Total Synthesis of Gelsemoxonine through a Spirocyclopropane Isoxazolidine Ring Contraction

Article abstract of DOI:10.1021/jacs.5b02574

Plants of the species Gelsemium have found application in traditional Asian medicine for over a thousand years. Gelsemoxonine represents a novel constituent of this plant incorporating a highly functionalized azetidine at its core. We herein report a full

Full text of DOI:10.1021/jacs.5b02574

Subscriber access provided by SUNY DOWNSTATE
Article
Total Synthesis of Gelsemoxonine through a
Spirocyclopropane Isoxazolidine Ring Contraction
Stefan Diethelm, and Erick M Carreira
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 13 Apr 2015
Downloaded from http://pubs.acs.org on April 13, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 15
Journal of the American Chemical Society
1
2
3
4
5
6
Total Synthesis of Gelsemoxonine through a Spirocyclopropane
Isoxazolidine Ring Contraction
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Stefan Diethelm, Erick M. Carreira*
Eidgenössische Technische Hochschule Zürich, Vladimir-Prelog-Weg 3, HCI H335, 8093 Zürich, Switzerland
ABSTRACT: Plants of the species Gelsemium have found application in traditional Asian medicine for over a thousand
years. Gelsemoxonine represents a novel constituent of this plant incorporating a highly functionalized azetidine at its
core. We herein report a full account of our studies directed towards the total synthesis of gelsemoxonine that relies on a
conceptually new approach for the construction of the central azacyclobutane. A spirocyclopropane isoxazolidine ring
contraction was employed to access a key β-lactam intermediate, which could be further elaborated to the azetidine of
the natural product. In the course of our studies, we have gained detailed insight into this intriguing transformation.
Furthermore, we report on previously unnoticed oligomerization chemistry of gelsemoxonine. We also document an
enantioselective synthesis of a key precursor en route to gelsemoxonine.
Introduction
ethyl ketone appendage as well as a fully substituted ste-
reocenter as part of the azetidine add to the complexity of
the natural product.
The use of plant extracts to treat human diseases has a
long history in traditional folk medicine all around the
globe. This knowledge, accumulated over thousands of
years, harbors a still largely unexplored potential for
modern pharmaceutical chemistry. In this context, explo-
ration of the constituents of medicinally relevant plants is
of great interest for the discovery of novel bioactive natu-
1
ral products with potential application in pharmacy.
Plants from the species Gelsemium, endemic to China and
Japan, have been used as an analgesic and antispasmodic
agent as well as for the treatment of skin ulcers over a
2
thousand years in traditional Asian medicine. Detailed
chemical studies on Gelsemium plants led to the identifi-
cation of an intriguing class of alkaloids, characterized by
3
a range of biological activities. Several members of the
Gelsemium alkaloid class have been shown to exhibit
potent antitumor activity or are capable of attenuating
4
inflammatory and neuropathic pain in mouse models.
Moreover, these natural products comprise unusual struc-
tural features such as a densely functionalized and com-
pact core as exemplified by the two prominent members
gelsedine (1) and gelsemine (2) (Figure 1, top). In 1991,
Clardy and co-workers isolated the novel alkaloid gelse-
Figure 1. Gelsemium alkaloids and azetidine synthesis.
The biosynthesis of gelsemoxonine (3) has been sug-
6
gested to start from gelsedine (1). After initial oxidation
steps, nucleophilic displacement of the resulting C(15)
tertiary alcohol by an amine was invoked to explain for-
mation of the azacyclobutane (Figure 1, a). Similarly, in
their pioneering total synthesis of gelsemoxonine, Fuku-
yama and co-workers employed a biomimetic strategy for
5
moxonine (3) from Gelsemium elegans bentham. More
than a decade later, Aimi and co-workers corrected the
originally proposed structure based on X-ray crystallo-
6
graphic analysis and 2D NMR studies. The authors re-
7
the construction of the central azetidine. Our group has
ported that gelsemoxonine (3) incorporates an unusual
azetidine as a key structural element, which distinguishes
it from all other members of the Gelsemium family. The
four-membered heterocycle is completely embedded in a
polycyclic framework, which severely aggravates the
strain associated with the core of gelsemoxonine. Fur-
thermore, a spiro-fused N-methoxy oxindole ring, an
been interested for some time in the preparation and
characterization of small saturated heterocycles as build-
8
ing blocks for drug discovery. In the context of this re-
search program, we decided to embark on studies di-
rected towards the total synthesis of gelsemoxonine em-
ploying a conceptually new approach to the construction
of its unusual azetidine ring (Figure 1, b).
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 15
We herein report a full account of efforts leading to the
total synthesis of gelsemoxonine. The synthesis relies on
Focusing on the preparation of projected key intermedi-
9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ate 6, we realized that this unusually substituted β-lactam
would not be amenable to traditional construction strate-
gies, relying on established protocols for β-lactam synthe-
a strategic spirocyclopropane isoxazolidine ring contrac-
tion providing a β-lactam intermediate, which is further
elaborated into the azetidine of the natural product (Fig-
ure 1, b). The studies reported here have led to a more
detailed understanding of this intriguing transformation.
In the course of this work, we have observed previously
unnoticed oligomerization chemistry of gelsemoxonine.
Furthermore, we present an enantioselective approach to
a key intermediate en route to the natural product.
1
2
sis. However, a report by Cordero and Brandi describing
the acid mediated ring contraction of spirocyclopropane
isoxazolidines I to form β-lactam products II caught our
1
3
attention (Scheme 1, box). Although this transformation
is unexplored in the service of complex molecule synthe-
sis, it offered an intriguing opportunity for the construc-
tion of the projected key intermediate 6. Accordingly, a
cyclization substrate would first have to be prepared in-
corporating various functional groups, which could po-
tentially interfere with the desired reaction. Nonetheless,
we set out to prepare a collection of precursors for the
ring contraction.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
As outlined in Scheme 1, the synthetic strategy focused
on a late-stage construction of the central seven-
membered carbocycle of gelsemoxonine (3). To imple-
ment this plan, two general strategies were considered as
indicated in Scheme 1 (strategy A vs. B). Retrosynthetic
cleavage of the C(3)-C(7) bond would lead back to oxoni-
um intermediate 4, which would be susceptible to attack
Results and Discussion
10
by a suitable C(7) nucleophile (strategy A). Alternatively,
Synthesis of β-lactam 17: Our synthetic efforts com-
menced with the preparation of a suitable ring contrac-
tion precursor, enabling late-stage oxidative functionali-
zation of C(3). The required isoxazolidine, incorporating a
spiro-fused cyclopropane at C(5), could be conveniently
prepared by intramolecular [3+2] dipolar cycloaddition of
a corresponding alkylidenecyclopropane derivative. As
outlined in Scheme 2, known alkylidenecyclopropane
formation of the C(6)C(7) bond could be achieved by
exploiting the nucleophilic character at C(7) through
displacement of a leaving group at the C(6)-methylene
leading back to intermediate 5 (strategy B). Most conven-
iently, introduction of the oxindole ring system could in
either case be carried out before or after construction of
the gelsemoxonine core fragment. Additionally, we opted
for a unified approach that would allow access to both
intermediates 4 and 5 through a common precursor. β-
lactam 6 was thereby identified as a versatile platform,
which would enable access to both these intermediates
through manipulations including introduction of the
C(6)-methylene by extension at the amide carbonyl along
with oxidation of the C(3) position. The latter transfor-
mation could possibly be carried out relying on a CH
oxidation approach, whereby the inherent tendency of
etheral methylene groups to undergo CH bond cleavage
14
substituted aldehyde 7 was chosen as a starting point.
Preparation of 7 was carried out using a slightly modified
literature protocol that allowed for the synthesis of multi-
gram quantities of this volatile aldehyde. Commercial
15
cyclopropanone hemiacetal 8 was first subjected to reac-
tion with vinyl Grignard followed by transformation of
the resulting
a
Scheme 2. Synthesis of oxazoline 13.
11
would be exploited. Furthermore, the ethyl ketone ap-
pendage of the natural product was planned to be intro-
duced as an inert surrogate (-R in Scheme 1), which could
later be elaborated to the required ketone functionality.
Scheme 1. Alternative synthetic strategies towards
gelsemoxonine.
a
Reagents and conditions: (a) HCl, MeOH, rt, 10 min, 80-
9
0%. (b) CH =CHMgCl, THF, 80 °C, 30 min, 75%. (c) TsCl,
2
NEt , CH Cl , rt, 12 h, 67%. (d) ethyl glycolate, NaH, THF, 0
3
2
2
°
C, 30 min; then 9, Pd(dba) (1 mol%), dppe (2 mol%), THF,
2
rt, 12 h, 83%. (e) DIBAL-H, CH Cl , ‒78 °C, 2 h. (f) MeNO ,
2
2
2
LDA, THF, 78 °C, 30 min; then 7, ‒78 °C, 1 h; then TMSCl, ‒
78 °C to rt, 1 h, 78% (2 steps). (g) PhNCO, NEt , C H , rt, 12 h;
3
6
6
then HCl, rt, 5 min, 75%, dr 4:1.
ACS Paragon Plus Environment

Page 3 of 15
Journal of the American Chemical Society
ethylamine or TMEDA to the reaction mixture did not
1
result in any product formation (entry 5). We next tested
the use of different Lewis acid activators in the reaction.
tertiary alcohol into tosylate 9. Tsuji-Trost reaction using
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ethyl glycolate as the nucleophilic component delivered
ester 10 in 83% yield. Finally, DIBAL-H mediated reduc-
tion of the ester group cleanly furnished aldehyde 7. Gen-
eration of a reactive dipole for the cycloaddition was
planned to be carried out from a respective nitroalkene
20
Although TiCl was ineffective (entry 6), addition of BF3
4
etherate provided the desired product 15 in low yield of
22% (entry 7). Seebach and co-workers have investigated
the use of preformed titanium acetylides as mild nucleo-
21
16
philes in the addition to ketones. Following these re-
ports, we tested triisopropoxy(propyn-1-yl) titanium as a
potential nucleophile (entry 8). Again however, no prod-
uct was obtained under these conditions. A frequently
encountered problem for the addition of organometallic
nucleophiles to carbonyl compounds is competing enoli-
zation due the dual role of the organometal species as a
nucleophile as well as a strong base. Various strategies
precursor.
Henry addition of lithiated nitromethane provided, after
quenching with Me SiCl, silyl ether 11 in 78% yield (2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
steps). When 11 was exposed to phenyl isocyanate, dehy-
drative formation of putative nitrile oxide intermediate 12
occurred, followed by dipolar cycloaddition to provide
isoxazoline 13 as a 4:1 mixture of diastereomers. The rela-
tive configuration of this intermediate could be confirmed
by nOe analysis (see Supporting Information). Interest-
ingly, the major product observed is consistent with a
pseudo-axial arrangement of the silyl ether substituent, as
indicated in transition state 12. We speculate that when
the reactions proceeds through transition state IIIa the
rate is faster, or accordingly the activation energy is low-
est. Structure IIIa is preferred because it avoids position-
ing the ether C-O bond in the equatorial arrangement as
shown for IIIb, wherein its interaction with the dipole
HOMO would lead to lowering of its energy and thus an
increase of the activation energy for the dipolar cycload-
dition.
22
have been studied to alleviate the problem. In particular,
addition of lanthanide salts to these reactions was found
to decrease the basicity of the organometallic reagent,
23
thus disfavoring the competing deprotonation pathway.
24
Indeed, when anhydrous CeCl3 was added to propyn-1-yl
lithium prior to exposure to the electrophile 13, we were
able to isolate the desired product in good yield (68%)
and with complete diastereoselectivity for the desired
isomer. We next tested addition of cyanide to oxime ether
13. Various standard protocols known for the conversion
of ketones to cyanohydrins were tested to this effect.
However, only treatment of 13 with Et AlCN at 60 °C af-
2
25
forded the desired nitrile 16 in 63% yield.
We next turned to the introduction of a suitable ethyl
ketone surrogate at C(15) to access cis-fused isoxazolidine
a
Table 1. Addition of nucleophiles to isoxazoline 13.
1
4. As outlined in table 1, 13 was subjected to various nu-
cleophiles to effect diastereoface selective addition to the
dihydroisoxazole. We thereby chose coupling partners
that would enable elaboration of a ketone functionality at
C(20) at a late stage of the synthesis. It has been previous-
ly noted that isoxazolines are generally poorly reactive
towards nucleophilic reagents. We were therefore not
surprised to find that treatment of 13 with buten-2-yl
lithium at -78 °C did not yield any coupled product (data
not shown). Lewis acid activation has been previously
Entry Nucleophile
Additives
Conditions Yield
‒78 °C, THF
1
BF ∙OEt /ZnCl
2
-
3
2
1
7
2
3
BF ∙OEt
‒78 °C, THF 19%
3
2
reported to enhance the nucleophilicity of isoxazolidine.
However, addition of BF etherate or ZnCl as Lewis acid
1
2
8
3
BF ∙OEt
rt, THF
-
-
-
-
3
2
promoter did not change the reaction outcome (entry 1).
We reasoned that attack of any nucleophile from the
convex face of 13 might be sterically hampered by the
adjacent hydroxyl or alkoxy substituent. Accordingly,
smaller nucleophiles such as propen-2-yl lithium or vi-
nylmagnesium bromide were tested under similar condi-
tions (entries 2 and 3). Indeed, addition of propen-2-yl
lithium successfully provided the desired product albeit
with very low conversion (entry 2). We next tested addi-
tion of propyn-1-yl lithium to isoxazoline 13, whereby the
triple bond could later serve as a handle for ketone intro-
duction through regioselective hydration (entries 4-7).
Direct treatment of 13 with this reagent again led to com-
plete recovery of unreacted starting material (entry 4).
Reports on solution studies of lithium acetylides have
indicated that addition of amine bases to these reagents
can alter their aggregation behavior and attendant reac-
4
5
6
7
8
9
-
‒78 °C, THF
b
NEt or TMEDA ‒78 °C, Et O
3
2
TiCl4
‒78 °C, THF
BF ∙OEt
‒78 °C, THF 22%
‒78 °C, THF
‒78 °C, THF 68%
60 °C, THF 63%
3
2
TiCl4
-
c
d
BF ∙OEt
3
2
10
Et AlCN
2
-
a
Conditions: substrate (0.3 mmol), nucleophile (4.0
1
9
b
tivity. In our case however, the addition of either tri-
equiv.), additive (1.0 equiv.), 2 h. used as co-solvent (1:1).
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 15
c
d
2
.0 equiv. BF ∙OEt used. 49% yield (65% brsm) with 24 verted into the corresponding acylcarbamate by treat-
3
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mmol substrate.
ment with Boc O. Subjecting this intermediate to Petasis’
2
reagent (Cp TiMe ) in the presence of pyridine indeed
2
2
afforded the strained enecarbamate 19 in good yield
Interestingly, attempts to add any nucleophiles to O-
protected analogues of 13 (such as the O-trimethylsilyl or
O-benzyl derivative of 13) failed and unreacted starting
material was recovered. Moreover, using a starting mate-
rial with the opposite configuration at the C(14) hydroxyl
stereocenter failed to give the expected product (data not
shown). These observations, along with the data present-
ed above, led us to suggest that the hydroxyl group in 13
plays a key role in the reaction, probably by acting as a
directing group for the incoming organometalic reagent.
31
(
74%). Notably, this intermediate proved stable to
chromatography on silica gel and could be handled with-
out significant hydrolysis of the sensitive methyleneaza-
cyclobutane. Next, the C(5) stereocenter was installed by
hydroboration of the olefin using 9-BBN dimer followed
by oxidative workup (H O /NaOH). Diol 20 was obtained
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
as a single diastereomer at C(5), indicating exclusive at-
tack of the hydroborating agent from the convex face of
the molecule.
Installation of the oxindole moiety was first attempted
by nucleophilic displacement at C(6) starting from vari-
ous derivatives of primary alcohol 20 (tosylate, mesylate,
iodide). Surprisingly however, all of these compounds
proved inert towards nucleophiles, reflecting the dense
steric environment around C(6). Successful introduction
of the C(7) oxindole fragment was finally achieved by
Having a short, efficient route to oxazolidines 15 and 16
in hand, we set out to test the key ring contraction. As
outlined in Scheme 3, nitrile 16 was subjected to reported
reaction conditions. To our surprise, no product for-
mation could be observed using various different protic
acid additives. We reasoned that the electron withdraw-
ing effect of the nitrile might be responsible for the im-
paired reactivity of the system. We next treated alkyne 15
with trifluoroacetic acid (TFA) at 80 °C. To our delight,
after 6 hours of reaction time β-lactam 17 was isolated as
the only product in good yield (67%). The structure of 17
was confirmed by X-ray crystallographic analysis of Boc-
protected derivative 18 (Scheme 3, box). Notably, in con-
trast to the original report, no nitrogen protecting group
was needed for the ring contraction to proceed. Moreo-
ver, an unprotected alcohol as well as an electron rich
alkyne were well tolerated under the reaction conditions.
1
3
conversion of alcohol 20 to aldehyde 21 (TEMPO,
32
26
PhI(OAc)
)
2
and subsequent aldol condensation with N-
33
methoxy oxindole 22 using conditions previously report-
34
ed by Evans. Olefin 23 was obtained as an E/Z mixture of
2:1 favoring the E isomer as confirmed by nOe analysis
(
see Supporting Information for further details).
a
Scheme 4. Synthesis of oxindole derivative 23.
Scheme 3. Ring contraction of isoxazolidines 16 and
15.
a
Reagents and conditions: (a)Boc O, NEt , DMAP, CH Cl ,
2
3
2
2
rt, 30 min, 58%. (b) Cp TiMe , pyridine, toluene, 70 °C, 5 h,
2
2
7
4%. (c) 9-BBN dimer, THF, rt, 1 h; then H O /NaOH, rt, 4 h,
2 2
7
9%. (d) TEMPO (30 mol%), PhI(OAc) , CH Cl , rt, 2 h, 80%.
2
2
2
(
e) 22, MgBr (15 mol%), TMSCl, NEt , THF, rt, 20 min; then
2
3
21, rt, 15 min, 82% (E/Z 2:1).
Directed CH oxidation at C(3): To gain access to the
full carbon framework of gelsemoxonine starting from 17,
installation of the C(6) methylene group by manipulation
of the β-lactam carbonyl was required. However, any
manipulation of the azacyclobutanone would be accom-
panied by the risk of undesired ring opening. We sought
conditions which would avoid such side reactions leaving
the four-membered heterocycle intact. Initial experiments
involving Corey-Chaykovsky-type epoxide formation on
For the closure of the seven-membered carbocycle of
gelsemoxonine by construction of the C(3)C(7) bond, we
initially opted for the generation of an oxonium ion at
C(3), which would eventually be trapped by an oxindole
enolate (Scheme 1, strategy A). To implement this plan,
adjustment of the oxidation state at the C(3) ether of
intermediate 23 was required. Etheral CH bonds are
known to be susceptible to direct oxidation due to their
reduced CH bond strength. Numerous protocols have
been developed in the past to achieve CH oxidations in
2
7
28
the amide carbonyl or addition of carbenes to the re-
29
spective β-thiolactam, failed to give any of the desired
products. Relying on the well-known chemistry of amides
to participate in olefination reactions, we were hoping to
access a respective enamine intermediate by methylena-
3
5
highly complex settings. However, model studies with
our system revealed that targeting of the C(3)H bond
using such protocols was unsuccessful, instead resulting
in the preferential oxidation of other CH bonds or the
3
0
tion of β-lactam 17. Following this plan, 17 was first con-
ACS Paragon Plus Environment

Page 5 of 15
Journal of the American Chemical Society
the electron rich alkyne in substrates such as 24 was not
complete decomposition of the respective substrates (see
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Supporting Information for further details). Moreover,
attempted elimination of the secondary alcohol in 23
resulted in rapid decomposition of the starting material.
tolerated under the reaction conditions. Various other
procedures for the generation of a nitrogen-centered
radical from carbamate derivatives were evaluated but
remained without success (see Supporting Information
We next set out to evaluate directed oxidation of the
C(3) methylene relying on the adjacent C(14) hydroxyl
4
1
for
further
detail).
36
group as a handle to achieve selectivity. In this context,
Du Bois and co-workers have reported an elegant ap-
Scheme 6. Hofmann-Löffler-Freytag reaction on
a
proach for the functionalization of unbiased CH bonds
model system 28.
37
in the vicinity of hydroxyl groups. Their strategy is based
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
on the generation of a reactive nitrene species covalently
linked to an OH through a sulfonamide or carbamate
group. Accordingly, we prepared carbamate 24 as de-
scribed in Scheme 5. Subsequent treatment of this inter-
38
2
mediate with Rh (esp) in the presence of PhI(OAc) as
2
2
a
the stoichiometric oxidant cleanly afforded oxazolidinone
5 in excellent yield (80%). It is noteworthy that the elec-
Reagents and conditions: hν (UV), PhI(OAc)
rt, 45 min, 35% (29).
, I , MeCN,
2 2
2
tron rich alkyne adjacent to the reactive nitrene is not
touched and selective insertion into the C(3) methylene
CH occurs as the exclusive reaction. Although we had
achieved the desired oxidation of C(3), we were unable to
generate precursors such as 26 suitable for the generation
of putative oxonium ion 27 as the oxazolidinone in 25
resisted any attempts directed towards its cleavage (see
Supporting Information for further details).
Early oxidation of C(3): Based on the previously de-
scribed results, it became apparent that C(3) functionali-
zation at a late stage of the synthesis would be met with
major hurdles due to the high functionalization of any
substrate. Accordingly, we turned our attention to ad-
justment of the C(3) oxdiation state early in the synthesis.
Scheme 7 describes our strategy to realize this notion.
The revised strategy entailed initial formation of second-
ary alcohol 31 through a Henry reaction starting with
previously employed aldehyde 7 (70%). When this inter-
a
Scheme 5. Directed CH oxdiation of carbamate 24.
mediate was treated with Boc O in the presence of
2
DMAP, rapid formation of an E/Z mixture of nitroolefin
32 was observed (15 min). However, when the reaction
was run for a prolonged period of time (12 h), intermedi-
ate 32 reacted further to ultimately form enol ether 33.
We hypothesize that dehydration of the nitro group in 32
mediated by Boc O leads to the production of nitrile ox-
2
ide 34, whereby the double bond transposes by one car-
bon unit thus functionalizing C(3). However, only the
thermodynamically less favored Z-isomer 35 can undergo
subsequent intramolecular cycloaddition to form the final
product. We surmise that an equilibrium between iso-
mers 34 and 35 exists, which most likely results from an
addition/elimination sequence of the DMAP additive to
the unsaturated nitrile oxide. In fact, it proved crucial to
add 2 equivalents of DMAP to the reaction mixture in
a
Reagents and conditions: (a)NaBH , THF/MeOH (1:1), 0
4
°
C, 20 min, 63%. (b) trichloroacetyl isocyanate, CH Cl , rt, 15
2 2
min; then NaHCO , MeOH, rt, 1 h, 80%. (c) Rh (esp) (4
3
2
2
mol%), PhI(OAc) , MgO, CH Cl , 60 °C (sealed tube), 4 h,
2
2
2
81%.
As outlined in Scheme 6, an alternative C(3) oxidation
approach involving a Hofmann-Löffler-Freytag reaction
was investigated next. This transformation relies on the
generation of a highly reactive nitrogen-centered radical
species, which engages in an intramolecular 1,5-hydrogen
42
order to achieve a good yield of 33 (79%). With enol
ether 33 in hand, we next turned to further functionaliza-
tion of C(14) and C(3). As outlined in Scheme 7, this goal
was achieved by initial epoxidation of the olefin in 33
followed by nucleophilic opening of the highly unstable
oxirane. Among the various reagent systems tested for
3
9
abstraction. In a modification of the original protocol,
Barton and co-workers have reported the transformation
oxidation, only dimethyldioxirane (DMDO) provided the
40
of primary amides to yield lactone products. Following
43,44
desired intermediate 36 in quantitative yield.
The
this strategy, we evaluated primary carbamate 28 as pre-
cursor for C(3) oxidation. Irradiation of this model system
under UV light and in the presence of PhI(OAc) and I
unstable product was obtained as a 2:1 mixture of dia-
stereomers favoring the desired isomer as shown in
Scheme 7. The stereochemistry could be confirmed later
in the synthesis by X-ray crystallographic analysis. Epox-
ide 36 proved highly sensitive to water or chromatograph-
2
2
delivered cyclic carbonate 29 in 35% yield along with a
minor amount of oxazolidinone 30. Unfortunately how-
ever, application of this protocol to substrates harboring a
more complex substitution pattern failed. In particular,
ic purification (SiO ) and was therefore directly treated
2
with benzyl alcohol to give acetal 37 in 73% yield, again as
ACS Paragon Plus Environment

Journal of the American Chemical Society
a 2:1 mixture of diastereomers (major isomer shown). The
newly installed benzyl ether functionality could later
serve to conveniently access the projected oxonium in-
termediate.
Page 6 of 15
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 7. Synthesis and oxidation of enol ether 33.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (a) 1-bromo-1-propene, n-BuLi,
THF/hexanes (1:1), ‒78 °C, 1.5 h; then CeCl , ‒78 °C, 30 min;
3
then 37, BF ∙OEt , ‒78 °C, 2.5 h, 35% (77% brsm). (b)
3
2
CF CO H, MeCN, 80 °C, 2 h, 78%. (c) Boc O, NEt , DMAP,
3
2
2
3
CH Cl , rt, 30 min, 77%. (d) Cp TiMe , pyridine, toluene, 70
a
2
2
2
2
Reagents and conditions: (a) MeNO , LDA, THF, 78 °C,
2
°
C, 6 h, 80%. (e) 9-BBN dimer, THF, rt, 3 h; then
30 min; then 7, ‒78 °C to rt, 2 h, 70%. (b) Boc O, DMAP,
toluene, rt, 12 h, 79%. (c) DMDO, 4Å MS, CH Cl /acetone
1:1), 0 °C, 2 h; then BnOH, CHCl , rt to 60 °C, 12 h, 73%, dr
:1.
2
H O /NaOH, rt, 1 h, 88% (3:1 mixture of diastereomers at
2
2
2
2
C(5)).
Isoxazolidine 38 was subjected to the conditions em-
ployed previously for acid-mediated ring-contraction
CF CO H, 80 °C). The desired β-lactam was thereby ob-
(
2
3
Having achieved efficient oxidation of C(3), we next fo-
(
3
2
cused on the elaboration of a suitable substrate for the
projected C(7)C(3) ring closure to assemble the gelse-
moxonine core. As presented in Scheme 8, application of
the previously developed chemistry was successfully ap-
plied to benzyl acetal 37. Addition of the propynyl ap-
pendage to 37 delivered oxazolidine 38 in 35% yield (77%
brsm). Notably, we did not observe any by-products re-
sulting from the reaction of the acid-labile acetal group in
tained in 78% yield. It is important to note, that the acid-
sensitive acetal in substrate 38 was not affected by the
reaction conditions. The structure of the β-lactam prod-
uct was confirmed by X-ray crystallographic analysis of
carbamate 40. Access to primary alcohol 41 was gained
through the previously established sequence of Petasis
olefination followed by enecarbamate hydroboration.
Interestingly, alcohol 41 was obtained as a 3:1 mixture of
diastereomers at C(5) as confirmed by derivatization and
3
7. Interestingly, when the reaction was performed in the
presence of more than 2 equivalents of BF etherate, a
significant amount (10-20%) of ring expanded dihydro-
3
46
subsequent nOe analysis (see Supporting Information).
With compound 41 in hand, we embarked on studies
aimed towards the addition of an oxindole nucleophile to
C(3) via the generation of an oxonium intermediate acces-
sible from the benzyl acetal. Preliminary experiments
revealed that α-fluoroacetal intermediates served opti-
mally for the generation of an oxonium ion at C(3). α-
Fluoroacetals have been successfully employed as donors
in glycosylation chemistry based on their stability and the
pyridine 39 was isolated along with the desired product
(
Scheme 8, box). Brandi and Salaün had previously de-
scribed the ring expansion of spirocyclopropane isoxa-
4
5
zolines. However, the conditions originally employed
involved high temperature (130-200 °C), whereas for-
mation of by-product 39 was observed at 78 °C. This
circumstance suggests a polar mechanism for the for-
mation of 39, opposed to the radical pathway proposed
for the high temperature ring expansion.
4
7
straightforward mode of activation. Accordingly, various
glycosyl donor-type substrates were prepared starting
from alcohol 41 and evaluated for their ability to undergo
intermolecular reaction with oxindole nucleophiles
a
Scheme 8. Synthesis of alcohol 41.
4
8
(
Scheme 9).
Scheme 9. Investigations towards C(7) functionaliza-
tion.
ACS Paragon Plus Environment

Page 7 of 15
Journal of the American Chemical Society
addition strategy. Motivated by this result, we set out to
1
2
3
4
5
6
7
8
9
prepare various substrates suited for intramolecular ring
closure by attack of a carbon nucleophile to C(3). Unfor-
tunately, subjecting α-fluoroacetals such as 50, incorpo-
rating an oxindole nuclephile at C(6), to various cycliza-
tion conditions, produced a complex mixture of dimeric
products (equation 4). Substrates incorporating smaller
nucleophiles at C(6) were not able to undergo ring clo-
sure. We hypothesized that the congested environment
around C(3) did not allow for carbon nucleophiles to
attack from the concave face of the molecule.
5
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The failure of the cyclization reactions detailed above
demanded drastic changes to the synthetic strategy. An
alternative approach for the construction of the central
ring of gelsemoxonine would entail early introduction of
C(7) and subsequent ring closure between a C(7) nucleo-
phile and a leaving group at C(6) (Scheme 1, strategy B).
According to this plan, we investigated functionalization
of the C(3) position in enol ether 33 by a carbon append-
age (Scheme 10). Again relying on the chemistry devel-
oped earlier, DMDO mediated epoxidation of enol ether
33 delivered a highly reactive oxirane, which was then
subjected to various carbon nucleophiles to effect epoxide
opening. We found that the use of ketene silyl acetal 51
was most successful providing ethyl ester 52 in 56% yield.
It is thereby noteworthy that careful tuning of the reac-
As shown in Scheme 9, our efforts started with acetate
protected diol 42. We reasoned that the C(14) acetate
could be engaged as a participating group in parallel with
its ability to undergo anchiomeric assistance as observed
in glycosylation reactions. In such a scenario, the incom-
ing nucleophile would be posed to attack opposite the
acetate group. In the event, treatment of 42 with N-
methoxy oxindole derived silyl enol ether 43 in the pres-
tion conditions and Lewis acid additive (InBr ) allowed
3
for the differential reaction of only the major epoxide
diastereomer, whereas the minor undesired isomer re-
mained unreactive and decomposed during workup. We
were therefore able to isolate 52 cleanly without the need
for tedious chromatographic separation of diastereomers.
Addition of the propynyl side chain proceeded readily to
produce isoxazolidine 53 in 78%. Notably, the ethyl ester
in 52 was not affected under these conditions and only
addition to the dihydroisoxazole was observed. Ring con-
4
9
ence of BF etherate as an activator delivered acetal 44
3
as the only product (equation 1). This outcome closely
parallels the chemistry of glycosyl donors incorporating a
sterically hindered anomeric center, whereby the acetate
participating group is attacked by the nucleophile rather
traction
of
53
proceeded
in
45%
yield
5
0
than the anomeric carbon. In order to circumvent such
issues, we next tested substrate 45, devoid of a protecting
group on the C(14) hydroxyl (equation 2). Indeed, we
observed the exclusive formation of C(3)-coupled oxin-
dole 46 under the conditions used before. However, clos-
er spectroscopic analysis as well as follow-up transfor-
mations of this product indicated that the newly formed
stereocenter at C(3) had the wrong configuration as indi-
cated in Scheme 9. We reasoned that installation of a
bulky protecting group on the C(14) alcohol could shield
the outer face of the putative oxonium intermediate and
thereby allow attack only from the opposite face. Follow-
ing this hypothesis, silyl ether 47 was prepared and treat-
a
Scheme 10. Synthesis of ester 55.
a
ed with silyl enol ether 48 in the presence of BF etherate.
Reagents and Conditions: (a) DMDO, 4Å MS,
3
To our surprise, cyclic acetal 49 was obtained as the only
product in moderate yield (57%) indicating intramolecu-
lar trapping of the highly hindered oxonium by the C(6)
oxygen. Although we ultimately opted for the addition of
carbon nucleophile to C(3), this transformation repre-
sented a proof of concept that formation the gelsemoxo-
nine core might indeed be possible through an oxonium
CH Cl /acetone (1:1), 0 °C, 2 h; then 51, InBr
2 2
3
(5 mol%),
CH Cl , 60 °C to rt, 2 h, 56% (dr > 20:1). (b) 1-bromo-1-
2 2
propene, n-BuLi, THF/hexanes (1:1), ‒78°C, 1.5 h; then CeCl
78 °C, 30 min; then 52, BF ∙OEt , ‒78 °C, 2 h, 78%. (c)
,
3
‒
3
2
CF CO H, MeCN, 80 °C, 6 h, 40-45%; or CF CO H, MeCN, 80
3
2
3
2
°
C, 2 h; then NEt , 35% (70% brsm). (d) Boc O, NEt , DMAP,
3
2
3
CH Cl , rt, 30 min, 85%.
2
2
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 15
under the previously used conditions to produce 54. the quaternary stereocenter at C(7). We initially envi-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
However, careful quenching of the reaction with NEt3
proved essential to achieve acceptable product yield.
Subsequent protection of the amide and the alcohol using
sioned the use of enolate arylation chemistry to achieve
this goal. For this purpose, saturated aldehyde 60 was
prepared by reduction of enal 58 using Stryker’s reagent.
Remarkably, we found that this reaction best proceeds at
78 °C to give the product as a 2:1 mixture of diastere-
omers after only 20 min of reaction time. Generally, ele-
vated temperatures and reaction times up to 24 h are
required to achieve the reduction of unsaturated alde-
Boc O delivered acylcarbamate 55. X-ray crystallographic
2
analysis confirmed the structure of 55.
As outlined in Scheme 11, introduction of the C(6)
methylene could again be achieved using the Petasis ole-
52
fination/hydroboration sequence employed previously
5
4
hydes using Stryker’s reagent. The unusually high reac-
tivity of substrate 58 can be attributed to a highly strained
double bond as a result of the tricyclic scaffold.
delivering alcohol 56. No reaction of the ester was ob-
served during the olefination step. We next attempted
closure of the cycloheptane ring of gelsemoxonine
through conversion of the C(6) hydroxyl group into a
potent leaving group (e.g. OMs or OTs) followed by SN2
displacement by a C(7) nucleophile. However, all the
reactions tested towards this end involving enolate or
enol chemistry, proved unsuccessful. We next prepared
dialdehyde 57, which would be amenable to an intramo-
lecular aldol cyclization. Intramolecular aldol reactions of
dialdehydes are often hampered by regioselectivity is-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 12. Attempts for enolate arylation.
53
a
sues. However, we surmised that the C(7) aldehyde in 57
Reagents and conditions: (a) [(PPh )CuH] , toluene, 78
3
6
might preferentially act as a nucleophile due to favorable
steric factors. Indeed, when 57 was exposed to 20 mol% of
pyrrolidine at 40 °C, condensation product 58 was ob-
tained in 73% yield as the sole product. Interestingly,
when DL-proline (20 mol%) was used as the promoter, we
selectively obtained secondary alcohol 59 (82%) as a sin-
gle diastereomer as confirmed by nOe and J-coupling
analysis. The ability to control the outcome of the aldol
cyclization step later proved essential in the synthesis as
detailed below.
°C, 20 min, 68% (dr 2:1). (b) NaClO , NaH PO ∙2H O, 2-
2 2 4 2
methyl-2-butene, t-BuOH/H
O (4:1), rt, 1 h; then TMSCHN ,
2
2
CH Cl /MeOH (9:1), rt, 15 min, 65%.
2
2
With saturated aldehyde 60 and its ester derivative 61 in
hand, various approaches for the introduction of an aryl
group at C(7) were tested. Very efficient protocols for
enolate arylation have recently been developed by Buch-
55
wald, Hartwig and Miura relying on addition of arylpal-
56
57
58
ladium reagents to esters, amides or aldehydes. How-
ever, all protocols tested on either aldehyde 60, ester 61
or amide derivatives thereof, failed to give the desired
products. Moreover, more traditional approaches such as
5
9
a
Fischer indolization or addition of electrophiles to eno-
lates of 60 or 61 similarly failed. We attribute the difficul-
ties for functionalization of C(7) to the steric hindrance
around this center as well as on difficulties of generating
Scheme 11. Closure of the gelsemoxonine core.
60
the respective enolate of 61.
An alternative strategy for the installation of the C(7)
oxindole group would entail arylation of the olefin in
unsaturated aldehyde 58 or its derivatives. In this respect,
intramolecular enolate arylation of an anilide would pro-
vide the requisite regioselectivity. Unfortunately, all the
conditions we evaluated for the oxidation of aldehyde 58
to its corresponding unsaturated carboxylic acid were
unsuccessful due to complications with the highly reac-
tive double bond in 58. Luckily, as depicted in Scheme 13,
we were able to perform Pinnick oxidation/esterification
of saturated aldehyde 59 obtained from the proline-
mediated aldol condensation shown in Scheme 11. Subse-
quent elimination of the secondary alcohol in the inter-
mediate methyl ester was achieved employing TFAA to
produce unsaturated ester 62. The structure of 62 could
be confirmed by X-ray crystallographic analysis. Ester
hydrolysis proved difficult as nucleophilic reagents such
a
Reagents and conditions: (a) Cp TiMe , pyridine, toluene,
2
2
7
0 °C, 8 h, 77% (85% brsm). (b) 9-BBN dimer, THF, rt, 45
min; then NaBO ∙4H O, THF/H O (1:1), rt, 1 h, 92%. (c)
DIBAL-H, THF/CH Cl (2:1), 78 °C, 75 min, 81%. (d)
COCl) , DMSO, CH Cl , 78 °C, 15 min; then diol substrate,
78 °C, 45 min; then NEt , 78 °C to rt, 1 h, 73%. (e) pyrroli-
dine (20 mol%), toluene, 40 °C, 2 h, 73%. (f) DL-proline (20
mol%), DMSO, rt, 12 h, 82%.
3
2
2
2
2
(

2 2 2
3
61
as LiOH or KOTMS reacted with the strained double
bond through conjugate addition. Only the use of
Oxindole formation: With the carbocyclic framework
of gelsemoxonine established, we next turned to the in-
stallation of the oxindole of the natural product including
Me SnOH at 80 °C provided the desired unsaturated car-
3
62
boxylic acid in excellent yield. Transformation into the
ACS Paragon Plus Environment

Page 9 of 15
Journal of the American Chemical Society
cyclization, the resulting alkyl palladium intermediate
corresponding acid chloride followed by amide bond
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
formation using 2-iodo-N-methylaniline as coupling part-
ner delivered N-methylamide 63 in 70% yield over 2 steps.
could undergo various side reactions competing with the
desired reductive quenching step, including β-hydride
67
a
elimination, opening of the adjacent azetidine or oxin-
dole ring opening leading back to the starting aryl palla-
dium species. Regardless of these expected difficulties, 63
Scheme 13. Construction of the oxindole.
was treated with Pd(OAc) in the presence of potassium
2
formate as hydride donor. To our delight, oxindole 64 was
isolated in 68% yield in a diastereomeric ratio of 10:1 fa-
voring the desired isomer as confirmed by nOe analysis.
We suspect that the face selectivity results from coordina-
tion of the palladium species to the adjacent oxygen of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
68
the ether bridge in 63. We next focused on the introduc-
tion of the methoxy group found on the amide nitrogen of
glesemoxonine (3). After initial attempts for oxidation of
an amide NH in either uncyclized aryl amides or oxindole
derivatives had failed, we turned our attention to the
69
a
synthesis of hydroxamic acid 65. However, N-selective
coupling of the gelsemoxonine core fragment with various
N-hydroxyaniline derivatives proved unexpectedly diffi-
Reagents and conditions: (a) NaClO , NaH PO ∙2H O, 2-
2
2
4
2
methyl-2-butene, t-BuOH/H O (4:1), rt, 20 min; then
2
TMSCHN , CH Cl /MeOH (9:1), rt, 10 min, 91%. (b) TFAA,
2
2
2
70
cult. We finally succeeded in synthesizing 65 through
DBU, THF, rt, 30 min, 94%. (c) Me SnOH, 1,2-
3
71
the combination of an etheral solution of the acid chlo-
dichloroethane, 80 °C, 24 h. (d) (COCl) , DMF (cat.), CH Cl ,
2
2
2
ride derived from 62 with 2-bromo-N-hydroxyaniline in
rt, 1.5 h; then 2-iodo-N-methylaniline, NEt , CH Cl , rt, 45
3
2
2
72
the presence of solid NaHCO3. Reductive Heck cycliza-
min, 70% (2 steps) (for 63). (e) (COCl) , DMF (cat.), CH Cl ,
2
2
2
tion of 65 was achieved under slightly modified condi-
rt, 1 h; then N-(2-bromophenyl)hydroxylamine, NaHCO3,
Et O/CH Cl (3:1), 0 °C, 45 min, 58% (85% brsm, 2 steps) (for
tions using PdCl (MeCN) (10 mol%) and formic acid as
2 2
65b
2
2
2
65). (f) SmI , N,N,N’,N’-tetramethylguanidine, H O (3.0
reductant. Oxindole 66 was thereby obtained in 72%
yield and as a single diastereomer. After methylation
(MeI, K CO ) the relative stereochemistry at C(7) of oxin-
2
2
euqiv.), THF, 20 °C. (g) 63, Pd(OAc) (10 mol%), KHCO , n-
2
2
Bu NBr, DMF, rt, 12 h, 68% (dr 10:1) (64). (h) 65,
4
2
3
PdCl (MeCN) (10 mol%), 1,2,2,6,6-pentamethylpiperidine,
dole 67 was again confirmed by nOe analysis.
2
2
HCO H, DMF, 60 °C, 1.5 h, 72% (dr > 20:1) (66). (i) NaH, MeI,
2
The final remaining task to elaborate the natural prod-
uct gelsemoxonine involved conversion of the alkyne in
67 to an ethyl ketone moiety. To this end, various proto-
cols for triple bond hydration were evaluated on 67 and
DMF, 0 °C to rt, 45 min, 92%.
Several reports have appeared documenting radical me-
diated cyclization of unsaturated N-arylamides to form
63
73
oxindole products. Following these procedures, we ex-
derivatives thereof (Scheme 14). However, all of the
posed iodoarene 63 to various conditions for generation
of the corresponding radical. Whereas the use of
tested conditions failed to deliver the desired product.
Instead, exposure of substrates incorporating a Boc pro-
tecting group on the azetidine to carbophilic reagents
75
AIBN/Bu SnH only resulted in protodeiodination, em-
3
74
2
ploying SmI as radical generator in the presence of vari-
such as (Ph
3
P)AuNTf
or [Cl Pt(C H )] (Zeise’s dimer)
2 2 4
2
ous different additives led to decomposition of the start-
ing material.
resulted in the formation of products such as 68 through
attack of the carbamate carbonyl group to C(19) (see Sup-
porting Information for further details). We therefore
opted for an alternative strategy involving a directed hy-
Overman and coworkers have extensively studied the
Heck reaction for the synthesis of oxindole rings harbor-
ing quaternary stereocenters, starting from haloarenes
76
drosilylation of the alkyne. As outlined in Scheme 14,
selective deprotection of the C(14) hydroxyl was carried
out using K CO /MeOH (87%). Subsequent treatment of
64
such as 63. However, the transiently formed alkyl palla-
dium intermediate generally undergoes β-hydride elimi-
nation to deliver an olefin as the final product, as dictated
by the Heck mechanism. In contrast, engaging amide 63
in a Heck cyclization would need to be followed by in situ
reductive quenching of a respective alkyl palladium in-
2
3
the resulting alcohol 69 with (Me SiH) NH (neat, 50 °C)
2
2
delivered an unstable siloxane, which was directly sub-
jected to {[RuCl (C H )] } (20 mol%) providing vinylsilane
2
6
6
2
70 as the only product in 58% yield. Interestingly, 70 was
obtained as a mixture of E/Z double bond isomers. This
observation is in agreement with previous reports on
alkyne hydrosilylation reactions using {[RuCl (C H )] } as
termediate to directly deliver a saturated product such as
65
6
4. Moreover, a few delicate obstacles would have to be
2
6
6
2
overcome to access 64 via a reductive palladium mediated
cyclization. An initially formed aryl palladium species
would have to attack the electron deficient olefin in 63
selectively at the highly congested C(7) with an approach
from the top face of the double bond. Such a diastereose-
lective olefin functionalization would not be expected on
7
7
catalyst. Ethyl ketone 71 could be accessed in 65% yield
from vinylsilane 70 by Tamao-Flemming oxidation using
78,79
KHF /H O /Ac O.
2
2
2
2
a
Scheme 14. Synthesis of ketone 71.
66
steric grounds. Furthermore, following any successful
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 15
EtOAc delivered the natural product gelsemoxonine (3) in
7% after aqueous workup of the reaction mixture. NMR
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
9
spectroscopic data (Figure 2, a) as well as IR and MS char-
acterization of the synthetic material were in complete
agreement with the reported data of the natural product.
Interestingly though, attempted chromatographic purifi-
8
0
cation of synthetic gelsemoxonine (3) or repeated con-
centration of the material from solution (CHCl or EtOAc)
3
resulted in the formation of a complex mixture of com-
1
pounds, as indicated by the H NMR spectrum obtained
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
after these attempts (Figure 2, b). The same result was
obtained when using d -pyridine as the NMR solvent,
6
a
Reagents and conditions: (a) K CO , MeOH, 50 °C, 30
excluding the possibility that residual acid promotes the
2
3
8
1
min, 87%. (b) (Me SiH) NH, 50 °C, 2.5 h; then
{[RuCl (C H )] } (20 mol%), CH Cl , rt, 17 h, 58%, E/Z 1:1. (c)
KHF , Ac O, H O , DMF, rt, 12 h, 65%. (d) 3 M HCl, EtOAc, 0
°C, 10 min, 97%.
observed chemistry. The original report documenting
the isolation and characterization of gelsemoxonine did
not mention any difficulties associated with the handling
2
2
2
6
6
2
2
2
2
2
2
2
of
the
natural
product
Various Brønsted and Lewis acids were tested to effect
final removal of the Boc carbamate. After some experi-
mentation we found that treatment of 71 with 3 M HCl in
(a)
(b)
(c)
*
*
*
*
*
1
Figure 2. Comparision of H NMR spectra in CDCl (5.2-3.0 ppm range) of (a) crude gelsemoxonine (3); (b) compound mixture
3
obtained after column chromatography (CH Cl /MeOH) or concentration of 3 from solution (CHCl or EtOAc); (c) product of
2
2
3
acetylation of the mixture shown in (b) corresponding to bisacetoxy gelsemoxonine 72 (after chromatography). *peaks from
rotamers (see also 2D NMR spectra of 72 in the Supporting Information).
6
under similar conditions. However, the authors reported
a bisacetylated derivative of gelsemoxonine obtained after
treatment of the isolated alkaloid with Ac O/pyridine.
Accordingly, we subjected the complex product mixture
obtained after concentration of synthetic gelsemoxonine
to the acetylation conditions. To our surprise, we ob-
tained a single product 72 from this reaction. Spectro-
scopic analysis indicated that this compound was identi-
cal to the bisacetylated gelsemoxonine derivative reported
by the isolation group (Figure 2, c). Based on these results
along with a closer spectroscopic analysis of the com-
pound mixture obtained from synthetic gelsemoxonine
(see Supporting information for further details) we pro-
pose that the natural product is prone to the formation of
2
ACS Paragon Plus Environment

Page 11 of 15
Journal of the American Chemical Society
a
oligomeric species upon concentration or chromato-
Reagents and conditions: (a) 73, NaH, THF, 0 °C, 30 min;
(5 mol%), dppe (7 mol%), THF, rt, 12 h, 52%
(62% brsm). (b) H NOH∙HCl, MeCN/H O (1:1), 80 °C, 12 h,
8
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
then 9, Pd(dba)
2
graphic purification. As indicated in Scheme 15, we hy-
pothesize that a dynamic equilibrium of various aggregat-
ed species exists, likely
2
2
86%. (c) (n-Bu Sn) O, CH Cl , rt, 1 h; then t-BuOCl, 30 °C to
rt, 30 min, 88% (dr 10:1). (d) FeCl , CH Cl , 0 °C to rt, 1 h,
3
2
2
2
3
2
2
87%. (e) PPh , I , imidazole, toluene/MeCN (5:1), rt, 12 h. (f)
3
2
a
Scheme 15. Proposed chemistry of gelsemoxonine.
DBU, toluene/MeCN (4:1), 100 °C, 6 h, 82% (2 steps). (g)
PPh , I , imidazole, toluene/MeCN (5:1), rt, 12 h; then DBU,
3
2
100 °C, 12 h, 62%.
treatment of 76 with FeCl3 delivered enantioenriched
alcohol ()-13 in 87% yield. Elimination of the secondary
alcohol using standard protocols resulted in very low
product yield or complete decomposition of the starting
material. Fortunately, we found that conversion into the
corresponding alkyl iodide (PPh , I ) followed by DBU
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (a) repeated concentration from
solution. (b) Ac O, pyridine, rt, 12 h, 60-65%.
2
3
2
resulting from the addition of the nucleophilic azeditine
mediated elimination delivered enol ether ()-33 in 82%
yield (2 steps). Alternatively, the substitution/elimination
sequence could be performed in one single step delivering
the product in slightly lower yield (62%).
8
3
nitrogen to the electrophilic ketone. This analysis is
further supported by the observation that the presence of
either the ketone or a free azetidine functionality alone is
not sufficient to produce the observed chemistry of the
natural product. We surmise that treatment of this com-
Conclusion
plex mixture of oligomers with Ac O results in the exclu-
sive reaction of the monomeric species, delivering 72 as
2
In summary, we have achieved the total synthesis of gel-
semoxonine (3) in 21 steps starting from known aldehyde
7. Construction of the central azetidine of the natural
product was possible through the use of an unusual spiro-
cyclopropane isoxazolidine ring contraction. In the course
of our synthetic studies, we have gained insight into the
electronic requirement as well as the functional group
tolerance of this transformation. Our group has also re-
cently reported a mechanistic study of this reaction re-
the sole product.
Enantioselective synthesis of enol ether 33: With an
efficient synthesis to (±)-gelsemoxonine (3) established,
we envisioned the development of an enantioselective
strategy to a key intermediate en route to the natural
product. Enol ether 33 thereby offered various opportuni-
ties for enantioselective preparation. As outlined in
Scheme 16, we implemented this plan by addition of en-
antioenriched alcohol 73, obtained in three steps from
to allylic tosylate 9. Subsequent
conversion of the resulting methylene cyclopropane de-
rivative 74 to oxime 75 could be carried out in one step
86%) without any erosion of the stereochemistry at C(14)
as confirmed by SFC analysis later in the synthesis. Dipo-
lar cycloaddition was effected by the generation of a reac-
tive nitrile oxide from 75 using (n-Bu Sn) O/t-BuOCl.
Oxazolidine 76 was obtained in 88% and as a 10:1 mixture
of diastereomers favoring the isomer depicted in Scheme
6 as confirmed by nOe analysis (Scheme 16, box). Re-
moval of the benzyl group was not possible by hydrogen-
olysis, as the NO bond underwent reductive cleavage
8
6
vealing a concerted nature of the reaction mechanism.
8
4
Furthermore, we have discovered previously unnoticed
chemistry of gelsemoxonine. The experimental data sug-
gests that the natural product can form oligomers
through condensation of its ketone functionality with the
nucleophilic azetidine nitrogen. Finally, we documented
an enantioselective approach to a key intermediate of the
presented synthesis.
(
+)-diethyl tartrate,
(
8
5
3
2
ASSOCIATED CONTENT
1
Experimental procedures and characterization data for all
1
13
reactions and products, including H and C NMR spectra.
This material is available free of charge via the Internet at
http://pubs.acs.org.
under
these
conditions.
However,
a
Scheme 16. Enantioselective synthesis of ()-33.
AUTHOR INFORMATION
Corresponding Author
erickm.carreira@org.chem.ethz.ch
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
This work was supported by a grant of the Swiss National
Science Foundation. We are grateful to Dr. B. Schweizer and
M. Solar for X-ray crystallographic analysis, and to Dr. M.-O.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 15
Ebert, R. Arnold, R. Frankenstein and P. Zumbrunnen for
NMR measurements. S.D. acknowledges the Stipendienfonds
Schweizerische Chemische Industrie (SSCI)
(16) For an overview covering nitrile oxide chemistry, see: (a)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Torssell, K. B. G. In Nitrile Oxides, Nitrones and Nitronates in
Organic Synthesis; Organic Nitro Chemistry Series; VCH Pub-
lisher Inc.: Deerfield Beach; Vol. 2, 1988, pp. 14-74. (b) Nitrile
Oxides, Nitrones, and Nitronates in Organic Synthesis, 2nd Ed.;
Feuer, H., Ed.; Wiley-Interscience: Hoboken; 2008, pp. 3-117. (c)
Ono, N. In The Nitro Group in Organic Synthesis; Wiley-VCH:
New York; 2001, pp. 231-301.
REFERENCES
(1) Baker, D. D.; Chu, M.; Oza, U.; Rajgarhia, V. Nat. Prod. Rep.
2007, 24, 1225.
(
2) Kitajima, M.; Arai, Y.; Takayama, H.; Aimi, N. Proc. Japan
Acad., Ser. B 1998, 74, 159.
3) For reviews on Gelsemium alkaloids: (a) Saxton, J. E. In The
(
17) Uno, H.; Terakawa, T.; Suzuki, H. Chem. Lett. 1989, 1079.
(18) Hatano, M.; Suzuki, S.; Ishihara, K. J. Am. Chem. Soc.
(
2006, 128, 9998.
Alkaloids; Manske, R. H. F., Ed.; Academic Press: New York, 1965,
Vol. 8, pp. 93-117. (b) Liu, Z.-J.; Lu, R.-R. In The Alkaloids; Brossi,
A., Ed.; Academic Press: San Diego, 1988; Vol. 33, pp. 83-140; (c)
Takayama, H.; Sakai, S.-I. In The Alkaloids; Cordell, G. A., Ed.;
Academic Press: San Diego, 1997; Vol. 49, pp. 1-78.
(4) For recent investigations of the biological activity of vari-
ous Gelsemium alkaloids, see: (a) Kitajima, M.; Nakamura, T.;
Kogure, N.; Ogawa, M.; Mitsuno, Y.; Ono, K.; Yano, S.; Aimi, N.;
Takayama, H. J. Nat. Prod. 2006, 69, 715. (b) Kitajima, M.; Naka-
mura, T.; Kogure, N.; Ogawa, M.; Mitsuno, Y.; Ono, K.; Yano, S.;
Aimi, N.; Takayama, H. J. Nat. Prod. 2007, 70, 142. (c) Zhao, Q.-
C.; Hua, W.; Zhang, L.; Guo, T.; Zhao, M.-H.; Yang, M.; Shi, G.-B.;
Wu, L.-J. J. Asian Nat. Prod. Res. 2010, 12, 731. (d) Liu, M.; Shen,
J.; Liu, H.; Xu, Y.; Su, Y.-P.; Yang, J.; Yu, C.-X. Biol. Pharm. Bull.
2011, 34, 1877.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(19) (a) Fraenkel, G.; Pramanik, P. J. Chem. Soc., Chem.
Commun. 1983, 1527. (b) Fraenkel, G. Polym. Prepr., Am. Chem.
Soc. Div. Polym. Chem. 1986, 27, 132. (c) Carreira, E. M.; Du Bois,
J. J. Am. Chem. Soc. 1995, 117, 8106.
(
4
20) TiCl has been reported to promote addition of organo-
metallic nucleophiles to β-hydroxy ketones: (a) Bartoli, G.;
Bosco, M.; Di Martino, E.; Marcantoni, E.; Sambri, L. Eur. J. Org.
Chem. 2001, 2901. (b) Reetz, M. T.; Kyung, S. H.; Hüllmann, M.
Tetrahedron 1986, 42, 2931. (c) Reetz, M. Angew. Chem. Int. Ed.
Engl. 1984, 23, 556.
(
21) (a) Krause, N.; Seebach, D. Chem. Ber. 1987, 120, 1845. (b)
Ukaji, Y.; Kanda, H.; Ymamoto, K.; Fujisawa, T. Chem. Lett. 1990,
9, 597. (c) Weidmann, B.; Seebach, D. Angew. Chem. Int. Ed.
Engl. 1983, 22, 31.
22) (a) Gocmen, M.; Soussan, G. J. Organomet. Chem. 1974,
0, 303. (b) Canonne, P.; Foscolos, G.; Caron, H.; Lemay, G.
1
(
(5) Lin, L.-Z.; Cordell, G. A.; Ni, C.-Z.; Clardy, J. Phytochemis-
try 1991, 30, 1311.
8
Tetrahedron 1982, 38, 3563. (c) Swain, C. G.; Boyles, H. B. J. Am.
Chem. Soc. 1951, 73, 870. (d) Chastrette, M.; Amouroux, R. Bull.
Soc. Chim. Fr. 1970, 4348. (e) Chastrette, M.; Amouroux, R. J.
Chem. Soc., Chem. Commun. 1970, 470.
(
6) Kitajiama, M.; Kogure, N.; Yamaguchi, K.; Takayama, H.;
Aimi, N. Org. Lett. 2003, 5, 2075.
7) Shimokawa, J.; Harada, T.; Yokoshima, S.; Fukuyama, T. J.
Am. Chem. Soc. 2011, 133, 17634.
8) For representative examples, see: (a) Wuitschik, G.; Rog-
(
(
23) (a) Imamoto, T.; Takiyama, N.; Nakamura, K. Tetrahedron
(
Lett. 1985, 26, 4763. (b) Imamoto, T.; Takiyama, N.; Nakamura,
K.; Hatajima, T.; Kamiya, Y. J. Am. Chem. Soc. 1989, 111, 4392. (c)
Krasovskiy, A.; Kopp, F.; Knochel, P. Angew. Chem. Int. Ed. 2006,
ers-Evans, M.; Müller, K.; Fischer, H.; Wagner, B.; Schuler, F.;
Polonchuk, L.; Carreira, E. M. Angew. Chem. Int. Ed. 2006, 45,
7736. (b) Wuitschik, G.; Rogers-Evans, M.; Buckl, A.; Bernasconi,
M.; Märki, M.; Godel, T.; Fischer, H.; Wagner, B.; Parrilla, I.;
Schuler, F.; Schneider, J.; Alker, A.; Schweizer, W. B.; Müller, K.;
Carreira, E. M. Angew. Chem. Int. Ed. 2008, 47, 4512. (c)
Burkhard, J. A.; Wagner, B.; Fischer, H.; Schuler, F.; Müller, K.;
Carreira, E. M. Angew. Chem. Int. Ed. 2010, 49, 3524.
4
5, 497.
3
24) Anhydrous CeCl was prepared according to: Dimitrov,
(
V.; Kostova, K.; Genov, M. Tetrahedron Lett. 1996, 37, 6787.
(25) (a) García Ruano, J. L.; Tito, A.; Culebras, R. Tetrahedron
1
996, 52, 2177. (b) García Ruano, J. L.; Martín, A. M.; Rodriguez, J.
H. Tetrahedron Lett. 1991, 32, 3195.
26) Various different protic acid promotors (TFA, p-TsOH, HCl,
TfOH, Cl CCO H, MsOH, 2-chloropyridium hydrochloride) were
(
9) Diethelm, S.; Carreira, E. M. J. Am. Chem. Soc. 2013, 135,
8500.
(10) This step was supported by similar reactions reported in
(
3
2
tested but TFA gave the product in the highest yield (see Supporting
Information, Table S1, for further details).
(27) β-Lactams are prone to ring opening due to their inher-
ent ring strain of 28 kcal/mol: Roux, M. V.; Jiménez, P.; Dávalos,
J. Z.; Castaño, O.; Molina, M. T.; Notario, R.; Herreros, M.; Ab-
boud, J.-L. M. J. Am. Chem. Soc. 1996, 118, 12735.
the context of model studies towards the total synthesis of
gelsedine as well as a recent total synthesis of gelsemine: (a)
Kende, A. S.; Luzzio, M. J.; Mendoza, J. S. J. Org. Chem. 1990, 55,
9
18. (b) Zhou, X.; Xiao, T.; Iwama, Y.; Qin, Y. Angew. Chem. Int.
Ed. 2012, 51, 4909.
11) For an overview on CH oxidation in complex molecule
synthesis, see: (a) Davies, H. M. L.; Manning, J. R. Nature, 2008,
51, 417. (b) McMurray, L.; O’Hara, F.; Gaunt, M. J. Chem. Soc.
(
(28) For the use of the Corey-Chaykosvky reaction on amides
with reduced CN double bond character, see: (a) Szostak, M.;
Aubé, J. J. Am. Chem. Soc. 2009, 131, 13246. (b) Reduced amide
resonance in β-lactams has been postulated: Glover, S. A.;
Rosser, A. A. J. Org. Chem. 2012, 77, 5492.
(29) (a) Fang, F. G.; Danishefsky, S. J. Tetrahedron Lett. 1989,
30, 2747. (b) Kim, G.; Chu-Moyer, M. Y.; Danishefsky, S. J. J. Am.
Chem. Soc. 1990, 112, 2003. (c) Maggini, M.; Prato, M.; Ranelli,
M.; Scorrano, G. Tetrahedron Lett. 1992, 33, 6537. (d) Kim, G.;
Kang, S.; Keum, G. Tetrahedron Lett. 1994, 35, 3747.
4
Rev. 2011, 40, 1885. (c) Gutekunst, W. R.; Baran, P. S. Chem. Soc.
Rev. 2011, 40, 1976. (d) Yamaguchi, J.; Yamaguchi, A. D.; Itami, K.
Angew. Chem. Int. Ed. 2012, 51, 8960. (e) for a recent discussion
on the selectivity of C–H bond oxidations, see: Newhouse, T.;
Baran, P. S. Angew. Chem. Int. Ed. 2011, 50, 3362.
(
12) For selected reviews on β-lactam and azetidine synthesis,
see: (a) Brandi, A.; Chicchi, S.; Cordero, F. M. Chem. Rev. 2008,
08, 3988. (b) The Chemistry of β-Lactams; Page, M. I., Ed.;
1
(30) (a) Tebbe, F. N.; Parshall, G. W.; Reddy, G. S. J. Am.
Chapman & Hall: London, 1992.
Chem. Soc. 1978, 100, 3611. (b) Petasis, N. A.; Bzowej, E. I. J. Am.
Chem. Soc. 1990, 112, 6392. (c) Petasis, N. A.; Lu, S.-P.
Tetrahedron Lett. 1995, 36, 2393.
(31) Martínez, I.; Howell, A. R. Tetrahedron Lett. 2000, 41,
5607.
(13) Cordero, F. M.; Pisaneschi, F.; Goti, A.; Ollivier, J.; Salaün,
J.; Brandi, A. J. Am. Chem. Soc. 2000, 122, 8075.
(
14) Ferrara, M.; Cordero, F. M.; Goti, A.; Brandi, A.; Estieu, K.;
Paugam, R.; Ollivier, J.; Salaün, J. Eur. J. Org. Chem. 1999, 2725.
15) 8 can also be prepared in one step according to: Salaün, J.;
Marguerite, J. Org. Synth. 1985, 63, 147.
(
(32) De Mico, A.; Margarita, R.; Parlanti, L.; Vescovi, A.;
Piancatelli, G. J. Org. Chem. 1997, 62, 6974.
ACS Paragon Plus Environment

Page 13 of 15
Journal of the American Chemical Society
(
33) 22 was prepared according to: (a) Somei, M.; Kawasaki, T.
S. Carbohyd. Res. 2004, 339, 2407. (e) Plante, O. J.; Palmacci, E.
R.; Andrade, R. B.; Seeberger, P. H. J. Am. Chem. Soc. 2001, 123,
9545. (f) Szpilman, A. M.; Carreira, E. M. Org. Lett. 2009, 11, 1305.
(51) Ester and amide substrates were tested such as:
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Heterocycles 1989, 29, 1251. (b) Kawasaki, T.; Kodama, A.;
Nishida, T.; Shimizu, K.; Somei, M. Heterocycles 1991, 32, 221.
(34) (a) Evans, D. A.; Tedrow, J. S.; Shaw, J. T.; Downey, C. W.
J. Am. Chem. Soc. 2002, 124, 392. (b) Evans, D. A.; Downey, C.
W.; Shaw, J. T.; Tedrow, J. S. Org. Lett. 2002, 4, 1127.
(
35) For selected examples of innate CH oxidations at etheral
positions, see: (a) Djerassi, C.; Engle, R. R. J. Am. Chem. Soc.
953, 75, 3838. (b) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.;
3
(52) NaBO was used for oxidative quenching of the hydrobo-
ration reaction: Kabalka, G. W.; Shoup, T. M.; Goudgaon, N. M. J.
Org. Chem. 1989, 54, 5930.
1
Sharpless, K. B. J. Org. Chem. 1981, 46, 3936. (c) McNeill, E.; Du
Bois, J. J. Am. Chem. Soc. 2010, 132, 10202. (d) Heyns, K.;
Buchholz, H. Chem. Ber. 1976, 109, 3707. (e) Sen, A.; Lin, M.; Kao,
L.-C.; Hutson, A. C. J. Am. Chem. Soc. 1992, 114, 6385. (f) Olah, G.
A.; Gupta, B. G. B.; Fung, A. P. Synthesis 1980, 897. (g) Harrison,
I. T.; Harrision, S. Chem. Commun. 1966, 752.
(53) (a) For a comprehensive overview covering intramolecu-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
lar aldol reactions, see: Heathcock, C. H. In Comprehensive Or-
ganic Synthesis; Trost, B. M.; Flemming, I., Eds; Pergamon: Ox-
ford, 1991; Vol. 2, pp. 156-170. See also: (b) Snyder, S. A.; Corey, E.
J. Tetrahedron Lett. 2006, 47, 2083.
(36) For an overview on directed CH oxidation, see: Brückl,
(54) a) Mahoney, W.S.; Brestensky, D.M.; Stryker, J.M. J. Am.
T.; Baxter, R. D.; Ishihara, Y.; Baran, P. S. Acc. Chem. Res. 2012,
5, 826.
Chem. Soc. 1988, 110, 291-293; b) Mahoney, W.S.; Stryker, J.M. J.
Am. Chem. Soc. 1989, 111, 8818-8823; c) Brestensky, D.M.; Stryker,
J.M. Tetrahedron Lett. 1989, 30, 5677-5680.
4
4
(37) (a) Espino, C. G.; Du Bois, J. Angew. Chem. Int. Ed. 2001,
0, 598. (b) Espino, C. G.; Wehn, P. M.; Chow, J.; Du Bois, J. J.
Am. Chem. Soc. 2001, 123, 6935.
(38) (a) Espino, C. G.; Fiori, K. W.; Kim, M.; Du Bois, J. J. Am.
Chem. Soc. 2004, 126, 15378. (b) Zalatan, D. N.; Du Bois, J. J. Am.
Chem. Soc. 2009, 131, 7558.
(55) For an overview covering palladium catalysed α-arylations
of carbonyl compounds, see: Culkin, D. A.; Hartwig, J. F. Acc.
Chem. Res. 2003, 36, 234.
(
56) (a) Moradi, W. A.; Buchwald, S. L. J. Am. Chem. Soc. 2001,
23, 7996. (b) Hama, T.; Hartwig, J. F. Org. Lett. 2008, 10, 1549.
c) Hama, T.; Liu, X.; Culkin, D. A.; Hartwig, J. F. J. Am. Chem.
1
(39) (a) Hofmann, A. W. Ber. 1883, 16, 558. (b) Löffler, K.;
(
Freytag, C. A. Ber. 1909, 42, 3427. (c) Löffler, K.; Kober, S. Ber.
909, 42, 3431. (d) Corey, E. J.; Hertler, W. R. J. Am. Chem. Soc.
960, 82, 1657.
Soc. 2003, 125, 11176. (d) Liu, W.; Hartwig, J. F. J. Am. Chem. Soc.
2004, 126, 5182.
1
1
(57) (a) Hama, T.; Culkin, D. A.; Hartwig, J. F. J. Am. Chem.
Soc. 2006, 128, 4976. (b) Lee, S.; Hartwig, J. F. J. Org. Chem. 2001,
(40) (a) Barton, D. H. C.; Beckwith, A. L. Proc. Chem. Soc.
1963, 335. (b) Barton, D. H. C.; Beckwith, A. L.; Goosen, A. J.
Chem. Soc. 1965, 181. (c) Petterson, R. C.; Wambsgans, A. J. J. Am.
Chem. Soc. 1964, 86, 1648.
6
6, 3402. (c) Shaughnessy, K. H.; Hamann, B. C.; Hartwig, J. F. J.
Org. Chem. 1998, 63, 6546.
58) For pioneering work, see: (a) Satoh, T.; Kawamura, Y.;
(
(
41) For reviews on the generation and chemistry of amidyl
radicals, see: (a) Mackiewicz, P.; Furstoss, R. Tetrahedron 1978,
4, 3241. (b) Zard, S. Z. Chem. Soc. Rev. 2008, 37, 1603.
42) Although we believe that the oxygen of the enol ether can
Miura, M.; Nomura, M. Angew. Chem. Int. Ed. 1997, 36, 1740. (b)
Palucki, M.; Buchwald, S. L. J. Am. Chem. Soc. 1997, 119, 11108. (c)
Hamann, B. C.; Hartwig, J. F. J. Am. Chem. Soc. 1997, 119, 12382.
(59) (a) Fischer, E.; Jourdan, F. Ber. Deutsch. Chem. Ges. 1883,
16, 2241. (b) Fischer, E.; Hess, O. Ber. Deutsch. Chem. Ges. 1884,
3
(
also assist double bond isomerization through oxonium for-
mation, DMAP is most likely the major contributor for the isom-
erization reaction. Using only catalytic amounts of DMAP (10
mol%) drastically decreased the yield of the reaction (10-15%)
and led to the formation of a complex mixture of polymeric
decomposition products.
1
7, 559. For reviews covering the Fischer indole synthesis, see: (c)
The Fischer Indole Synthesis, Downing, R. S.; Kunkeler, P. J.;
Wiley-VCH: Weinheim, 2001. (d) Hughes, D. L. Org. Prep. Pro-
ced. Int. 1993, 25, 607.
(60) Treatment of 61 with strong bases (e.g. LiHMDS, 78 °C
(
43) Halcomb, R. L.; Danishefsky, S. J. J. Am. Chem. Soc. 1989,
11, 6661.
44) Epoxide 36 proved highly sensitive to water. Carful exclu-
or rt) followed by quenching with D O did not lead to incorpora-
2
1
tion of deuterium at the α-position of the ester, thus indicating
that enolate formation does not occur.
(
sion of water from the reaction by addition of molecular sieves
was therefore crucial.
(45) (a) Guarna, A.; Brandi, A.; Goti, A.; De Sarlo, F. J. Chem.
Soc., Chem. Commun. 1985, 1518. (b) Brandi, A.; Guarna, A.; Goti,
A.; Pericciuoli, F.; De Sarlo, F. Tetrahedron Lett. 1986, 27, 1727.
(61) Laganis, E. D.; Chenard, B. L. Tetrahedron Lett. 1984, 25,
5
831.
62) (a) Furlán, R. L. E.; Mata, E. G.; Mascaretti, O. A.
(
Tetrahedron Lett. 1996, 37, 5229. (b) Nicolaou, K. C.; Nevalainen,
M.; Zak, M.; Bulat, S.; Bella, M.; Safina, B. S. Angew. Chem. Int.
Ed. 2003, 42, 3418.
(46) The reduced diastereoselectivity of this hydroboration might
be a result of a conformational change of the pyran ring brought about
by the OBn group that leads to opening of the endo-face of the ene-
carbamate.
(47) For an overview on glycosyl fluoride chemistry, see:
Shoda, S. In Handbook of Chemical Glycosylation; Demchenko,
A.V., Ed.; Wiley-VCH: Weinheim, 2008; pp. 29-59.
(63) (a) Jones, K.; Thompson, M.; Wright, C. J. Chem. Soc.,
Chem. Commun. 1986, 115. (b) Jones, K.; McCarthy, C. Tetrahe-
dron Lett. 1989, 30, 2657. (c) Hart, D. J.; Wu, S. C. Tetrahedron
Lett. 1991, 32, 4099. (d) Clark, A. J.; Jones, K. Tetrahedron Lett.
1989, 30, 5485. (e) Cabri, W.; Candiani, I.; Colombo, M.; Franzi,
L.; Bedeschi, A. Tetrahedron Lett. 1995, 36, 949. (f) Iwasaki, H.;
Eguchi, T.; Tsutsui, N.; Ohno, H.; Tanaka, T. J. Org. Chem. 2008,
73, 7145.
(64) (a) Abelman, M. M.; Oh, T.; Overman, L. E. J. Org. Chem.
1987, 52, 4130. (b) Madin, A.; O’Donnell, C. J.; Oh, T.; Old, D. W.;
Overman, L. E.; Sharp, M. J. J. Am. Chem. Soc. 2005, 127, 18054.
(65) For selected examples for the use of reductive Heck reac-
tions in total synthesis, see: (a) Clayton, S. C.; Regan, A. C. Tet-
rahedron Lett. 1993, 34, 7493. (b) Trost, B. M.; Thiel, O. R.; Tsui,
H.-C. J. Am. Chem. Soc. 2003, 125, 13155. (c) Ichikawa, M.;
(
48) For the preparation of α-fluoroethers 42/45/47 see Sup-
porting Information.
49) Various reagents were tested to activate the α-fluoro
(
ethers including TMSOTf, ZnCl
proved to be the best reagent.
(
Razon, P.; Wirtz, C.; Mynott, R. Chem. Eur. J. 2003, 9, 320. (b)
Gung, B. W.; Fox, R. M. Tetrahedron 2004, 60, 9405. (c) Nico-
laou, K. C.; Danies, R. A.; Ogawa, Y.; Chakraborty, T. K. J. Am.
Chem. Soc. 1988, 110, 4696. (d) Kuszmann, J.; Medgyes, G.; Boros,
2
and BF
3
etherate. BF
3
etherate
50) For selected examples, see: (a) Fürstner, A.; Jeanjean, F.;
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 15
Takahashi, M.; Aoyagi, S.; Kibayashi, C. J. Am. Chem. Soc. 2004,
126, 16553. (d) Gao, P.; Cook, S. P. Org. Lett. 2012, 14, 3340. (e)
Ekebergh, A.; Börje, A.; Mårtensson, J. Org. Lett. 2012, 14, 6274.
(75) Hiscox, W.; Jennings, P. W. Organometallics 1990, 9,
1997.
(76) For a recent review on olefin hydrosilylation, see: Trost,
B. M.; Ball, Z. T. Synthesis 2005, 853.
(77) Denmark, S. E.; Pan, W. Org. Lett. 2002, 4, 4163.
(78) (a) Tamao, K.; Ishida, N.; Tanaka, T.; Kumada, M. Organ-
ometallics 1983, 2, 1694. (b) Fleming, I.; Henning, R.; Plaut, H. J.
Chem. Soc. Chem. Commun. 1984, 29. (c) Jones, G. R.; Landais,
Y. Tetrahedron 1996, 52, 7599.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(66) In the crystal structure of ester 62, no significant steric
difference between the two faces of the olefin are apparent (see
figure below). We assume a comparable situation for amides
such as 63 or 65.
(79) (a) Trost, B. M.; Ball, Z. T. J. Am. Chem. Soc. 2003, 125, 30.
(b) Trost, B. M.; Ball, Z. T.; Laemmerhold, K. M. J. Am. Chem.
Soc. 2005, 127, 10028.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(80) Column chromatography was tried with several different
solvent systems but the oligomerization chemistry was observed
under any of these conditions (see Supporting Information for
further details).
(67) β-hydride elimination would be energetically costly due to the
highly strained olefin generated.
(68) For recent examples of ethers as directing groups in pal-
ladium catalyzed reactions, see: (a) Li, G.; Leow, D.; Wan, L.; Yu,
J.-Q. Angew. Chem. Int. Ed. 2013, 52, 1245. (b) Jamieson, A. G.;
Sutherland, A. Org. Biomol. Chem. 2005, 3, 735.
6
(81) Upon dissolving the crude material in d -pyridine, gelse-
moxonine slowly (within about 10-20 min) underwent the de-
scribed oligomerization chemistry as observed for other solvents.
(82) We believe that residual acid (either from CHCl or from
3
the workup) as the promotor of the observed oligomerization
chemistry can be excluded, as already concentration of clean
gelsemoxonine from EtOAc or measuring the NMR spectra in
pyridine produces the observed chemistry. Again, these oligo-
mers could be resolved by acetylation.
(
69) For the oxidation of amides to hydroxamic acids, see: (a)
Matlin, S. A.; Sammes, P. G. J. Chem. Soc., Chem. Commun. 1972,
222. (b) Matlin, S. A.; Sammes, P. G.; Upton, R. M. J. Chem. Soc.,
1
Perkin Trans. 1 1979, 2481. (c) Rigby, J. H.; Qabar, M. J. Org.
Chem. 1989, 54, 5852. (d) Brewer, G. A.; Sinn, E. Inorg. Chem.
1981, 20, 1823.
(70) A complex mixture of products including significant
amounts of O-coupling product was obtained.
(83) Retro-aldol reactions involving the ketone, hydroxyl,
ether or oxindole moieties can be ruled out based on experi-
ments with various derivatives of 3, the observed clean acetyla-
tion of the compound mixture presented in Figure 2 as well as
spectroscopic analysis of this mixture (see Supporting Infor-
mation).
(
71) Employing other solvents such as CH
led to decomposition of the N-hydroxyaniline starting material.
72) (a) Gupta, V. K.; Tandon, S. G. J. Indian. Chem. Soc. 1969,
6, 831. (b) Smissman, E. E.; Corbett, M. D. J. Org. Chem. 1972,
2 2
Cl in the reaction
(
4
(84) 73 has been previously prepared using a slightly different
3
7, 1847.
procedure: Suami, T.; Tadano, K.-I.; Suga, A.; Ueno, Y. J. Carbo-
hydr. Chem. 1984, 3, 429. For further details see Supporting
Information.
(73) For a comprehensive review covering alkyne hydration,
see: Hintermann, L.; Labonne, A. Synthesis 2007, 1121.
(74) (a) Buzas, A.; Gagosz, F. Org. Lett. 2006, 8, 515. (b) Buzas,
A. K.; Istrate, F. M.; Gagosz, F. Tetrahedron 2009, 65, 1889. For
similar gold catalyzed cyclizations, see: (c) Kang, J.-E.; Shin, S.
Synlett 2006, 717. (d) Shin, S. Bull. Korean Chem. Soc. 2005, 26,
(85) Moriya, O.; Takenaka, H.; Iyoda, M.; Urata, Y.; Endo, T. J.
Chem. Soc., Perkin Trans. 1 1994, 413.
(86) Diethelm, S.; Schoenebeck, F.; Carreira, E. M. Org. Lett.
2014, 16, 960.
1925. (e) Das, A.; Chang, H.-K.; Yang, C.-H.; Liu, R. S. Org. Lett.
2
008, 10, 4061. (f) Das, A.; Liao, H.-H.; Liu, R.-S. J. Org. Chem.
2007, 72, 9214.
ACS Paragon Plus Environment

Page 15 of 15
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
5
ACS Paragon Plus Environment