Derivatization of 1-phenyl substituted 4-amino-2-benzazepin-3-ones: evaluation of Pd-catalyzed coupling reactions

Article abstract of DOI:10.1016/j.tet.2007.02.084

Several Pd-catalyzed reactions were explored to further functionalize the bromo-substituted 4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-one scaffold (Aba). We report in this paper suitable reaction conditions for Suzuki, Buchwald-Hartwig, and Heck reactions

Full text of DOI:10.1016/j.tet.2007.02.084

Tetrahedron 63 (2007) 3718–3727
Derivatization of 1-phenyl substituted 4-amino-2-benzazepin-
3-ones: evaluation of Pd-catalyzed coupling reactions
a
Steven Ballet, Rien De Wachter, Bert U. W. Maes and Dirk Tourwe
a
b
ꢀa,
*
aDepartment of Organic Chemistry, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium
bOrganic Synthesis, University of Antwerp, Groenenborgerlaan 171, B-2020 Antwerp, Belgium
Received 5 January 2007; revised 16 February 2007; accepted 20 February 2007
Available online 23 February 2007
Abstract—Several Pd-catalyzed reactions were explored to further functionalize the bromo-substituted 4-amino-1,2,4,5-tetrahydro-2-benza-
zepin-3-one scaffold (Aba). We report in this paper suitable reaction conditions for Suzuki, Buchwald–Hartwig, and Heck reactions. The sub-
stitution pattern of the starting aminobenzazepinone turned out to be crucial for the success of these transition metal-catalyzed reactions,
which often required modifications of standard literature procedures. The Pd-catalyzed methods provide access to novel substitution patterns
of the Aba scaffold.
Ó 2007 Elsevier Ltd. All rights reserved.
R5
R1
1. Introduction
The 4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-one (Aba)
skeleton 1 is present in several bioactive compounds
(Fig. 1).1–5 This template is the basic unit in ACE-,1 NEP-,2
and farnesyl transferase inhibitors,3 fibrinogen receptor
antagonists,4 and analgesics.5 Synthetic methods, that allow
the efficient introduction of a variety of substituents at differ-
ent positions, are therefore of high value from a medicinal
chemistry point of view.
5
4
1
R4
2N
R2
N
R3
O
1
Figure 1. The 4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-one scaffold 1.
use Pd chemistry. In this way new C–C and C–N bonds
should be easily formed in the late stage of the synthesis
route of the target compounds.
Our research group has reported several methods to prepare
disubstituted,6–8 trisubstituted,9,10 and tetrasubstituted11 4-
amino-2-benzazepin-3-ones with a wide range of substitu-
ents, but until now we never explored the possibilities to
further derivatize the scaffold. Since drug development is
often based on further ‘functionalization’ of an already exist-
ing lead compound, a convergent method for modifying
a common intermediate would increase the efficiency of
the synthesis. An interesting strategy to introduce new
vinyl-, aryl-, arylamino- or alkylamino-type substituents,
that would serve as eventual pharmacophoric groups, is to
The substrates for the evaluation of the Pd-catalyzed
methods were the phthaloyl-protected bromo-substituted
Aba analogs 2 and 3 obtained through two different, earlier
reported, N-acyliminium ion based cyclizations A and B, de-
picted in Scheme 1.11 Method A allows the simultaneous in-
troduction of a substituent at positions 1 (R1 in 1) and 2 (R2
in 1), although there were limitations for the choice of the
2-substituent, caused by steric hindrance. No cyclohexyl
group could, for example, be introduced as an R2 substituent.
The method yields a 1:1 mixture of stereoisomers. Method
B, a procedure based on Katritzky’s benzotriazole chemis-
try,12 allowed the introduction of different 1-substituents
and yields selectively the cis-stereoisomer. Only R2¼H
was however allowed.11 Despite the lower reactivity of
the aromatic ring in 6 (due to the bromine atom) for ring clo-
sure of type B (Scheme 1), we were able to prepare com-
pound 3. This 8-bromo-substituted analog is unavailable
through method A as reported earlier.11 The aryl bromide
moiety in scaffolds 2 and 3 allows to further elaborate
Keywords: 4-Amino-1,2,4,5-tetrahydro-2-benzazepin-3-ones; Pd[0] cataly-
sis; Suzuki reaction; Buchwald–Hartwig reaction; Heck reaction; N-Aryla-
tion.
Abbreviations: Aba, 4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-one; ACE,
angiotensin converting enzyme; BINAP, 2,20-bis(diphenylphosphino)-1,10-
binaphthyl; Cy2MeN, N,N-dicyclohexylmethylamine; dba, dibenzylidene-
acetone; DMA, dimethylacetamide; DME, ethylene glycol dimethyl ether;
NEP, neutral endopeptidase; XANTPHOS, 9,9-dimethyl-4,5-bis(diphenyl-
phosphine)-9H-xanthene.
* Corresponding author. Tel.: +32 2 6293295; fax: +32 2 6293304; e-mail:
datourwe@vub.ac.be
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.02.084

S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
3719
N
HN
Bt =
N
R8
Br
4: R8 = H, X = Br
6: R8 = Br, X = H
O
Cl
Bt
N
R2
N
O
R8
N
O
N
N
O
N
H
O
X
X
X
O
O
R8
B
Br
A
R2
N
O
R2 = H
O
N
R2
Cl
O
O
8
O
X = Br, R8 = H: 2
N
O
X = H, R8 = Br: 3
N
H
5: R8 = H, X = Br
7: R8 = Br, X = H
O
Bt
5
O
Scheme 1. Synthesis of 1-(4-bromophenyl)- and 1-phenyl-8-bromo-Aba scaffolds 2 and 3.
functionalization at these positions via Pd-catalyzed reac-
tions. For this purpose, we explored Buchwald–Hartwig
aminations, Heck alkenylations, and Suzuki arylation reac-
tions on 2 and 3.
the starting material was observed in this case and an iso-
lated yield of 60% was achieved.
We then examined whether the presence of an unprotected
secondary amide in benzazepinone 11 (Scheme 3) would in-
fluence the C–N bond formation. Compound 11 was ob-
tained through method B in Scheme 1. Taking into account
the success of the amination reaction with 2-aminopyridine
(entry 2, Table 1), we first chose to investigate its coupling
with 11. The reaction did, however, not occur. No conversion
to the desired compound was observed. Due to the fact that
the reactivity of an amide might be comparable to that of the
amidine 2-aminopyridine, we subsequently tried p-toluidine
and morpholine as, respectively, examples of aniline type
amines and cyclic aliphatic amines. This assumption of sim-
ilar reactivity is based on a recent paper of Buchwald and co-
workers in which they suggested that the chemoselectivity of
palladium catalysts follows the rough amine reactivity order:
aryl amines[primary and secondary aliphatic amines>2-
aminoheteroaromatics>primary amideswHN heterocy-
cles.17 Of course this order is based on the use of substituted
dialkylphosphinobiphenyl ligands that act as monodentate
P-ligands, while we used the bidentate XANTPHOS for
the coupling of 2-aminopyridine. Unfortunately, also p-
toluidine (reaction conditions of entry 1, Table 1) and mor-
pholine (reaction conditions of entry 8, Table 1) did not
result in any conversion to the desired compound 12. From
these results, we can conclude that position 2 in scaffold 1
has to be substituted/protected before Pd-catalyzed amina-
tion reactions are performed.
2. Results and discussion
2.1. Buchwald–Hartwig reactions
Based on our earlier work on dihalopyridines,13 XANT-
PHOS was chosen as a ligand and Pd2dba3 as Pd(0) source
for the catalyst to couple anilines and heteroarylamines
with 9 (see Scheme 2 and Table 1). Compound 9 was synthe-
sized through method A in Scheme 1 as a 1:1.2 mixture of
epimers at C1.11 With this Pd/L (ratio 1:1.1) system, we
were able to smoothly couple p-toluidine (entry 1) and 2-
aminopyridine (entry 2). When we tried to introduce a cyclic
aliphatic amine (entry 3), however, the target compound was
only formed in a small amount. Therefore, the ligand of the
catalytic system was changed. With racemic BINAP, only
10% conversion to the desired reaction product could be
obtained (entry 4).14 The use of biphenyl based ‘Buchwald
ligands’, 2-dicyclohexylphosphinobiphenyl (entry 5) and
2-di-tert-butylphosphinobiphenyl (entry 6), in combination
with K3PO4 induced the best conversions and isolated
yields.15 Increasing the catalyst loading from 5 to 7.5 mol %
did not further improve conversion using 2-di-tert-butyl-
phosphinobiphenyl as ligand. Interestingly, the recently
introduced ligand 2-di-tert-butylphosphino-20,40,60-triiso-
propylbiphenyl led to a reduced conversion of only 50%
after 24 h reaction time (entry 7).16 To verify if this incom-
plete conversion (90%) of entry 6 was not just due to the very
specific amine selected as model, we repeated the same
procedure with morpholine as another example of a cyclic
aliphatic amine (entry 8). Gratifyingly, full conversion of
2.2. Heck reaction
Three different literature systems were examined (see Table
2) for the reaction of 9 with ethyl acrylate (Scheme 4).
The combination Pd2dba3/(t-Bu)3P (Pd/L ratio 1:1) with
Pd2dba3/ligand
Base
Br
NRR'
RR'NH
O
O
N
N
COOEt
N
COOEt
N
O
O
9
10
O
O
Scheme 2. Buchwald–Hartwig reaction of Pht-1-(R,S)-(4-bromophenyl)-(S)-Aba-Gly-OEt 9.

3720
S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
Table 1. Overview of the Buchwald–Hartwig reaction conditions
Entry
1
Ligand (1.1 equiv)
Base (equiv)
Cs2CO3 (4)
Amine (1.2 equiv)
Solvent
DME
Time (h)/Temp. (ꢀC)
20/95
Conversion (%), Yield (%)
10a, C: 100, Y: 60
H2N
O
PPh2
PPh2
PPh2
PPh2
PPh2
PPh2
N
2
3
Cs2CO3 (4)
Cs2CO3 (4)
DME
DME
20/95
20/95
10b, C: 100, Y: 65
10c, C: 100, Y: 7a
H2N
O
O
O
HN
O
O
4
Cs2CO3 (2)
HN
O
DME
20/95
10c, C: 10
PPh2
PPh2
O
5
6
K3PO4 (2)
K3PO4 (2)
HN
O
Toluene
Toluene
20/100
20/100
10c, C: 88
(Cy)2P
O
HN
O
10c, C: 90, Y: 54
(t-Bu)2P
R
O
R
7
K3PO4 (2)
HN
O
Toluene
Toluene
20/100
20/100
10c, C: 50
R
P(t-Bu)2
R = iPr
8
K3PO4 (2)
10d, C: 100, Y: 60
HN
Br
O
(t-Bu)2P
Mostly side products were obtained.
a
Pd2dba3/ligand
Base
RR'NH
NRR'
O
O
NH
NH
N
N
O
O
O
11
O
12
Scheme 3. Attempted Buchwald–Hartwig reaction of Pht-1-(R)-(4-bromophenyl)-(S)-Aba 11.
Cy2MeN as a base, introduced by Fu et al., gave a conversion
of only 5% (entry 1).18 The shift from a phosphine to a car-
bene ligand (Table 2, entry 2), as reported by Nolan et al., did
not improve the yields.19 Although both catalytic systems
have been successfully used on a substantial set of
substituted bromobenzene derivatives it is clear that the
Table 2. Overview of the Heck reaction conditions
Entry Pd catalyst (mol %) Base (equiv)
Added ligand
(mol %)
Solvent
Time (h)/Temp. (ꢀC) Conversion (%), Yield (%)
1
2
Pd2dba3 (2.5)
Pd(OAc)2 (5)
Cy2MeN (1.1) (t-Bu)3P (5)
Dioxane 24/105
13, C: 5
13, C: 5
Cl -
N
N
Cs2CO3 (2)
NaHCO3 (2)
(10) DMA
24/120
24/120
3
Pd(PPh3)4 (30)
None
DMF
13, C: 100, Y: 66

S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
3721
O
OEt
Br
Pd [0]/ligand
Base
O
O
N
N
ethyl acrylate
N
COOEt
N
COOEt
O
O
9
O
13
O
Scheme 4. Heck reaction of Pht-1-(R,S)-(4-bromophenyl)-(S)-Aba-Gly-OEt 9.
Table 3. Reaction conditions for Suzuki couplings
complexity of the starting material, as exemplified by sub-
strate 9, imposes serious limits on the use of standard, care-
fully optimized, literature protocols in medicinal chemistry
programs.
Entry Base
(3 equiv)
Boronic acid
(equiv)
Solvent Time (h)/ Conversion
Temp. (ꢀC) (%), Yield
(%)
HO
HO
Remarkably, the best catalyst for the Heck reaction on 9
was standard tetrakis(triphenylphosphine)palladium (entry
3).20,21 However, a very high catalyst loading was required
(30%) to get full conversion in 24 h. This is probably due
to the interference with other functionalities present on sub-
strate 9. However, this result demonstrated that benzazepine
9 can be further functionalized with a a,b-unsaturated ester
function, thus meeting our primary medicinal chemistry goal.
1
2
KF
(1.5) THF 24/50
15a, C: 80
B
B
F
F
HO
HO
15a, C: 100,
Y: 71
KF
KF
(1.5) DME 24/95
(1.5) DME 24/95
HO
15b, C: 100,
Y: 66
3
B
OMe
HO
HO
S
4
5
KF
KF
(1.5) DME 24/95
15c, C: 50
15c, C: 85
B
B
2.3. Suzuki reaction
HO
HO
S
S
The functionalizations via Suzuki reaction were based on
a methodology developed by Buchwald et al.22 who reported
that a mixture of Pd(OAc)2 and 2-(di-tert-butylphosphino)-
biphenyl 14 is a good precatalytic system for Suzuki aryla-
tion of aryl bromides with KF as base (Scheme 5). We were
able to confirm these findings on our scaffold. Table 3 gives
an overview of the experiments with 4-fluorophenylboronic
acid as an organometallic partner. A loading of 5% Pd and
10% ligand 14 (Pd/L ratio 1:2) was chosen. Heating at
50 ꢀC in THF (entry 1) resulted in an incomplete conversion
of 9. Switching to DME as a solvent (entry 2) permitted to
increase the reaction temperature and resulted in complete
conversion of starting material 9. The electron rich p-
methoxyphenyl group was also smoothly introduced by
means of p-methoxyphenyl boronic acid (entry 3). Also het-
eroaromatic moieties can be introduced as exemplified by the
3-thienyl moiety (entries 4–6). In this case, an excess of
3.5 equiv arylboronic acid was required for a full conversion
in 24 h.
(2.5) DME 24/95
(3.5) DME 24/95
HO
HO
HO
15c, C: 100,
Y: 64
6
KF
B
As we observed the formation of a precipitate during the re-
action, we considered solubility issues to be responsible for
the incomplete conversion. Therefore DMA was added as
a co-solvent. Gratifyingly, employing a 1:1 solvent mixture
of DME/DMA yielded 70% of 16. We assume that deproto-
nation of the lactam moiety by KF is responsible for the
precipitation of 11 and consequently for the failure to fully
convert 11 by Suzuki reaction in DME.
Identical conditions as found optimal for substrate 11
(Scheme 6) were also applied to the regioisomeric 8-bromo
analog 3. We were able to obtain 17 in a good yield (Scheme
7). This represents an additional functionalization opportu-
nity of position 8 in the Aba scaffold. Analogs of type 17
would otherwise only be available from expensive unnatural
amino acid precursors.
We also investigated how cross-coupling with arylboronic
acids would proceed in the presence of the unprotected/
unsubstituted secondary amide in benzazepinone 11. Inter-
estingly, in contrast to the amination reactions, Suzuki
cross-coupling on this substrate with 4-fluorophenylboronic
acid gave, initially, partial conversion to the desired
compound 16.
2.4. N-Arylation of the lactam nitrogen
A further diversification of the N2-unsubstituted Aba scaffold
11 can be obtained by N-arylation (Scheme 8). Le Diguarher
Pd(OAc)2
Br
R
ArB(OH)2
O
O
N
N
N
COOEt
N
COOEt
O
O
9
15
O
(t-Bu)2P
O
14
Scheme 5. Suzuki reaction of Pht-1-(R,S)-(4-bromophenyl)-(S)-Aba-Gly-OEt 9.

3722
S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
F
Br
Pd(OAc)2
4-F-PhB(OH)2
O
O
NH
NH
N
N
O
O
O
O
11
(t-Bu)2P
70%
14
16
Scheme 6. Suzuki couplings of Pht-1-(R)-(4-bromophenyl)-(S)-Aba 11.
F
earlier by Buchwald for intermolecular coupling of aryl
halides and amides.24 To confirm this theoretical consider-
ation, we repeated the arylation reaction using the same
Pd2dba3/XANTPHOS precatalytic system on a compound
where sterical hindrance was substantially reduced by re-
moving the 1-substituent of 11 (Scheme 9). Interestingly,
compound 19 smoothly coupled with the electron deficient
aryl halide 4-iodobenzonitrile. Unfortunately, trying to ex-
tend the reaction to inactivated (electronically neutral) aryl
halides, such as 2-iodotoluene and 4-iodotoluene, did not
lead to the desired reaction products. As an amide group is
not very nucleophilic, good results can only be achieved
with sufficiently electron deficient aryl halides. This is based
on the knowledge that reductive elimination of Pd(II)–amide
complexes will go easier with electron deficient aryl groups
and more nucleophilic amides.25 Although the scope of this
Pd-catalyzed N-arylation reaction is obviously hampered by
steric as well as electronic factors, we were pleased to see
that electron deficient aryl halides, exemplified by 4-iodo-
benzonitrile, are suitable substrates. To the best of our
knowledge, the arylation of 19 is the first example in which
a 4-amino-tetrahydro-2-benzazepin-3-one has been success-
fully used in a Pd-catalyzed N-arylation reaction.
Pd(OAc)2
Br
8
Ligand 14
7
4F-PhB(OH)2
68%
6
5a
1
5
4
O
O
2NH
3
NH
N
N
O
O
17
3
O
O
Scheme 7. Suzuki reaction of Pht-1-(R)-phenyl-8-bromo-(4S)-Aba 3 with
4-fluorophenylboronic acid.
Catalyst
Ligand
Br
Br
O
O
NH
N
I
CN
N
N
O
O
O
O
CN
11
18
Scheme 8. AttemptedN-arylationofPht-1-(R)-(4-bromophenyl)-(S)-Aba11.
already reported N-arylation of a 4-amino-2-benzazepin-3-
one for the preparation of a farnesyl transferase inhibitor.3
This procedure was however not catalytic and based on the
use of an equimolar amount of Cu(OAc)2. Moreover, triaryl-
bismuth reagents were used as aryl source, limiting the func-
tional group compatibility of the process. In addition, these
organometallic partners are not easily accessible.
I
CN
O
O
Pd2dba3 (5 mol%)
NH
N
N
N
XANTPHOS
(11 mol%)
Cs2CO3
O
O
O
O
CN
19
20
71%
As N-arylation of amides has recently been described under
Cu-catalysis,23 we first investigated the reaction of 11 with
4-iodobenzonitrile, using a combination of CuI and trans-
1,2-cyclohexanediamine in dioxane (Table 4, entry 1) as
well as in toluene. This attempt was unsuccessful since no
desired compound could be observed by LC–MS analysis.
Switching back to an older literature procedure based on pal-
ladium catalysis, with XANTPHOS as ligand, also did not
provide the desired compound 18 (Table 4, entry 2).24 These
failures were attributed to the bulky substituents on both
sides of the lactam function. This leads to a problematic con-
version due to a steric factor in the binding of the deproto-
nated amide to the Pd(II) intermediate, as also reported
Scheme 9. N-Arylation of Pht-(S)-Aba 19.
3. Conclusions
A variety of new substituents were introduced via Pd-cata-
lyzed reactions. Buchwald–Hartwig as well as Suzuki reac-
tions were successfully optimized by analyzing different Pd/
L systems in a variety of solvents, using the N2-substituted
benzazepinone 9. The presence of an unsubstituted second-
ary amide in the scaffold, i.e. using compound 11, prevented
the formation of C–N bonds through Pd-catalyzed amination
Table 4. N-Arylation of Pht-1-(R)-(4-bromophenyl)-(S)-Aba 11
Entry
Catalyst (5 mol %)
Base (equiv)
K3PO4 (1.6)
Cs2CO3 (4)
Added ligand (mol %)
Solvent
Dioxane
Dioxane
Time (h)/Temp. (ꢀC)
24/110
Conversion (%), Yield (%)
No conversion
NH2
1
2
CuI
(15)
(+/-)
NH2
Pd2dba3
XANTPHOS (11)
24/100
No conversion

S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
3723
reactions. Suzuki cross-coupling reactions were, however,
still successfully performed on this type of substrate if a
dipolar aprotic solvent was used as a co-solvent. ‘Classical’
Heck reaction conditions, based on tetrakis(triphenylphos-
phine)palladium as precatalytic system, lead to full conver-
sion and consequently to a good isolated yield. Additionally,
we examined steric as well as electronic factors for the
N-arylation of N2-unsubstituted benzazepin-3-ones, and con-
cluded that the scope of this reaction is limited to electron
deficient aryl halides and a benzazepinone scaffold without
the 1-phenyl substitution. The investigated Pd-catalyzed
conversions led to the preparation of several benzazepinones
bearing new types of functionalities. This allows a consider-
able expansion of the range of substituent diversity available
for the priviledged template. Moreover, since the phthaloyl
protection of N4 is easily removed, further functionalization
at N4 can easily be performed.7
2-yl)acetate (10a, entry 1, Table 1). An oven dried 25 mL
round-bottomed flask was charged with Pd2dba3 (9.2 mg,
0.010 mmol, 5 mol %) and XANTPHOS (12.8 mg,
0.022 mmol, 11 mol %). The flask was flushed with N2
and subsequently DME (6 mL) was added. This solution
was stirred for 10 min at rt. During these 10 min, 9
(109.5 mg, 0.20 mmol), p-toluidine (25.7 mg, 0.24 mmol,
1.2 equiv), and Cs2CO3 (261 mg, 0.8 mmol, 4 equiv) were
weighed into another 25 mL flask. This flask was also
flushed with N2 and 3 mL of Pd(0)/XANTPHOS solution
(0.005 mmol Pd2dba3 and 0.011 mmol XANTPHOS) was
added. The reaction mixture was heated (oil bath tempera-
ture: 95 ꢀC) and magnetically stirred for 20 h under N2 at-
mosphere. After evaporation of the solvent, the crude
product was dissolved in a minimum amount of acetoni-
trile/water (1.2:1), filtered, and purified by preparative
HPLC. The desired compound 10a (68.8 mg, 0.12 mmol,
60%) was obtained as a mixture of diastereoisomers in the
form of a light brown solid. TLC: Rf 0.66 (EtOAc/hexanes
1:1), mp 125–127 ꢀC. MS (ESP+) found m/z¼574
([M+H]+), C35H31N3O5 requires 573.64. HPLC: retention
4. Experimental
4.1. General
1
time¼19.2 and 19.3 (min). H NMR 250 MHz (CDCl3):
dH¼1.08 and 1.22 (3H, 2t, HCH , J1¼3J2¼6.3 Hz), 2.30
3
3
2
0
Thin layer chromatography (TLC) was performed on a plas-
tic sheet precoated with silica gel 60F254 (Merck). Melting
(3H, s, HArꢂCH ), 2.54 and 3.20 (1H, 2dd, Hb, J1(Hb,Hb )¼
3
2
0
14.5 Hz,
J2(Hb,Hb )¼17.2 Hz,
3J1(Hb,Ha)¼3.2 Hz,
2
points (mp) were determined on a Buchi B540 Melting Point
3J2(Hb,Ha)¼4.6 Hz), 3.66 (1H, d, Ha Gly, J(Ha,Ha )¼
0
€
2
Apparatus with a temperature gradient of 1 ꢀC/min. Mass
spectra (MS) were recorded on a VG Quattro II spectrometer
using electrospray (ESP) ionization (positive or negative ion
mode). Data collection was done with MassLynx software.
1H NMR and 13C NMR spectra were recorded at 250 and
63 MHz, respectively, on an AC 250 Bruker spectrometer.
Solvents and chemical shifts (d), using TMS as an internal
standard, are reported for each compound. When mixtures
of diastereoisomers were analyzed, chemical shifts of iden-
tical protons or carbons in both isomers were grouped. The
16.7 Hz), 3.87 (0.6H, pseudo t, Hb , J(Hb ,Hb)z
0
0
3
0
0
J(Hb ,Ha)¼14.1 Hz), 4.20 (2.4H, m, Hb þ HCH ), 4.86
2
and 5.43 (1H, 2dd, Ha Aba, 3J1(Ha,Hb)¼3.2 Hz,
3
3
3J2(Ha,Hb)¼4.5 Hz, J1(Ha,Hb )¼13.4 Hz, J2(Ha,Hb )¼
0
0
2
0
12.2 Hz), 5.08 (1H, d, Ha Gly, J¼17 Hz), 5.60 and 5.64
(1H, 2s, H3), 6.95–7.58 (13H, M, H arom), 7.63–7.78 (4H,
M, HPht). 13C NMR 63 MHz (CDCl3): dC¼14.1 and 14.2
(CH3 Et), 20.6 (CH3Ar), 34.1 and 35.1 (Cb), 52.5 and 53.3
(Ca Gly), 53.5 and 54.3 (Ca Aba), 61.5 (CH2 Et), 69.3 and
70.1 (C3), 116.8 (CH arom), 117.3 (CH arom), 119.4 (CH
arom), 124.0 (CH arom), 127.5 (CH arom), 130.0 (CH
arom), 130.7 (CH arom), 131.1 (CH arom) 132.5
(CH arom), 133.3 (CH arom) 134.8 (Cquat arom), 134.9
(Cquat arom), 135.2 (Cquat arom), 137.0 (Cquat arom), 139.0
(Cquat arom), 139.9 (Cquat arom), 144.4 (Cquat arom), 168.3
(C]O), 168.8 (C]O), 170.2 (C]O).
n
n
notations of J1(Hx,Hy) and J2(Hx,Hy) were used for the
geminal (n¼2) and vicinal (n¼3) couplings constants of
isomer 1 and isomer 2, respectively. Reverse phase (RP)-
HPLC was performed using an Agilent 1100 Series system
(Waldbronn, Germany) with a RP C-18 column (Supelco
Discovery BIO Wide PoreÒ, 25 cmꢁ4.6 mm, 5 mm). The
mobile phase (water/acetonitrile) contained 0.1% TFA.
The standard gradient consisted of a 20 min run from 3%
to 97% acetonitrile at a flow of 1 mL minꢂ1 with UV detec-
tion at 215 nm. Preparative RP-HPLC was performed on
a Gilson apparatus and controlled by the software package
Unipoint. The reverse phase C-18 column (Discovery
Wide Pore, 25 cmꢁ21.2 mm, 10 mm) was used under the
same conditions as the analytical RP-HPLC, but at a flow
rate of 20 mL minꢂ1. The following compounds were bought
from commercial sources: XANTPHOS (Aldrich), 2-di-tert-
butylphosphino-20,40,60-triisopropylbiphenyl (Aldrich), rac.
4.2.2. Synthesis of ethyl {(4S)-4-(1,3-dioxo-1,3-dihydro-
2H-isoindol-2-yl)-3-oxo-1(R,S)-[4-(pyridine-2-ylamino)-
phenyl]1,3,4,5-tetrahydro-2H-2-benzazepin-2-yl}acetate
(10b, entry 2, Table 1). Identical reaction conditions as for
the synthesis of 10a were used.
The desired product 10b (72.5 mg, 0.13 mmol, 65%) was ob-
tained asamixtureofdiastereoisomers intheformofayellow
solid. TLC: Rf 0.35 (EtOAc/hexanes 1:1), mp 132–134 ꢀC.
MS (ESP+) found m/z¼561 ([M+H]+), C35H31N3O5 requires
560.60. HPLC: retention time¼13.2 and 13.5 (min). 1H
NMR 250 MHz (CDCl3): dH¼1.12 and 1.23 (3H, 2t,
BINAP
(Rhodia),
2-dicyclohexylphosphinobiphenyl
(Acros), 2-di-tert-butylphosphinobiphenyl (Aldrich), (t-
Bu)3P (Aldrich), 1,3-bis(2,4,6-trimethylphenyl)-imidazo-
lium chloride (Aldrich), Pd(OAc)2, and Pd2dba3 (Acros).
HCH
,
3J1¼3J2¼7.5 Hz), 2.55 and 3.18 (1H, 2dd, Hb,
3
J1(Hb,Hb )¼13.5 Hz, J2(Hb,Hb )¼16.5 Hz, 3J1(Hb,Ha)¼
2
2
0
0
3
2
2.5 Hz, J2(Hb,Ha)¼4.5 Hz), 3.61 (1H, d, Ha Gly, J(Ha,
2
0
0
0
4.2. Buchwald–Hartwig amination of compound 9
Ha )¼17.5 Hz), 3.73 (0.6H, pseudo t, Hb , J(Hb ,Hb)z
3
0
0
J(Hb ,Ha)¼14.0 Hz), 4.08–4.35 (2.8H, m, Hb þ HCH ),
2
2
0
0
4.2.1. Synthesis of ethyl ((4S)-4-(1,3-dioxo-1,3-dihydro-
2H-isoindol-2-yl)-1(R,S)-{4-[(4-methylphenyl)amino]-
phenyl}-3-oxo-1,3,4,5-tetrahydro-2H-2-benzazepin-
4.77 and 5.10 (1H, 2d, Ha Gly, J(Ha ,Ha)¼18.0 Hz),
4.86 and 5.26 (1H, 2dd, Ha Aba, 3J1(Ha,Hb)¼3.0 Hz,
3
3
3
0
0
J2(Ha,Hb)¼4.0 Hz, J1(Ha,Hb )¼ J2(Ha,Hb )¼14.0 Hz), 5.65

3724
S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
and 5.69 (1H, 2s, H3), 6.82 (1H, m, NH), 7.16–7.56 (10H, M,
H arom), 7.63–7.75 (6H, M, H arom). 13C NMR 63 MHz
(CDCl3): dC¼14.2 and 14.3 (CH3 Et), 32.1 and 33.5 (Cb),
51.6 and 53.0 (Ca Gly), 53.1 and 54.1 (Ca Aba), 62.0
(CH2 Et), 68.5 and 69.3 (C3), 111.5 (CH arom), 112.1 (CH
arom), 114.3 (CH arom), 125.1 (CH arom), 126.5
(CH arom), 128.5 (CH arom), 129.0 (CH arom), 130.0 (CH
arom), 130.2 (CH arom), 131.2 (CH arom), 134.0 (CH
arom), 135.1 (Cquat arom), 137.0 (Cquat arom), 137.3 (Cquat
arom), 137.5 (Cquat arom), 138.0 (Cquat arom), 144.0 (C6H
pyridine), 154.5 (C2quat pyridine), 168.3 (C]O), 169.0
(C]O), 170.1 (C]O).
(10d, entry 8, Table 1). Identical reaction conditions as for
the synthesis of 10c were used.
The desired product 10d (66.4 mg, 0.12 mmol, 60%) was
obtained as a mixture of diastereoisomers in the form of
a white solid. TLC: Rf 0.32 (EtOAc/hexanes 1:1), mp
95.8–98.0 ꢀC. MS (ESP+) found m/z¼554 ([M+H]+),
C32H31N3O6 requires 553.61. HPLC: retention time¼15.1
1
and 15.3 (min). H NMR 250 MHz (CDCl3): dH¼1.12 and
1.22 (3H, 2t, HCH , J1¼3J2¼7.5 Hz), 2.53 and 3.16 (1H,
3
3
2
2
0
0
2dd, Hb, J1(Hb,Hb )¼14.0 Hz, J2(Hb,Hb )¼17.0 Hz,
3J1(Hb,Ha)¼3.0 Hz, 3J2(Hb,Ha)¼4.2 Hz), 3.41 and 3.49
(4H, 2m, 2NCH2), 3.61 and 4.31 (1H, 2d, Ha Gly,
2
3
2
0
0
0
4.2.3. Synthesis of ethyl [(4S)-1(R,S)-[4-(1,4-dioxa-8-
azaspiro[4.5]dec-8-yl)phenyl]-4-(1,3-dioxo-1,3-dihydro-
2H-isoindol-2-yl)-3-oxo-1,3,4,5-tetrahydro-2H-2-benz-
azepin-2-yl]acetate (10c, entry 6, Table 1). An oven dried
25 mL round-bottomed flask was charged with Pd2dba3
(9.2 mg, 0.010 mmol, 5 mol %) and 2-(di-tert-butylphosphi-
no)biphenyl (12.0 mg, 0.040 mmol, 20 mol %). The flask
was flushed with N2 and subsequently toluene (6 mL)
was added. This solution was stirred for 10 min at rt.
During these 10 min, 9 (109.5 mg, 0.20 mmol), 1,4-dioxa-
8-azaspiro[4,5]decane (0.030 mL, 0.24 mmol, 1.2 equiv),
and K3PO4 (85 mg, 0.4 mmol, 2 equiv) were weighed into
another 25 mL flask. This flask was also flushed with N2
and 3 mL of the Pd(0)/ligand solution (0.005 mmol Pd2dba3
and 0.010 mmol 2-(di-tert-butylphosphino)biphenyl) was
added. The reaction mixture was heated (oil bath tempera-
ture: 100 ꢀC) and magnetically stirred for 20 h under N2
atmosphere. After evaporation of the solvent, the crude prod-
uct was dissolved in a minimum amount of acetonitrile/
water (1.2:1), filtered, and purified by preparative HPLC.
The desired product 10c (65.7 mg, 0.11 mmol, 54%) was ob-
tained as a mixture of diastereoisomers in the form of a white
solid. TLC: Rf 0.33 (EtOAc/hexanes 1:1), mp 128.7–
131.1 ꢀC. MS (ESP+) found m/z¼610 ([M+H]+),
C35H35N3O7 requires 609.67. HPLC: retention time¼13.8
J(Ha,Ha )¼17.0 Hz), 3.67 (1H, pseudo t, Hb , J(Hb ,Hb)z
0
J(Hb ,Ha)¼13.8 Hz), 4.07 and 4.10 (4H, 2m, –CH2OCH2–),
0
4.17 (2H, m, HCH ), 4.69 and 5.07 (1H, 2d, Ha Gly,
2
2
0
J(Ha ,Ha)¼17.0 Hz), 4.86 and 5.16 (1H, 2dd, Ha Aba,
3J1(Ha,Hb)¼3.0 Hz, 3J2(Ha,Hb)¼4.2 Hz,
J1(Ha,Hb )¼
3
0
3
0
13.0 Hz, J2(Ha,Hb )¼13.3 Hz), 5.62 and 5.68 (1H, 2s, H3),
7.18–7.55 (8H, M, H arom), 7.65–7.75 (2H, M, H arom),
7.85–7.91 (2H, M, H arom). 13C NMR 63 MHz (CDCl3):
dC¼13.8 and 14.1 (CH3 Et), 33.1 and 34.2 (Cb), 52.4
and 52.7 (Ca Gly), 52.8 (Ca Aba), 53.5 (NCH2), 61.5 (CH2
Et), 65.3 (CH2OCH2), 68.4 and 69.6 (C3), 118.9 (CH
arom), 123.4 (CH arom), 127.8 (CH arom), 128.4 (CH
arom), 129.9 (CH arom), 130.1 (CH arom), 130.4
(CH arom), 131.7 (Cquat arom), 134.0 (CH arom), 134.2
(Cquat arom), 136.6 (Cquat arom), 137.2 (Cquat arom), 144.8
(Cquat arom), 168.9 (C]O), 169.8 (C]O), 170.0 (C]O).
4.3. Heck coupling between 9 and ethyl acrylate
4.3.1. Synthesis of ethyl (2E)-3-{4-[(4S)-4-(1,3-dioxo-1,3-
dihydro-2H-isoindol-2-yl)-2-(2-ethoxy-2-oxoethyl)-3-
oxo-2,3,4,5-tetrahydro-1H-2-benzazepin-1-yl]phenyl}-
prop-2-enoate (13, entry 3, Table 2). An oven dried 25 mL
round-bottomed flask was charged with 9 (109.5 mg,
0.20 mmol), NaHCO3 (33.6 mg, 0.40 mmol, 2 equiv), and
ethyl acrylate (0.050 mL, 0.4 mmol, 2 equiv). The flask
was flushed with Ar and DMF (3 mL) was added. Subse-
quently, addition of Pd(PPh3)4 (70 mg, 0.06 mmol,
30 mol %) to the flask was followed by stirring the reaction
mixture for 24 h at 130 ꢀC (oil bath temperature). After
evaporation of the solvent, the crude product was dissolved
in a minimum amount of acetonitrile/water (1.2:1), filtered,
and purified by preparative HPLC. The desired product
(74.7 mg, 0.13 mmol, 66%) was obtained as a mixture of
diastereoisomers in the form of a white solid. TLC: Rf
0.64 (EtOAc/hexanes 1:1), mp 122.6–125.1 ꢀC. MS
(ESP+) found m/z¼567 ([M+H]+), C33H30N2O7 requires
566.60. HPLC: retention time¼18.1 and 18.5 (min). 1H
1
and 14.0 (min). H NMR 250 MHz (CDCl3): dH¼1.12 and
1.22 (3H, 2t, HCH ,3J1¼3J2¼7.6 Hz), 2.11 (4H, m, 2CH2),
3
2
0
2.53 and 3.16 (1H, 2dd, Hb, J1(Hb,Hb )¼15.2 Hz,
J2(Hb,Hb )¼17.7 Hz, 3J1(Hb,Ha)¼3.0 Hz, 3J2(Hb,Ha)¼
2
0
0
5.0 Hz), 3.60 (6H, m, 2NCH2+Ha Gly+Hb ), 4.01 and 4.03
(4H, 2s, –OCH2CH2O–), 4.17 (2H, m, HCH ), 4.68 and
0
2
2
0
5.06 (1H, 2d, Ha Gly, J(Ha ,Ha)¼17.7 Hz), 4.85 and
3
3
5.16 (1H, 2dd, Ha Aba, J1(Ha,Hb)¼3.0 Hz, J2(Ha,Hb)¼
3
3
0
0
4.5 Hz, J1(Ha,Hb )¼13.0 Hz, J2(Ha,Hb )¼12.5 Hz), 5.60
and 5.68 (1H, 2s, H3), 7.20–7.55 (8H, M, H arom), 7.65–
7.75 (2H, M, H arom), 7.85–7.91 (2H, M, H arom). 13C
NMR 63 MHz (CDCl3): dC¼13.8 and 14.1 (CH3 Et), 33.1
and 33.2 (Cb), 34.2 ðCNCH CH Þ, 52.2 and 52.6 (Ca Gly),
2
2
52.4 (Ca Aba), 53.5 (NCH2), 61.5 (CH2 Et), 64.7 (OCH2-
CH2O), 68.5 and 69.6 (C3), 104.9 (Cquat azaspirodecane),
119.7 (CH arom), 124.0 (CH arom), 127.8 (CH arom),
128.3 (CH arom), 129.9 (CH arom), 130.0 (CH arom),
130.1 (CH arom), 131.15 (Cquat arom), 130.4 (CH arom),
134.9 (Cquat arom), 137.0 (Cquat arom), 138.2 (Cquat arom),
144.8 (Cquat arom), 168.9 (C]O), 169.8 (C]O), 170.0
(C]O).
NMR 250 MHz (CDCl3): dH¼1.11 and 1.22 (3H, 2t,
3
HCH ðGly esterÞ, J1¼3J2¼7.5 Hz), 1.33 and 1.34 (3H, 2t,
3
HCH ðacrylate esterÞ, 3J1¼3J2¼7.3 Hz), 2.53 and 3.18
3
2
2
0
0
(1H, 2dd, Hb, J1(Hb,Hb )¼13.6 Hz, J2(Hb,Hb )¼16.6 Hz,
3J1(Hb,Ha)¼3.3 Hz, J2(Hb,Ha)¼4.0 Hz), 3.61 (0.6H, 2d,
3
2
0
0
Ha Gly J(Ha,Ha )¼17.5 Hz), 3.72 (0.6H, pseudo t, Hb ,
2
3
0
0
J(Hb ,Hb)z J(Hb ,Ha)¼13.8 Hz), 4.20 ð5:0H; m; HCH
2
0
acrylate ester þ HCH Gly ester þ Hb þ HaGlyÞ, 4.73 and
2
2
0
0
5.11 (1H, 2d, Ha Gly, J(Ha ,Ha)¼17.1 Hz), 4.85 and 5.26
4.2.4. Synthesis of ethyl [(4S)-4-(1,3-dioxo-1,3-dihydro-
2H-isoindol-2-yl)-1(R,S)-(4-morpholin-4-ylphenyl)-3-
oxo-1,3,4,5-tetrahydro-2H-2-benzazepin-2-yl]acetate
(1H, 2dd, Ha Aba 3J1(Ha,Hb)¼3.0 Hz, 3J2(Ha,Hb)¼4.5 Hz,
J1(Ha,Hb )¼13.3 Hz, 3J2(Ha,Hb’)¼12.6 Hz), 5.64 and
3
0
5.70 (1H, 2s, H3), 6.42 and 6.46 (1H, d, HCHC]O acrylate,

S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
3725
3J(HCHC]O,HPhCH]CH)¼16.1 Hz), 7.09–7.89 (13H, M, H
arom). 13C NMR 63 MHz (CDCl3): dC¼13.8 and 14.1
(CH3 Et), 14.2 and 14.3 (CH3 Et), 33.1 and 34.4 (Cb),
53.4 (Ca Gly), 54.6 (Ca Aba), 61.5 (CH2 Et), 62.5
(CH2 Et), 69.2 and 70.4 (C3), 118.2 (CHC]O acrylate),
123.2 (CH arom), 127.0 (CH arom), 127.8 (CH arom),
128.2 (CH arom), 128.8 (CH arom), 129.9 (CH arom),
130.1 (CH arom), 130.5 (CH arom), 133.8 (Cquat arom),
134.2 (CH arom), 136.9 (Cquat arom), 138.4 (Cquat arom),
141.6 (PhCH]CH), 143.9 (Cquat arom), 163.1 (C]O acry-
late), 168.8 (C]O), 169.8 (C]O), 170.0 (C]O).
The desired product 15b (75.7 mg, 0.132 mmol, 66%) was
obtained as a mixture of diastereoisomers in the form
of a white solid. TLC: Rf 0.65 (EtOAc/hexanes 1:1), mp
137.5–141.6 ꢀC. MS (ESP+) found m/z¼575 ([M+H]+),
C35H30N2O6 requires 574.62. HPLC: retention time¼
19.4 (min). 1H NMR 250 MHz (CDCl3): dH¼1.11 and
1.22 (3H, 2t, HCH , J1¼3J2¼7.5 Hz), 2.55 and 3.21 (1H,
3
3
2
2
0
0
2dd, Hb, J1(Hb,Hb )¼14.0 Hz, J2(Hb,Hb )¼12.0 Hz,
3J1(Hb,Ha)¼3.6 Hz, J2(Hb,Ha)¼4.6 Hz), 3.65 (1H, d, Ha
3
2
0
Gly, J(Ha,Ha )¼17.5 Hz), 3.83–3.85 (4H, 2s+pseudo t,
0
OMe+Hb ), 4.20 (2H, m, HCH ), 4.87 and 5.37 (1H, 2dd,
2
Ha
Aba,
0
3J1(Ha,Hb)¼3.6 Hz,
3J2(Ha,Hb)¼5.0 Hz,
3
3
0
4.4. Suzuki couplings on compound 9
J1(Ha,Hb )¼14.0 Hz, J2(Ha,Hb )¼12.0 Hz), 4.88 and 5.13
2
0
0
(1H, 2d, Ha Gly, J(Ha –Ha)¼17.0 Hz), 5.68 and 5.74
(1H, 2s, H3), 6.95 (2H, m, H arom), 7.20–7.75 (14H, M, H
arom). 13C NMR 63 MHz (CDCl3): dC¼13.8 and 14.1
(CH3 Et), 34.3 and 33.3 (Cb), 52.1 and 52.7 (Ca Gly), 53.6
(Ca Aba), 55.4 (OCH3), 61.4 (CH2 Et), 68.8 and 69.9 (C3),
114.2 (CH arom), 123.3 (CH arom), 126.8 (CH arom),
127.6 (CH arom), 128.0 (CH arom), 128.2 (CH arom),
129.7 (CH arom), 129.9 (CH arom), 130.4 (CH arom),
132.8 (Cquat arom), 133.9 (Cquat arom), 134.2 (CH arom),
137.0 (Cquat arom), 137.4 (Cquat arom), 138.8 (Cquat arom),
139.9 (Cquat arom), 159.0 (Cquat arom), 169.1 (C]O),
169.9 (C]O), 170.1 (C]O).
4.4.1. Synthesis of ethyl [(4S)-4-(1,3-dioxo-1,3-dihydro-
2H-isoindol-2-yl)-1(R,S)-(40-fluoro-1,10-biphenyl-4-yl)-3-
oxo-1,3,4,5-tetrahydro-2H-2-benzazepin-2-yl]acetate
(15a, entry 2, Table 3). An oven dried 25 mL round-
bottomed flask was charged with Pd(OAc)2 (7.3 mg,
0.033 mmol, 16.65 mol %) and 2-(di-tert-butylphosphino)-
biphenyl (20.0 mg, 0.067 mmol, 33.5 mol %). The flask
was flushed with N2 and subsequently DME (10 mL) was
added. This solution was stirred for 10 min at rt. During
these 10 min 9 (109.5 mg, 0.20 mmol), 4-fluorophenylbor-
onic acid (42.0 mg, 0.30 mmol, 1.5 equiv), and KF
(34.9 mg, 0.6 mmol, 3 equiv) were weighed into another
25 mL flask. This flask was also flushed with N2 and 3 mL
of the Pd(0)/2-(di-tert-butylphosphino)biphenyl solution
(0.010 mmol Pd(OAc)2 and 0.020 mmol 2-(di-tert-butyl-
phosphino)biphenyl) was added. The reaction mixture was
heated (oil bath temperature: 95 ꢀC) and magnetically stirred
for 24 h under N2 atmosphere. After evaporation of the sol-
vent, the crude product was dissolved in a minimum amount
of acetonitrile/water (1.2:1), filtered, and purified by
preparative HPLC. The desired product 15a (79.8 mg,
0.142 mmol, 71%) was obtained as a mixture of diastereo-
isomers in the form of a white solid. TLC: Rf 0.75 (EtOAc/
hexanes 1:1), mp 152.2–154.3 ꢀC. MS (ESP+) found
m/z¼563 ([M+H]+), C34H27FN2O5 requires 562.59. HPLC:
4.4.3. Synthesis of ethyl [(4S)-4-(1,3-dioxo-1,3-dihydro-
2H-isoindol-2-yl)-3-oxo-1(R,S)-(4-thien-3-ylphenyl)-
1,3,4,5-tetrahydro-2H-2-benzazepin-2-yl]acetate (15c,
entry 6, Table 3). An oven dried 25 mL round-bottomed
flask was charged with Pd(OAc)2 (7.3 mg, 0.033 mmol,
16.65 mol %)
and
2-(di-tert-butylphosphino)biphenyl
(20.0 mg, 0.067 mmol, 33.5 mol %). The flask was flushed
with N2 and subsequently DME (10 mL) was added. This
solution was stirred for 10 min at rt. During these 10 min
9
(109.5 mg, 0.20 mmol), 3-thiophene boronic acid
(90.0 mg, 0.70 mmol, 3.5 equiv), and KF (34.9 mg,
0.6 mmol, 3 equiv) were weighed into another 25 mL flask.
This flask was also flushed with N2 and 3 mL of the Pd(0)/
ligand solution (0.010 mmol Pd(OAc)2 and 0.020 mmol 2-
(di-tert-butylphosphino)biphenyl) was added. The reaction
mixture was heated (oil bath temperature: 95 ꢀC) and mag-
netically stirred for 24 h under N2 atmosphere. After evapo-
ration of the solvent, the crude product was dissolved in
a minimum amount of acetonitrile/water (1.2:1), filtered,
and purified by preparative HPLC. The desired product 15c
(70.5 mg, 0.128 mmol, 64%) was obtained as a mixture of di-
astereoisomers in the form of a white solid. TLC: Rf 0.47
(EtOAc/hexanes 1:1), mp 140.2–142.0 ꢀC. MS (ESP+) found
m/z¼551 ([M+H]+), C32H26N2O5S requires 550.63. HPLC:
1
retention time¼19.6 and 19.7 (min), H NMR 250 MHz
(CDCl3): dH¼1.12 and 1.22 (3H, 2t, HCH , 3J1¼3J2¼7.3 Hz),
3
2
0
2.54 and 3.21 (1H, 2dd, Hb, J1(Hb,Hb )¼13.8 Hz,
J2(Hb,Hb )¼18.0 Hz, 3J1(Hb,Ha)¼3.0 Hz, 3J2(Hb,Ha)¼
2
0
2
0
4.4 Hz), 3.65 (1H, d, Ha Gly, J(Ha,Ha )¼17.0 Hz), 3.83
2
3
0
0
0
(1H, pseudo t, Hb , J(Hb ,Hb)z J(Hb ,Ha)¼13.5 Hz), 4.18
3
(2H, m, HCH ), 4.88 (1H, dd, Ha Aba, J(Ha,Hb)¼3.0 Hz,
2
3
2
0
0
0
J(Ha,Hb )¼13.2 Hz), 5.13 (1H, d, Ha Gly, J(Ha ,Ha)¼
0
17.0 Hz), 5.32 (0.3H, m, Ha ) 5.69 and 5.72 (1H, 2s, H3),
7.16–7.76 (16H, M, H arom). 13C NMR 63 MHz (CDCl3):
dC¼14.0 and 14.2 (CH3 Et), 34.3 (Cb), 52.7 (Ca Gly),
53.6 (Ca Aba), 61.4 (CH2 Et), 69.8 and 71.1 (C3), 116.1
(CH arom), 116.2 (CH arom), 123.2 (CH arom), 127.1 (CH
arom), 128.5 (CH arom), 129.5 (CH arom), 129.6 (CH
arom), 130.0 (CH arom), 130.3 (Cquat arom), 130.4 (Cquat
arom), 131.0 (CH arom), 134.1 (Cquat arom), 137.4
(Cquat arom), 138.6 (Cquat arom), 139.0 (Cquat arom), 139.7
(Cquat arom), 169.7 (C]O), 169.8 (C]O), 170.1 (C]O).
1
retention time¼19.2 and 19.4 (min). H NMR 250 MHz
(CDCl3): dH¼1.10 and 1.22 (3H, 2t, HCH , 3J1¼3J2¼6.5 Hz),
3
2
0
2.54 and 3.20 (1H, 2dd, Hb, J1(Hb,Hb )¼14.0 Hz,
J2(Hb,Hb )¼15.2 Hz, 3J1(Hb,Ha)¼3.5 Hz, 3J2(Hb,Ha)¼
2
0
2
0
4.0 Hz), 3.65 (1H, d, Ha Gly, J(Ha,Ha )¼17.3 Hz), 3.82
2
3
0
0
0
(1H, pseudo t, Hb , J(Hb ,Hb)z J(Hb ,Ha)¼13.0 Hz), 4.18
(2H, m, HCH ), 4.87 and 5.36 (1H, 2dd, Ha Aba,
2
3J1(Ha,Hb)¼3.5 Hz, 3J2(Ha,Hb)¼4.5 Hz,
J1(Ha,Hb )¼
3
0
3
0
4.4.2. Synthesis of ethyl [(4S)-4-(1,3-dioxo-1,3-dihydro-
2H-isoindol-2-yl)-1(R,S)-(40-methoxy-1,10-biphenyl-4-
yl)-3-oxo-1,3,4,5-tetrahydro-2H-2-benzazepin-2-yl]ace-
tate (15b, entry 3, Table 3). Identical reaction conditions as
for the synthesis of 15a were used.
14.0 Hz, J2(Ha,Hb )¼12.0 Hz), 5.65 and 5.72 (1H, 2s, H3),
7.20 (2H, m, H arom), 7.26–7.70 (11H, M, H arom), 7.72
(2H, m, H arom). 13C NMR 63 MHz (CDCl3): dC¼13.8
and 14.1 (CH3 Et), 34.3 and 33.3 (Cb), 52.2 and 52.7 (Ca
Gly), 53.6 (Ca Aba), 61.4 (CH2 Et), 68.8 and 69.9 (C3),

3726
S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
120.5 (CH arom), 123.3 (CH arom), 126.2 (CH arom), 126.3
(CH arom), 126.6 (CH arom), 127.0 (CH arom), 127.6 (CH
arom), 129.6 (Cquat arom), 129.7 (CH arom), 129.9 (CH
arom), 130.4 (CH arom), 134.1 (CH arom), 134.9 (Cquat
arom), 136.9 (Cquat arom), 137.9 (Cquat arom), 138.8 (Cquat
arom), 142.0 (Cquat arom), 169.1 (C]O), 169.8 (C]O),
170.1 (C]O).
solution was refluxed over night. The reaction mixture was
cooled down to room temperature after which H2O (20 mL)
was added to quench the reaction. Washing the organic layer
with brine (20 mL), drying over MgSO4, and evaporation of
the solvent gave the crude product, which was crystallized in
EtOH. The desired compound 3 (0.067 g, 0.15 mmol, 28%)
was obtained. TLC: Rf 0.30 (EtOAc/hexanes 1:1), mp 300–
302 ꢀC. MS (ESP+) found m/z¼461 and 463 ([M+H]+),
C24H17BrN2O3 requires 461.31. HPLC: retention time¼16.2
4.5. Suzuki coupling of compound 11
1
(min). H NMR 250 MHz (DMSO-d6): dH¼2.74 (1H, dd,
2
4.5.1. Synthesis of 2-[(1R,4S)-1-(40-fluoro-1,10-biphenyl-
4-yl)-3-oxo-2,3,4,5-tetrahydro-1H-2-benzazepin-4-yl]-
1H-isoindole-1,3(2H)-dione 16 (Scheme 6). An oven dried
25 mL round-bottomed flask was charged with Pd(OAc)2
(7.3 mg, 0.033 mmol, 16.65 mol %) and 2-(di-tert-butyl-
phosphino)biphenyl (20.0 mg, 0.067 mmol, 33.5 mol %).
The flask was flushed with N2 and subsequently DME
(5 mL) and DMA (5 mL) were added. This solution was
stirred for 10 min at rt. During these 10 min 11 (92.0 mg,
0.20 mmol), 4-fluorophenylboronic acid (42.0 mg, 0.30
mmol, 1.5 equiv), and KF (34.9 mg, 0.6 mmol, 3 equiv)
were weighed into another 25 mL flask. This flask was
also flushed with N2 and 3 mL of the Pd(0)/ligand solution
(0.010 mmol Pd(OAc)2 and 0.020 mmol 2-(di-tert-butyl-
phosphino)biphenyl) was added. The reaction mixture was
heated (oil bath temperature: 95 ꢀC) and magnetically stirred
for 24 h under N2 atmosphere. After evaporation of the sol-
vent, the crude product was dissolved in a minimum amount
of acetonitrile/water (1.2:1), filtered, and purified by prepar-
ative HPLC. The desired product 16 (66.7 mg, 0.14 mmol,
70%) was obtained as a white solid. TLC: Rf 0.45 (EtOAc/
hexanes 1:1), mp 179.2–182.1 ꢀC. MS (ESP+) found m/z¼
477 ([M+H]+), C30H21FN2O3 requires 476.50. HPLC: re-
tention time¼17.3 (min). 1H NMR 250 MHz (CDCl3):
Hb, J(Hb,Hb )¼14.0 Hz, 3J(Hb,Ha)¼3.0 Hz), 3.61 (1H,
0
2
3
0
0
0
pseudo t, Hb , J(Hb ,Hb)z J(Hb ,Ha) J¼13.0 Hz), 4.84
3
3
0
(1H, dd, Ha Aba, J(Ha,Hb)¼3.0 Hz, J(Ha,Hb )¼12.8 Hz),
3
5.65 (1H, d, H3, J(H3,NH)¼7.0 Hz), 7.15–7.55 (8H, M H
arom), 7.70–7.75 (4H, M, H Pht), 8.93 (1H, d, NH,
3J(NH,H3)¼7.0 Hz). 13C NMR 63 MHz (DMSO-d6):
dC¼34.5 (Cb), 52.8 (Ca Aba), 61.7 (C3), 115.6 (CH arom),
120.5 (Cq arom), 123.9 (CH arom), 126.3 (CH arom),
128.1 (CH arom), 128.6 (CH arom), 129.9 (CH arom), 130.1
(CH arom), 131.9 (CH arom), 134.6 (Cquat arom), 136.7
(Cquat arom), 140.5 (Cquat arom), 142.5 (Cquat arom), 171.8
(C]O), 183.8 (C]O).
4.6.2. Synthesis of 2-[(1R)-8-(4-fluorophenyl)-3-oxo-1-
phenyl-2,3,4,5-tetrahydro-1H-2-benzazepin-4-yl]-1H-
isoindole-1,3(2H)-dione 17 (Scheme 7). Identical reaction
conditions as for the synthesis of 16 were used.
The desired product 17 (64.8 mg, 0.136 mmol, 68%) was
obtained as a white solid. TLC: Rf 0.58 (EtOAc/hexanes
1:1), mp 244.0–246.6 ꢀC. MS (ESP+) found m/z¼477
([M+H]+), C30H21FN2O3 requires 476.50. HPLC: retention
1
time¼17.4 (min). H NMR 250 MHz (CDCl3): dH¼2.67
2
(1H, dd, Hb, J(Hb–Hb )¼14.4 Hz, 3J(Hb–Ha)¼3.3 Hz),
0
2
3
2
3
0
2
0
0
0
dH¼2.66 (1H, 2dd, Hb, J(Hb,Hb )¼14.6 Hz, J(Hb,Ha)¼
3.74 (1H, pseudo t, Hb , J(Hb ,Hb)z J(Hb ,Ha), J¼
3
3.0 Hz), 3.76 (1H, pseudo t, Hb , J(Hb ,Hb)z J(Hb ,Ha)¼
13.3 Hz), 4.90 (1H, dd, Ha Aba, 3J(Ha,Hb)¼3.2 Hz,
0
0
0
13.3 Hz), 4.89 (1H, dd, Ha Aba, 3J(Ha,Hb)¼3.0 Hz,
J(Ha,Hb )¼13.0 Hz), 5.68 (1H, d, H3, J(H3,NH)¼7.0 Hz),
3
3
0
3
3
0
J(Ha,Hb )¼13.3 Hz), 5.65 (1H, d, H3, J(H3,NH)¼7.3 Hz),
7.20–7.65 (12H, M, H arom), 7.75 (2H, M, H Pht), 7.85 (2H,
M, H Pht). 13C NMR 63 MHz (CDCl3): dC¼33.9 (Cb), 52.4
(Ca Aba), 61.9 (C3), 115.6 (CH arom), 115.7 (CH arom),
123.6 (CH arom), 126.5 (CH arom), 127.8 (CH arom),
128.6 (CH arom), 128.7 (CH arom), 128.8 (CH arom), 129.6
(Cquat arom), 131.2 (CH arom), 132.0 (CH arom), 134.1
(Cquat arom), 135.5 (Cquat arom), 139.5 (Cquat arom), 139.6
(Cquat arom), 140.8 (Cquat arom), 160.7 (Cquat arom), 168.2
(C]O), 171.2 (C]O).
7.10–7.75 (17H, M, H arom). 13C NMR 63 MHz (CDCl3):
dC¼34.3 (Cb), 52.4 (Ca Aba), 61.6 (C3), 115.5 (CH arom),
115.8 (CH arom), 123.5 (CH arom), 127.0 (CH arom),
127.8 (CH arom), 128.6 (CH arom), 128.7 (CH arom), 129.3
(CH arom), 129.6 (Cquat arom), 130.7 (CH arom), 132.0 (CH
arom), 134.1 (Cquat arom), 136.5 (Cquat arom), 138.9 (Cquat
arom), 139.3 (Cquat arom), 140.0 (Cquat arom), 160.7 (Cquat
arom), 171.2 (C]O), 183.8 (C]O).
4.6. Synthesis of/and Suzuki coupling of compound 3
4.7. N-Arylation of 19 with 4-iodobenzonitrile
4.6.1. Synthesis of 2-(8-bromo-3-oxo-1-phenyl-2,3,4,5-
tetrahydro-1H-benzo[c]azepin-4-yl)-isoindole-1,3-dione
17 (Scheme 7). N-Phthaloyl-pBr-(S)-phenylalanine amide
(1 g, 2.75 mmol), benzotriazole (1 equiv, 0.306 g,
2.75 mmol), and p-TsOH (0.1 equiv, 0.056 g, 0.275 mmol)
were added to an oven dried 100 mL flask and dissolved in
dry benzene (35 mL). After addition of benzaldehyde
(1 equiv, 0.261 mg, 2.75 mmol), the flask, mounted by
a Dean–Stark apparatus, was heated by means of an oil
bath (115 ꢀC) and the solution was refluxed over night. After
evaporation of the solvent, trituration with Et2O gave com-
pound 6 quantitatively. The benzotriazole adduct 6 (0.300 g,
0.52 mmol) was dissolved in dry CH2Cl2 (40 mL). Then,
AlCl3 (0.420 g, 3.1 mmol, 6 equiv) was added and the
4.7.1. Synthesis of 4-[(4S)-4-(1,3-dioxo-1,3-dihydro-2H-
isoindol-2-yl)-3-oxo-1,3,4,5-tetrahydro-2H-2-benzaze-
pin-2-yl]benzonitrile 20 (Scheme 9). An oven dried 25 mL
round-bottomed flask was charged with Pd2dba3 (9.2 mg,
0.010 mmol, 5 mol %) and XANTPHOS (12.8 mg,
0.022 mmol, 11 mol %). The flask was flushed with Ar and
subsequently dioxane (6 mL) was added. This solution was
stirred for 10 min at rt. During these 10 min 19 (61.0 mg,
0.20 mmol), 4-iodobenzonitrile (55.0 mg, 0.24 mmol,
1.2 equiv), and Cs2CO3 (261 mg, 0.8 mmol, 4 equiv) were
weighed into another 25 mL flask. This flask was also
flushed with N2 and 3 mL of the Pd(0)/XANTPHOS solution
(0.005 mmol Pd2dba3 and 0.011 mmol XANTPHOS) was
added. The reaction mixture was heated (oil bath

S. Ballet et al. / Tetrahedron 63 (2007) 3718–3727
3727
temperature: 100 ꢀC) and magnetically stirred for 24 h under
Ar atmosphere. After evaporation of the solvent, the crude
product was dissolved in a minimum amount of acetoni-
trile/water (1.2:1), filtered, and purified by preparative
HPLC. The desired product (53.7 mg, 0.132 mmol, 66%)
was obtained as a white solid. TLC: Rf 0.78 (EtOAc/hexanes
1:1), mp 113.5–115.4 ꢀC. MS (ESP+) found m/z¼408
([M+H]+), C25H17N3O3 requires 407.42. HPLC: retention
M. F.; Southhal, L. S.; Uzinskas, I.; Vasko-Moser, J. A.;
Venslavsky, J. W.; Wong, A. S.; Huffman, W. F. J. Med.
Chem. 1999, 42, 545–559.
5. Sawa, Y.; Takeshi, K.; Toru, M.; Mikio, H.; Hajime, F. Chem.
Pharm. Bull. 1975, 23, 1917–1927.
ꢀ
6. Van Rompaey, K.; Van den Eynde, I.; De Kimpe, N.; Tourwe,
D. Tetrahedron 2003, 59, 4421–4432.
7. Van den Eynde, I.; Van Rompaey, K.; Lazzaro, F.; Tourwe, D.
J. Comb. Chem. 2004, 6, 468–473.
ꢀ
1
time¼15.1 (min). H NMR 250 MHz (CDCl3): dH¼3.27
2
(1H, dd, Hb, J(Hb,Hb )¼15.9 Hz, 3J(Hb,Ha)¼4.6 Hz),
ꢀ
8. Tourwe, D.; Verschueren, K.; Frycia, A.; Davis, P.; Porreca, F.;
Hruby, V. J.; Toth, G.; Jaspers, H.; Verheyden, P.; Van Binst, G.
0
2
3
0
0
0
4.31 (1H, dd, Hb , J(Hb ,Hb)¼16.4 Hz, J(Hb ,Ha)¼
2
0
Biopolymers 1995, 38, 1–12.
ꢀ
13.0 Hz), 4.98 (1H, d, Ha Gly, J(Ha–Ha )¼16.0 Hz), 5.12
2
0
0
9. Casimir, J. R.; Tourwe, D.; Iterbeke, K.; Guichard, G.; Briand,
J.-P. J. Org. Chem. 2000, 65, 6487–6492.
10. Ballet, S.; Frycia, A.; Piron, J.; Chung, N. N.; Schiller, P. W.;
(1H, d, Ha Gly J(Ha –Ha)¼16.1 Hz), 5.59 (1H, dd, Ha
Aba, 3J(Ha,Hb)¼4.9 Hz, J(Ha,Hb )¼13.1 Hz), 7.22–7.41
3
0
(6H, M, H arom), 7.65 (2H, d, H arom (2CHCCN),
3J¼8.7 Hz), 7.76 (2H, M, H Pht), 7.91 (2H, M, H Pht).
13C NMR 63 MHz (CDCl3): dC¼35.4 (Cb), 50.6 (C3), 52.8
(Ca Aba), 110.1 (Cquat arom), 116.5 (Cquat CN), 122.2 (CH
arom), 128.0 (CH arom), 128.4 (CH arom), 129.0 (CH
arom), 129.9 (CH arom), 130.1 (CH arom), 131.7 (Cquat
arom), 132.0 (CH arom), 132.2 (CH arom), 134.1 (Cquat
arom), 137.2 (Cquat arom), 144.8 (Cquat arom), 167.8
(C]O), 174.4 (C]O).
ꢀ
Kosson, P.; Lipkowski, A. W.; Tourwe, D. J. Pept. Res. 2005,
66, 222–230.
11. Ballet, S.; Urbanczyk-Lipkowska, Z.; Tourwe, D. Synlett 2005,
ꢀ
2791–2795.
12. Katritzky, A. R.; Lan, X.; Zhang, Z. J. Heterocycl. Chem. 1993,
30, 381–387.
13. Loones, K. T. J.; Maes, B. U. W.; Meyers, C.; Deruytter, J.
J. Org. Chem. 2006, 71, 260–264; For a recent paper providing
new insights in Xantphos/Pd-catalyzed C–N bond formation,
see: Klingensmith, L. M.; Strieter, E. R.; Barder, T. E.;
Buchwald, S. L. Organometallics 2006, 25, 82–91.
14. Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 2000, 65, 1144–
1157.
Acknowledgements
This work was supported by the Institute for the Promotion
of Innovation by Science and Technology in Flanders (IWT,
Belgium) and the Fund for Scientific Research-Flanders
(FWO-Belgium). We would like to thank Dr. G. Mignani
of Rhodia for rac. BINAP.
15. Wolfe, J. P.; Tomori, H.; Sadighi, J. P.; Yin, J.; Buchwald, S. L.
J. Org. Chem. 2000, 65, 1158–1174.
16. Barder, T. E.; Walker, S. D.; Martinelli, J. R.; Buchwald, S. L.
J. Am. Chem. Soc. 2005, 127, 4685–4696.
17. Anderson, K. W.; Tundel, R. E.; Ikawa, T.; Altman, R. A.;
Buchwald, S. L. Angew. Chem., Int. Ed. 2006, 45, 6523–
6527.
References and notes
18. Littke, A. F.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 6989–
7000.
19. Yang, C.; Nolan, S. P. Synlett 2001, 1539–1542.
20. Wang, Z.; Elokdah, H.; McFarlene, G.; Pan, S.; Antane, M.
Tetrahedron Lett. 2006, 47, 3365–3369.
21. Battace, A.; Zair, T.; Doucet, H.; Santelli, M. Tetrahedron Lett.
2006, 47, 459–462.
22. Wolfe, J. P.; Singer, R. A.; Yang, B. H.; Buchwald, S. L. J. Am.
Chem. Soc. 1999, 121, 9550–9561.
1. Flynn, G. A.; Giroux, E. L.; Dage, R. C. J. Am. Chem. Soc.
1987, 109, 7914–7915.
2. Warchawsky, A. M.; Flynn, G. A.; Koehl, J. R.; Mehdi, S.; Vaz,
R. J. Bioorg. Med. Chem. Lett. 1996, 6, 957–962.
3. Le Diguarher, T.; Ortono, J.-C.; Shanks, D.; Guilbaud, N.;
ꢀ
ꢁ
Pierre, A.; Raimbaud, E.; Fauchere, J.-L.; Hickman, J. A.;
Tucker, G. C.; Casara, P. J. Bioorg. Med. Chem. Lett. 2004,
14, 767–771.
4. Keenan, R. M.; Callahan, J. F.; Samanen, J. M.; Bondinell,
W. E.; Calvo, R. R.; Chen, L.; DeBrosse, C.; Eggleston,
D. S.; Haltiwanger, R. C.; Hwang, S.-M.; Jakas, D. R.; Ku,
T. W.; Miller, W. H.; Newlander, K. A.; Nichols, A.; Parker,
23. Clapars, A.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc.
2002, 124, 7421–7428.
24. Yin, J.; Buchwald, S. L. Org. Lett. 2000, 2, 1101–1104.
25. Hartwig, J. F. Synlett 2006, 1283–1294.