Regioselective Bromination of Benzocycloheptadienones for the Synthesis of Substituted 3,4-Benzotropolones Including Goupiolone A

Article abstract of DOI:10.1002/chem.201700622

Type-18 or -23 benzocycloheptadienones are readily prepared by ring-closing olefin metatheses. Adding Br₂ to 23 and eliminating HBr gave the bromoolefin 28 using DBU or its isomer iso-28 using DABCO, both with near-perfect regiocontrol. Both 28

Full text of DOI:10.1002/chem.201700622

A Journal of
Accepted Article
Title: Regioselective Bromination of Benzocycloheptadienones for
the Synthesis of Substituted 3,4-Benzotropolones Including
Goupiolone A
Authors: Reinhard Brückner, Deniz Arican, and Stefan Braukmüller
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201700622
Link to VoR: http://dx.doi.org/10.1002/chem.201700622
Supported by

10.1002/chem.201700622
Chemistry - A European Journal
Entry for the Table of Contents
Nonbenzenoid Aromatic
Compounds
OMe
HO
HO
HO
for
1. Br2
1. Br2
MeO
O
2. DABCO
(g 9-bromo-1)
2. DBU
R =
O
O
O
R
CO2Et
8
9
8
8
(g 8-bromo-1)
CO2Et:
BBr3
9
MeO
MeO
MeO
HO
HO
MeO
MeO
D. Arican, S. Braukmüller, R. Brückner*
________________ Page – Page
3. C,C-coupling
4. H3O
3. C,C-coupling
4. H3O
R
MeO
9-R-1
1
8-R-1
goupiolone A
Regioselective Bromination of
Benzocycloheptadienones for the Synthesis
of Substituted 3,4-Benzotropolones
Including Goupiolone A
Carbofunctionalizing a C=C bond regioselectively: Adding Br2 to the RCM product 1
and eliminating HBr gave 8-bromo-1 using DBU and 9-bromo-1 using DABCO. Both
compounds underwent Sonogashira, Suzuki, Negishi, and Heck couplings and formed esters
with CO in ROH. Hydrolysis of the ketal group liberated the enol moiety of
benzotropolones type 8- or 9-R-1. One of them was bis(demethylated) providing goupiolone
A in fewer steps (9) than hitherto (19).
1
This article is protected by copyright. All rights reserved.

10.1002/chem.201700622
Chemistry - A European Journal
DOI: 10.1002/anie.201((will be filled in by the editorial staff))
Nonbenzenoid Aromatic Compounds
Regioselective Bromination of Benzocycloheptadienones for the Syn-
thesis of Substituted 3,4-Benzotropolones Including Goupiolone A**
Deniz Arican,†,‡ Stefan Braukmüller,† and Reinhard Brückner†,*
Abstract: Type-18 or -23 benzocycloheptadienones are readily
prepared by ring-closing olefin metatheses. Adding Br2 to 23 and
eliminating HBr gave the bromoolefin 28 using DBU or its isomer
iso-28 using DABCO, both with near-perfect regiocontrol. Both 28
and iso-28 underwent Sonogashira, Suzuki, Negishi, and Heck
couplings as well as Pd-catalyzed alkoxycarbonylations. Hydrolysis
pyrogallols (→ tropolone moiety). The method was first reported in
1869 [2 × pyrogallol → purpurogallin (3)6] and, varying the oxidant,
has remained the most frequently used approach till now.7,8 Recent-
ly, it provided the 3,4-benzotropolones 47b − the first “small mole-
cule“ inhibitor of a toll-like receptor heterodimer7b and hence a lead
for the chemotherapy of acne9 or sepsis10 − and 5,7a namely thea-
flavin.11
of the resulting α-ketoketals and enolization of the liberated
α-
diketones delivered portfolio of hitherto unknown 3,4-
a
benzotropolones. The 8-ethoxycarbonylated dimethyl-3,4-benzotro-
polone 50 obtained by this route was demethylated to give goupio-
lone A (52). This synthesis encompasses 9 steps from 22, i. e., half
as many as the only previous synthesis (19 steps). A variant of our
route afforded the 1,8-dibromide 54. Coupling with excess phenyl-
boronic acid and ketal hydrolysis provided the diphenylated
benzotropolone 56 and suggests a strategy, by which the natural
bispulvinone aurantricholone (7) might be reached.
HO
HO
HO
HO
7
O
O
O
O
8
R3
CO2Hex
H
O
H
O
H
O
9
O
H
R2
O
H
O
H
R1
OH
OMe
4: CU-CDT227b
1
3
2
O
OH
Ph
O
O
HO
HO
HO
CO2H
O
O
2-Hydroxycyclohepta-2,4,6-trienon-1-one (= tropolone) was the
first isocyclic non-benzenoid aromatic compound recognized.1 Tro-
polones may tautomerize to give “transposed” tropolones,2 which
may be separable by crystallization.3 In contrast, 3,4-benzotropol-
ones (1, Fig. 1) do not tautomerize (→ less stable4 6,7-benzotropol-
ones). At least 36 3,4-benzotropolones including one bis(3,4-benzo-
tropolone)5 were identified in nature. Without exception their sub-
stitution patterns equal 2. In the laboratory, such benzotropolones
result from oxidizing mixtures of catechols (→ benzene moiety) and
O
O
O
OH
OH
O
H
O
H
H
O
O
Ca2
Ph
OH
O
O
H
CO2H
O
H
O
H
O
HO
OH
5: theaflavin7a
6: crocipodin7c
7: aurantricholone12
Figure 1. 3,4-Benzotropolone (1), 3,4-benzotropolones 2 from nature or, in the
case of purpurogallin (3), from the oxidation of catechol/ pyrogallol mixtures.
Selected C-functionalized synthetic (4) and naturally occurring 3,4-benzotropo-
lones (5-7).
[†] Dr. D. Arican, Dr. S. Braukmüller, Prof. Dr. R. Brückner
Institut für Organische Chemie
It appears as if not each benzotropolone of the substitution pat-
tern 2 is accessible by co-oxidizing catechol/pyrogallol mixtures or
oxidizing pyrogallols alone. Crocipodin (6) is a reported excep-
tion.7c Aurantricholone (7)12 might be another one.13 Such limita-
tions invite to extend existing methodology − as en route to 67c − or
to contribute novel methodology − like our metathesis route to ben-
zotropolones 10. 14 It is outlined in Scheme 1. It perceives 10 as the
enol of an α-diketone and reaches the latter by hydrolyzing the re-
spective monoketal 9. Representing cycloalkenes as well, com-
pounds like 9 can be made by ring-closing type-8 dienes in the pres-
ence of the 2nd generation Grubbs catalyst15 (“Grubbs II catalyst”).
Albert-Ludwigs-Universität Freiburg
Albertstraße 21, D-79104 Freiburg (Germany)
Fax: (+) 49 761 203 6100
E-Mail: reinhard.brueckner@ocbc.uni-freiburg.de
Homepage: http://www.brueckner.uni-freiburg.de
[‡] New address: Carbogen-Amcis AG
Schachenallee 29, 5001 Aarau, Switzerland
[
]
The authors thank Dr. Jens Geyer (at the time Institut für Organ-
ische Chemie, Albert-Ludwigs-Universität Freiburg) for the X-
ray analyses. They are grateful to Clara Heimburger (at the time
Institut für Organische Chemie, Albert-Ludwigs-Universität Frei-
burg) and Tamara Huck (Institut für Organische Chemie, Albert-
Ludwigs-Universität Freiburg) for their skilled assistance. They
express their gratitude to the Deutsche Forschungsgemeinschaft
for financial support.
Supporting information (Experimental procedures,
characterization data, copies of NMR spectra, and
crystallographic details) for this article is available on the WWW
under http://www.angewandte.org.
2
This article is protected by copyright. All rights reserved.

10.1002/chem.201700622
Chemistry - A European Journal
OMe
MeO
1) BuLi, Et2O, rt; −78°C; 17, rt
2) Dess-Martin reagent, pyridine, CH2Cl2,
0°C; HCl, rt
O
OMe
OMe
MeO
MeO
MeO
OMe
3) Ph3P=CH2, THF, rt
4) Grubbs II catalyst, CH2Cl2, rt
HO
MeO OMe
O
MeO
O
Grubbs II
pTsOH,
H2O,
MeO
O
catalyst,15
8
22
23 (86%
9
over the 4 steps)
Rx
Rx
Rx
CH2Cl2,
rt or up
MeCN,
75°C
to 100°C
OMe
OMe
Br2 (2.2 eq),
pyridine,
MeO
O
MeO
O
8
9
10
8
8
Br
Br
Et2O,
OMe
HO
RO
HO
MeO
O
MeO
MeO
+
9
9
0°C;
immediate
workup
O
O
O
Br
Br
8
Me
8
9
9
MeO
MeO
MeO
MeO
Me
40 : 60
−
60 : 40
trans-24
cis-24
(56% by crystal-
lization)
MeO
OMe
15
11
12: R = Me
13: R = H
14
(89%)
(slow isomerization on silica gel column)
Scheme 1. Existing ring-closing metathesis route to 3,4-benzotropolones 10.14
C-Functionalizations realized by this strategy (11,14 1414) or by analogous enyne
metatheses (13,14 1516).
OMe
OMe
OMe
MeO
O
MeO
O
MeO
O
Br
Br
Br
MeO
MeO
Most benzotropolones 10 prepared as sketched in the preceding
paragraph14 were unsubstituted at C-8 and C-9 (Scheme 1). How-
ever, subjecting the α-methylstyrene analog of diene 8, Rx = H, to
the same transformations gave the 9-methylated benzotropolone
11.14 Likewise, diene 8, Rx = 2 × OMe, with a methallyl rather than
allyl group furnished the 8-methylated benzotropolone 14.14 Replac-
ing, alternatively, one vinyl group of type-8 dienes by an ethynyl
group allowed ring-closures by enyne metatheses. This led to the 9-
vinylated benzotropolone 1214 (yet not to the O-unprotected parent
compound 13) as well as to the 8-vinylated ketal 1516 (but no closer
to an 8-vinylated benzotropolone). Here we disclose − exemplarily
− how manipulating the unsubstituted C8=C9 bond of the previously
described14 ring-closure products 18 (Scheme 2) and 23 (Scheme 3)
leads to a manifold of 9- (Scheme 5) and 8- functionalized 3,4-
benzotropolones (Scheme 6).
MeO
MeO
25
(+ Br g cis-24
and/or g trans-24)
26
(+ Br g trans-24)
27
(+ Br g trans-19)
Scheme 3. Synthesis of benzocycloheptadienone 2314 and its reaction with
elemental bromine. ORTEP plot of the crystal structure of dibromide cis-24 at
115 K.19− For the role of the precursors 25/26 of dibromides cis- and trans-24
and 27 of the precursor 27 of dibromide trans-19 see footnote20
.
The mentioned benzocycloheptadienone 23 was prepared from
the substituted benzaldehyde acetal 22 in 4 steps14 (Scheme 3).
Adding bromine to the C8=C9 bond of 23 furnished mixtures (up to
89% yield) of similar amounts of the dibromide diastereomers cis-
24 (3JH−CBr−CBr−H = 1.9 Hz) and trans-24 (3JH−CBr−CBr−H = 4.8 Hz).
Working up such mixtures by crystallization from cyclohex-
ane/AcOEt (5:1) allowed to isolate pure cis-24 [56% yield from 23,
m.p. 158-159°C (dec.)]. Its 3D structure was established by single
crystal X-ray analysis19 (Scheme 3).20,21 The isomeric dibromide
trans-24 was enriched by subjecting mixtures of cis- and trans-24 to
flash-chromatography on silica gel.22 However, we could not sur-
pass trans-contents of 85:15. This was because on silica trans-24
tended to isomerize giving cis-24 anew.
OMe
MeO
O
Br
MeO OMe
O
1) BuLi, THF, −78°C;
(17), rt
2) Dess-Martin reagent, CH2Cl2, rt
3) Grubbs II catalyst, CH2Cl2, rt
16
18 (79%
over 3 steps)
Br2,
pyridine,
Et2O,
rt
MeO
OMe
MeO
MeO
7
7
Ph, CuI,
Pd(PPh3)2Cl2,
DBU,
THF, rt
8
O
O
O
8
Br
8
NEt3, 50°C
3.5 h
9
9
9
OMe
OMe
OMe
Br
Br
DBU
(1.1 eq),
THF,
DABCO
(1.1 eq),
THF,
MeO
O
MeO
O
MeO
O
7
7
8
Ph
8
Br
Br
Br
21 (90%)
20 (69%)
trans-19 (95%;
94:6 mixture
with cis-isomer)
MeO
MeO
MeO
9
rt,
60 min
45°C,
24 h
9
Br
MeO
iso-28 (96%)
MeO
MeO
Scheme 2. Synthesis of benzocycloheptadienone 18,14 functionalization with
cis-24
28 (72%)
Br2, and a proof-of-principle C,C-coupling.
HO
O
HO
7
7
8
O
8
Br
pTsOH, H2O,
pTsOH, H2O,
MeO
MeO
MeO
MeO
The benzocycloheptadienone 18 was made from ortho-bromo-
styrene (16) in 3 steps14 (Scheme 2). It picked up elemental bromine
in 95% yield with a 94:6 trans:cis-selectivity17. The product mixture
and 5 equiv. of DBU underwent β-eliminations both of HBr and
MeOH. The benzotropolone methyl ether 20 resulted in 69% yield,
the bromine substituent residing at C-9.18 It Sonogashira-coupled
phenylacetylene in triethylamine. This gave 90% of the correspond-
ing benzotropolone ether 21.
MeCN, 75°C,
MeCN, reflux,
9
9
Br
5 h
24 h
29 (60%)
iso-29 (39%)
Scheme 4. Regioselective dehydrobromination of dibromide cis-24 furnishing
the bromoolefin isomers 28 and iso-28. Hydrolyses to the corresponding bromo-
benzotropolones.
We found that the dibromide cis-24 could be dehydrobrominated
in the two conceivable directions with perfect regiocontrol (Scheme
4). Treatment of a THF solution of cis-24 with DBU at ambient
3
This article is protected by copyright. All rights reserved.

10.1002/chem.201700622
Chemistry - A European Journal
OMe
HO
O
MeO
O
Et2Zn/ZnCl2 (1:1),
Pd(PPh3)2Cl2,
pTsOH,
9
MeO
aq. MeCN,
MeO
MeO
Et
Et
THF,
rt, 5 h
reflux, 2 h;
workup
temperature for 2 h furnished a single bromoolefin 28 in 72% yield.
It contained the motive Cquat−7CH2−8CH=9CBr−Cquat in the cy-
cloheptadienone ring as shown 1H-NMR spectroscopically. 23 In
contrast, treatment of a THF solution of dibromide cis-24 with
DABCO at 45°C for 24 h furnished the bromoolefin iso-28 in 96%
MeO
30 (85%)
with aq.
NaHCO3
31 (72%)
OMe
HO
(HO)2BPh,
PdCl2(dppf),
Na2CO3,
MeO
same as
above
O
O
yield. Its 1H-NMR spectrum revealed the presence of
quat−7CH2−8CBr=9CH−Cquat unit in the cycloheptadienone ring.24
a
9
MeO
MeO
MeO
MeO
Ph
Ph
aq. DMF,
90°C, 90 min
C
32 (93%)
33 (69%)
Unpurified mixtures (~1:1) of dibromides cis- and trans-24 de-
livered the bromoolefins 28 or iso-28 as single regioisomers, too,
when dehydrobrominated with DBU or DABCO, respectively. The
respective yields were 65% and 81% and hence 7% and 15% lower
than when to the dehydrobrominating the pure dibromide cis-24.
Because of the higher overall yields, employing 24 as an isomeric
mixture was preferable.
OMe
MeO
O
SiMe3
,
same as
above
Pd(PPh3)2Cl2,
CuI,
MeO
no benzo-
tropolone
NEt3 (neat),
50°C, 5 h
MeO
SiMe3
34 (94%)
OMe
MeO
O
Acid-catalyzed hydrolyses of the α-ketoketal moieties of bromo-
olefins 28 or iso-28 in hot aqueous acetonitrile gave the 9-bromo-
benzotropolone 29 in 60% yield and the 8-bromobenzotropolone
iso-29 in 39% yield (Scheme 4). We did not attempt to cross-couple
these compounds although the Sonogashira coupling of Scheme 2
suggests some feasibility. However, the free OH groups of the ben-
zotropolones 29 and iso-29 might interfere with the intended
Negishi or Suzuki couplings due to their acidity. We circumvented
such problems by cross-coupling the preceding bromoolefins 28 (de-
tails: Scheme 5) and iso-28 (details: Scheme 6). Most coupling
products delivered the corresponding benzotropolone when hydro-
lyzed under the previously employed conditions.14
9
28
MeO
MeO
Br
OMe
CO2iPr ,
Pd(OAc)2,
MeO
O
P(oTol)3,
Bu4NBr,
same as
above,
MeO
MeO
no benzo-
tropolone
CO2iPr
NaHCO3
DMF, aq. MeCN,
70°C, 26 h
but workup
with aq.
NaHCO3
35 (48%)
OMe
HO
CO (1 bar),
MeOH,
Pd(PPh3)4,
iPr2NEt,
MeO
O
O
same as
above,
9
The bromoolefin 28 allowed to engage its C9−Br bond in repre-
sentative couplings of the Negishi (Scheme 5; → 85% 30), Suzuki
(→ 93% 32), Sonogashira (→ 94% 34), and Heck types (→ 48%
35). The C-nucleophiles incorporated during these reactions were
ethylzinc chloride, phenylboronic acid, (trimethylsilyl)ethyne, and
isopropyl acrylate, respectively. In another coupling, pioneered by
Tsuji, cat. Pd(PPh3)4 allowed to alkoxycarbonylate the bromoolefin
28 adding methanol or ethanol as a co-solvent and an atmosphere of
CO. The latter was supplied at atmospheric pressure.25 This afforded
the methyl ester 36 in 87% yield and the ethyl ester 38 in 79% yield.
The Negishi coupling, Suzuki, Sonogashira coupling, and
alkoxycarbonylation products hydrolyzed under our standard condi-
tions (pTsOH, aq. MeCN, prolonged heating). This provided the re-
spective 9-substituted benzotropolones 31, 33, 37, and 39 in 31-72%
yield. The enyne moiety formed in the Sonogashira coupling and the
dienoic ester moiety formed in the Heck coupling did not stand up to
these hydrolysis conditions. By consequence, the respective com-
pounds (34, 35) degraded without rendering benzotropolones.
MeO
MeO
MeO
CO2Me
CO2Me
THF/CH2Cl2,
70°C, 24 h
but 6 h
MeO
36 (87%)
37 (55%)
OMe
HO
CO (1 bar),
EtOH,
Pd(PPh3)4,
iPr2NEt,
MeO
O
O
same as
above,
9
MeO
MeO
MeO
MeO
CO2Et
CO2Et
but 20 h
THF/CH2Cl2,
70°C, 29 h
38 (79%)
39 (31%)
Scheme 5. Exemplary C,C-couplings with the 9-brominated precursor 28 of 9-
C-functionalized benzotropolones..
The bromoolefin iso-28 underwent identical cross-couplings at C-
8 (Scheme 6) as its isomer 28 underwent at C-9 (Scheme 5). I. e.,
iso-28 Negishi-coupled with ethylzinc chloride at room temp. (→
86% 40), Suzuki-coupled with phenylboronic acid at 90°C (→ 82%
42), Sonogashira-coupled with (trimethylsilyl)ethyne at 50°C (→
67% 44) and Heck-coupled with excess isopropyl acrylate at 70°C
(→ 83% 46). Moreover, the bromoolefin iso-28 was methoxy-
carbonylated by CO at 1 bar and methanol. This furnished the
methyl ester 47 in 75% yield. Analogously, iso-28 was ethoxy-
carbonylated by CO and ethanol. This rendered the ethyl ester 49 in
84% yield. 5 of the 6 coupling products of Scheme could be
hydrolyzed in yields ranging from 49 to 90%. The Sonogashira
coupling product 44 was amenable to such a hydrolysis, too.
However, its enyne motive was hydrated concomitantly.
Accordingly, the resulting benzotropolone 45 displays an acetyl
group at C-8 instead of 2-(trimethylsilyl)ethyn-1-yl. The Heck-
coupling product 46 did not tolerate hydrolysis conditions and
therefore rendered no benzotropolone.
4
This article is protected by copyright. All rights reserved.

10.1002/chem.201700622
Chemistry - A European Journal
OMe
OMe
OMe
MeO
MeO
O
MeO
O
Br2 (5 eq),
pyridine,
CH2Cl2,
rt, 24 h
Br2 (5.0 eq),
pyridine,
CH2Cl2,
rt, 20 h;
8
O
8
Br
Br
Br
9
MeO
MeO
MeO
9
Br
OMe
MeO
O
HO
1
1
1
separa-
MeO
MeO
MeO
Br
O
O
O
tion22 from
Et2Zn/ZnCl2 (1:1),
Pd(PPh3)2Cl2,
pTsOH,
aq. MeCN,
Et
Et
8
cis-24 (19%)
MeO
MeO
MeO
MeO
cis-24
23
cis-53
(72% from 23,
65% from cis-24;
THF,
rt, 2 h
reflux, 2 h;
workup
with aq. KOH
Ca2
Ph
DABCO (2.0 eq),
O
trans-53 not observed)
41 (59%)
40 (86%)
THF,
50°C, 18 h
O
HO
OMe
HO
O
MeO
O
O
(HO)2BPh,
PdCl2(dppf),
Na2CO3,
8
O
OMe
same as
above,
Ph
Ph
8
MeO
HO
7
O
MeO
MeO
MeO
O
8
1
Br
Ph
aq. DMF,
90°C, 90 min
but 1.5 h
HO
MeO
MeO
9
MeO
O
43 (90%)
42 (82%)
aurantricholone (712
)
Br
54 (62%)
OMe
HO
OMe
HO
MeO
MeO
O
O
SiMe3
,
O
SiMe3
same as
above,
O
(HO)2BPh,
PdCl2(dppf),
Na2CO3,
8
8
Pd(PPh3)2Cl2,
CuI,
pTsOH,
MeO
MeO
MeO
MeO
MeO
MeO
MeO
MeO
aq MeCN,
1
but 2.5 h
NEt3 (neat),
50°C, 3 h
reflux,
4 h
aq. DMF,
90°C, 1 h
45 (63%)
44 (98%)
55 (82%)
56 (68%)
OMe
MeO
O
Br
8
Scheme 7. Model study for a total synthesis of the benzotropolone natural
iso-28
MeO
MeO
product aurantricholone (712) by the bromination/cross-coupling strategy of this
work: synthesis of dibromide 54 via tribromide cis-53, one-pot bis(Suzuki
coupling), and hydrolysis/tautomerism to the benzotropolone 56.
CO2iPr ,
OMe
MeO
O
CO2iPr
Pd(OAc)2,
P(oTol)3,
Bu4NBr,
same as
above,
MeO
In an extension of our strategy we treated the benzocycloheptadi-
enone 23 with 5 equiv. of bromine in CH2Cl2 rather than with 2.2
equiv. of bromine in diethyl ether as before (Scheme 3). Now the di-
bromide cis-24 resulted to some extent (19%) but the tribromide 53
predominantly (72%; Scheme 7). X-ray crystallography revealed
that 53 was cis-configured.27 This compound originates from sub-
strate 23 by a cis-addition/para-substitution sequence. Indeed, bro-
mine and the pure dibromide cis-24 reacted to give the same tribro-
mide cis-53 in 65% yield. DABCO dehydrobrominated this tribrom-
ide with the same regioselectivity, with which it dehydrobrominated
the dibromide cis-24. In the case at hand (Scheme 7), the bromoar-
ene/bromoolefin 54 formed as the sole product (62% yield).28
no benzo-
tropolone
NaHCO3
but workup
with aq.
NaHCO3
MeO
DMF, aq. MeCN,
70°C, 24 h
46 (83%)
OMe
HO
O
CO (1 bar),
MeOH,
Pd(PPh3)4,
iPr2NEt,
MeO
O
CO2Me
CO2Me
8
same as
above,
MeO
MeO
MeO
THF/CH2Cl2,
70°C, 26 h
but 24 h
MeO
48 (53%)
47 (75%)
OMe
HO
CO (1 bar),
EtOH,
Pd(PPh3)4,
iPr2NEt,
MeO
O
O
CO2Et
CO2Et
8
same as
above,
MeO
MeO
MeO
At 90°C the bromoarene/bromoolefin 54 and phenylboronic acid
Suzuki-coupled under PdCl2(dppf)-catalysis within one hour both at
the C1−Br and the C8−Br bond (Scheme 7). 29 The bis(coupling
product) 55 resulted in 82% yield. Its α-ketoketal moiety could be
hydrolyzed as usually (pTsOH, aq. MeCN). This provided 68% of
the diphenylated benzotropolone 56. The bicyclic cores of the latter
compound, of crocipodin (6;6 Scheme 1), and of aurantricholone
(712) exhibit identical O- and C-functionalization sites. With regard
to aurantricholone (7) this hints at how a total synthesis might be
approached − provided one knew a coupling partner for introducing
unprotected 4-methylidene-2-phenyltetronic acid groups. Efforts to-
wards developing such a reagent and the respective cross-coupling
conditions are underway in our laboratory.
but 48 h
THF/CH2Cl2,
70°C, 24 h
MeO
50 (49%)
49 (84%)
HO
HO
O
pTsOH,
aq. MeCN,
reflux, 24 h;
BBr3,
CH2Cl2,
−78°C,
O
CO2H
CO2Et
8
8
MeO
MeO
HO
HO
aq. KOH,
r.t., 1 h
45 min;
0°C, 12 h;
EtOH
51 (95%)
52 (98%;
goupiolone A)
Scheme 6. Illustrative C,C-couplings with the 8-brominated precursor iso-28 of
8-C-functionalized benzotropolones.
The benzotropolone syntheses presented here should be readily
extendable. The benzene moiety in our key intermediate 23 need not
be methoxylated as proved by the bromination 18→20 in the entire
absence of methoxy groups (Scheme 2). In fact this benzene moiety
might tolerate a variety of other substituents. Moreover, compounds
like the bromoarene/bromoolefin 5429 or analogous iodoarene/bro-
moolefins look like precursors of bis(carbofunctionalized) benzo-
tropolones.
Hydrolyzing the CO2Et-containing product 49 in acidic aqueous
acetonitrile at reflux temperature for 48 h led to the CO2Et-
containing benzotropolone 50 in 49% yield (Scheme 6). The ester
moiety survived the ensuing cleavage of the methoxy groups with
BBr3, too. It delivered 98% of the CO2Et-containing catechol 52.
This compound equals synthetic goupiolone A. In the current study,
it emerges from a 9-step synthesis. It establishes the substituted
tropolone moiety in a straightforward manner. The only previous
synthesis of goupiolone A (52) required 19 steps.26
Experimental Section
5
This article is protected by copyright. All rights reserved.

10.1002/chem.201700622
Chemistry - A European Journal
Keywords: bromination • C,C coupling • dehydrobromination • natural
products • nonbenzenoid aromatics • ring-closing olefin metathesis •
Full experimental details are given in the Supporting Information.
Received: ((will be filled in by the editorial staff))
Published online on ((will be filled in by the editorial staff))
[22] W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 2923-2925.
[23] The respective 1H-NMR resonances (CDCl3, 400 MHz) were: δ7-H
= 2.59 (d, 3J7,8 = 7.7 Hz), δ8-H = 6.26 ppm (t, 3J8,7 = 7.7 Hz).
[24] Pertinent 1H-NMR resonances (CDCl3, 400 MHz) were: δ7-H = 3.15
(d, 4J7,9 = 1.4 Hz), δ9-H = 6.26 ppm (t, 4J9,7 = 1.4 Hz).
[25] Pd-catalyzed ester formations under related conditions are known,
e.g., in the presence of the diphosphanes Xantphos (J. R.
Mantinelli, D. A. Watson, D. M. M. Freckmann, T. E. Barder, S. L.
Buchwald, J. Org,. Chem. 2008, 73, 7102-7107) or BINAP (W.
Yang, W. Han, W. Zhang, L. Shan, J. Sun, Synlett 2011, 2253-
2255).
[1] M. J. S. Dewar, Nature 1945, 155, 50-51.
[2] R. L. Redington, J. Chem. Phys. 2000, 113, 2319-2335.
[3] K. Tanaka, R. Nagahiro, S. Ohba, M. Eishima, Tetrahedron Lett.
2001, 42, 925-929.
[4] M. Randić, Chem. Rev. 2003, 103, 3449-3605.
[5] S. Sang, S. Tian, X. Meng, R. E. Stark, R. T. Rosen, C. S. Yang,
C.-T. Ho, Tetrahedron Lett. 2002, 43, 7129-7134.
[6] a) First chemical synthesis: M. A. Girard, C. R. Hebd. Séances
Acad. Sci. 1869, 69, 865-868; b) first enzymatic synthesis: M. G.
Bertrand, C. R. Hebd. Séances Acad. Sci. 1895, 120, 267-269.
[7] E. g. a) Y. Kawabe, Y. Aihara, Y. Hirose, A. Sakurada, A. Yoshida,
M. Inai, T. Asakawa, Y. Hamashima, T. Kan, Synlett 2013, 479-
482; b) K. Cheng, X. Wang, S. Zhang, H. Yin, Angew. Chem. Int.
Ed. 2012, 51, 12246-12249; c) L. Kerschensteiner, F. Löbermann,
W. Steglich, D. Trauner, Tetrahedron 2011, 67, 1536-1539; d) C.
Huang, H. Bai, C. Li, G. Shi, Chem. Commun. 2011, 47, 4962-4964.
[8] Analogues of the surmised 1,9-dihydroxy-4a,5-dihydro-9H-5,9-
methanobenzocycloheptene-2,8,10-trione and 3,4,5-trihydroxy-5,9-
dihydro-5,9-methanobenzocycloheptene-6,10-dione intermediates
of such oxidations were isolated by a) W. Dürckheimer, E. F. Pau-
lus, Angew. Chem. Int. Ed. Engl. 1985, 24, 224-225 and b) E.
Yanase, K. Sawaki, S.-i. Nakatsuka, Synlett 2005, 17, 2661-2663,
respectively.
[26] N. Fukui, K. Ohmori, K. Suzuki, Helv. Chim. Acta 2012, 95, 2194-
2202.
[27] The crystallographic data of tribromide cis-53 are contained in
CCDC 1505661. They can be obtained free of charge from the
Cambridge Crystallographic Data Centre via the link
www.ccdc.cam.ac.uk/data_request/cif.
[28] Pertinent 1H-NMR resonances (CDCl3, 400 MHz): δ7-H = 3.16 (d,
4J7,9 = 0.5 Hz), δ9-H = 7.08 ppm (t, 4J9,7 = 0.5 Hz).
[29] Interestingly, the bromoarene/bromoolefin 54 revealed a potential
for being amenable to differential sequential biscouplings: It
picked up a C1-substituted propargyl alcohol selectively at C8−Br
(76% yield).
[9] E. A. Kurt-Jones, M. Chan, S. Zhou, J. Wang, G. Reed, R. Bronson,
M. M. Arnold, D. M. Knipe, R. W. Finberg, Proc. Natl. Acad. Sci.
USA 2004, 101, 1315-1320.
[10] S. Zhou, A. M. Cerny, G. Bowen, M. Chan, D. M. Knipe, E. A.
Kurt-Jones, R. W. Finberg, Antiviral Res. 2010, 87, 295-306.
[11] a) Isolation: E. A. H. Roberts, M. J. Myers, Sci. Food Agric. 1959,
10, 176-179; b) structure: A. L. Davis, Y. Cai, A. P. Davies, Magn.
Res. Chem. 1995, 33, 549-552.
[12] D. Klostermeyer, L. Knops, T. Sindlinger, K. Polborn, W. Steglich,
Eur. J. Org. Chem. 2000, 4, 603-609.
[13] D. Arican, Diplomarbeit, Universität Freiburg, 2008.
[14] D. Arican, R. Brückner, Org. Lett. 2013, 15, 2582-2585.
[15] [1,3-Bis-(2,4,6-trimethylphenyl)-2-imidazolidinylidene]dichloro-
(phenylmethylene)(tricyclohexylphosphane)ruthenium.
[16] D. Arican, Dissertation, Universität Freiburg, 2013.
[17] The configurational assignments of dibromides trans-19 vs. cis-19
were based on 1H-NMR comparisons with dibromides trans-24
and cis-24; the configuration of dibromide cis-24 was clear from
the X-ray analysis (Scheme 3). For example, the coupling constant
3J8,9 differed characteristically in trans-19 (4.6 Hz) and trans-24
(4.5 Hz) from the 1.9 Hz measured for both cis-19 and cis-24.
[18] 1H NMR20, tropolone moiety (CDCl3, 400 MHz): δOMe = 3.85 (d, 5JOMe,7
=
0.4 Hz), δ7-H = 6.20 (dq, 3J7,8 = 9.8 Hz, 5J7,OMe = 0.4 Hz), δ8-H = 7.39
ppm (d, 3J8,7 = 9.8 Hz).
[19] The crystallographic data of dibromide cis-24 are contained in
CCDC 1505660. They can be obtained free of charge from the
Cambridge Crystallographic Data Centre via the link
www.ccdc.cam.ac.uk/data_request/cif.
[20] The intermediacy of bromonium-ion 27 (Scheme 3) in the bromi-
nation of benzocycloheptadienone 18 explains why dibromide
trans-19 results with a 94:6 preference (Scheme 2). The cis-
addition of bromine to benzocycloheptadienone 23 implies the
intermediacy of carbenium-ion 25 (Scheme 3). It remains unknown
whether the latter forms directly from the reactants or indirectly
from a trans→cis isomerization of initially formed dibromide
trans-24. Carbenium-ion 25 looks stabilized by the +M effect of
one methoxy group.
[21] Benzocycloheptadienone 23 shares a styrene motif with certain
olefins, which add chlorine cis-selectively but not bromine: a) S. J.
Cristol, F. R. Stermitz, P. S. Ramey, J. Am. Chem. Soc. 1956, 78,
4939-4941; b) G. E. Heasley, T. R. Bower, K. W. Dougharty, J. C.
Easdon, V. L. Heasley, S. Arnold, T. L. Carter, D. B. Yaeger, B. T.
Gipe, D. F. Shellhamer, J. Org. Chem. 1980, 45, 5150-5155.
6
This article is protected by copyright. All rights reserved.