Synthesis of 2,3,4,7-tetrahydro-1H-azepines as privileged ligand scaffolds for the design of aspartic protease inhibitors via a ring-closing metathesis approach

Article abstract of DOI:10.1016/j.jorganchem.2006.09.031

We have developed a short and highly efficient synthetic strategy towards the hitherto hardly known 3,5- and 3,6-disubstituted 2,3,4,7-tetrahydro-1H-azepine scaffold via a ring-closing metathesis approach utilizing inexpensive and readily available starti

Full text of DOI:10.1016/j.jorganchem.2006.09.031

Journal of Organometallic Chemistry 691 (2006) 5406–5422
www.elsevier.com/locate/jorganchem
Synthesis of 2,3,4,7-tetrahydro-1H-azepines as privileged
ligand scaffolds for the design of aspartic protease inhibitors via a
ring-closing metathesis approach
Sascha Brass, Nan-Si Chan, Christof Gerlach, Torsten Luksch,
*
Jark Bo¨ttcher, Wibke E. Diederich
Institut fu¨r Pharmazeutische Chemie, Fachbereich Pharmazie, Philipps-Universita¨t Marburg, Marbacher Weg 6, D-35037 Marburg, Germany
Received 3 July 2006; received in revised form 15 September 2006; accepted 15 September 2006
Available online 23 September 2006
Abstract
We have developed a short and highly efficient synthetic strategy towards the hitherto hardly known 3,5- and 3,6-disubstituted 2,3,4,7-
tetrahydro-1H-azepine scaffold via a ring-closing metathesis approach utilizing inexpensive and readily available starting material such as
methyl acrylate and allylamine. Both seven-membered azacycle scaffolds bearing suitable functional groups, which can easily be modified
by means of standard synthetic chemistry, serve as non-peptidic heterocyclic core structures for the further design and synthesis of aspar-
tic protease inhibitors. Through specific decoration with appropriate side chains, individual inhibitors can be tailored with respect to
selectivity towards particular family members. A first generation of this class of non-peptidic inhibitors have been tested against the
aspartic proteases Plasmepsin II and HIV-I protease, respectively, showing promising activity as well as selectivity with IC50 values
in the micromolar range.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Azacycles; RCM; Aspartic proteases; Privileged structures; Azepines
1. Introduction
ous substituents [3], we recently engaged in ring-closing
metathesis reactions in continuation of our efforts on the
synthesis of biologically active nitrogen-containing hetero-
cycles from a medicinal chemist’s point of view [4].
Throughout the last two decades, ring-closing metathe-
sis (RCM) has become one of the most important tools
in organic chemistry and has therefore been reviewed
extensively [1]. This elegant approach allows the conve-
nient preparation of a plethora of new molecules and
‘‘soon imagination will be the only limit to what molecules
can be built’’ as it was articulated in the press release
regarding the award of the Nobel Price in Chemistry
2005 by The Royal Swedish Academy of Science [2].
At the beginning of the 21st century, Malaria and HIV
infections are still among the most serious global health
problems causing more than four million deaths per year.
Infections by Plasmodium falciparum, the most dangerous
form of malaria, are responsible or contribute to an esti-
mate of more than two million deaths yearly. Among them,
particularly children below an age of five are affected [5].
Infections with the HI-Virus, which inevitably lead to the
development of AIDS, claimed the lives of some 2.2 million
people alone in sub-Saharan Africa in 2003 [6].
Since RCM has also proven to be a precious tool for the
synthesis of five- and six-membered azacycles as well as
other nitrogen-containing fused ring systems bearing vari-
In the patho-physiological processes of both pandemics,
aspartic proteases are involved in crucial pathways. During
its life cycle, P. falciparum degrades the hemoglobin of the
host cell to cover its energy and nutrition demands. It has
*
Corresponding author. Tel.: +49 6421 2825810; fax: +49 6421
2828994.
E-mail address: diederic@staff.uni-marburg.de (W.E. Diederich).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.09.031

S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
5407
a
b
S2
P2
S1'
S3'
O
O
O
H
H
N
N
OH
N
N
H
N
O
O
P1'
O
P3'
H
H
H
N
H
OH
O
OH
O
O
N
N
H
N
N
H
H
P3
O
P1
O
P2'
S2'
O
O
-
O
S1
O
S3
Asp 214
Asp 34
Fig. 1. (a) Nomenclature for substrate residues and their corresponding binding sites. (b) Pepstatine in the active site of Plasmepsin II.
been shown that the inhibition of this metabolic pathway is
lethal for the parasite [7]. The aspartic proteases Plasmep-
sin I, II, and IV, the fundamental hemoglobin degrading
enzymes of the parasite, are therefore believed to be attrac-
tive targets for the development of an anti-malarial drug
therapy [8]. In case of the HI-Virus, HIV protease has pro-
ven to be an invaluable drug target due to its essential role
in the virus’ replication process. However, the continuously
increasing resistance towards existing drugs calls for the
development of novel anti-infectives favorably possessing
a new mechanism of action which is pivotal to combat
these afflictions of mankind [9].
Most inhibitors that have been developed so far for the
class of aspartic proteases are transition state analogues
such as statine and norstatine derivatives addressing via a
hydroxyl or hydroxyl-like moiety the two aspartate resi-
dues present in the enzyme’s active site and thus replacing
the native substrate in the peptide’s cleavage site as shown
in Fig. 1 [10]. The catalytic mechanism of aspartic prote-
ases is virtually the same, however, differences in specificity
result from structural evolution of the sub-pockets (e.g. S3–
S30) for substrate side chain binding. In general, aspartic
proteases recognize 6–10 amino acids of their polypeptide
substrate (e.g. P3–P30) very specifically.
Recently, substituted secondary amines, in particular
substituted pyrrolidines, have proven to be micromolar
inhibitors of the HIV-1 protease [11,12]. These results
prompted us to investigate in greater detail the suitability
of five- as well as seven-membered azacycles as privileged
ligand scaffolds designed to address the key interactions
of the conserved recognition pattern of the aspartic prote-
ase family. In this article, however, we would like to focus
primarily on the metal-catalyzed synthesis and biological
evaluation of substituted 2,3,4,7-tetrahydro-1H-azepines
utilizing this seven-membered azacycle scaffold as non-pep-
tidic heterocyclic core structure for the further design and
synthesis of selective aspartic protease inhibitors.
groups for the later introduction of various substituents
in the positions 3 and 5 and/or 6, respectively, would
be the most promising starting point. As depicted in
Fig. 2, the azacycle ring system addresses via its basic nitro-
gen the conserved catalytic dyad. Through specific decora-
tion with appropriate side chains, individual inhibitors can
be tailored with respect to selectivity towards particular
family members.
The main synthetic challenge associated with this
approach of using a privileged ligand scaffold was the
development of a synthetic strategy towards an advanced
intermediate for each of the azepine core-structures 1–3.
They should bear at least two suitable anchoring groups
which enable us to introduce a variety of different side
chains at a later stage of the synthesis by means of standard
synthetic chemistry as outlined in Scheme 1. Furthermore,
we envisaged that the convergent syntheses of the desired
core structures should preferably comprise the same syn-
thetic intermediates accessible through mainly inexpensive
and readily available starting materials. We sought to syn-
thesize our azepine-scaffolds 1–3 via an RCM approach.
The RCM precursors of type 4 should be obtainable via
alkylation of an appropriately functionalized allylamine
derivative of type 5 with bromoallyl derivative 6, each of
them readily available through short synthetic sequences
commencing with methyl acrylate 7.
S2'
S2
R2
R1
R3
S1'
S1
+
N
H
H
2. Results and discussion
O
O
2.1. Synthesis
-
O
O
Asp 34
Asp 214
Preliminary modelling studies indicated that the 2,3,4,7-
tetrahydro-1H-azepine scaffold bearing suitable anchoring
Fig. 2. Putative schematic binding mode of the azepine core structure in
the active site of Plasmepsin II.

5408
S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
O
PG
PG
N
PG
N
Br
a,b,c,d
OMe
N
1
OTBS
3
R3
R3
R1
R1
R1
5
7
9
R2
R2
a,g
1
2
3
e, f
RCM
Br
PG
N
R3
OMe
BOC
N
O
13
R1
MeO
R2
4
O
10
alkylation
R2
h
PG
N
i
+
BOC
N
BOC
N
Br
R1
R3
5
6
MeO
MeO
2 steps
4 steps
O
O
OTBS
OMe
O
O
11
14
OMe
7
j
k
for PG: BOC,
R1: CO2Me
R2: H, CH2OH, CH2OTBS, CO2Me
R3: H, CH2OH, CH2OTBS
BOC
N
BOC
N
Scheme 1. Retrosynthetic analysis of the target structures 1–3.
MeO
MeO
O
O
OTBS
OMe
For the key step of our synthetic approach only the
commercially available Grubbs catalyst 2nd generation
was employed (see Fig. 3).
O
12
15
The N-protected 2,3,4,7-tetrahydro-1H-azepine scaffold
of type 12 bearing two anchoring groups, the methyl ester
moiety in position 3 as well as the TBS-protected hydroxy-
methyl functionality in position 5, could easily be synthe-
sized via a straightforward convergent strategy utilizing
inexpensive and readily available starting material such as
methyl acrylate 7 and allylamine as depicted in Scheme 2.
Baylis–Hillmann reaction of methyl acrylate 7 with formal-
dehyde afforded the a-hydroxymethylated ester. Conver-
Scheme 2. Synthesis of 12 and 15. Reagents and conditions: (a) H2CO,
DABCO, dioxane/H2O, 15 h, rt, 68%; (b) TBSCl, TEA, DMAP, DCM,
0 ꢁC to rt, 15 h, 91%; (c) DIBAL-H, THF, ꢀ78 to 0 ꢁC, 2 h, 80%; (d) TPP/
Br2, imidazole, DCM, 0 ꢁC, 30 min, 84%; (e) allylamine, MeOH, 40 ꢁC,
4 h, 75%; (f) (BOC)2O, TEA, cat. DMAP, DCM, rt, 14 h, 93%; (g) TPP/
Br2, imidazole, DCM, 0 ꢁC, 20 min, 68%; (h) 13, LDA, DMPU, THF,
ꢀ40 ꢁC, 4 h, 50%; (i) 9, LDA, HMPA, THF, ꢀ40 ꢁC, 5 h, 84%; (j) Grubbs
2nd generation catalyst 5 mol%, DCM, 40 ꢁC, 8 h, 78%; (k) Grubbs 2nd
generation catalyst 5 mol%, DCM, 40 ꢁC, 5 h, 87% [4].
sion of the latter into its TBS-ether followed by
subsequent reduction with DIBAL-H gave rise to the
TBS-protected allylic alcohol, which then was transferred
into the corresponding bromide 9. Addition of allylamine
to methyl acrylate 7 and subsequent protection of the
amine functionality led to the corresponding b-amino ester
10 in excellent yield. a-Substitution of ester 10 with bromo
allyl derivative 9 to form the RCM-precursor 11 turned out
to be a crucial step in this reaction sequence. Nevertheless,
Mes
Ph
Mes
N
Cl
N
Ru
Cl
PCy3
8
Fig. 3. Grubbs catalyst 2nd generation (8).

S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
5409
the use of HMPA or DMPU as carbanion-stabilizing addi-
tive and freshly prepared LDA furnished the desired a-
substituted b-amino ester 11 in good yield. Although the
usage of HMPA in general proceeds with slightly higher
yields, we mainly used the far less toxic DMPU as additive
particularly for large scale preparations. Subjecting this
diene 11 to Grubbs 2nd generation catalyst 8 readily ren-
dered the corresponding substituted azepine core structure
12 via a six step procedure in an overall yield of 30% as
reported previously [4].
The corresponding N-protected 2,3,4,7-tetrahydro-1H-
azepine 15 featuring two methyl ester functionalities in
position 3 and 5 was accessible by means of a similar reac-
tion sequence as shown in Scheme 2. The bromo allyl pre-
cursor 13, which is also commercially available, was
synthesized via a two step sequence again employing a Bay-
lis–Hillmann reaction in the first step. The resulting a-
hydroxymethylated ester was then converted into bromo
derivative 13 [13]. Subsequent reaction with b-amino ester
10 applying the above reaction conditions gave rise to diene
14 which was analogously subjected to Grubbs 2nd gener-
ation catalyst yielding 15 in moderate overall yield.
Having developed a straightforward and high yielding
synthetic route towards the 3,5-disubstituted N-protected
2,3,4,7-tetrahydro-1H-azepines of type 12 and 15, respec-
tively, we next concentrated on the reduction of the Grubbs
catalyst load, as it is known that incomplete removal of the
catalyst may not only cause side reactions or even decom-
position of the reaction products, but, if not removed by
the final stage of the synthesis, could influence the outcome
of the biological evaluation. Moreover, ruthenium species
are also considered as potentially toxic [14]. Furthermore,
it was of utmost importance to us to run this key reaction
step on a multi-gram scale as the RCM reaction turned out
to be the bottleneck in our synthetic sequence. Usually,
RCM reactions are run in high dilution to minimize forma-
tion of dimeric alkene side products arising from, e.g.
cross-metathesis reactions [15]. In contrast, we not only
intended to reduce the catalyst load but also to run the
reaction in a more concentrated fashion. Therefore, we
subjected diene 11 to Grubbs 2nd generation catalyst 8
under standardized conditions concerning reaction time,
reaction temperature as well as workup procedure
(DCM, reflux, 8 h).
The reaction was still found to proceed smoothly when
the amount of the catalyst was decreased from 5 mol%
(Table 1, entry 1) to 2.5 mol% (Table 1, entry 2). In both
cases the product was obtained in equally high yields after
column chromatography. In a next run (Table 1, entry 3)
we kept the amount of catalyst at 2.5 mol%, but now the
volume of solvent was halved resulting in about the same
concentration of catalyst as compared to the original reac-
tion conditions (Table 1, entry 1). As we did not notice any
significant change regarding the product yield obtained in
the reaction, we investigated to what extend the amount
of catalyst could be lowered while keeping the overall con-
centration of the catalyst constant (Table 1, entries 4 and
5). Even with 0.5 mol% of Grubbs catalyst the reaction
progressed equally well.
For the synthesis of the 3,6-disubstituted N-protected
azepines of type 20 we sought to alter our synthetic
sequence as little as possible but rather rely on our hitherto
developed synthetic procedures for the 3,5-disubstituted
derivatives of type 12 according to Scheme 3. Alkylation
of b-alanine methyl ester hydrochloride 16 with allylic pre-
cursor 9 followed by protection of the resulting secondary
amine with di-tert-butyl-dicarbonate gave rise to the N-
BOC-protected b-amino ester 18 in moderate overall yield.
However, attempts to increase the rather poor yield of the
alkylation step by varying the reaction conditions have up
to now been to no avail. In a different approach to generate
allylamino ester 17 via a reductive amination of 16 with
aldehyde 21, easily generated in good yield starting from
methyl acrylate 7 (Scheme 4) [16], remained likewise fruit-
less. As 16 is commercially available at low price and pre-
cursor 9 can easily be prepared in large scale, we therefore
decided to keep to our initial reaction step. Substitution of
18 with allyl bromide smoothly rendered diene 19, which,
after being subjected to Grubbs 2nd generation catalyst 8
employing the above-mentioned reaction conditions, fur-
nished the corresponding azepine 20 in a reasonable overall
yield.
Table 1
Ring-closing metathesis of 11
BOC
BOC
N
N
8
MeO
MeO
O
O
OTBS
OTBS
12
11
Entry
Amount of 8 (mol%)
Concentration of 11 (mM)
Concentration of 8 (lM)
Yield (%)
1
2
3
4
5
5
11.8
11.8
25
50
100
588
294
625
500
500
87
81
90
85
78
2.5
2.5
1
0.5

5410
S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
OTBS
9 + 18
R
N
O
a
a
OMe
H2N
MeO
* HCl
O
16
OTBS
OTBS
17: R = H
18: R = BOC
b
BOC
N
BOC
N
b
OTBS
MeO
MeO
OTBS
BOC
N
O
OTBS
O
d
c
22
23
MeO
O
c
19
OH
BOC
N
BOC
N
BOC
N
b
OH
MeO
MeO
MeO
O
O
OH
OTBS
OH
O
24
25
20
Scheme 5. Attempts to synthesize 3,5,6-trisubstituted azepines 23 and 25.
Reagents and conditions: (a) LDA, HMPA, THF, ꢀ40 ꢁC, 4 h, 52%; (b)
Grubbs 2nd generation 5 mol%, DCM, 40 ꢁC, 8 h, 0%; (c) HCl in THF, rt,
25 min, 80%.
Scheme 3. Synthesis of 20. Reagents and conditions: (a) 9, TEA, DCM, rt,
16 h, 42%; (b) (BOC)2O, TEA, DMAP, DCM, rt, 12 h, 86%; (c) allyl
bromide, LDA, HMPA, THF, ꢀ40 ꢁC, 3 h, 81%; (d) Grubbs 2nd
generation 1 mol%, DCM, 40 ꢁC, 6.5 h, 82%.
the desired sterically less demanding bis-hydroxymethyl-
ated precursor 24, but to our disappointment we even
now did not notice any transformation of 24 into 25 in
the following RCM using 5 mol% of Grubbs 2nd genera-
tion catalyst in DCM at 40 ꢁC for 8 h. Further studies
modifying the reaction conditions as well as employing
other commercially available catalysts are currently
ongoing.
Having conceived straightforward convergent synthetic
strategies towards the class of 3,5-disubsituted and 3,6-
disubstituted 2,3,4,7-tetrahydro-1H-azepines, respectively,
we then focussed on the validation of our concept of using
a privileged ligand scaffold for the design and synthesis of
selective aspartic protease inhibitors. Both non-peptidic
azepine core structures (12 and 20) are fitted with suitable
functional groups that can easily be modified by means of
standard synthetic chemistry thus allowing specific side
chain decoration as outlined in Schemes 6 and 7.
Initial modelling studies suggested different carboxylic
ester moieties such as acetyl, benzoyl, or indolylacetyl for
the side chain decoration of the azepine core structures
26, 30, 34, and 38 in order to induce favourable protein
interactions. Selective cleavage of the TBS group in 12 fur-
nishes the corresponding hydroxy methyl derivative 26,
esterification of which with suitable carboxylic acids under
standard coupling conditions followed by cleavage of the
BOC-protecting group under non-aqueous conditions gives
O
O
H
a, b
OMe
OTBS
7
21
Scheme 4. Synthesis of aldehyde 21 [16]. Reagents and conditions: (a) (i)
H2CO, DABCO, dioxane/H2O, 15 h, rt, 68%; (ii) TBSCl, TEA, DMAP,
DCM, 0 ꢁC to rt, 15 h, 91%; (iii) DIBAL-H, THF, ꢀ78 to 0 ꢁC, 2 h, 80%;
(b) DMSO, (COCl)2, DIPEA, ꢀ78 ꢁC to rt, 1.5 h, 82%.
With a reasonable route to the N-BOC-protected b-
amino ester 18 in hand, our next challenge was the synthe-
sis of 3,5,6-trisubstituted 2,3,4,7-tetrahydro-1H-azepines of
type 3, generating a tetrasubstituted double bond upon ring
closure. The formation of tetrasubstituted double bonds
via RCM employing Grubbs 2nd generation catalyst is
already documented in the literature, however, the out-
come seems to be strongly depended on the nature of the
utilized diene [17]. In our case, the required precursor 22
is readily available using our previously developed stan-
dard protocol via a-substitution of 18 with bromo allyl
derivative 9 as depicted in Scheme 5. However, subjecting
diene 22 to Grubbs 2nd generation catalyst did not result
in formation of 23, which is probably not surprising in light
of the sterically demanding TBS groups. The selective
cleavage of both TBS groups proceeded smoothly to form

S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
5411
BOC
N
BOC
N
BOC
N
d
g
MeO
HO
a
R2OCO
O
OR
OR
OR
H
N
* HCl
12: R = TBS
26: R = H
29: R = TBS
30: R = H
33: R = TBS
34: R = H
a
a
k
R2OCO
OH
b
e
h
37
l
BOC
N
BOC
N
BOC
N
BOC
N
R2OCO
MeO
R1OCO
R2OCO
OCOR2
OCOR1
OCOR3
OCOR4
O
27
31
35 a
35 b
f
m
c
i
H
N
H
N
H
N
H
N
* HCl
R1OCO
R2OCO
* HCl
* HCl
R2OCO
* HCl
MeO
OCOR2
OCOR1
OCOR4
O
OCOR3
28
32
36 a
36 b
R1 = benzoyl
R2 = indolylacetyl
R3 = 4-propylbenzoyl
R4 = acetyl
Scheme 6. Reaction pathway towards 3,5-disubstituted inhibitors of type 28, 32, 36, and 37, respectively. Reagents and conditions: (a) HCl in THF, rt, for
26: 30 min, 87%, for 30: 20 min, 87%, for 34: 40 min, 85%; (b) indole-3-acetic acid, THF/DCM, DIC, DMAP, 14 h, 87%; (c) HCl in diethyl ether, rt, 1.5 h,
89%; (d) DIBAL-H, THF, ꢀ78 to 0 ꢁC, 3 h, 76%; (e) benzoyl chloride, TEA, DMAP, DCM, 0 ꢁC to rt, 14 h, 75%; (f) HCl in diethyl ether, rt, 16 h, 75%;
(g) indole-3-acetic acid, THF/DCM, DIC, DMAP, 14 h, 82%; (h) 4-propyl-benzoyl chloride, DCM, DIPEA, DMAP, rt, 14 h, 61%; (l) acetyl chloride,
DCM, TEA, DMAP, rt, 4 h, 79%; (i) HCl in diethyl ether, rt, 3 h, 92%; (m) HCl in diethyl ether, rt, 8 h, 66%; (k) HCl in diethyl ether, rt, 1.5 h, 93%.
rise to the 3-methoxy-carbonyl azepines 28 as their hydro-
chloride salts being the first inhibitor type. Reduction of
methylester 12 with DIBAL-H to form 29 followed by
cleavage of the TBS group renders the bis-hydroxymethyl-
ated azepine 30, which, after bis-esterification and BOC-
deprotection, yields the second inhibitor type 32. The
introduction of two divergent side chains aimed to appro-
priately address the differences in the enzymes’ specificity
pockets is easily accomplished via the following synthetic
pathway: Esterification of the unprotected hydroxyl group
in 29 gives rise to 33, which, after selective deprotection of
the TBS group, renders 34. Further esterification of 34
leads to diester 35, which upon final cleavage of the
BOC-protecting group gives rise to inhibitors of type 36.
Inhibitor core structure 37, which in contrast to the diesters
28, 32, and 36 bears only one ester-functionality as well as a
modifiable hydroxymethyl substituent at position 5 is easily
accessible from azepines 33 by complete deprotection.
The corresponding 3,6-disubstituted azepines of type 40
are likewise easily accessible following the above synthetic
procedure: Cleavage of the TBS group to 38, followed by
esterification to 39 and final deprotection of the BOC-
group straightforwardly renders the azepine inhibitors of
type 40. Further functionalizations of this inhibitor type
at the methyl ester moiety are currently ongoing.
For synthetic accessibility reasons we considered ester
formation. However, we are aware of the fact that for met-
abolic stability in the further development the ester bonds
have to be replaced by some more stable structural ele-
ments. This hitherto outlined synthetic approach not only
allows the generation of various esters as just described,
but, starting from appropriate azepines with unprotected
hydroxyl substituents via Swern-oxidation followed by
reductive amination, also the preparation of a variety of
substituted amines, the synthesis of which will be reported
in due course.

5412
S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
assess whether the inhibitory effect is not only attributed
BOC
N
BOC
N
to this moiety. As reported in the literature, medium sized
aromatic or aliphatic residues are suitable moieties to
address the S1/S10-pocket of HIV-I protease [10a]. We thus
introduced two benzoate moieties in positions 3 and 5 of
our azepine core structure giving rise to 32, which exhibited
inhibitory activity and selectivity in a comparable range
within this series against HIV-I protease. These results
clearly indicate that the 3,5- as well as 3,6-disubstituted
2,3,4,7-tetrahydro-1H-azepines are suitable scaffolds for
the design and synthesis of selective aspartic protease
inhibitors, whereupon individual family members can be
addressed through specific decoration with appropriate
side chains. First structure–activity relationships derived
from these results indicate the necessity to introduce fur-
ther substituents addressing additional sub-pockets of the
enzymes’ active sites to increase binding affinity, a project
currently ongoing in our group.
b
MeO
MeO
OCOR2
OR
O
O
20: R = TBS
38: R = H
39
a
*HCl
H
N
c
MeO
O
OCOR2
R2 = indolylacetyl
40
Scheme 7. Synthesis of 3,6-disubstituted inhibitor type 40. Reagents and
conditions: (a) HCl in THF, rt, 40 min, 80%; (b) indole-3-acetic acid,
DCC, DMAP, THF/DCM, 16 h, 96%; (c) HCl in diethyl ether, 3 h, 92%.
3. Conclusion
2.2. Biological evaluation
In summary, we have developed an efficient synthetic
route towards 3,5- as well as 3,6-disubstituted azepines
via a ring-closing metathesis approach mainly utilizing
inexpensive and readily available starting materials. These
azepines have been proven to be suitable core structures
for the further design and synthesis of aspartic protease
inhibitors. Through implementation of appropriate side
chains in a first series of inhibitors, we could demonstrate
that single family members of the aspartic protease family
can be addressed specifically. Our inhibitor scaffolds bear
suitable functional groups for further structural modifica-
tions by means of standard synthetic procedures giving rise
to a variety of second generation inhibitors and thus
enables us to probe structure–activity relationships in more
detail.
We now directed our efforts on the synthesis of a first
series of inhibitors of types 28, 32, 36, 37, and 40, respec-
tively, following the above reaction schemes. In case of
HIV-I protease, inhibition constants were determined using
a competitive binding assay, wherein the initial rate of
enzymatic cleavage of the fluorogenic substrate Abz-Thr-
Ile-p-nitrophenylalanine-Phe-Gln-Arg-NH2 was measured
in presence of varying concentrations of the inhibitors
[18]. Analogously, inhibitory constants for Plasmepsin II
were determined using Haemoglobin labelled with the flu-
orogen 7-amino-4-methylcoumarin-3-acetic acid (AMCA)
as substrate [19]. IC50-values for selected inhibitors are
given in Table 2.
Out of a small set of initially designed and synthesized
inhibitors aimed to selectively address Plasmepsin II, com-
pounds decorated with a 1H-indolyl-3-acetic acid moiety
turned out to be the most active inhibitors (28, 36b, 37,
and 40) showing promising activity as well as selectivity
with IC50 values in the micromolar range against Plasmep-
sin II and HIV-I protease, respectively. Intriguingly, as
these inhibitors are just decorated with one larger substitu-
ent and therefore only capable to address one of the spec-
ificity pockets of the enzyme’s active site together with the
fact that all inhibitors are racemic mixtures, the obtained
IC50 values are most promising. Introduction of an acetyl
moiety at the hydroxymethyl function in position 5 (37 ver-
sus 36b) for instance results in a 1.5-fold decrease in
potency against Plasmepsin II. Through implementation
of a 4-propyl-benzoate moiety instead of the acetyl-substi-
tuent in 36, the specificity can easily be shifted towards
HIV-I protease with a 32-fold selectivity against Plasmep-
sin II (36a versus 36b).
4. Experimental
4.1. Kinetic assays
4.1.1. Determination of inhibition constants toward
Plasmepsin II
Plasmepsin II activity assays were performed in 96-well
microtiter plates with a Tecan Spectra Fluor spectrometer
at excitation wavelength 360 nm and emission wavelength
465 nm. Haemoglobin labelled with the fluorogen 7-
amino-4-methylcoumarin-3-acetic acid (AMCA) was used
as substrate. For IC50 determinations, inhibitors were
allowed to equilibrate with enzyme at 37 ꢁC for 5 min.
The final substrate concentration was 2.6 lM. The Plas-
mepsin II concentration, measured by UV/Vis, was
13 nM in an acetate/formate buffer (0.06 M, pH 4.65).
DMSO (2.5%) was used to guarantee complete dissolution
of the inhibitors. IC50 values were taken from plots of vi/v0
as a function of inhibitor concentration, for which vi and v0
are the initial rates of reaction in presence and absence of
As the most active inhibitors out of our first series all
possess a 1H-indolyl-3-acetic acid moiety, we wanted to

S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
5413
Table 2
IC50 values of inhibitors based on the azepine scaffold against HIV-I protease and Plasmepsin II, respectively
Compound
IC50 (lM)
Plasmepsin II
40
HIV-I-protease
ꢁ1000
H
N
* HCl
MeO
O
O
O
28
N
H
>300
72
H
N
* HCl
O
O
O
O
32
>1250
40
H
N
H
N
* HCl
O
O
O
O
36 a
170
>1000
H
N
H
N
* HCl
O
O
O
O
36 b
125
>1000
H
N
H
N
* HCl
OH
O
O
37
125
ꢁ500
*HCl
H
N
MeO
O
O
N
H
O
40

5414
S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
the inhibitor, respectively. Substrate hydrolysis was
recorded as an increase in fluorescence intensity over a per-
iod of 3 min, during which the intensity increased linearly
with time [19].
further purification unless otherwise noted. Tetrahydrofu-
ran was dried by distillation from sodium/benzophenone.
All moisture-sensitive reactions were carried out using
oven-dried glassware under a positive stream of argon. If
necessary, solvents were deoxygenated by standard proce-
dures. Grubbs’ 2 was purchased from Sigma–Aldrich.
Compounds 9, 10, 11, 12 and 26 have been prepared
according to the literature.
4.1.2. Determination of inhibition constants toward HIV-I
protease
IC50 values were taken from plots of vi/v0 as a function
of inhibitor concentration, for which vi and v0 are the
catalytic rates in presence and absence of the inhibitor,
respectively. The fluorogenic substrate Abz-Thr-Ile-para-
nitrophenylalanine-Phe-Gln-Arg-NH2 was purchased from
Bachem. Recombinant HIV-I protease was expressed from
Escherichia coli and purified as previously described [18a].
Enzymatic assays were performed in 172 ll of assay buffer
(100 mM MES, 300 mM KCl, 5 mM EDTA, 1 mg/mL
BSA, pH 5.5) by adding substrate dissolved in 4 ll DMSO;
distinct inhibitor concentrations dissolved in 4 ll DMSO,
and 20 ll HIV-I protease solution to give a final volume
of 200 ll (final DMSO concentration 4%). The final
enzyme concentration was 6.25 nM and the substrate con-
centration was 20 lM. The hydrolysis of the substrate was
recorded as an increase in fluorescence intensity (excitation
wavelength 337 nm, emission wavelength 410 nm) over a
period of 10 min during which the rate increased linearly
with time [18b].
4.2.1. 5-(tert-Butyl-dimethyl-silanyloxymethyl)-2,3,4,7-
tetrahydro-1H-azepine-1,3-dicarboxylic acid-1-tert-butyl
ester 3-methyl ester (12)
The title compound has been prepared according to a
modified literature procedure [4].
A solution of 11 in 190 mL of thoroughly degassed
DCM was heated to 40 ꢁC and benzyliden-[1,3,-bis-(2,4,
6-trimethylphenyl)-2-imidazolidinylidene]-dichloro-(tricy-
clohexylphosphine)-ruthenium
8
(Grubbs
2 catalyst,
1 mol%), dissolved in 10 mL of degassed DCM, was added
to the reaction mixture. After stirring for 8 h at 40 ꢁC, the
reaction mixture was allowed to reach room temperature.
After quenching the reaction through addition of 20 mL
of DMSO [20] and subsequent stirring for 12 h, the reac-
tion mixture was washed three times with diluted NaCl-
solution, twice with brine, dried over MgSO4, and filtered.
Removal of the solvent under reduced pressure followed by
column chromatography (hexane/t-BuOMe: 8:1) of the oily
residue gave rise to 3.584 g (83%) of 12. The isolated prod-
uct exhibited identical spectroscopic data to those reported
previously.
4.2. General synthetic
Reported yields refer to the analytically pure product
obtained by distillation or column chromatography. All
proton and carbon nuclear magnetic resonance spectra
were recorded on a Jeol Eclipse +500 MHz spectrometer
(1H NMR: 500 MHz, 13C NMR: 125 MHz). Chemical
shifts are stated in parts per million (ppm) and were refer-
enced to TMS at 0.00 ppm (1H) except for compounds con-
taining a silyl protecting group which were referenced to
the residual CHCl3 in CDCl3 at 7.24 ppm and to CDCl3
at 77.0 ppm (13C), respectively. NMR spectra were
recorded in CDCl3 unless otherwise indicated. Abbrevia-
tions: bdd = broad doublet of doublets, bm = broad multi-
plet, bs = broad singlet, d = doublet, m = multiplet,
sm = symmetric multiplet, q = quartet, s = singlet,
t = triplet, ps = pseudo. Mass spectra were obtained from
a double focussing sectorfield spectrometer typ 7070 H by
Vacuum Generators or by a double focussing sectorfield
spectrometer typ VG-AutoSpec by Micromass. Combus-
tion analyses were determined on a CH analyzer by Labor-
matic/Woesthoff or a CHN autoanalyzer 185 (only
nitrogen) by Hewlett–Packard. Flash column chromatog-
raphy was performed using silica gel 60 (0.05–0.1 mm) or
silica gel 60 (0.04–0.063 mm) purchased from Macherey-
Nagel. TLC was carried out using 0.2 mm aluminium
plates coated with silica gel 60 F254 by Merck and the prod-
ucts were visualized by UV detection, iodine or by utiliza-
tion of phosphormolybdic acid (‘‘blue stain’’). Solvents and
reagents that are commercially available were used without
4.2.2. 2-Bromomethyl-acrylic acid methyl ester (13) [13]
To a solution of triphenylphoshine (TPP, 5.120 g,
19.53 mmol, 1.1 equiv.) in DCM (30 mL), a solution of
bromine (1 mL, 19.52 mmol, 1.1 equiv.) in 10 mL of
DCM was added dropwise at 0 ꢁC. The reaction mixture
was stirred until a colourless precipitate indicated the for-
mation of the desired phosphonium salt, which was then
added slowly to
a solution of imidazole (1.450 g,
21.30 mmol, 1.2 equiv.) and 2-hydroxymethyl-acrylic acid
methyl ester [4] (2.060 g, 17.75 mmol, 1.0 equiv.) in
30 mL DCM at 0 ꢁC. After stirring for 20 min maintaining
the temperature at 0 ꢁC, the reaction mixture was poured
into an ice–water mixture and extracted twice with t-
BuOMe. The combined organic layers were washed with
brine, dried over MgSO4, filtered, and concentrated under
reduced pressure. The residual slurry was re-dissolved in a
small amount of DCM and added dropwise into hexane
giving rise to a suspension, which was filtered, and washed
thoroughly. After concentration of the filtrate under vac-
uum, hexane was added to the residual suspension, which
was filtered again and concentrated under reduced pres-
sure. Bulb to bulb distillation (40 ꢁC/2.2 · 10ꢀ2 mbar) of
the remaining residue afforded 2.152 g (68%) of 13 as a col-
1
ourless oil: H NMR d 6.34 (s, 1H), 5.97 (s, 1H), 4.19 (s,
2H), 3.82 (s, 3H); 13C NMR d 165.2, 137.2, 129.1, 52.2,

S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
5415
29.2; MS (EI) m/z 180 (53, M+[Br81]), 178 (54, M+[Br79]),
149 (30), 147 (33), 121 (29), 119 (18), 99 (100).
NMR (rotamers) d 173.5, 167.3, 155.0, 139.2, 139.1,
131.6, 130.5, 80.3, 80.2, 52.0, 51.8, 47.5, 47.2, 47.1, 42.6,
42.2, 28.2, 25.7, 25.0; MS (ES+) m/z 336 (100,
[M+Na]+), 649 (86, [2M+Na]+); HRMS (ES+) m/z calcd
for C15H23NO6Na: 336.142307; found: 336.140934.
4.2.3. 2-[(Allyl-tert-butoxycarbonyl-amino)-methyl]-4-
methylene-pentane dioic acid dimethyl ester (14)
To a solution of n-BuLi (2.5 M in hexane, 1.6 mL,
4.0 mmol, 2.0 equiv.) in THF, neat diisopropyl amine
(0.56 mL, 4.0 mmol, 2.0 equiv.) was added at ꢀ78 ꢁC. The
reaction was allowed to reach 0 ꢁC, stirred for additional
10 min, and cooled again to ꢀ78 ꢁC. A solution of b-amino
ester 10 [4] (0.972 g, 4.0 mmol, 2.0 equiv.) dissolved in
THF was added slowly. After stirring for 30 min, 4 mL of
DMPU were added, followed by the addition of a solution
of allyl bromide derivative 13 (0.328 g, 2.0 mmol, 1.0 equiv.)
in THF. The reaction mixture was warmed to ꢀ40 ꢁC and
kept at this temperature until TLC indicated completion of
the reaction (ꢂ4 h). After subsequent addition of diethyl
ether and a saturated NH4Cl-solution, the organic layer
was separated and the aqueous layer was extracted twice
with diethyl ether. The combined organic layers were washed
with brine, dried over MgSO4, filtered, and concentrated
under reduced pressure. The remaining product was purified
by flash chromatography (hexane/t-BuOMe: 4:1) and gave
rise to 0.341 g (50%) of 14 as colourless oil: 1H NMR d (rota-
mers, ratio: ꢁ1:1) 6.19 (d, 1H, J = 1.2 Hz), 5.80–5.70 (sm,
1H), 5.61 (psd, 1H, J = 18.3 Hz), 5.12 (bs, 2H), 3.98 (d,
0.5H, J = 14.9 Hz), 3.88–3.80 (m, 0.5H), 3.73 (s, 3H), 3.73–
3.67 (m, 1H), 3.64 (s, 3H), 3.43–3.30 (bm, 2H), 3.08–3.02
(sm, 1H), 2.63–2.45 (bm, 2H), 1.47 (s, 4.5H), 1.44 (s, 4.5H);
13C NMR d (rotamers) 174.4, 166.9, 166.8, 155.4, 155.2,
137.2, 133.8, 133.7, 127.1, 116.6, 116.1, 80.0, 79.7, 51.9,
51.6, 50.4, 49.8, 48.6, 44.0, 43.6, 32.8, 32.3, 28.3; MS (ES+)
m/z 364 (100, [M+Na]+), 705 (23, [2M+Na]+); HRMS
(ES+) m/z calcd for C17H28NO6: 342.191663; found:
342.191758.
4.2.5. 3-[2-(tert-Butyl-dimethyl-silanyloxymethyl)-
allylamino]-propionic acid methyl ester (17)
To a stirred suspension of b-alanin methyl ester hydro-
chloride (0.698 g, 5.00 mmol, 1.0 equiv.) in DCM
(50 mL), TEA (1.48 mL, 10.50 mmol, 2.1 equiv.) was
added followed by the dropwise addition of (2-Bromo-
methyl-allyloxy)-tert-butyl-dimethylsilane 9 [4] (1.390 g,
5.25 mmol, 1.05 equiv.) in DCM (10 mL) at 0 ꢁC. The reac-
tion mixture was stirred at this temperature for 30 min,
warmed to ambient temperature and stirred for an addi-
tional 16 h. After subsequent addition of diethyl ether
and saturated NH4Cl-solution, the organic layer was sepa-
rated and the aqueous layer was extracted twice with
diethyl ether. The combined organic layers were washed
with brine, dried over MgSO4, filtered and concentrated
under reduced pressure. Column chromatography (hex-
ane/ethyl acetate: 1:1) gave rise to 0.595 g (42%) of 17 as
1
colourless oil: H NMR d 5.09 (s, 1H), 4.97 (s, 1H), 4.13
(s, 2H), 3.66 (s, 3H), 3.23 (s, 2H), 2.83 (t, 2H,
J = 6.4 Hz), 2.49 (t, 2H, J = 6.4 Hz), 1.58 (bs, 1H), 0.88
(s, 9H), 0.04 (s, 6H); 13C NMR d 173.0, 146.5, 110.2,
64.8, 51.6, 51.3, 44.2, 34.4, 25.8, 18.2, ꢀ5.5; MS (ES+)
m/z 288 (100, [M+H]+); HRMS (ES+) m/z calcd for
C14H30NO3Si: 288.199498; found: 288.200005.
4.2.6. 3-{tert-Butoxycarbonyl-[2-(tert-butyl-dimethyl-
silanyloxymethyl)-allyl]-amino}-propionic acid methyl ester
(18)
To a solution of 17 (1.010 g, 3.50 mmol, 1.0 equiv.),
TEA (0.591 mL, 4.20 mmol, 1.2 equiv.) and a catalytic
amount of DMAP in DCM at 0 ꢁC, a solution of (BOC)2O
(0.920 g, 4.20 mmol, 1.2 equiv.) in DCM was added drop-
wise over a period of 30 min. The reaction mixture was
allowed to reach room temperature and stirred for an addi-
tional 12 h. After subsequent addition of t-BuOMe and a
saturated NH4Cl-solution, the organic layer was separated
and the aqueous layer was extracted twice with t-BuOMe.
The combined organic layers were washed with brine, dried
over MgSO4, filtered, and concentrated under reduced
pressure. Flash chromatography (hexane/ethyl acetate:
4.2.4. 2,3,4,7-Tetrahydro-azepine-1,3,5-tricarboxylic acid 1-
tert-butyl ester 3,5-dimethyl ester (15)
A solution of 14 (0.341 g, 1.0 mmol, 1.0 equiv.) in 90 mL
of thoroughly degassed DCM was heated to 40 ꢁC and
benzyliden-[1,3,-bis-(2,4,6-trimethylphenyl)-2-imidazolid-
inylidene]-dichloro-(tricyclohexylphosphine)-ruthenium
(0.043 g, 0.05 mmol, 5 mol%, Grubbs 2 catalyst), dissolved
in 5 mL of degassed DCM, was added to the reaction mix-
ture. After stirring for 8 h at 40 ꢁC, the reaction mixture
was allowed to reach room temperature and stirred for
12 h at ambient temperature. The reaction mixture was
concentrated under reduced pressure followed by column
chromatography (hexane/t-BuOMe: 2:1) of the oily residue
1
9:1) gave rise to 1.170 g (86%) of 18 as colourless oil: H
NMR (rotamers, ratio ꢁ1:1) d 5.15 (s, 0.5H), 5.11 (s,
0.5H), 4.88 (s, 0.5H), 4.86 (s, 0.5H), 4.05 (s, 2H), 3.83 (s,
1H), 3.80 (s, 1H), 3.64 (s, 3H), 3.49–3.37 (bs, 2H), 2.60–
2.48 (bs, 2H), 1.43 (s, 4.5H), 1.41 (s, 4.5H), 0.88 (s, 9H),
0.04 (s, 6H); 13C NMR (rotamers) d 172.3, 172.1, 155.3,
144.6, 144.4, 110.6, 110.2, 79.7, 64.0, 63.9, 51.5, 49.9,
49.0, 42.9, 42.6, 33.3, 32.9, 28.2, 25.8, 25.6, 18.2, 14.1,
ꢀ3.7, ꢀ5.5; MS (ES+) m/z 410 (100, [M+Na]+), 797
(100, [2M+Na]+); HRMS (ES+) m/z calcd for C19H37NO5-
SiNa: 410.233872; found: 410.233324.
1
which gave rise to 0.244 g (78%) of 15 as colourless oil: H
NMR (rotamers, ratio: ꢁ1.4:1) d 6.92 (s, 1H), 4.33 (d,
0.60H, J = 16.7 Hz), 4.27 (d, 0.43H, J = 18.6 Hz), 4.13
(dd, 0.59H, J = 17.2, 3.9 Hz), 4.00 (dd, 0.43H, J = 18.3,
3.6 Hz), 3.93 (dd, 0.45H, J = 14.0, 6.4 Hz), 3.80–3.77 (m,
0.40 Hz), 3.77 (s, 1.33H), 3.75 (s, 1.68H), 3.71 (s, 1.70H),
3.69 (s, 1.29H), 3.69–3.60 (m, 1H), 3.05–2.94 (m, 1H),
2.88–2.78 (m, 2H), 1.47 (s, 5.39H), 1.46 (s, 3.97H); 13C

5416
S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
4.2.7. 2-({tert-Butoxycarbonyl-2[-(tert-butyl-dimethyl-
silanyloxymethyl)-allyl]-amino}-methyl)-pent-4-enoic acid
methyl ester (19)
0.79H), 3.86–3.71 (sm, 1.25H), 3.67 (s, 1.62H), 3.65 (s,
1.42H), 3.52–3.46 (sm, 1H), 2.89–2.79 (m, 1H), 2.51–2.35
(m, 2H), 1.42 (s, 9H), 0.88 (s, 4.7H), 0.87 (s, 4.3H), 0.04
(s, 3.2H), 0.03 (s, 2.8H); 13C NMR (rotamers) d 173.9,
155.1, 155.0, 140.9, 140.5, 122.3, 121.1, 79.76, 79.72,
66.34, 66.32, 51.7, 48.8, 48.5, 47.9, 47.6, 43.3, 42.8, 28.4,
27.3, 26.3, 25.8, 18.2, ꢀ5.4; MS (ES+) m/z 400 (100,
[M+H]+); HRMS (ES+) m/z calcd for C20H38NO5Si:
400.251927; found: 400.250423.
To a solution of n-BuLi (2.5 M in hexane, 1.84 mL,
4.60 mmol, 1.0 equiv.) in THF (8 mL), neat diisopropyl
amine (0.35 mL, 2.50 mmol, 1.0 equiv.) was added at
ꢀ78 ꢁC. The reaction mixture was allowed to reach 0 ꢁC,
stirred for additional 15 min, and cooled again to
ꢀ78 ꢁC. A solution of 18 (1.800 g, 4.60 mmol, 1.0 equiv.),
dissolved in THF, was added slowly. After stirring for
30 min, a solution of allyl bromide (0.840 g, 0.6 mL,
6.90 mmol, 1.5 equiv.) in HMPA (1.2 mL, 6.90 mmol, 1.5
equiv.) was then added slowly. The reaction mixture was
warmed to ꢀ40 ꢁC and kept at this temperature until
TLC indicated the completion of reaction (ꢂ3 h). After
subsequent addition of diethyl ether and saturated
NH4Cl-solution, the organic layer was separated and the
aqueous layer was extracted twice with diethyl ether. The
combined organic layers were washed with brine, dried
over MgSO4, filtered, and concentrated under reduced
pressure. Column chromatography (hexane/t-BuOMe:
10:1) afforded 1.590 g (81%) of 19 as colourless oil: 1H
NMR (rotamers, ratio: ꢁ1:1) d 5.75–5.64 (sm, 1H), 5.14
(bs, 0.5H), 5.12 (bs, 0.5H), 5.04 (dd, 1H, J = 17.1,
1.5 Hz), 5.00 (d, 1H, J = 9.8 Hz), 4.82 (bs, 1H), 4.02 (s,
2H), 3.94 (d, 0.5H, J = 15.9 Hz), 3.85 (d, 0.5H,
J = 16.2 Hz), 3.68 (d, 1H, J = 16.2 Hz), 3.64 (s, 3H),
3.38–3.25 (bs, 2H), 2.90–2.75 (bs, 1H), 2.36–2.25 (sm,
1H), 2.23–2.10 (bs, 1H), 1.43 (s, 4.5H), 1.40 (s, 4.5H),
0.88 (s, 9H), 0.03 (s, 6H); 13C NMR (rotamers) d 174.6,
155.4, 144.5, 144.2, 134.6, 117.2, 117.0, 110.5, 110.1, 79.9,
79.7, 64.1, 64.0, 51.6, 50.1, 49.2, 48.6, 48.0, 44.6, 44.1,
34.3, 28.3, 28.8, 18.3, ꢀ5.5; MS (ES+) m/z 428 (100,
[M+H]+), 878 (43, [2M+H+Na]+); HRMS (ES+) m/z
calcd for C22H42NO5Si: 428.283227; found: 428.285523.
Anal. Calc. for C22H41NO5Si: C, 61.79; H, 9.66; N, 3.28.
Found: C, 61.88; H, 9.60; N, 3.23%.
4.2.9. 2-(tert-Butyl-dimethyl-silanyloxymethyl)-propenal
(21) [16]
DMSO (2.13 mL, 30.00 mmol, 3.0 equiv.) was added
dropwise to a stirred solution of oxalyl chloride (1.29 mL,
15.00 mmol, 1.5 equiv.) in DCM (20 mL) at ꢀ78 ꢁC. After
stirring for additional 10 min, a solution of 2-hydroxy-
methyl-acrylic acid methyl ester (2.024 g, 10.00 mmol, 1.0
equiv.) [4] in 20 mL DCM was added slowly. After stirring
for 1 h, N-ethyl diisopropylamine (10.27 mL, 60.00 mmol,
6.0 equiv.) was added slowly. The reaction mixture was
warmed to room temperature and stirred for 10 min. After
quenching the reaction through addition of 40 mL H2O
and diluting with DCM, the organic layer was separated
and the aqueous layer was extracted twice with DCM. The
combined organic layers were washed three times with 1 N
HCl, three times with saturated NaHCO3-solution, twice
with brine, dried over MgSO4, filtered, and concentrated
under reduced pressure. Column chromatography (hexane/
t-BuOMe: 100:3) gave rise to 1.640 g (82%) of 21 as colour-
1
less oil: H NMR d 9.59 (s, 1H), 6.50 (dd, 1H, J = 3.2,
2.1 Hz), 6.08 (dd, 1H, J = 3.2, 1.8 Hz), 4.37 (t, 2H,
J = 2.1 Hz), 0.90 (s, 9H), 0.06 (s, 6H); 13C NMR d 193.4,
149.5, 132.5, 59.5, 25.6, 18.2, ꢀ5.6; MS (ES+) m/z 201
(100, [M+H]+), 231 (28, [M+MeOH]+).
4.2.10. 2-({tert-Butoxycarbonyl-[2-(tert-butyl-dimethyl-
silanyloxymethyl)-allyl]-amino}-methyl)-4-(tert-butyl-
dimethyl-silanyloxymethyl)-pent-4-enoic acid methyl ester
(22)
4.2.8. 6-(tert-Butyl-dimethyl-silanyloxymethyl)-2,3,4,7-
tetrahydro-azepine-1,3-dicarboxylic acid 1-tert-butyl ester
3-methyl ester (20)
To a solution of n-BuLi (2.5 M in hexane, 1.0 mL,
2.50 mmol, 1.06 equiv.) in THF (5 mL), neat diisopropyl
amine (0.35 mL, 2.50 mmol, 1.06 equiv.) was added at
ꢀ78 ꢁC. The reaction mixture was allowed to reach 0 ꢁC, stir-
red for additional 15 min, and cooled again to ꢀ78 ꢁC. A
solution of 18 (0.910 g, 2.35 mmol, 1.0 equiv.) dissolved in
THF was added slowly. After stirring for 30 min, a solution
of bromo allyl derivative 9 (0.660 g, 2.50 mmol, 1.06 equiv.)
in HMPA (0.44 mL, 2.50 mmol, 1.06 equiv.) was then added
slowly. The reaction mixture was warmed to ꢀ40 ꢁC and
kept at this temperature until TLC indicated the completion
of reaction (ꢂ4 h). After subsequent addition of diethyl ether
and saturated NH4Cl-solution, the organic layer was sepa-
rated and the aqueous layer was extracted twice with diethyl
ether. The combined organic layers were washed with brine,
dried over MgSO4, filtered, and concentrated under reduced
pressure. Column chromatography (hexane/ethyl acetate:
10:1) afforded 0.694 g (52%) of 22 as colourless oil: 1H
A solution of 19 (1.300 g, 3.00 mmol) in 50 mL thor-
oughly degassed DCM was heated to 40 ꢁC and benzyl-
iden-[1,3-bis-(2,4,6-trimethylphenyl)-2-imidazolidinylidene]-
dichloro-(tricyclohexyl-phosphine)-ruthenium (Grubbs 2
catalyst, 1 mol%, 25.5 mg, 0.03 mmol), dissolved in 5 mL
of degassed DCM, was added to the reaction mixture.
After stirring for 6.5 h at 40 ꢁC, the reaction mixture was
allowed to reach room temperature. After quenching the
reaction through addition of 20 mL of DMSO [20] and stir-
ring for 12 h, the reaction mixture was washed three times
with diluted NaCl-solution, twice with brine, dried over
MgSO4, filtered, and concentrated under reduced pressure.
Flash chromatography (hexane/t-BuOMe: 10:1) of the oily
residue gave rise to 0.980 g (82%) of 20 as colourless oil: 1H
NMR (rotamers, ratio: ꢁ1.1:1) d 5.70–5.64 (sm, 1H), 4.11–
4.04 (m, 1.46H), 4.04–3.99 (m, 1.56H), 3.99–3.90 (m,

S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
5417
NMR d (rotamers, ratio: ꢁ1:1) 5.15 (bs, 0.5H), 5.10 (bs,
0.5H), 5.04 (s, 1H), 4.82 (s, 2H), 4.03 (s, 4H), 3.95 (d, 0.5H,
J = 16.0 Hz), 3.84 (d, 0.5H, J = 16.5 Hz), 3.67 (d, 1H,
J = 17.2 Hz), 3.62 (s, 3H), 3.31 pst (2H), 3.03–2.90 (bs,
1H), 2.35–2.20 (m, 1H), 2.16–2.05 (m, 1H), 1.44 (s, 4.5H),
1.40 (s, 4.5H), 0.88 (s, 9H), 0.87 (s, 9H), 0.04 (s, 12H); 13C
NMR d (rotamers) 174.8, 155.4, 145.3, 144.5, 144.2, 110.8,
110.6, 110.2, 80.0, 79.7, 65.6, 64.1, 64.0, 51.7, 50.1, 49.2,
49.0, 48.7, 43.5, 43.0, 33.2, 28.3, 25.9, 18.3, ꢀ5.4; MS
(ES+) m/z 572 (30, [M+H]+), 594 (65, [M+Na]+), 1165
(100, [2M+Na]+); HRMS (ES+) m/z calcd for C29H57NO6-
NaSi2: 594.362216; found: 594.358649.
(15 mL) were added 26 (0.984 g, 3.45 mmol, 1.0 equiv.) dis-
solved in a 1:1 mixture of THF/DCM and DMAP (0.021 g,
0.17 mmol, 5 mol%) followed by the dropwise addition of a
solution of DIC (0.70 mL, 4.49 mmol, 1.3 equiv.) in a 1:1
mixture of THF/DCM. The reaction mixture was stirred
at room temperature for 14 h, then carefully quenched by
addition of a saturated NaHCO3-solution and the layers
were separated. The aqueous layer was extracted three
times with diethyl ether and the combined organic layers
were washed twice with brine, dried over MgSO4, filtered,
and concentrated under reduced pressure. The residue
was re-dissolved in a small amount of acetone p.a., insolu-
ble material was filtered off, and the resulting residue was
thoroughly washed with acetone p.a. The combined
acetone layers were concentrated under reduced pressure.
Column chromatography of the oily residue (hexane/t-
BuOMe: 1:1) afforded 1.326 g (87%) of 27 as a colourless
oil: 1H NMR d (rotamers, ratio: ꢁ1.2:1) 8.19 (s, 1H),
7.60 (d, 1H, J = 7.3 Hz), 7.35 (d, 1H, J = 8.3 Hz), 7.19 (t,
1H, J = 7.2 Hz), 7.16 (d, 1H, J = 2.3 Hz), 7.12 (pst, 1H,
J = 7.1 Hz), 5.67 (s, 0.47H), 5.62 (s, 0.56H), 4.49 (s, 2H),
4.14–4.01 (sm, 1H), 3.97–3.84 (sm, 1H), 3.80 (s, 2H),
3.77–3.70 (sm, 1H), 3.69 (s, 1.67H), 3.66 (s, 1.33H), 3.60–
3.46 (sm, 1H), 2.80–2.77 (m, 1H), 2.45–2.31 (m, 2H), 1.45
(s, 4.93H), 1.43 (s, 4.09H); 13C NMR d (rotamers) 173.7,
171.7, 155.3, 155.1, 136.1, 135.0, 133.6, 127.1, 126.8,
126.1, 123.2, 122.0, 119.4, 118.6, 111.2, 108.0, 80.1, 79.9,
69.1, 51.8, 48.0, 47.6, 46.6, 46.4, 42.7, 42.1, 31.3, 28.6,
28.32, 28.25, 27.8; MS (ES+) m/z 465 (100, [M+Na]+),
907 (49, [2M+Na]+); HRMS (ES+) m/z calcd for
C24H30N2O6Na: 465.200157; found: 465.197041.
4.2.11. 2-{[tert-Butoxycarbonyl-(2-hydroxymethyl-allyl)-
amino]-methyl}-4-hydroxymethyl-pent-4-enoic acid methyl
ester (24)
0.516 g (0.90 mmol) of 22 was dissolved in a 30:1 mix-
ture of THF and aqueous HCl 32% and stirred at room
temperature for 25 min. After addition of a saturated
NaHCO3-solution, the reaction mixture was extracted
three times with ethyl acetate. The combined organic layers
were washed with brine, dried over MgSO4, filtered, and
concentrated under reduced pressure. Column chromato-
graphy (hexane/t-BuOMe: 1:3) of the oily residue gave rise
to 0.235 g (80%) of 24 as colourless oil: 1H NMR (DMSO-
d6, rotamers) d 5.05 (s, 1H), 4.98 (s, 1H), 4.82 (pst, 2H),
4.77–4.72 (m, 2H), 3.90–3.78 (m, 5H), 3.70–3.60 (bs, 1H),
3.56 (bs, 3H), 3.28–3.20 (m, 2H), 2.97–2.88 (m, 1H), 2.21
(dd, 1H, J = 14.7, 9.2 Hz), 2.07 (dd, 1H, J = 14.7,
5.7 Hz), 1.38 (s, 9H); 13C NMR (rotamers) d 175.0,
174.7, 156.3, 155.5, 145.4, 144.5, 113.4, 112.3, 112.0,
111.3, 80.8, 80.2, 65.2, 63.4, 51.7, 50.2, 49.4, 48.8, 43.4,
43.0, 33.6, 28.1; MS (ES+) m/z 366 (100, [M+Na]+), 709
(70, [2M+Na]+); HRMS (ES+) m/z calcd for
C17H29NO6Na: 366.189258; found: 366.187395.
4.2.14. 5-(2-1H-Indol-3-yl-acetoxymethyl)-3-
methoxycarbonyl-2,3,4,7-tetrahydro-1H-azepinium chloride
(28)
A suspension of 27 (0.100 g, 0.23 mmol, 1.0 equiv.) in a
2 M solution of hydrogen chloride in diethyl ether
(2.26 mL, 4.52 mmol, 20.0 equiv.) was stirred for 15 min at
0 ꢁC. The reaction mixture was allowed to reach room tem-
perature and stirred for an additional 1.5 h. The reaction
mixture was filtered under a positive stream of argon and
the resulting residue was thoroughly washed with diethyl
ether. Column chromatography (DCM/MeOH: 92:8) of
the brownish residue gave rise to 0.076 g (89%) of 28 as a
white powder: m.p. 50–60 ꢁC (hygroscopic); 1H NMR
(DMSO-d6) d 11.0 (s, 1H), 7.49 (d, 1H, J = 7.8 Hz), 7.36
(d, 1H, J = 8.0 Hz), 7.26 (d, 1H, J = 2.3 Hz), 7.09 (t, 1H,
J = 7.0 Hz), 6.99 (t, 1H, J = 7.0 Hz), 5.70 (t, 1H,
J = 5.7 Hz), 4.52 (s, 2H), 3.79 (s, 2H), 3.64 (s, 3H), 3.60 (d,
1H, J = 5.5 Hz), 3.51 (dd, 1H, J = 15.8, 5.7 Hz), 3.40 (dd,
1H, J = 13.4, 4.4 Hz), 3.33 (dd, 1H, J = 13.4, 7.9 Hz),
3.05–2.98 (m, 1H), 2.45 (d, 2H, J = 5.7 Hz); 13C NMR
(DMSO-d6) d 172.1, 171.1, 139.9, 136.1, 127.0, 124.1,
121.5, 121.0, 118.4, 118.3, 111.4, 106.7, 67.2, 52.0, 48.6,
43.2, 39.4, 30.7, 29.3; MS (EI) m/z 342 (5, M+), 167 (100),
154 (18), 130 (86), 108 (46); HRMS (EI) m/z calcd for
C19H22N2O4: 342.157957; found: 342.158371.
4.2.12. 5-Hydroxymethyl-2,3,4,7-tetrahydro-azepine-1,3-
dicarboxylic acid 1-tert-butyl ester 3-methyl ester (26)
Compound 26 has already been prepared by us via a dif-
ferent route [4].
Compound 12 (1.600 g, 4.00 mmol, 1.0 equiv.) was dis-
solved in a 30:1 mixture of THF and aqueous HCl 32%
and stirred at room temperature for 30 min. After addition
of a saturated NaHCO3-solution, the reaction mixture was
extracted three times with ethyl acetate. The combined
organic layers were washed with brine, dried over MgSO4,
filtered, and concentrated under vacuum. Column chroma-
tography (hexane/t-BuOMe: 1:1) of the oily residue affor-
ded 0.987 g (87%) of 26 as a colourless oil, which
exhibited identical spectroscopic data to those obtained
using the method described earlier [4].
4.2.13. 5-(2-1H-Indol-3-yl-acetoxymethyl)-2,3,4,7-
tetrahydro-azepine-1,3-dicarboxylic acid 1-tert-butyl ester 3-
methyl ester (27)
To a stirred solution of indole-3-acetic acid (0.725 g,
4.14 mmol, 1.2 equiv.) in a 1:1 mixture of THF/DCM

5418
S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
4.2.15. 5-(tert-Butyl-dimethyl-silanyloxymethyl)-3-
hydroxymethyl-2,3,4,7-tetrahydro-azepine-1-carboxylic acid
tert-butyl ester (29)
[M+Na]+), 537 (70, [2M+Na]+); HRMS (ES+) m/z calcd
for C13H23NO4Na: 280.152478; found: 280.152423.
DIBAL-H (8.9 mL of a 1.5 M solution in toluene,
13.20 mmol, 2.2 equiv.) was added dropwise over a period
of 45 min to a stirred solution of 12 (2.398 g, 6.00 mmol,
1.0 equiv.) in THF (15 mL) at ꢀ78 ꢁC. After 3 h the reac-
tion mixture was slowly warmed to 0 ꢁC, stirred for
30 min and then carefully quenched by addition of 1 mL
H2O in 5 mL of THF. Subsequent addition of 25 mL
diethyl ether and 20 mL of saturated Rochelle’s solution
(potassium sodium L-tartrate tetrahydrate in water) pro-
duced a gelatine-like solid, which, after addition of
10 mL of a saturated NH4Cl-solution, was stirred at room
temperature until the slurry re-dissolved and a separation
of the layers was observed (30 min). The aqueous layer
was extracted three times with diethyl ether and the com-
bined organic layers were washed twice with brine, dried
over MgSO4, filtered, and concentrated under reduced
pressure. Column chromatography (hexane/ethyl acetate:
4.2.17. 3,5-Bis-benzoyloxymethyl-2,3,4,7-tetrahydro-
azepine-1-carboxylic acid tert-butyl ester (31)
To a stirred solution of 30 (0.732 g, 2.84 mmol, 1.0
equiv.) in DCM (20 mL) were added TEA (1.03 mL,
7.38 mmol, 2.6 equiv.) and a catalytic amount of DMAP
at 0 ꢁC. The reaction mixture was allowed to reach room
temperature followed by the dropwise addition of a solu-
tion of benzoylchloride (0.726 mL, 6.25 mmol, 2.2 equiv.)
and stirred for 14 h at ambient temperature. After addition
of t-BuOMe the reaction mixture was quenched with a sat-
urated NH4Cl-solution and the aqueous layer was
extracted twice with t-BuOMe. The combined organic lay-
ers were washed with brine, dried over MgSO4, filtered, and
concentrated under reduced pressure. Column chromatog-
raphy (hexane/t-BuOMe: 8:2) of the remaining oily residue
afforded 0.985 g (75%) of 31 as a colourless oil: 1H NMR d
8.02 (d, 2H, J = 7.3 Hz), 8.0 (d, 2H, J = 7.1 Hz), 7.58–7.50
(m, 2H), 7.44–7.34 (m, 4H), 5.95–5.87 (m, 1H), 4.73 (dd,
2H, J = 12.8, 4.8 Hz), 4.36–4.27 (m, 1H), 4.23 (dd, 1H,
J = 11.0, 6.9 Hz), 4.14–3.96 (m, 2H), 3.80–3.65 (m, 1H),
3.50 (dd, 1H, J = 14.3, 7.3 Hz), 2.57–2.42 (m, 2H), 2.34
(dd, 1H, J = 15.6, 8.7 Hz), 1.46 (s, 9H); 13C NMR d (rota-
mers) 166.2, 166.1, 155.4, 155.2, 135.7, 134.8, 132.9, 129.9,
129.47, 129.42, 128.31, 128.25, 127.2, 126.9, 80.0, 79.7,
69.4, 66.7, 66.5, 49.2, 46.2, 37.1, 37.0, 30.0, 29.5, 28.3;
MS (ES+) m/z 488 (100, [M+Na]+), 504 (30, [M+K]+);
HRMS (ES+) m/z calcd for C27H31NO6Na: 488.204908;
found: 488.203494.
1
2:1) gave rise to 1.684 g (76%) of 29 as colourless oil: H
NMR d (rotamers, ratio: 2.4:1) 5.77–5.71 (m, 0.7H),
5.69–5.66 (m, 0.3H), 4.18 (dd, 0.7H, J = 16.3, 6.0 Hz),
4.05 (dd, 0.8H, J = 14.4, 3.7 Hz), 4.00 (bs, 0.3H), 3.98–
3.95 (m, 1H), 3.94 (s, 1.2H), 3.77 (s, 0.7H), 3.60–3.54 (m,
1.5H), 3.57–3.43 (m, 1.5H), 3.36–3.29 (m, 0.7H), 3.16
(dd, 0.8H, J = 14.3, 4.0 Hz), 3.01 (pst, 0.2H), 2.41 (d,
0.3H, J = 14.4 Hz), 2.27 (d, 0.7H, J = 14.4 Hz), 2.21–2.00
(m, 2.1H), 1.44 (s, 2.7H), 1.43 (s, 6.3H), 0.89 (s, 9H),
0.07 (s, 2.1H), 0.05 (s, 4.8H); 13C NMR d (rotamers)
156.2, 155.1, 141.2, 138.9, 123.2, 121.1, 79.7, 79.4, 68.9,
67.5, 63.8, 62.3, 49.3, 49.0, 46.4, 39.6, 38.8, 29.6, 28.6,
28.3, 28.2, 25.7, 18.3, 18.2, ꢀ5.5; MS (ES+) m/z 394 (14,
[M+Na]+), 765 (100, [2M+Na]+), 1136 (15, [3M+Na]+);
HRMS (ES+) m/z calcd for C19H38NO4Si: 372.257012;
found: 372.256321.
4.2.18. 3,5-Bis-benzoyloxymethyl-2,3,4,7-tetrahydro-1H-
azepinium chloride (32)
A suspension of 31 (0.266 g, 0.57 mmol, 1.0 equiv.) in a
2 M solution of hydrogen chloride in diethyl ether
(5.71 mL, 11.42 mmol, 20.0 equiv.) was stirred for 16 h at
room temperature. The resulting precipitate was separated
by filtration under a positive stream of argon, washed several
times with dry diethyl ether and dried to give rise to 0.172 g
4.2.16. 3,5-Bis-hydroxymethyl-2,3,4,7-tetrahydro-azepine-1-
carboxylic acid tert-butyl ester (30)
Compound 29 (0.615 g, 1.66 mmol, 1.0 equiv.) was dis-
solved in a 30:1 mixture of THF and aqueous HCl 32%
and stirred at room temperature for 20 min. After addition
of a saturated NaHCO3-solution, the reaction mixture was
extracted three times with ethyl acetate. The combined
organic layers were washed with brine, dried over MgSO4,
filtered, and concentrated under vacuum. Column chroma-
tography (ethyl acetate) of the oily residue afforded 0.370 g
1
(75%) of 32 as a colourless powder: m.p. 175–180 ꢁC; H
NMR (DMSO-d6) d 9.60–9.43 (bs, 1H), 9.34–9.08 (bs, 1H),
8.01–7.95 (sm, 4H), 7.70–7.63 (sm, 2H), 7.55–7.47 (sm,
4H), 5.96–5.91 (pst, 1H), 4.80 (s, 2H), 4.31–4.24 (sm, 2H),
3.79 (dd, 1H, J = 15.6, 6.0 Hz), 3.73 (dd, 1H, J = 15.5,
5.6 Hz), 3.51–3.46 (sm, 1H), 3.26, 3.19 (sm, 1H), 2.65–2.57
(sm, 1H), 2.49–2.45 (m, 2H); 13C NMR (DMSO-d6) d
165.4, 165.2, 141.2, 133.4, 133.3, 129.3, 129.1, 128.7, 128.6,
120.7, 67.7, 66.1, 49.4, 42.5, 33.3, 30.4; MS (ES+) m/z 366
(100, [M+H]+); HRMS (ES+) m/z calcd for C22H24NO4:
366.170534; found: 366.172069.
1
(87%) of 30 as colourless oil: H NMR (rotamers, ratio:
ꢁ1.6:1) d 5.80 (bs, 0.39H), 5.75 (bs, 0.61H), 4.19 (dd,
0.61H, J = 16.5, 5.7 Hz), 4.07–3.95 (m, 2.96H), 3.84 (bd,
0.83H, J = 16.9 Hz), 3.66–3.47 (m, 2.40H), 3.45–3.37 (m,
1H), 3.22 (dd, 0.86H, J = 14.3, 4.2 Hz), 2.47 (d, 0.37H,
J = 15.1 Hz), 2.39–2.29 (m, 1H), 2.25 (dd, 1.2H, J = 15.0,
6.7 Hz), 2.18–2.06 (bs, 1H), 1.46 (s, 9H); 13C NMR (rota-
mers) d 156.1, 155.4, 141.4, 140.1, 124.3, 122.6, 80.0,
79.8, 68.2, 67.7, 63.3, 63.1, 49.7, 49.3, 46.4, 46.0, 39.4,
38.8, 29.5, 29.0, 28.3, 28.2; MS (ES+) m/z 280 (100,
4.2.19. 5-(tert-Butyl-dimethyl-silanyloxymethyl)-3-(2-1H-
indol-3-yl-acetoxymethyl)-2,3,4,7-tetrahydro-azepine-1-
carboxylic acid tert-butyl ester (33)
To a stirred solution of indole-3-acetic acid (0.420 g,
2.40 mmol, 1.2 equiv.) in a 1:1 mixture of THF and

S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
5419
DCM were added 29 dissolved in a 1:1 mixture of THF
and DCM (0.743 g, 2.00 mmol, 1.0 equiv.) and DMAP
(0.012 g, 0.10 mmol, 5 mol%) followed by the addition
of DCC (0.536 g, 2.60 mmol, 1.3 equiv.) in a 1:1 mixture
of THF and DCM. The reaction mixture was stirred at
room temperature for 14 h and then carefully quenched
by addition of a saturated NaHCO3-solution. After addi-
tion of diethyl ether the layers were separated. The
organic layer was filtered and the residue was washed
three times with diethyl ether. The combined filtrates
were washed twice with brine, dried over MgSO4, fil-
tered, and concentrated under reduced pressure. Column
chromatography (hexane/t-BuOMe: 1.5:1) afforded
0.865 g (82%) of 33 as a colourless oil: 1H NMR d
8.12 (bs, 1H), 7.60 (d, 1H, J = 7.9 Hz), 7.34 (d, 1H,
J = 7.9 Hz), 7.22–7.10 (m, 3H), 5.67 (d, 1H,
J = 14.7 Hz), 4.05–3.91 (m, 3H), 3.85 (s, 3H), 3.77 (s,
2H), 3.59 (d, 0.5H, J = 12.2 Hz), 3.49 (d, 0.5H,
J = 12.8 Hz), 3.27 (dd, 1H, J = 14.1, 7.9 Hz), 2.19 (bs,
1H), 2.13–1.90 (m, 2H), 1.42 (s, 9H), 0.88 (s, 9H), 0.03
(s, 6H); 13C NMR (rotamers) d 172.0, 155.4, 155.2,
140.3, 138.9, 136.1, 127.1, 123.2, 123.1, 122.0, 121.9,
121.6, 121.5, 119.4, 118.7, 118.6, 111.23, 111.18, 107.9,
79.7, 79.5, 67.4, 67.0, 66.8, 66.4, 49.5, 49.1, 46.2, 46.0,
36.8, 36.3, 31.2, 29.3, 28.6, 28.4, 25.8, 18.3, ꢀ5.35,
ꢀ5.41; MS (ES+) m/z 546 (90, [M+NH4]+), 1075 (100,
[2M+H+NH4]+); HRMS (ES+) m/z calcd for
C29H48N3O5Si: 546.336326; found: 546.333278.
4.2.21. 3-(2-1H-Indol-3-yl-acetoxymethyl)-5-(4-propyl-
benzoyloxymethyl)-2,3,4,7-tetrahydro-azepine-1-carboxylic
acid tert-butyl ester (35a)
To a solution of 34 (0.480 g, 1.16 mmol, 1.0 equiv.), N-
ethyl diisopropylamine (0.236 mL, 1.51 mmol, 1.3 equiv.)
and a catalytic amount of DMAP in DCM at 0 ꢁC, a solu-
tion of 4-propyl-benzoyl chloride in DCM (0.214 mL,
1.28 mmol, 1.1 equiv.) was added dropwise over a period
of 30 min. The reaction mixture was allowed to slowly
reach room temperature and stirred for additional 14 h.
After careful addition of a saturated NH4Cl-solution, the
reaction mixture was extracted three times with t-BuOMe.
The combined organic layers were washed twice with brine,
dried over MgSO4, filtered, and concentrated under
reduced pressure. Column chromatography (hexane/t-
BuOMe: 1:1) gave rise to 0.395 g (61%) of 35a as colourless
1
oil: H NMR d 8.14 (s, 1H), 7.95 (d, 2H, J = 8.0 Hz), 7.59
(d, 1H, J = 8.0 Hz), 7.32 (d, 1H, J = 8.0 Hz), 7.24 (d, 2H,
J = 8.3 Hz), 7.17 (t, 1H, J = 7.4 Hz), 7.13–7.08 (m, 2H),
5.79 (s, 1H), 4.51 (s, 2H), 4.13–4.07 (m, 1H), 3.96 (dd,
2H, J = 11.0, 7.8 Hz), 3.92–3.84 (m, 1H), 3.74 (s, 2H),
3.62–3.46 (m, 1H), 3.32 (dd, 1H, J = 14.1, 7.9 Hz), 2.63
(t, 2H, J = 7.7 Hz), 2.33–2.15 (m, 2H), 2.13–2.03 (m, 1H),
1.65 (sex, 2H, J = 7.5 Hz), 1.44 (s, 9H), 0.94 (t, 3H,
J = 7.3 Hz); 13C NMR (rotamers) d 171.9, 166.3, 155.4,
155.2, 148.4, 136.04, 135.97, 134.7, 129.6, 128.5, 127.4,
127.1, 126.6, 126.1, 123.1, 122.1, 122.0, 119.5, 118.7,
111.2, 108.1, 80.0, 79.8, 69.1, 66.3, 49.2, 49.0, 46.2, 46.0,
37.9, 36.9, 36.5, 31.3, 29.8, 29.2, 28.4, 24.2, 13.7; MS
(ES+)
m/z 583 (100, [M+Na]+), 1144
(71,
[2M+H+Na]+); HRMS (ES+) m/z calcd for C33H44N3O6
4.2.20. 5-Hydroxymethyl-3-(2-1H-indol-3-yl-
acetoxymethyl)-2,3,4,7-tetrahydro-azepine-1-carboxylic
acid tert-butyl ester (34)
([M+NH4]+): 578.323012; found: 578.322479.
Compound 33 (0.633 g, 1.20 mmol, 1.0 equiv.) was
dissolved in a 30:1 mixture of THF and aqueous HCl
32% and stirred at room temperature for 40 min. After
addition of a saturated NaHCO3-solution, the reaction
mixture was extracted three times with ethyl acetate.
The combined organic layers were washed with brine,
dried over MgSO4, filtered, and concentrated under
reduced pressure. Column chromatography (hexane/t-
BuOMe: 1:4) afforded 0.424 g (85%) of 34 as a colourless
oil: 1H NMR d 8.23 (s, 1H), 7.62 (d, 1H, J = 7.8 Hz),
7.63 (d, 1H, J = 8.0 Hz), 7.20 (t, 1H, J = 7.4 Hz), 7.17
(d, 1H, J = 1.8 Hz), 7.14 (d, 1H, J = 7.2 Hz), 5.63 (t,
1H, J = 4.4 Hz), 4.08–4.01 (m, 1H), 3.97 (dd, 1H,
J = 14.8, 7.6 Hz), 3.94–3.85 (bm, 2H), 3.79 (s, 2H),
3.75 (d, 2H, J = 7.10 Hz), 3.61–3.45 (bm, 1H), 3.30
(dd, 1H, J = 14.0, 7.8 Hz), 2.30–2.18 (bs, 1H), 2.18–
2.07 (bm, 1H), 1.99 (dd, 1H, J = 15.0, 8.1 Hz), 1.58
(bs, 1H), 1.44 (s, 9H); 13C NMR d (rotamers) 172.0,
155.3, 140.5, 139.6, 136.1, 127.1, 123.31, 123.25, 122.8,
122.0, 121.9, 119.4, 118.7, 118.6, 111.3, 107.9, 79.9,
79.7, 67.6, 67.4, 66.4, 49.3, 49.2, 46.1, 36.8, 36.3, 31.3,
29.3, 29.0, 28.4; MS (ES+) m/z 437 (100, [M+Na]+),
851 (56, [2M+Na]+); HRMS (ES+) m/z calcd for
C23H31N2O5: 415.223297; found: 415.221057.
4.2.22. 3-(2-1H-Indol-3yl-acetoxymethyl)-2,3,4,7-
tetrahydro-azepine-1-carboxylic acid tert-butyl-ester (35b)
To a solution of 34 (0.843 g, 2.03 mmol, 1.0 equiv.), tri-
ethylamine (0.4 mL, 2.85 mmol, 1.4 equiv.) and a catalytic
amount of DMAP in DCM at room temperature, a solu-
tion of acetyl chloride in DCM (0.174 mL, 2.44 mmol,
1.2 equiv.) was added dropwise. The reaction mixture
was stirred for 4 h at ambient temperature. After careful
addition of a saturated NH4Cl-solution, the reaction mix-
ture was extracted three times with diethyl ether. The com-
bined organic layers were washed twice with brine, dried
over MgSO4, filtered, and concentrated under reduced
pressure. Column chromatography (hexane/t-BuOMe:
1
1:1) gave rise to 0.730 g (79%) of 35b as colourless oil: H
NMR d 8.24 (s, 1H), 7.61 (d, 1H, J = 8.0 Hz), 7.35 (pst,
1H), 7.23–7.10 (m, 3H), 5.73–5.67 (bs, 1H), 4.29 (s, 2H),
4.09–4.01 (m, 1H), 4.00–3.92 (m, 2H), 3.88–3.82 (m, 1H),
3.79 (s, 2H), 3.61–3.47 (sm, 1H), 3.34–2.95 (sm, 1H),
2.33–2.22 (bm, 1H), 2.21–2.10 (sm, 1H), 2.04 (s, 3H),
2.07–1.97 (bm, 1H), 1.44 (s, 9H); 13C NMR (rotamers) d
171.9, 170.7, 155.3, 155.2, 136.1, 135.6, 134.5, 127.1,
126.7, 126.3, 123.1, 122.1, 122.0, 119.5, 118.73, 118.65,
111.25, 111.18, 108.1, 79.9, 79.7, 68.9, 66.4, 49.0, 46.1,
46.0, 36.7, 36.3, 31.3, 29.7, 29.6, 29.2, 28.4, 28.1, 20.8;

5420
S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
MS (ES+) m/z 479 (86, [M+Na]+), 935 (100, [2M+Na]+);
HRMS (ES+) m/z calcd for C25H32N2O6Na: 479.215807;
found: 479.213201.
4.2.25. 5-Hydroxymethyl-3-(2-1H-indol-3-yl-
acetoxymethyl)-2,3,4,7-tetrahydro-1H-azepinium chloride
(37)
A suspension of 34 (0.450 g, 1.10 mmol, 1.0 equiv.) in a
2 M solution of hydrogen chloride in diethyl ether (8.2 mL,
16.30 mmol, 15.0 equiv.) was stirred for 15 min at 0 ꢁC.
The reaction mixture was allowed to reach room tempera-
ture and stirred for an additional 1.5 h. The suspension was
filtered under a positive stream of argon and the resulting
residue was thoroughly washed with diethyl ether. Column
chromatography (DCM/MeOH: 8:2) of the brownish resi-
due yielded 0.360 g (93%) of 37 as a white, hygroscopic
4.2.23. 3-(2-1H-Indol-3-yl-acetoxymethyl)-5-(4-propyl-
benzoyloxymethyl)-2,3,4,7-tetrahydro-1H-azepinium
chloride (36a)
A solution of 35a (0.250 g, 0.45 mmol, 1.0 equiv.) in a
2 M solution of hydrogen chloride in diethyl ether
(4.5 mL, 8.92 mmol, 20.0 equiv.) was stirred for 3 h at
room temperature. The reaction mixture was filtered under
a positive stream of argon, and the resulting residue was
washed four times with diethyl ether to give rise to
0.205 g (92%) of 36a as white powder: m.p. 135–140 ꢁC;
1H NMR (DMSO-d6) d 10.99 (s, 1H), 9.50–9.00 (bs,
1.5H), 7.91 (d, 2H, J = 7.3 Hz), 7.48 (d, 1H, J = 7.3 Hz),
7.40–7.30 (m, 3H), 7.23 (s, 1H), 7.07 (t, 1H, J = 6.9 Hz),
6.98 (t, 1H, J = 7.1 Hz), 5.86 (s, 1H), 4.67 (s, 2H), 4.01
(d, 2H, J = 4.8 Hz), 3.73 (s, 3H), 3.68–3.56 (m, 1H),
3.43–3.20 (m, 1H, overlapping with H2O), 3.04 (sm, 1H),
2.62 (t, 2H, J = 7.2 Hz), 2.45–2.35 (m, 1H), 2.30 (psd,
2H, J = 14.4 Hz), 1.63–1.53 (m, 2H), 0.88 (t, 3H,
J = 7.0 Hz); 13C NMR (DMSO-d6) d 169.2, 163.0, 146.1,
139.4, 133.9, 127.1, 126.6, 124.8, 121.9, 118.9, 118.1,
116.3, 116.2, 109.3, 104.5, 65.4, 63.3, 47.4, 40.6, 35.0,
31.1, 28.4, 28.1, 21.5, 11.4; MS (ES+) m/z 461 (100,
[M+H]+), 921 (5, [2M+H]+); HRMS (ES+) m/z calcd for
C28H33N2O4: 461.244033; found: 461.240402. Anal. Calc.
for C28H32N2O4 Æ HCl Æ 1.5H2O: C, 64.17; H, 6.83; N,
5.35; found: C, 64.52; H, 6.55; N, 5.26%.
1
powder: m.p. 40-50 ꢁC; H NMR (DMSO-d6) d 11.01 (s,
1H), 9.40–8.90 (bs, 2H), 7.51 (d, 1H, J = 7.8 Hz), 7.36 (d,
1H, J = 8.0 Hz), 7.26 (d, 1H, J = 2.5 Hz), 7.08 (t, 1H,
J = 7.0 Hz), 7.00 (t, 1H, J = 7.1 Hz), 5.68 (t, 1H,
J = 5.5 Hz), 5.10–5.00 (bs, 1H), 3.98 (d, 2H, J = 4.8 Hz),
3.78 (s, 2H), 3.76 (s, 2H), 3.66 (dd, 1H, J = 15.4, 6.2 Hz),
3.54 (dd, 1H, J = 15.4, 5.7 Hz), 2.98 (dd, 1H, J = 13.1,
9.9 Hz), 2.26–2.13 (m, 2H), 2.08 (d, 1H, J = 14.0 Hz); 13C
NMR (DMSO-d6) d 171.4, 147.2, 136.1, 127.0, 124.0,
121.0, 118.5, 118.4, 115.4, 111.4, 106.7, 65.5, 64.5, 49.7,
42.9, 33.3, 30.6, 29.9; MS (EI) m/z 314 (10, M+), 130
(55); HRMS (EI) m/z calcd for C18H22N2O3: 314.163043;
found: 314.163128.
4.2.26. 6-Hydroxymethyl-2,3,4,7-tetrahydro-azepine-1,3-
dicarboxylic acid 1-tert-butyl ester 3-methyl ester (38)
Compound 20 (0.850 g, 2.13 mmol) was dissolved in a
30:1 mixture of THF and aqueous HCl 32% and stirred
at room temperature for 40 min. After addition of a satu-
rated NaHCO3-solution, the reaction mixture was
extracted three times with ethyl acetate. The combined
organic layers were washed with brine, dried over MgSO4,
filtered, and concentrated under vacuum. Column chroma-
tography of the oily residue afforded 0.481 g (80%) of 38 as
4.2.24. 5-Acetoxymethyl-3-(2-1H-indol-3-yl-
acetoxymethyl)-2,3,4,7-tetrahydro-1H-azepinium chloride
(36b)
1
A solution of 35b (0.345 g, 0.76 mmol, 1.0 equiv.) in a
2 M solution of hydrogen chloride in diethyl ether
(7.6 mL, 15.11 mmol, 20.0 equiv.) was stirred for 8 h at
room temperature. The reaction mixture was filtered
under a positive stream of argon and the resulting residue
was thoroughly washed with diethyl ether. Column chro-
matography (DCM/MeOH: 92:8) of the brownish residue
gave rise to 0.197 g (66%) of 36b as a white, hygroscopic
colourless oil: H NMR d (rotamers, ratio: ꢁ3:1) 5.83 (t,
0.75H, J = 6.0 Hz), 5.79–5.74 (bs, 0.24H), 4.12–4.01 (m,
4H), 3.95 (d, 0.74H, J = 12.1 Hz), 3.75 (s, 0.6H), 3.73–
3.67 (m, 3.6H), 3.59–3.51 (sm, 0.21H), 3.45 (dd, 0.69H,
J = 14.0, 9.6 Hz), 2.72–2.64 (m, 1.26H), 2.57–2.48 (m,
1.28H), 2.48–2.39 (m, 1H), 1.46 (s, 9H); 13C NMR d (rota-
mers) 173.9, 173.6, 155.5, 155.1, 142.9, 141.4, 125.9, 124.2,
80.6, 80.0, 66.3, 66.1, 51.85, 51.79, 50.8, 48.9, 47.5, 47.4,
43.0, 42.7, 28.25, 28.20, 27.4; MS (ES+) m/z 308 (56,
[M+Na]+), 593 (100, [2M+Na]+); HRMS (ES+) m/z calcd
for C14H23NO5Na: 308.147393; found: 308.148530.
1
powder: m.p. 35–40 ꢁC; H NMR (DMSO-d6) d 10.99 (s,
1H), 9.35 (bs, 1H), 9.05 (bs, 1H), 7.50 (d, 1H,
J = 7.8 Hz), 7.36 (d, 1H, J = 8.3 Hz), 7.26 (d, 1H,
J = 2.3 Hz), 7.08 (t, 1H, J = 7.5 Hz), 7.00 (t, 1H,
7.0 Hz), 5.74 (t, 1H, J = 5.3 Hz), 4.40 (s, 2H), 3.99 (d,
2H, J = 6.0 Hz), 3.78 (s, 2H), 3.73–3.65 (m, 1H), 3.64–
3.54 (m, 1H), 3.07–2.96 (m, 1H), 2.35–2.16 (m, 3H),
2.09–1.99 (m, 1H), 2.04 (s, 3H). 13C NMR (DMSO-d6) d
171.4, 169.8, 141.4, 136.1, 127.0, 124.0, 121.0, 119.8,
118.5, 118.4, 111.4, 106.7, 66.8, 65.3, 49.4, 42.5, 33.1,
30.6, 30.0, 20.5; MS (ES+) m/z 357 (100, [M+H]+);
HRMS (ES+) m/z calcd for C20H25N2O4: 357.181433;
found: 357.181830.
4.2.27. 6-(2-1H-Indol-3-yl-acetoxymethyl)-2,3,4,7-
tetrahydro-azepine-1,3-dicarboxylic acid 1-tert-butyl ester 3-
methyl ester (39)
To a stirred solution of indole-3-acetic acid (0.252 g,
1.44 mmol, 1.2 equiv.) in a 1:1 mixture of THF and
DCM were added 38 dissolved in above described mixture
(0.342 g, 1.20 mmol, 1.0 equiv.) and DMAP (0.008 g,
0.06 mmol, 5 mol%) followed by the addition of DCC
(0.322 g, 1.56 mmol, 1.3 equiv.) in a 1:1 mixture of THF

S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
5421
and DCM. The reaction mixture was stirred at room tem-
perature for 16 h. After careful addition of a saturated
NaHCO3-solution, the reaction mixture was extracted
three times with diethyl ether, the combined organic layers
were washed twice with brine, dried over MgSO4, filtered,
and concentrated under reduced pressure. Column chro-
matography (hexane/ethyl acetate: 3:1) afforded 0.510 g
(96%) of 39 as a colourless oil: 1H NMR d (rotamers, ratio:
ꢁ1:1) 8.20 (psd, 1H, J = 7.6 Hz), 7.61 (d, 1H, J = 7.8 Hz),
7.35 (t, 1H, J = 7.8 Hz), 7.22–7.11 (m, 3H), 5.75 (dd, 1H,
J = 12.3, 6.0 Hz), 4.53 (m, 2H), 4.06 (d, 0.5H,
J = 17.2 Hz), 3.95–3.87 (m, 1H), 3.84 (d, 0.5H,
J = 17.2 Hz), 3.79 (d, 2H, J = 2.8 Hz), 3.76–3.71 (m, 1H).
3.69 (s, 1.57H), 3.67 (s, 1.44H), 3.56–3.50 (sm, 1H), 2.88–
2.81 (m, 1H), 2.46–2.38 (m, 2H), 1.46 (s, 4.41H), 1.43 (s,
4.57H); 13C NMR d (rotamers) 173.84, 173.76, 171.6,
155.2, 155.1, 136.7, 136.1, 135.9, 127.7, 127.14, 127.08,
126.7, 123.2, 123.1, 122.1, 122.0, 119.54, 119.50, 118.8,
118.7, 111.2, 111.1, 108.1, 80.1, 67.9, 51.9, 48.6, 48.3,
48.1, 47.7, 43.2, 42.6, 31.30, 31.26, 28.35, 28.30, 27.3,
26.1; MS (ES+) m/z 465 (100, [M+Na]+), 907 (74,
[2M+Na]+); HRMS (ES+) m/z calcd for C24H34N3O6
([M+NH4]+): 460.244761; found: 460.244144.
rung des wissenschaftlichen Nachwuchses’’ for W.E.D. The
authors in particular would like to thank Prof. Dr. G.
Klebe, Philipps-Universita¨t Marburg, for his generous sup-
port of their research.
References
[1] (a) For recent reviews on ring closing metathesis see: M. Schuster, S.
Blechert, Angew. Chem., Int. Ed. 36 (1997) 2036–2056;
(b) R.H. Grubbs, S. Chang, Tetrahedron 54 (1998) 4413–4450;
(c) A. Furstner, Top. Organomet. Chem. 1 (1998) 37–72;
¨
(d) A. Furstner, Angew. Chem., Int. Ed. 39 (2000) 3012–3043;
¨
(e) T.M. Trnka, R.H. Grubbs, Acc. Chem. Res. 34 (2001) 18–29.
[2] Available from: <http://nobelprize.org/nobel_prizes/chemistry/laure-
ates/2005/press.html>.
[3] (a) For reviews see: A.J. Phillips, A.D. Abell, Aldrichim. Acta 32
(1999) 75–89;
(b) A. Nadin, J. Chem. Soc., Perkin Trans. 1 (1998) 3493–3513;
(c) A. Mitchinson, A. Nadin, J. Chem. Soc., Perkin Trans. 1 (1999)
2553–2581;
(d) A. Mitchinson, A. Nadin, J. Chem. Soc., Perkin Trans. 1 (2000)
2862–2892;
(e) F.-X. Felpin, J. Lebreton, Eur. J. Org. Chem. (2003) 3693–3712;
(f) A. Deiters, S.F. Martin, Chem. Rev. 104 (2004) 2199–2238.
[4] S. Brass, H.-D. Gerber, S. Do¨rr, W.E. Diederich, Tetrahedron 62
(2006) 1777–1786.
[5] (a) B. Greenwood, T. Mutabingwa, Nature 415 (2002) 670–672;
(b) J. Sachs, P. Malaney, Nature 415 (2002) 680–685;
(c) R.W. Snow, M. Craig, U. Deichmann, K. Marsh, Bull. World
Health Organ. 77 (1999) 624–640;
4.2.28. 6-(2-1H-Indol-3-yl-acetoxymethyl)-3-
methoxycarbonyl-2,3,4,7-tetrahydro-1H-azepinium chloride
(40)
(d) World Health Organisation/UNICEF The Africa Malaria report
(2003) 1–112.
A solution of 39 (0.372 g, 0.84 mmol, 1.0 equiv.) in a
2 M solution of hydrogen chloride in diethyl ether
(8.41 mL, 1.82 mmol, 20.0 equiv.) was stirred for 20 min
at 0 ꢁC. The reaction mixture was allowed to reach room
temperature and stirred for an additional 3 h. The reaction
mixture was filtered under a positive stream of argon and
thoroughly washed with diethyl ether. Column chromatog-
raphy (DCM/MeOH: 92:8) of the brownish residue affor-
ded 0.292 g (92%) of 40 as a white, hygroscopic powder:
m.p. 50–60 ꢁC; 1H NMR (DMSO-d6) d 11.0 (s, 1H),
9.50–9.10 (bs, 2H), 7.49 (d, 1H, J = 8.0 Hz), 7.36 (d, 1H,
J = 8.0 Hz), 7.25 (d, 1H, J = 2.3 Hz), 7.08 (d, 1H,
J = 7.1 Hz), 7.00 (d, 1H, J = 7.1 Hz), 6.03 (t, 1H,
J = 6.2 Hz), 4.52 (s, 2H), 3.78 (s, 2H), 3.67 (d, 1H,
J = 15.6 Hz), 3.64 (s, 3H), 3.60 (d, 1H, J = 15.6 Hz), 3.47
(dd, 1H, J = 13.4, 4.1 Hz), 3.40–3.33 (m, 1H), 3.06–2.98
(sm, 1H), 2.64–2.58 (m, 2H); 13C NMR (DMSO-d6) d
171.9, 171.2, 136.0, 132.1, 130.9, 127.0, 124.1, 121.0,
118.44, 118.37, 111.4, 106.7, 66.8, 52.1, 18.4, 44.6, 39.1,
30.6, 27.4; MS (EI) m/z 342 (20, M+), 168 (68), 130
(100), 108 (72); HRMS (ES+) m/z calcd for C19H22N2O4:
342.157957; found: 342.157728. Anal. Calc. for
C19H22N2O4 Æ HCl Æ 1H2O: C, 57.50; H, 6.35; N, 7.06.;
found: C, 57.20; H, 6.34; N, 7.09%.
[6] C. Laurent, A. Bourgeois, M. Mpoudi, C. Butel, M. Peeters, E.
Mpoudi-Ngole’, E. Delaporte, Emerg. Infect. Dis. 10 (2004) 1953–
1956.
[7] (a) S.E. Francis, D.J. Sullivan Jr., D.E. Goldberg, Annu. Rev.
Microbiol. 51 (1997) 97–123;
(b) G.H. Coombs, J.C. Mottram, Parasitology 114 (Suppl.) (1997)
S61–S80;
(c) D. Leung, G. Abbenante, D.P. Fairlie, J. Med. Chem. 43 (2000)
305–341;
(d) K.A. Werbovetz, Curr. Med. Chem. 7 (2000) 835–860;
(e) P.J. Rosenthal, Emerg. Infect. Dis. 4 (1998) 49–57.
[8] (a) S.E. Francis, I.Y. Gluzman, A. Oksman, A. Knickerbocker, R.
Mueller, M.L. Bryant, D.R. Sherman, D.G. Russell, D.E. Goldberg,
EMBO J. 13 (1994) 306–317;
(b) R. Banerjee, J. Liu, W. Beatty, L. Pelosof, M. Klemba, D.E.
Goldberg, Proc. Natl. Acad. Sci. USA 99 (2002) 990–995;
(c) K. Ersmark, I. Feierberg, S. Bjelic, E. Hamelink, F. Hackett,
M.J. Blackman, J. Hulten, B. Samuelsson, J. Aqvist, A. Hallberg, J.
Med. Chem. 47 (2004) 110–122;
(d) J. Liu, I.Y. Gluzman, M.E. Drew, D.E. Goldberg, J. Biol. Chem.
280 (2005) 1432–1437.
[9] E. De Clercq, J. Med. Chem. 48 (2005) 1297–1313.
[10] (a) D. Leung, G. Abbenante, D.P. Fairlie, J. Med. Chem. 43 (2000)
305–341;
(b) D.H. Rich, J. Med. Chem. 28 (1985) 263–273;
(c) J.B. Cooper, Current Drug Targets 3 (2002) 155–173.
[11] E. Specker, J. Bo¨ttcher, S. Brass, A. Heine, H. Lilie, A. Schoop, G.
Muller, N. Griebenow, G. Klebe, ChemMedChem 1 (2006) 106–117.
¨
[12] E. Specker, J. Bo¨ttcher, H. Lilie, A. Heine, A. Schoop, G. Muller, N.
¨
Griebenow, G. Klebe, Angew. Chem., Int. Ed. 44 (2005) 3140–3144.
[13] (a) Compound 13 has already been synthesized via a different
sequence: A. Khazaei, R.G. Vaghei, E. Karkhanei, Synth. Commun.
32 (2002) 2107–2113;
Acknowledgements
We gratefully acknowledge the financial support of our
research by the Deutsche Pharmazeutische Gesellschaft
(DPhG) within the fellowship ‘‘DPhG Stiftung zur Fo¨rde-
(b) P.B. Anzeveno, J.A. Campbell, W.L. White, Synth. Commun. 16
(1986) 387–392;

5422
S. Brass et al. / Journal of Organometallic Chemistry 691 (2006) 5406–5422
(b) U. Majumder, J.M. Cox, J.D. Rainier, Org. Lett. 5 (2003) 913–
(c) H. Saimoto, K. Nishio, H. Yamamoto, M. Shinoda, T. Hiyama,
H. Nozaki, Bull. Chem. Soc. Jpn. 56 (1983) 3093–3098;
916;
(d) H. Duettmann, P. Weyerstahl, Chem. Ber. 112 (1979) 3480–3485.
[14] H.D. Maynard, R.H. Grubbs, Tetrahedron Lett. 40 (1999) 4137–
4140.
[15] For a representative procedure see: T.A. Kirkland, R.H. Grubbs, J.
Org. Chem. 62 (1997) 7310–7318.
[16] (a) Compound 21 has already been described in the literature: A.B.
Smith III, P.A. Levenberg, P.J. Jerris, R.M. Scarborough, P.M.
Wovkulich, J. Am. Chem. Soc. 103 (1981) 1501–1513, the described
method seems to be less convenient;
(c) A.K. Chatterjee, F.D. Toste, S.D. Goldberg, R.H. Grubbs, Pure
Appl. Chem. 75 (2003) 421–425;
(d) M. Marhold, A. Buer, H. Hiemstra, J.H.v. Maarseveen, G.
Haufe, Tetrahedron Lett. 45 (2004) 57–60.
[18] (a) A. Taylor, D.P. Brown, S. Kadam, M. Maus, W.E. Kohlbrenner,
D. Weigl, M.C. Turon, L. Katz, Appl. Microbiol. Biotechnol. 37
(1992) 205–210;
(b) M.V. Toth, G.R. Marshall, Int. J. Pept. Prot. Res. 36 (1990) 544–
550.
(b) S. Brandau, R. Fro¨hlich, D. Hoppe, Tetrahedron Lett. 46 (2005)
6709–6711.
[17] (a) M. Scholl, S. Ding, C.W. Lee, R.H. Grubbs, Org. Lett. 1 (1999)
953–956;
[19] S. Weik, T. Luksch, A. Evers, J. Bo¨ttcher, C.A. Sotriffer, A. Hasilik,
H.-G. Lo¨ffler, G. Klebe, J. Rademann, ChemMedChem 1 (2006)
445–457.
[20] Y.M. Ahn, K.-L. Yang, G.I. Georg, Org. Lett. 3 (2001) 1411–1413.