Stereoselective synthesis of dibenzoxapine containing tetrasubstituted exocyclic alkenes via cascade methylboronic acid coupling reactions

Article abstract of DOI:10.1016/j.tetlet.2008.01.099

A stereoselective method for the preparation of dibenzoxapine containing tetrasubstituted exocyclic E-alkenes has been developed. The key reaction involves an intramolecular cyclocarbopalladation of alkynes to form a vinylpalladium species and subsequentl

Full text of DOI:10.1016/j.tetlet.2008.01.099

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 1915–1918
Stereoselective synthesis of dibenzoxapine containing
tetrasubstituted exocyclic alkenes via cascade
methylboronic acid coupling reactions
a,
a
b
b
b
*
Hannah Yu , Rachel N. Richey , Javier Mendiola , Marta Adeva , Carmen Somoza ,
Scott A. May a, Matthew W. Carson c, Michael J. Coghlan c
a Chemical Product Research and Development, Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN 46285, USA
b Discovery Chemistry Synthesis Group, Lilly Research Laboratories, Eli Lilly and Company, Alcobendas, Spain
c Discovery Chemistry Research, Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN 46285, USA
Received 13 December 2007; revised 22 January 2008; accepted 24 January 2008
Available online 30 January 2008
Abstract
A stereoselective method for the preparation of dibenzoxapine containing tetrasubstituted exocyclic E-alkenes has been developed.
The key reaction involves an intramolecular cyclocarbopalladation of alkynes to form a vinylpalladium species and subsequently
coupling the vinylpalladium with methylboronic acid. The approach provides a straightforward method for the synthesis of tetrasubsti-
tuted exocyclic E-alkenes.
Ó 2008 Elsevier Ltd. All rights reserved.
The dibenzoxapine framework is commonly found in
important pharmaceuticals. A novel series of dibenzoxa-
pine containing tetrasubstituted exocyclic alkene deriva-
tives have been identified at Eli Lilly and Company as
nuclear hormone receptor modulators.1
prepared from strategy B by carbometallation of alkynes
8a–d followed by terminating the vinylpalladium species
with methylboronic acid 7. Alkynes 8a–d could be prepared
from benzyl alcohols 9a–b and phenols 10a–c through either
a one-step Mitsunobu reaction5 or a two-step alkylation
procedure.
Recently, we published a stereoselective method for the
preparation of these tetrasubstituted alkenes by intramole-
cular cyclocarbopalladation of alkynes to form a vinylpalla-
dium species and subsequently coupling the vinylpalladium
with arylboronic acids (Scheme 1, strategy A).2 The strategy
was proved to be a highly efficient method for the synthesis
of tetrasubstituted exocyclic Z-alkenes. To synthesize
E-alkenes 1b–d using strategy A, as shown in Scheme 1,
quantities of substituted 2-propynylphenols 6a–c would be
needed. To gain access to alkynes 6a–c, the phenols had
to be protected to enable the Sonogashira coupling.3,4 To
avoid the protection and deprotection steps, we envisioned
that the tetrasubstituted exocyclic E-alkenes could also be
The palladium-catalyzed cross-coupling reaction of aryl
halides with organoboron compounds has emerged as a
powerful method for carbon–carbon bond formation.6,7
While many types of organoboron compounds8 are suit-
able donors, boronic acids are especially popular due to
their ease of synthesis. Aryl- and alkenylboronic acids typ-
ically afford good yields of cross-coupled products with a
wide range of electrophiles. In contrast, the coupling of
n-alkylboronic acids is much more challenging9 and thal-
lium or silver bases are often needed.10 Furthermore, to
the best of our knowledge, the use of n-alkylboronic acids
in the palladium-catalyzed cascade reactions has not been
reported. Herein, we reported a stereoselective method
for the synthesis of dibenzoxapine containing tetrasub-
stituted exocyclic alkenes through a palladium-catalyzed
*
Corresponding author.
E-mail address: yu_hannah@lilly.com (H. Yu).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.01.099

1916
H. Yu et al. / Tetrahedron Letters 49 (2008) 1915–1918
R2
OH
1'
O
B(OH)2
HO
Strategy A
+
R2
+
2'
1
2
R1
I
R1
7
NO2
4
O
O
I
6a, R2 = H
5a, R1 = H
5b, R1 = F
3
4a, R1 = H, R2 = H
R2
6b, R2 = 5-F
R1
R1
R2
6c, R2 = 5-OnBu
10
4b, R1 = H, R2 = 5'-F
1
4c, R1 = H, R2 = 5'-OnBu
4d, R1 = 5-F, R2 = 5'-F
R2
O
HN
R1
S
NO2
1'
O
OH
O
2a, R1 = H, R2 = H
2'
1a, R1 = H, R2 = H
HO
1
2b, R1 = H, R2 = 3-F
Strategy B
I
R2
1b, R1 = H, R2 = 3-F
+
2c, R1 = H, R2 = 3-OnBu
2d, R1 = 8-F, R2 = 3-F
CH3B(OH)2
+
R1
1c, R1 = H, R2 = 3-OnBu
1d, R1 = 8-F, R2 = 3-F
I
2
NO2
7
10a, R2 = H
10b, R2 = 5-F
NO2
8a, R1 = H, R2 = H
10c, R2 = 5-nBu
8b, R1 = H, R2 = 5'-F
8c, R1 = H, R2 = 5'-OnBu
8d, R1 = 5-F, R2 = 5'-F
9a, R1 = H
9b, R1 = F
Scheme 1. Synthetic strategies to 1a–d.
Table 1
Optimization studiesa
O
O
O
O
catalyst, base, solvent
90 oC, 12 h
I
+
+
CH3B(OH)2
12
11
NO2
NO2
8a
2a
NO2
NO2
Entry
Catalyst
Base
Solvent
Yield (%) of 8ab
Yield (%) of 11b
Yield (%) of 12b
Yield (%) of 2ab,c
1
2
3
4
5
6
7
8
9
10
11
12
13
a
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
PdCl2(dppf)ꢀCH2Cl2
Pd2(dba)3/dppf
K2CO3
K2CO3
K2CO3
K2CO3
KF
THF
14
3
45
13
49
5
71
0
0
0
0
35
9
33
71
6
10
3
8
3
65
5
5
1
0
0
1
0
13
4
0
9
13
0
52
21
47
65
43
67
17
25
83
75
93 (76)
55
78 (58)
THF/waterd
Dioxane
Dioxane/waterd
Dioxane
Dioxane
DMF
CsF
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Toluene
THF
Dioxane
Dioxane/waterb
DME
2
4
3
0
1
Pd(PPh3)4
DME
All reactions were run using 3 mol % catalyst, 1.5 equiv of methylboronic acid 7, 3 equiv of base for 12 h with total concentration at 0.1 M.
b
c
Reversed-phase HPLC area percents.
Isolated yields are reported in parentheses.
4:1 v/v.
d
carbometallation-cross-coupling reaction with the use of
incomplete reactions were observed under a variety of
solvents (entries 1–4). In addition, significant amounts of
direct coupling byproduct 11 was also detected under these
conditions. Using Cs2CO3 as the base provided better
reactivity in general. In solvents such as toluene, THF,
dioxane and dioxane/water, reactions proceeded to
completion at 90 °C within several hours (entries 8–11).
In particular, when dioxane/water was used, both direct
coupling byproduct 11 and des-iodo byproduct 12 were
well controlled (entry 11). Therefore, under this optimized
condition, 3 mol % PdCl2(dppf)ꢀCH2Cl2, 1.5 equiv of
methylboronic acid as the terminating trapping species.
As shown in Table 1, alkyne 8a was initially chosen to
test the feasibility of this sequence. Therefore, exploration
of tetrasubstituted dibenzoxapine formation was probed
by heating 1.5 equiv of methylboronic acid, 3 mol % of pal-
ladium catalyst and 3.0 equiv of base in 0.1 M solvent at 90
°C for 12 h. PdCl2(dppf)ꢀCH2Cl2 was used as the catalyst in
most of our screenings, since it has been described as an
effective catalyst in promoting coupling reactions involving
alkylboronic acids.9a When K2CO3 was used as the base,

H. Yu et al. / Tetrahedron Letters 49 (2008) 1915–1918
1917
Table 2
methylboronic acid 7 as the terminating species. As illus-
trated in Table 2, several different functional groups could
be incorporated into the aromatic rings to afford the tetra-
substituted exocyclic E-alkenes 2a–d as a single stereo-
isomer11 using optimized condition A (entries 1–4).12 2a–d
were further converted to the desired nuclear hormone
modulators 1a–d via a nitro group reduction followed by
a sulfonation procedure.2
Surprisingly, when the same condition was applied to
the cyclization of alkyne 8e, in which a methylsulfonamide
group was present, significant amounts of wrong stereoiso-
mer was observed13 and a poor yield was obtained as a
result (entry 5). To overcome the issue, an alternative con-
dition B was developed for substrate 8e. Thus, when
K2CO3 was used as the base and dioxane as the solvent,
the formation of the wrong isomer was reduced to less than
4% (entry 6). It is worth noting that when condition B was
applied to the cyclization of 8a, significant amounts of
starting material remained (Table 1, entry 3). Unfortu-
nately, when ethylboronic acid was used as the terminating
species, very low yield of desired product 13 was isolated
(Table 2, entry 7). No product was isolated when n-propyl-
boronic acid was used (Table 2, entry 8).14
Synthesis of dibenzoxapine derivatives
Entry
1
ArI
Product
Yieldc (%)
76a
O
O
I
8a
2a
NO2
NO2
F
O
F
O
I
2
3
4
5
6
7
71a
69a
65a
38a
58b
11a
0a
8b
2b
NO2
NO2
OnBu
n
O
O
Bu
O
I
8c
2c
NO
NO2
2
F
O
F
F
O
F
I
8d
The methodology was not limited to aryl iodide sub-
strates. We had also obtained excellent results using aryl
bromide 15 as the starting substrate (Eq. 1). An improved
yield was observed when trimethylboroxine instead of
methylboronic acid was used as the terminating species
for aryl bromide 15.15
2d
NO2
NO2
O
O
I
O
8e
S
S
O
NH
N
H
O
1a
S
O
Pd(OAc)2 (5 mol %)
tri- -tolyphosphine (10 mol %)
O
o
O
O
O
Na2CO3 (3 equiv), dioxane/water (4:1)
100 oC, 12 h
Br
I
B
O
NO2
8e
O
B
O
B
(1.5 equiv)
yield: 79%
yield: 62%
O
NO2
NH
N
H
O
1a
2a
S
O
15
O
CH3B(OH)2 (1.5 equiv)
O
O
O
ð1Þ
I
In summary, we have developed a palladium-catalyzed
cascade reaction for the synthesis of a series of nuclear hor-
mone modulators using methylboronic acid as the termi-
nating species. The synthesis provided a straightforward
approach to the dibenzoxapine containing tetrasubstituted
exocyclic E-alkenes.
8a
NO2
13
NO2
O
I
8
8a
References and notes
NO2
14
NO2
a
1. 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.
2. Yu, H.; Richey, R. N.; Carson, M. W.; Coghlan, M. J. Org. Lett.
2006, 8, 1685.
Condition A: PdCl2(dppf)ꢀCH2Cl2 (3 mol %), boronic acid (1.5 equiv),
Cs2CO3 (3 equiv), dioxane/water (4:1), 90 °C, 12 h.
b
Condition B: PdCl2(dppf)ꢀCH2Cl2 (3 mol %), methylboronic acid 7
(1.5 equiv), K2CO3 (3 equiv), dioxane, 90 °C, 12 h.
c
Isolated yield.
3. Arcadi, A.; Cacchi, S.; Rosario, M. D.; Fabrizi, G.; Marinelli, F. J.
Org. Chem. 1996, 61, 9280.
4. Hiroya, K.; Suzuki, N.; Yasuhara, A.; Egawa, Y.; Kasano, A.;
Sakamoto, T. J. Chem. Soc., Perkin Trans. 1 2000, 4339.
5. (a) Simon, C.; Hosztafi, S.; Makleit, S. J. Heterocycl. Chem. 1997, 34,
349; (b) Hughes, D. L. Org. Prep. Proced. Int. 1996, 28, 127; (c)
Mitsunobu, O. Synthesis 1981, 1.
methylboronic acid and 3.0 equiv of Cs2CO3, the desired
product 2a was isolated in 76% yield.
We next applied the optimized conditions to the forma-
tion of tetrasubstituted dibenzoxapine derivatives, using

1918
H. Yu et al. / Tetrahedron Letters 49 (2008) 1915–1918
6. Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
7. Suzuki, A. J. Organomet. Chem. 1999, 576, 147.
(10 mL) and water (10 mL). EtOAc layer was washed with
brine (10 mL) and dried with MgSO4. The residue after evaporation
was subjected to silica gel chromatography (9:1 hexane/CH2Cl2)
to give dibenzoxapine 2a (287 mg, 0.84 mmol, 76% yield) as a clear
oil. 1H NMR (500 MHz, CDCl3) d 8.01–7.96 (m, 2H), 7.37–7.29 (m,
4H), 7.23–7.18 (m, 2H), 6.89 (m, 3H), 6.67 (d, J = 7.5 Hz, 1H),
8. (a) Miyaura, N.; Ishiyama, T.; Sasaki, H.; Ishikawa, M.; Satoh, M.;
Suzuki, A. J. Am. Chem. Soc. 1989, 111, 314; (b) Soderquist, J. A.;
Santiago, B. Tetrahedron Lett. 1990, 31, 5541; (c) Furstner, A.; Seidel,
G. Tetrahedron 1995, 51, 11165; (d) Molander, G. A.; Ito, T. Org.
Lett. 2001, 3, 393.
9. (a) Molander, G. A.; Yun, C. Tetrahedron 2002, 58, 1465; (b)
Kataoka, N.; Shelby, Q.; Stambuli, J. P.; Hartwig, J. F. J. Org. Chem.
2002, 67, 5553; (c) Kondolff, I.; Doucet, H.; Santelli, M. Tetrahedron
2004, 60, 3813; (d) Walker, S. D.; Barder, T. E.; Martinelli, J. R.;
Buchwald, S. L. Angew. Chem., Int. Ed. 2004, 43, 1871.
10. (a) Sato, M.; Miyaura, N.; Suzuki, A. Chem. Lett. 1989, 1405; (b)
Chemler, S. R.; Trauner, D.; Danishefsky, S. J. Angew. Chem., Int.
Ed. 2001, 40, 4544; (c) Zou, G.; Reddy, Y. K.; Falck, J. R.
Tetrahedron Lett. 2001, 42, 7213.
11. The undesired diastereomers were not detectable by NMR in crude
reaction mixtures.
5.89 (d, J = 12.5 Hz, 1H), 4.99 (d, J = 12.5 Hz, 1H), 2.40 (s, 3H); 13
NMR (125 MHz, CDCl3) 154.9, 147.8, 145.0, 143.2, 137.3,
C
d
134.4, 133.2, 132.8, 131.1, 129.6, 128.8, 128.6, 128.2, 127.9, 127.6,
124.9, 123.2, 121.4, 120.2, 119.7, 70.1, 22.0; IR (KBr) m 3069, 3032,
2935, 1607, 1532 cmꢁ1; MS (EI) m/z (rel intensity) 344 (M+, 100);
HRMS (ES) m/z (M+1) calcd for C22H17NO3, 344.1287; found,
344.1294.
13. The E/Z isomerization could occur at either the vinylpalladium or the
product stage. When dibenzoxapine 1a was stirred under the reaction
conditions overnight, no isomerization was observed. Therefore, we
suspected that isomerization occurred at the vinylpalladium stage
under condition A.
12. General procedure for the preparation of 2a–d: To a round bottom
flask was added alkyne 8a (500 mg, 1.10 mmol), methylboronic acid 7
(99 mg, 1.65 mmol), Cs2CO3 (1.08 g, 3.30 mmol), 8.8 mL of dioxane
and 2.2 mL of water. The flask was purged with N2 and
PdCl2(dppf)ꢀCH2Cl2 (27 mg, 0.033 mmol) was added. The reaction
mixture was stirred at 90 °C under N2 for 12 h. The reaction mixture
was concentrated and the residue was partitioned between EtOAc
14. Examination of alternative ethylboronic acid and n-propylboronic
acid mediated reaction conditions also revealed poor results.
15. For substrate 15, replacing methylboronic acid with trimethylbor-
oxine reduced the amount of direct coupling byproduct 11. For aryl
iodide substrates 8a–e, <2% of direct coupling products were formed
under our optimized conditions and very similar results were obtained
when trimethylboroxine was used to replace methylboronic acid.