An intra/intermolecular suzuki sequence to benzopyridyloxepines containing geometrically pure exocyclic tetrasubstituted alkenes

Article abstract of DOI:10.1021/ol800834q

(Chemical Equation Presented) A route to enable the preparation of 5-benzylidenyl-benzopyridyloxepine analogues was developed to continue our research in the field of nuclear hormone receptor modulators. The key steps are1) a syn-stereoselective diboratio

Full text of DOI:10.1021/ol800834q

ORGANIC
LETTERS
2008
Vol. 10, No. 13
2701-2704
An Intra/Intermolecular Suzuki Sequence
to Benzopyridyloxepines Containing
Geometrically Pure Exocyclic
Tetrasubstituted Alkenes
Matthew W. Carson,* Matthew W. Giese, and Michael J. Coghlan
DiscoVery Chemistry Research and Technology, Lilly Research Laboratories,
Eli Lilly and Company, Indianapolis, Indiana 46285
carsonma@lilly.com
Received April 11, 2008
ABSTRACT
A route to enable the preparation of 5-benzylidenyl-benzopyridyloxepine analogues was developed to continue our research in the field of
nuclear hormone receptor modulators. The key steps are1) a syn-stereoselective diboration of a tethered aryl alkyne; 2) an intramolecular
Suzuki cross-coupling reaction, which forms in a stereo- and regiocontrolled fashion, the 5-exoalkylidenyl 7-membered ring imbedded within
the core of the scaffold and; 3) an intermolecular Suzuki to furnish the final tetra-substituted olefinic benzopyridyloxepines.
The highly substituted olefin motif is found in a number of
biologically active series including the triphenylethylenes,
most notably represented by the SERM tamoxifen,1 and
Lilly’s dibenzoxepine nuclear hormone receptor (NHR)
modulators.2 Recently there has been impressive progress
on the regio- and stereoselective syntheses of tetra-substituted
alkenes.3 At Lilly, we developed a one-pot procedure to the
dibenzoxepines, which contain a tri- or tetrasubstituted
exobenzylidene moiety, utilizing an intramolecular carbo-
metalation of an alkyne followed by cross-coupling with a
boronic acid (Heck-Suzuki).4,5 In this paper, we report on a
complimentary route involving a stereospecific dimetalation
of an alkyne followed by regioselective sequential intra/
intermolecular cross couplings, which allowed us to synthe-
size a series of benzopyridyloxepines containing stereode-
fined tetra-substituted exocyclic alkenes.
(4) (a) Yu, H.; Richey, R. N.; Carson, M. W.; Coghlan, M. J. Org. Lett.
2006, 8, 1685. (b) Yu, H.; Richey, R. N.; Mendiola, J.; Adeva, M.; Somoza,
C.; May, S. A.; Carson, M. W.; Coghlan, M. J. Tetrahedron. Lett 2008, 49,
1915.
(5) (a) For previous examples of intramolecular addition of arylpalladium
intermediates to tethered alkynes followed by cross coupling to give
exocyclic tetrasubstituted alkenes, see Burns, B.; Grigg, R.; Sridharan, V.;
Stevenson, P.; Sukirthalingam, S.; Worakum, T Tetrahedron Lett. 1989,
30, 1135. (b) Zhang, Y.; Negishi, E-I J. Am. Chem. Soc. 1989, 111, 3454.
(c) Negishi, E-I; Noda, Y.; Lamaty, F.; Vawter, E. J. Tetrahedron Lett.
1990, 31, 4393. (d) Wang, R-T; Chou, F-L; Luo, F-T J. Org. Chem. 1990,
55, 4846. (e) Cheung, W. S.; Patch, R.; Player, M. R. J. Org. Chem. 2005,
70, 3741. (f) Bour, C.; Blond, G.; Salem, B.; Suffert, J. Tetrahedron 2006,
62, 10567, and references therein.
(1) (a) Jordan, V. C. J. Med. Chem. 2003, 46, 884. (b) Jordan, V. C.
J. Med. Chem. 2003, 46, 1082.
(2) (a) Coghlan, M. J.; Green, J. E.; Grese, T. A.; Jadhav, P. K.;
Matthews, D. P.; Steinberg, M. I.; Fales, K. R.; Bell, M. G. WO 2004/
052847, A2, 2004. (b) Carson, M. W.; Coghlan, M. J. WO 2008/008882,
A2, 2008.
(3) Zhou, C.; Larock, R. C. J. Org. Chem. 2006, 71, 3184, and references
therein.
10.1021/ol800834q CCC: $40.75
Published on Web 06/07/2008
2008 American Chemical Society

To continue the study of the dibenzoxepine structure-
activity relationship (SAR) as NHR modulators, it was our
initial desire to apply a route utilizing the key Heck-Suzuki
step for the preparation of pyridine analogues. However, to
our disappointment, under the ligandless conditions optimal
for the phenyl analogues, the palladium-catalyzed reaction
of alkyne-tethered aryl iodide 1 in the presence of 3-nitro-
phenyl boronic acid afforded a low yield of desired pyridyl-
oxepine 2a(10%) along with direct coupling product 3
(Scheme 1). Based upon mass balance after column chro-
step without further purification.8 It should be noted that the
use of bromopyridine is required as an attempted diboration
of the aryl iodide 1 resulted in recovery of starting material
due to the likely C-I insertion of the platinum. With the bis
boronic ester 4a in hand, we investigated conditions for the
intramolecular Suzuki cross coupling to give vinyl boronic
ester 5a (Scheme 2). Optimization of this step revealed that
dilute (0.01 M 4a in anhydrous dioxane) conditions, using
PdCl2(dppf)•CH2Cl2 and K3PO4, delivered the desired vinyl
boronic ester 5a. To complete the synthesis of 2a, we were
faced with the challenge of coupling a bulky boronic ester
with an aryl iodide. Prior art reveals that cross coupling of
aryl halides with sterically hindered boronic esters under
standard Suzuki conditions (e.g., Pd(PPh3)4, aqueous weak
Scheme 1
Scheme 3
matography, apparent oligomerization was the main outcome
of the reaction.6 To circumvent the issue of a low yielding
cyclization step for the preparation of NHR modulator
targets, we then envisioned a route highlighted by an
intramolecular Suzuki coupling to synthesize the seven-
membered ring containing the geometrically pure exocyclic
alkene. A similar approach involving an intramolecular Stille
coupling to trisubstituted exocyclic alkenes of benzopyridyl-
oxepines was reported by Finch and co-workers.7 Scheme 2
base, polar aprotic solvents, 80 °C) results in slow reaction
times and low yields.9 However, both Armstrong and Begtrup
have shown that one could cross couple in a crowded
environment similar to 5a to afford tamoxifen derivatives
by utilizing the electron rich 3,5-dimethoxyphenol (3,5-DMP)
as an additive. Thus, under these conditions (Pd(PPh3)4,
aqueous KOH, 3,5-DMP, dioxane, 80 °C), the reaction of
vinyl boronic ester 5a (Scheme 2) with 3-iodonitrobenzene
gave 5-benzylidenyl- benzopyridyloxepine 2a as a geo-
metrically pure isomer in 45% isolated yield from tethered
alkynyl aryl pyridyl bromide 1a.11
Scheme 2
Upon optimization of the intra/intermolecular Suzuki route
for the preparation of pyridyloxepine 2a, we wanted to
expand the structural diversity of this series. In the Heck-
Suzuki route, 1 was prepared by conducting a Mitsunobu
reaction between iodopyridyl methanol 6 and alkynyl phenol
7, an intermediate that required three steps to prepare from
Scheme 4
illustrates the new sequence commencing with a platinum-
catalyzed syn diboration of tethered alkynyl aryl pyridyl
bromide 1awith bis(pinacolato)diboron to cleanly give bis
vinylpinacol boronic ester 4a, which was used in the next
(6) This result is reminiscent of our previous findings in which we
conducted exploratory Heck-Suzuki reactions of phenyl analogues in the
presence of a ligand. The fact that the pyridine resulted in much lower
yields than the phenyl analogue under the ligandless conditions suggests
that another molecule of 1or resultant intermediates function as a ligand.
In other words, the complexation of the pyridyl nitrogen to organopalladium
intermediates disfavors the desired intramolecular Heck-Suzuki compared
to side reactions such as 1) direct coupling to give 3 and; 2) intermolecular
Heck reaction leading to high molecular weight products.
2702
Org. Lett., Vol. 10, No. 13, 2008

Table 1. Synthesis of Benzopyridyloxepines 2a-h from
Tethered Alkynyl Aryloxy Pyridyl Bromides 1a-ha
Table 2. Synthesis of Benzopyridyloxepine Derivatives 2i-m
from 1aa
a Diboration: 1.1 equiv of bis(pinacolato)diboron and 8 mol % of
Pt(PPh3)4 in DMF (0.1 M) at 80 °C for 24 h. Intramolecular Suzuki: 10
mol % of PdCl2(dppf)·CH2Cl2 and 3 equiv of K3PO4 in dioxane (0.01 M)
at 80 °C for 24 h. Intermolecular Suzuki: 1.5 equiv of aryl iodide, 5 mol %
of Pd(PPh3)4, 6 M aqueous KOH (0.4 M), and 5 equiv of 3,5-dimethox-
yphenol in dioxane (0.15 M) at 80 °C for 18 h. b Isolated yields (1 to 2,
three steps) after column chromatography. c (E)-2l and (Z)-2l were isolated
as single isomers via column chromatography in 21% and 16% yield,
respectively.
commercially available iodophenol (Scheme 3). However,
the intra/intermolecular Suzuki route allowed us to target
bromopyidines as intermediates which shortened the syn-
theses and allowed for later structural divergence. As shown
in Scheme 4, Mitsunobu reaction between bromopyridyl
methanols 6a-c and iodophenols 8a-c afforded bromoben-
zyl iodoaryl ethers 9a-e which undergo regioselective
Sonogashira reactions at the C-I locus to give alkynes
1a-h.
a Diboration: 1.1 equiv of bis(pinacolato)diboron and 8 mol % of
Pt(PPh3)4 in DMF (0.1 M) at 80 °C for 24 h. Intramolecular Suzuki: 10
mol % PdCl2(dppf)·CH2Cl2 and 3 equiv of K3PO4 in dioxane (0.01 M) at
80 °C for 24 h. Intermolecular Suzuki: 1.5 equiv of 3-iodonitrobenzene, 5
mol % of Pd(PPh3)4, 6 M aqueous KOH (0.4 M), and 5 equiv of
3,5-dimethoxyphenol in dioxane (0.15 M) at 80 °C for 18 h. b Isolated yields
(1 to 2, three steps) after column chromatography.
It should be noted that bromopyridines with reactive C-Br
bonds (see 6d) were outside the scope of the selective
Sonogashira reaction, therefore Mitsunobu between 6d and
Org. Lett., Vol. 10, No. 13, 2008
2703

7 using trin-butylphine and 1,1′-(azodicarbonyl)-dipiperidine
(ADDP)afforded the desired bromopyridine 1d (Scheme 3).
alternative to the intramolecular carbometalation for the
stereoselective preparation of tetra-substituted exocyclic
alkenes.
With intermediates 1a-h in hand, we applied the intra/
intermolecular Suzuki technology to prepare compounds for
the SAR studies. The methodology worked in moderate
yields, calculated at the end of the three-step diboration, intra/
intermolecular coupling sequence, to afford the 1,2, and
3-pyridyl isomers 2a-c (Table 1, entries 1-3). To our
disappointment, the sequence failed in the case of the
4-pyridyl analogue 1d (Table 1, entry 4). Both the fluoro
and methoxy substituted examples (Table 1, entries 5-6)
resulted in similar yields suggesting a lack of electronic effect
in the alkynyl phenyl ring. We successfully expanded the
series to propylidenyl 2g and isobutylidenyl 2h analogues
as well (Table 1, entries 7-8) resulting in yields similar to
the ethylidenyl analogue 2a (Table 1, entry 1).
In summary, a diboration, inter/intramolecular Suzuki
cross-coupling sequence enabled the synthesis of pyridyl-
benzoxepines with a highly substituted stereodefined olefinic
moeity at the five-position. Highlights of the sequence
include (1) an intramolecular Suzuki cross coupling to give
a seven-membered ring and (2) an example of an intermo-
lecular Suzuki in a sterically congested environment. The
route allowed us to synthesize a diverse set of targets for
our NHR program, and it serves as complimentary route to
the Heck-Suzuki cascade. In future work, we will expand
the scope of this technology to biologically active exocyclic
alkenes outside the scope of dibenzoxepines.
Supporting Information Available: Detailed experimen-
tal procedures and NMR of products. This material is
available free of charge via the Internet at http://pubs.acs.org.
The intra/intermolecular Suzuki route allowed for access
to substituted aryl rings at the vinyl position with late
divergence in the synthesis. Analogues phenyl 2i, p-fluoro-
phenyl 2j, and p-aminophenyl 2k were prepared in moderate
yields (Table 2, entries 1-3). However, the conditions of
the coupling reaction in some cases gave undesired results.
For example, unlike the m-NO2 analogue 2a (Table 1, entry
1), the coupling of 5a with 4-iodonitrobenzene afforded both
the E- and Z- isomers of 2l (Table 2, entry 4). We attempted
to synthesize the p-cyanophenyl derivative, but could only
isolate the primary amide 2m in low yield (Table 2, entry
5). Despite the aforementioned limits, one could conclude
that the three-step diboration, inter/intramolecular Suzuki
cross coupling methodology has the potential to be a viable
OL800834Q
(7) Finch, H.; Pegg, N. A.; Evans, B. Tetrahedron Lett. 1993, 34, 8353.
(8) (a) Ishiyama, T.; Matsuda, N.; Miyaura, N.; Suzuki, A. J. Am. Chem.
Soc. 1993, 115, 11018. (b) Ishiyama, T.; Matsuda, N.; Murata, M.; Ozawa,
F.; Suzuki, A.; Miyaura, N. Organometallics 1996, 15, 713. (c) Brown,
S. D.; Armstrong, R. W. J. Am. Chem. Soc. 1996, 118, 6331. (d) Ramirez,
J.; Fernandez, E. Synthesis 2005, 1698.
(9) Watanabe, T.; Miyaura, N.; Suzuki, A. Synlett 1992, 207.
(10) (a) Brown, S. D.; Armstrong, R. W J. Org. Chem. 1997, 62, 7076.
(b) Wenckens, M.; Jakobsen, P.; VedsØ; Huusfeldt, P. O.; Gissel, B.;
Barfoed, M.; Brockdorff, B. L.; Lykkesfeldt, A. E.; Begtrup, M Bioorg.
Med. Chem. 2003, 11, 1883.
(11) Due to instability, boronic acid intermediates 4a and 5a as well as
analogous intermediates (Videinfra) were difficult to purify, therefore the
material was used in subsequent steps in crude form and the yield was
calculated over three steps.
2704
Org. Lett., Vol. 10, No. 13, 2008