohydr. Res. 1987, 171, 13–24.
[34] a) M. Dos Santos, X. Franck, R. Hocquemiller, B. Figadꢅre, J.-F. Peyrat, O.
Provot, J.-D. Brion, M. Alami, Synlett 2004, 2697–2700; b) for a related
but uncatalyzed alkylation of gem-dibromo-alkenes, see: H.-T. T. Thien,
A. Novoa, N. Pellegrini-Moise, F. Chrꢆtien, C. Didierjean, Y. Chapleur, Eur.
J. Org. Chem. 2011, 6939–6951.
218; b) P. J. Fagan, W. S. Mahoney, J. C. Calabrese, I. D. Williams, Organo-
metallics 1990, 9, 1843–1852.
[54] a) F. Lacombe, K. Radkowski, G. Seidel, A. Fꢀrstner, Tetrahedron 2004, 60,
7315–7324; b) A. Fꢀrstner, K. Radkowski, Chem. Commun. 2002, 2182–
2183.
[55] For previous applications, see: a) K. Micoine, A. Fꢀrstner, J. Am. Chem.
Soc. 2010, 132, 14064–14066; b) K. Micoine, P. Persich, J. Llaveria, M.-H.
Lam, A. Maderna, F. Loganzo, A. Fꢀrstner, Chem. Eur. J. 2013, 19, 7370–
7383; c) A. Fꢀrstner, M. Bonnekessel, J. T. Blank, K. Radkowski, G. Seidel,
F. Lacombe, B. Gabor, R. Mynott, Chem. Eur. J. 2007, 13, 8762–8783.
[56] Cossy and co-workers have already observed some sort of allylic rear-
rangement in the final stages of their synthesis of a diastereomer of tu-
learin A, although the structure of the resulting product was not report-
ed, cf. ref. [8].
[57] a) G. Stork, A. W. Burgstahler, J. Am. Chem. Soc. 1955, 77, 5068–5077;
b) G. Gamboni, H. Schinz, A. Eschenmoser, Helv. Chim. Acta 1954, 37,
964–971; c) A. Eschenmoser, L. Ruzicka, O. Jeger, D. Arigoni, Helv. Chim.
Acta 1955, 38, 1890–1904.
[58] For representative examples of hydroxyl-terminated cyclization cas-
cades of (poly)unsaturated substrates which are triggered by different
catalysts but seem to pass invariably through highly ordered intermedi-
ates (or transition states), see the following and literature cited therein:
a) K. Ishihara, H. Ishibashi, H. Yamamoto, J. Am. Chem. Soc. 2002, 124,
3647–3655; b) A. Fꢀrstner, L. Morency, Angew. Chem. Int. Ed. 2008, 47,
5030–5033; Angew. Chem. 2008, 120, 5108–5111; c) C. A. Mullen, A. N.
Campbell, M. R. Gagnꢆ, Angew. Chem. Int. Ed. 2008, 47, 6011–6014;
Angew. Chem. 2008, 120, 6100–6103.
[35] This particular alkynol derivative was also used for the synthesis of the
perfume ingredient Muscenone, see: K. Lehr, A. Fꢀrstner, Tetrahedron
2012, 68, 7695–7700.
[36] L. Hoffmeister, P. Persich, A. Fꢀrstner, Chem. Eur. J. 2014, 20, 4396–4402.
[37] a) L. K. P. Lam, R. A. H. F. Hui, J. B. Jones, J. Org. Chem. 1986, 51, 2047–
2050; b) L. Poppe, L. Novꢂk, P. Kolonits, ꢇ. Bata, C. Szꢂntay, Tetrahedron
1988, 44, 1477–1487.
[59] J. Heppekausen, R. Stade, A. Kondoh, G. Seidel, R. Goddard, A. Fꢀrstner,
Chem. Eur. J. 2012, 18, 10281–10299.
[38] Alternatively, non-enzymatic desymmetrization reactions of the corre-
sponding anhydride can be used, see: a) Y. Chen, S.-K. Tian, L. Deng, J.
Am. Chem. Soc. 2000, 122, 9542–9543; b) S. H. Oh, H. S. Rho, J. W. Lee,
J. E. Lee, S. H. Youk, J. Chin, C. E. Song, Angew. Chem. Int. Ed. 2008, 47,
7872–7875; Angew. Chem. 2008, 120, 7990–7993.
[39] a) D. Boyall, D. E. Frantz, E. M. Carreira, Org. Lett. 2002, 4, 2605–2606;
b) D. E. Frantz, R. Fꢄssler, E. M. Carreira, J. Am. Chem. Soc. 2000, 122,
1806–1807.
[40] S. E. Denmark, T. K. Jones, J. Org. Chem. 1982, 47, 4595–4597.
[41] a) B. Scheiper, M. Bonnekessel, H. Krause, A. Fꢀrstner, J. Org. Chem.
2004, 69, 3943–3949; b) A. Fꢀrstner, H. Krause, C. W. Lehmann, Angew.
Chem. Int. Ed. 2006, 45, 440–444; Angew. Chem. 2006, 118, 454–458.
[42] a) X. Zeng, M. Qian, Q. Hu, E. Negishi, Angew. Chem. Int. Ed. 2004, 43,
2259–2263; Angew. Chem. 2004, 116, 2309–2313; b) K. C. Nicolaou,
A. L. Nold, R. R. Milburn, C. S. Schindler, Angew. Chem. Int. Ed. 2006, 45,
6527–6532; Angew. Chem. 2006, 118, 6677–6682.
[43] The use of old batches that might contain traces of oxidation/hydrolysis
products led to erratic results; in view of the high sensitivity of Me2Zn
it is necessary to handle the reagent with utmost care using proper
Schlenk techniques.
[44] T. Ohkuma, M. Kitamura, R. Noyori, in Catalytic Asymmetric Synthesis
(Ed.: I. Ojima), 2nd ed., Wiley-VCH, New York, 2000, pp. 1–110.
[45] a) J. Madec, X. Pfister, P. Phansavath, V. Ratovelomanana-Vidal, J.-P.
GenÞt, Tetrahedron 2001, 57, 2563–2568; b) C. Roche, N. Desroy, M.
Haddad, P. Phansavath, J.-P. Genet, Org. Lett. 2008, 10, 3911–3914.
[46] a) G. Frꢂter, U. Mꢀller, W. Gꢀnther, Tetrahedron 1984, 40, 1269–1277;
b) D. Seebach, J. Aebi, D. Wasmuth, Org. Synth. 1985, 63, 109–120.
[47] E. Negishi, Bull. Chem. Soc. Jpn. 2007, 80, 233–257.
[48] a) Y. Tamaru, H. Ochiai, T. Nakamura, K. Tsubaki, Z. Yoshida, Tetrahedron
Lett. 1985, 26, 5559–5562; b) Y. Tamaru, H. Ochiai, T. Nakamura, Z. Yosh-
ida, Tetrahedron Lett. 1986, 27, 955–958; c) C. A¨ıssa, R. Riveiros, J.
Ragot, A. Fꢀrstner, J. Am. Chem. Soc. 2003, 125, 15512–15520.
[49] Z. Huang, E. Negishi, Org. Lett. 2006, 8, 3675–3678.
[60] For the cleavage of MOM-acetals with Me2BBr, see: a) Y. Guindon, H. E.
Morton, C. Yoakim, Tetrahedron Lett. 1983, 24, 3969–3972; b) Y. Guin-
don, C. Yoakim, H. E. Morton, J. Org. Chem. 1984, 49, 3912–3920; c) J.
Willwacher, N. Kausch-Busies, A. Fꢀrstner, Angew. Chem. Int. Ed. 2012,
51, 12041–12046; Angew. Chem. 2012, 124, 12207–12212.
[61] K. Takai, K. Nitta, O. Fujimura, K. Utimoto, J. Org. Chem. 1989, 54, 4732–
4734.
[62] A. Fꢀrstner, Chem. Rev. 1999, 99, 991–1045.
[63] This route via the aldehyde was chosen because extensive model stud-
ies had indicated that activated acid derivatives derived from alcohol
55 do not couple well with various organometallic reagents formed
from iodide 33b.
[64] For the stereochemical course of the L-Selectride reduction of a-silyloxy
ketones, see: a) J. M. Concellꢈn, P. L. Bernad, R. Alvarez, A. Rodrꢉguez, B.
BaragaÇa, Tetrahedron Lett. 1999, 40, 2863–2864; b) T. Takahashi, S.
¯
Kusaka, T. Doi, T. Sunazuka, S. Omura, Angew. Chem. Int. Ed. 2003, 42,
5230–5234; Angew. Chem. 2003, 115, 5388–5392.
[65] a) G. Valot, C. S. Regens, D. P. O’Malley, E. Godineau, H. Takikawa, A.
Fꢀrstner, Angew. Chem. Int. Ed. 2013, 52, 9534–9538; Angew. Chem.
2013, 125, 9713–9717; b) L. Brewitz, J. Llaveria, A. Yada, A. Fꢀrstner,
Chem. Eur. J. 2013, 19, 4532–4537; c) V. Hickmann, A. Kondoh, B. Gabor,
M. Alcarazo, A. Fꢀrstner, J. Am. Chem. Soc. 2011, 133, 13471–13480;
d) S. Benson, M.-P. Collin, A. Arlt, B. Gabor, R. Goddard, A. Fꢀrstner,
Angew. Chem. Int. Ed. 2011, 50, 8739–8744; Angew. Chem. 2011, 123,
8898–8903; e) W. Chaładaj, M. Corbet, A. Fꢀrstner, Angew. Chem. Int. Ed.
2012, 51, 6929–6933; Angew. Chem. 2012, 124, 7035–7039; f) K. Geba-
uer, A. Fꢀrstner, Angew. Chem. 2014, 126, 6511–6514; Angew. Chem. Int.
Ed. 2014, 53, 6393–6396.
ˇ
´
[66] P. Kocovsky, Tetrahedron Lett. 1986, 27, 5521–5524.
[67] K. Radkowski, B. Sundararaju, A. Fꢀrstner, Angew. Chem. Int. Ed. 2013,
52, 355–360; Angew. Chem. 2013, 125, 373–378.
[68] B. Sundararaju, A. Fꢀrstner, Angew. Chem. Int. Ed. 2013, 52, 14050–
14054; Angew. Chem. 2013, 125, 14300–14304.
[69] S. M. Rummelt, A. Fꢀrstner, Angew. Chem. 2014, 126, 3700–3704;
Angew. Chem. Int. Ed. 2014, 53, 3626–3630.
[70] A. Fꢀrstner, Angew. Chem. 2014, 126, 8728–8740; Angew. Chem. Int. Ed.
2014, 53, 8587–8598.
[50] R. A. Johnson, K. B. Sharpless, in Catalytic Asymmetric Synthesis (Ed.: I.
Ojima), 2nd ed., Wiley-VCH, New York, 2000, pp. 357–398.
[51] J. Heppekausen, R. Stade, R. Goddard, A. Fꢀrstner, J. Am. Chem. Soc.
2010, 132, 11045–11057.
[52] a) B. M. Trost, Z. T. Ball, T. Jçge, J. Am. Chem. Soc. 2002, 124, 7922–7923;
b) B. M. Trost, Z. T. Ball, J. Am. Chem. Soc. 2001, 123, 12726–12727;
c) B. M. Trost, Z. T. Ball, Synthesis 2005, 853–887.
[53] For representative examples of stable Cp*Ru-diene complexes, see: a) S.
Steines, U. Englert, B. Drießen-Hçlscher, Chem. Commun. 2000, 217–
Received: August 15, 2014
Published online on November 3, 2014
Chem. Eur. J. 2015, 21, 219 – 227
www.chemeurj.org
227
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim