Design, synthesis, and biological evaluation of simplified side chain hybrids of the potent actin binding polyketides rhizopodin and bistramide

Article abstract of DOI:10.1002/cmdc.201402508

The natural products rhizopodin and bistramide belong to an elite class of highly potent actin binding agents. They show powerful antiproliferative activities against a range of tumor cell lines, with IC₅₀ values in the low-nanomolar range. At the molecular level they disrupt the actin cytoskeleton by binding specifically to a few critical sites of G-actin, resulting in actin filament stabilization. The important biological properties of rhizopodin and bistramide, coupled with their unique and intriguing molecular architectures, render them attractive compounds for further development. However, this is severely hampered by the structural complexity of these metabolites. We initiated an interdisciplinary approach at the interface between molecular modeling, organic synthesis, and chemical biology to support further biological applications. We also wanted to expand structure-activity relationship studies with the goal of accessing simplified analogues with potent biological properties. We report computational analyses of actin-inhibitor interactions involving molecular docking, validated on known actin binding ligands, that show a close match between the crystal and modeled structures. Based on these results, the ligand shape was simplified, and more readily accessible rhizopodin-bistramide mimetics were designed. A flexible and modular strategy was applied for the synthesis of these compounds, enabling diverse access to dramatically simplified rhizopodin-bistramide hybrids. This novel analogue class was analyzed for its antiproliferative and actin binding properties.

Full text of DOI:10.1002/cmdc.201402508

DOI: 10.1002/cmdc.201402508
Full Papers
Design, Synthesis, and Biological Evaluation of Simplified
Side Chain Hybrids of the Potent Actin Binding
Polyketides Rhizopodin and Bistramide
Daniel Herkommer,[a] Sandra Dreisigacker,[a] Galina Sergeev,[b] Florenz Sasse,[b]
Holger Gohlke,[c] and Dirk Menche*[a]
The natural products rhizopodin and bistramide belong to an
elite class of highly potent actin binding agents. They show
powerful antiproliferative activities against a range of tumor
cell lines, with IC50 values in the low-nanomolar range. At the
molecular level they disrupt the actin cytoskeleton by binding
specifically to a few critical sites of G-actin, resulting in actin fil-
ament stabilization. The important biological properties of rhi-
zopodin and bistramide, coupled with their unique and intrigu-
ing molecular architectures, render them attractive compounds
for further development. However, this is severely hampered
by the structural complexity of these metabolites. We initiated
an interdisciplinary approach at the interface between molecu-
lar modeling, organic synthesis, and chemical biology to sup-
port further biological applications. We also wanted to expand
structure–activity relationship studies with the goal of access-
ing simplified analogues with potent biological properties. We
report computational analyses of actin–inhibitor interactions
involving molecular docking, validated on known actin binding
ligands, that show a close match between the crystal and
modeled structures. Based on these results, the ligand shape
was simplified, and more readily accessible rhizopodin–bistra-
mide mimetics were designed. A flexible and modular strategy
was applied for the synthesis of these compounds, enabling
diverse access to dramatically simplified rhizopodin–bistramide
hybrids. This novel analogue class was analyzed for its antipro-
liferative and actin binding properties.
Introduction
Along with tubulin, actin is one of the two major components
of the eukaryotic cell cytoskeleton. It is involved in separation
during cell division and in the organization of the cell shape
and polarity. It also plays an essential role in cell locomotion,
as well as intracellular transportation.[1] Actin structures are
constantly assembled and disassembled in a reversible manner
based on a dynamic and sophisticated polymerization/depoly-
merization equilibrium between monomeric globular actin (G-
actin) and double-stranded filamentous actin (F-actin). This
process is regulated by a series of actin binding proteins with
specific functions.[2] Malfunction in these kinetics has been as-
sociated with various diseases,[3] including cancer[4] as well as
bacterial and viral (e.g., HIV) infections.[5] This makes the devel-
opment and molecular-level understanding of potent actin
binding agents important research goals, both for potential
new therapeutic agents and as biological tools in advancing
the understanding and cellular function of actin.[6]
As shown in Figure 1, G-actin is composed of four main sub-
units: domains 1–4. Assembly of these monomers proceeds in
a bidirectional manner, with subunits 1 and 3 being at one
[a] D. Herkommer, Dr. S. Dreisigacker, Prof. Dr. D. Menche
Kekulꢀ-Institut fꢁr Organische Chemie und Biochemie
Universitꢂt Bonn, Gerhard-Domagk-Str. 1, 53121 Bonn (Germany)
E-mail: dirk.menche@uni-bonn.de
[b] Dr. G. Sergeev, Dr. F. Sasse
Chemische Biologie
Helmholtz-Zentrum fꢁr Infektionsforschung GmbH
Inhoffenstr. 7, 38124 Braunschweig (Germany)
Figure 1. Structure of molecular actin (globular actin; G-actin) with two
binding sites for actin polymerization (actin structure taken from the RCSB
Protein Data Bank (PDB) ID 1QZ5): a ‘pointed end’ and a ‘barbed end’. The
four subdomains are shown in different colors. The ATP binding site (fawn)
is located between domains 2 and 4; the binding site for the regulatory pro-
tein gelsolin lies between domains 1 and 3. The ‘barbed end’ inhibitors ad-
dress the binding site of gelsolin (exemplarily shown for kabiramide C (moss
green; PDB ID: 1QZ5)).
[c] Prof. Dr. H. Gohlke
Institut fꢁr Pharmazeutische und Medizinische Chemie
Heinrich-Heine-Universitꢂt Dꢁsseldorf
Universitꢂtsstr. 1, 40225 Dꢁsseldorf (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201402508.
ChemMedChem 0000, 00, 0 – 0
1
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
end (’barbed end’) and subunits 2 and 4 at the other side
(’pointed end’) of the elongating chain.[7] This process is con-
trolled by regulatory proteins, such as gelsolin, which binds be-
tween domains 1 and 3, and by hydrolysis of adenosine tri-
phosphate (ATP), which is essential for this process and takes
place in a cleft between subunits 2 and 4.
to the attractiveness for further structure–activity relationship
(SAR) studies and inhibitor design.
However, further promotion of these highly potent agents is
severely hampered by the extremely low natural supply of
these scarce natural products. Moreover, structural complexi-
ties impede large-scale synthetic access to enable further bio-
logical evaluation.[20–23] This renders the development of more
readily available analogues critical to enhancing further biolog-
ical evaluations of these highly promising lead structures. Such
development will also allow establishment of further
SARs[12f,17d,24–27] to enhance the preclinical advancement of
these agents.
Nature has provided an important source of structurally
complex and novel actin binding molecules with filament-sta-
bilizing agents that address the gelsolin binding site.[8,9] As
shown in Figure 2, these polyketide natural products include
the trisoxazoles (i.e., kabiramides 1),[10] jaspisamides 2, ulapua-
lides 3 and halichondramides (not shown),[11] the reidispongio-
lide/sphinxolide family 4–6,[12] the swinholides 7,[13] aplyronines
8,[14] bistramides 9,[15] lobophorolides 10,[16] and rhizopodin
11.[17] All these compounds have been isolated from marine
sources, except for rhizopodin, which was isolated from myxo-
bacteria by the research groups of Hçfle and Reichenbach at
the HZI in Braunschweig.[17a] At the molecular level, they act as
actin-filament-stabilizing agents.[8] They demonstrate highly
Herein, we report in silico studies of the noncovalent actin
interactions of these complex polyketides, as well as the
design, synthesis, and biological evaluation of a structurally
simplified novel analogue class, which is based on a hybrid
structure of bistramide and rhizopodin.
potent binding affinities for G-actin at low-nanomolar concen- Results and Discussion
trations. Furthermore, they exhibit impressive antiproliferative
Modeling noncovalent actin–inhibitor interactions
activities against a range of human cancer cell lines with IC50
values in the low-nanomolar range. The binding events and
molecular influence of these inhibitors on dynamic actin as-
sembly are increasingly well understood. Importantly, X-ray
analyses of the corresponding inhibitor–actin complexes have
been reported,[18,19] which allow a detailed perception of
target–inhibitor interactions at the molecular level. This adds
Analysis of the X-ray structures of actin–ligand complexes re-
vealed three possible binding sites in the cleft between subdo-
mains 1 and 3, which are mainly hydrophobic in nature.[18,19] As
shown in Figure 3, the first one is represented by the gelsolin
binding site and involves residues Y166, G168, Y169, Y143,
T148, T149, I345, L346, L349, T351, F352, and M355. The
Figure 2. Complex polyketides binding in the cleft between subunits 1 and 3 of G-actin. PDB IDs of the corresponding co-crystals with actin are shown in pa-
rentheses. Ac=acetate.
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
2
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
ly, the binding modes of these
structurally similar side chains in
the crystal structures are nearly
identical, although the binding
modes of the macrocyclic parts
of these compounds can be dif-
ferent.
The research groups of
Yamada and Marriott have re-
ported that the side chains
alone do not retain the activity
of the parent natural pro-
ducts.[12f,26,27] Marriott and co-
workers also reported the evalu-
ation of a library of tail-region
mimetics, demonstrating the sig-
nificance of an appending hy-
drophobic protecting group on
the terminal hydroxy groups for
binding affinity, while the side
chain alone is inactive.[12f,26] This
is in agreement with an observa-
tion reported by Nicolaou and
colleagues, who have shown
that monomeric rhizopodin re-
tains the actin binding proper-
ties of the authentic dimeric nat-
ural product.[24]
Figure 3. Different perspectives of the possible binding pockets in the cleft between subunits 1 and 3 of G-actin,
as derived by X-ray crystallography of actin–inhibitor complexes, as previously reported.[10b,12e,15f] a) Front perspec-
tive with bistramide A and gelsolin, b) left-side perspective with bistramide A and gelsolin, c) right side perspec-
tive with reidispongiolide A, and d) right side perspective with kabiramide C. The binding pocket for gelsolin is
shown in olive, the one for bistramide A in blue, and the one for macrocyclic ligands in green. The binding site of
gelsolin partially overlaps with the binding site of macrocyclic ligands as well as with the binding site of bistrami-
de A.[29]
Thus, the overlapping region
of bistramide A and the side
chain of the other barbed-end
inhibitors around residues G168,
second one partially overlaps with the gelsolin binding site
and also includes residues Y133, I136, V139, A170, L171, P172,
and F375 inside the hydrophobic pocket at the ‘left’ of the
cleft (Figure 3a,b). This binding site is addressed by the bislac-
tam polyether bistramide A (9).[15f] In contrast, all macrocyclic li-
gands shown in Figure 2, i.e., the trisoxazoles, the reidispon-
giolide/sphinxolide family, the swinholides, lobophorolides,
aplyronines, and rhizopodin, address a third binding domain
located at the ‘right’ of the cleft, also exhibiting partial overlap
with the gelsolin binding site (Figure 3c,d).[10b,12e,15f,18,19,28]
Furthermore, these ligands comprise a side chain, which par-
tially overlaps with bistramide A as well as with the gelsolin
binding site (exemplarily depicted for rhizopodin in Figure 4).
In contrast, the macrocycle of these compounds does not
seem to show significant overlap with the bistramide A ligand.
The side chains of the trisoxazoles, reidispongiolide/sphinxo-
lides, aplyronines, and rhizopodin bear remarkable similarities,
which has been well documented[11c,d] and explicitly not-
ed.[8,12d,f,17a,18] Importantly, the tail region has been defined as
the functional unit of the severing and capping activities of
the parent molecule.[8b,12f] In detail, all tail regions have an N-
methylvinylformamide group at the terminus, two oxygen sub-
stituents at C5 and C11, together with two methyl-bearing cen-
ters at C6 and C10, and a carbonyl at C7 (Figure 5). According-
Figure 4. X-ray-derived structural overlap of the binding sites of bistrami-
de A (grey) and the side chains of the macrocyclic ‘barbed end’ inhibitors,
exemplarily shown for rhizopodin (black).
Y169, Y143, T148, I345, L346, L349, T351, F352, and M355, as
well as Y133, I136, V139, A170, and F375, appears to be essen-
tial for the noncovalent interactions at the actin protein and
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
3
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Rhizopodin and swinholide A are dimeric C2-symmet-
ric macrolides; however, their constitutions, particu-
larly of the side chains, are different. The binding
energy obtained for lobophorolide A (3M6G) of
ꢀ11.8 kcalmolꢀ1 indicates a poor ligand–receptor in-
teraction for this small ligand. However, its binding
efficiency (ꢀ0.22 kcalmolꢀ1) is similar to that of rhizo-
podin (2VYP). This discrepancy most likely arises from
the greater number of target–ligand interactions
formed by larger ligands than smaller ones, and from
the scoring process neglecting changes in configura-
tional entropy upon binding. Therefore, in our study
we not only considered the overall binding energy,
but also the binding efficiency in relation to the
number of heavy atoms. Importantly, only small dif-
ferences between the calculated binding modes and positions
obtained from X-ray crystallographic data were observed,
which supports the reliability of this in silico model (see Sup-
porting Information for details).
Figure 5. Structural comparison of the vinylformamide side chains of the inhibitors.
Table 1. Binding energies of the highest populated clusters obtained
from re-docking of the ‘barbed end’ inhibitors.
Figure 6. X-ray-derived ligand–receptor interactions of the side chain of
‘barbed end’ inhibitors with actin. As an example, the side chain and part of
the macrocycle of rhizopodin are shown (blue).
Ligand
(PDB ID)
Ebind
[kcalmolꢀ1
Binding Eff.
No. Heavy Convergence[d]
Atoms[c]
[a]
[b]
]
[kcalmolꢀ1
]
therefore presents a promising starting point for the design of
effective inhibitors (Figure 6).
1 (1QZ5)
2 (1QZ6)
3 (1S22)
ꢀ18.9
ꢀ18.1
ꢀ19.0
ꢀ12.8
ꢀ31.9
ꢀ17.6
ꢀ13.1
ꢀ18.5
ꢀ17.8
ꢀ21.6
ꢀ11.8
ꢀ0.28
ꢀ0.30
ꢀ0.30
ꢀ0.17
ꢀ0.33
ꢀ0.26
ꢀ0.19
ꢀ0.28
ꢀ0.36
ꢀ0.21
ꢀ0.22
67
61
63
76
98
68
68
66
50
104
54
74
82
29
4
96
81
8
81
3
3
We first wanted to evaluate, whether docking by a combina-
tion of AutoDock 3.0[30] for binding mode generation, and the
DrugScore scoring function[31] for evaluating the binding
modes, may be useful for a predictive analysis of this system.
Therefore, all inhibitors shown in Figure 2 were re-docked into
the actin binding pockets of their corresponding crystal struc-
tures. The resulting binding modes were clustered with a root-
mean-square deviation (RMSD) value of 1.0 ꢁ (see the Experi-
mental Section and Supporting Information for details). The
calculated binding energies of the highest populated clusters
and the binding efficiencies with regard to the number of
heavy atoms are listed in Table 1.
8 (1WUA)
7 (1YXQ)
4 (2ASM)
6 (2ASO)
5 (2ASP)
9 (2FXU)
11 (2VYP)
10 (3M6G)
35
[a] Mean docked binding energy of all solutions in the highest populated
cluster; the search area included 90 grid points in each dimension,[32]
with a grid point distance of 0.375 ꢁ. [b] Binding efficiency: Ebind divided
by the number of heavy atoms. [c] Number of heavy atoms (C, N, O) in
the ligand structure. [d] Docking convergence: number of structures in
the highest populated cluster, clustered with an RMSD value of 1 ꢁ, out
of 100 GA runs.
Analysis of the mean docked energies (red line) revealed the
best ligand–receptor interaction for the dimeric ligands, such
as swinholide A (PDB ID: 1YXQ) and rhizopodin (2VYP), with re-
spective binding energies of ꢀ31.9 and ꢀ21.6 kcalmolꢀ1. How-
ever, as for the binding efficiencies (blue line) of these differ-
ently large ligands, rhizopodin is less potent, with ꢀ0.22
kcalmolꢀ1, whereas bistramide A (2FXU) was most effective,
showing a binding efficiency of ꢀ0.36 kcalmolꢀ1 in agreement
with the excellent activity data observed for this compound.
As discussed above (Figure 5), there are striking structural
similarities in the side chains of the macrolides shown in
Figure 2, all binding in an overlapping region with the gelsolin
binding site. Consequently, the significance of the essential
amino acids within this binding domain for the interactions
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
4
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
As shown in Figure 8, three main fragments were differenti-
ated during this analogue design. The first (fragment 1) would
address the bistramide binding site only, the middle (frag-
ment 2) would bind to the overlapping region, while the third
(fragment 3) would target the internal rhizopodin pocket. Con-
sequently, this strategy resulted in three different fragments:
a bistramide-derived fragment, a fragment that represents the
overlapping region, and a rhizopodin-derived fragment.
According to this strategy, different structural frameworks
were docked into the binding pocket of the crystal structure of
PDB ID: 1WUA. The position of the N1 atom in all crystal struc-
tures is highly conserved between the three tyrosines Y143,
Y169, and Y133. Therefore, an amidic nitrogen was retained at
this position (fragment 3, Figure 8). As chain extension, we
chose the authentic side chain of rhizopodin (11), which lacks
methyl groups at C4 and C12 relative to the other inhibitors
with vinylformamide side chains (building block D). It was envi-
sioned that fragments 1 and 2 should functionally address the
back part of the cavity of the bistramide and the overlapping
binding site, respectively (compare Figure 3a,b with Figure 4).
For this purpose, the middle part of the bistramide A ligand
structure was directly retained (building block B) and slightly
modified in order to evaluate the flexibility of this segment on
Figure 7. X-ray-derived position of the vinylformamide side chain of rhizopo-
din in the gelsolin binding pocket. The gelsolin binding pocket residues are
shown in olive, those of the bistramide binding pocket in blue, and all
others are in grey. The region of the docking site of the macrocyclic-type li-
gands is presented in green.
with the ligands from the crystal structures was ana-
lyzed in silico. For all structures, this revealed interac-
tions of the vinylformamide side chain with residues
G168, Y169, Y143, T148, I345, L346, L349, T351, F352,
and M355 (Figure 7). In contrast, residues Y166 and
T149 included in the gelsolin binding site are located
at the outside of the cavity and are therefore not ad-
dressed by the smaller ligand side chains. However,
the less bulky side chain of the ligands reaches fur-
ther inside the pocket than gelsolin, resulting in addi-
tional interactions with residues Y133, I136, V139,
A170, and F375[33] of the hydrophobic bistramide
binding pocket. Interestingly, analysis of the ligand
structures containing an amide function (either vinyl-
formamide or simply amide) also revealed polar inter-
actions with tyrosines Y143 and Y133, in agreement
with the crystal structures.
Analogue design
Given the excellent potencies of bistramide and rhi-
zopodin (see above), these two agents were selected
for further analogue studies. It was envisioned that
a particularly promising analogue would result from
a combination of these two agents, in a rationale to
address both the bistramide and the rhizopodin
binding sites. Notably, these compounds present
novel structural analogue types that are different
from those previously reported. The previous ana-
logues are based only on the macrocylic tail.[12f,26,27]
Consequently, a common pharmacophore model was
pursued, and simplified bistramide–rhizopodin hybrid
structures were targeted.
Figure 8. Design of simplified bistramide–rhizopodin hybrid structures based on
a common pharmacophore model: template structure and modular building blocks.
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
5
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
the binding affinity. Concerning fragment 1, the tetrahydropy-
ran was replaced by a simplified piperidine (building block A).
In the crystal structure of actin-bound bistramide A, p–p inter-
actions between the terminal enone and Y169 were observed.
To retain this interaction, this fragment was terminated with
a phenyl ring (building block A). Finally, we attempted to find
a promising replacement for the side chain of rhizopodin. An
overlay of the crystal structures of bistramide A and swinholi-
de A, containing side chains that are different from rhizopodin,
showed that a bulky aliphatic residue may effectively interact
with the hydrophobic pocket consisting of T351, L346, F352,
and M355. Therefore, in addition to the rhizopodin side
chain D, building block E was also designed. This resulted in
four rationally designed compounds with modular assemblies
ABD, ACD, ABE, and ACE. The actin interactions of all of these
compounds were calculated in a cross-docking approach with
all available protein structures (see the Supporting Information
for further details). The resulting binding energies of the high-
est populated clusters and the binding efficiencies with regard
to the number of heavy atoms are summarized in Table 2.
All four analogues were predicted to show significantly im-
proved binding efficiencies in the range of ꢀ0.41 to ꢀ0.46 kcal
molꢀ1 as compared with rhizopodin and bistramide in connec-
tion with a decrease in the number of atoms. Interestingly, the
sequences ABE and ACE with the simplified building blocks ex-
hibit the highest binding efficiencies calculated: ꢀ0.45 and
ꢀ0.46 kcalmolꢀ1. Replacement of the authentic side chain of
rhizopodin (building block D, fragment 3) with the piperidine
fragment (building block E, fragment 3) resulted in slightly im-
proved binding efficiencies, which may suggest that the com-
plex fragment of the rhizopodin structure may be replaced
without loss of—or possibly even an increase in—potency. Re-
placement of the middle fragment (fragment 2) results in no
significant energetic differences.
The reason behind the good binding efficiencies for the
modular sequences ABE and ACE may be explained by the
binding modes. As exemplarily shown in Figure 9 for 12, the
simplified analogue shows all the essential interactions for
both the rhizopodin and bistramide binding sites. As anticipat-
ed, the N1 nitrogen is fixed between the three tyrosines Y143,
Y169, and Y133. The piperidine with the appending aliphatic
group of fragment 3 appears to fill the hydrophobic pocket
consisting of L346, F352, T351, and M355. Fragment 1 likewise
fits well into the hydrophobic pocket of the bistramide binding
site and interacts with residues V139, P109, L110, and P112.
In summary, these modeling studies suggest that smaller
side chain analogues may result in similar binding efficiencies
as the ‘barbed end’ inhibitors if the template incorporates all
essential interactions with the binding pocket of the protein.
Based on the calculated effective increase in binding efficien-
cies, all analogues 12–15 should form good bases for the
design of ligands with simplified core structures relative to the
Table 2. Binding energies and efficiencies of the rationally designed analogues of ‘barbed end’ actin inhibitors.[a]
[b]
[c]
Ligand
Sequence
ABE
Ebind [kcalmolꢀ1
]
Binding Eff. [kcalmolꢀ1
]
No. Heavy Atoms[d]
ꢀ17.8
ꢀ0.45
40
ACE
ABD
ACD
ꢀ18.1
ꢀ17.2
ꢀ18.3
ꢀ0.46
ꢀ0.41
ꢀ0.42
39
42
43
[a] Results from docking into the actin protein crystal structure (PDB ID: 1WUA). [b] Mean docked energy of all solutions in the highest populated cluster;
the search area included 90 grid points in each dimension; 0.375 ꢁ grid point distance; clustering by 1 ꢁ RMSD. [c] Binding efficiency: Ebind divided by the
number of heavy atoms. [d] Number of heavy atoms (C, N, O) in the ligand structure.
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
6
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 10. Retrosynthetic analysis.
Scheme 1. Synthesis of western fragment 16a. Reagents and conditions: a) S-
(+)-mandelic acid, EtOAc, reflux, 36%; b) K2CO3, BnBr, acetone, RT, 87%;
c) 6m HCl, RT, 99%.
Figure 9. a) 2D and b) 3D presentations of the calculated position of the
simplified hybrid analogue 12 with the fragment sequence (ABE) with actin.
Shown is the cluster representative of the highest populated cluster.
stereomeric salts with S-(+)-mandelic acid according to pub-
lished procedure.[34] Liberation of the amine was followed by
benzylation[35] with benzyl bromide. Subsequent ester cleavage
gave the desired compound 16 as its hydrochloride in nearly
quantitative yield.
authentic natural products. In addition to the complete struc-
tures consisting of all three fragments, the evaluation of single
or double fragments may also be instructive to delineate the
minimum chain length required for biological potency. Impor-
tantly, this strategy is highly flexible and modular; each of
these subunits can be readily replaced, thereby adding consid-
erable flexibility to this approach.
Synthesis of the middle fragments
For construction of the anti-propionate motif in the middle
fragment 17, a Masamune anti-aldol reaction with an appropri-
ate aldehyde derived from amino alcohol 25 was chosen
(Scheme 2).[36,37] The coupling proceeded in high diastereose-
lectivities and yields.[37] To allow facile liberation of the amine
at a later stage of the synthesis by mild hydrogenolysis, instal-
lation of a benzyloxycarbonyl (Cbz) protecting group was nec-
essary. After removal of the auxiliary of 26, the tert-butoxycar-
bonyl (Boc) group was removed by using trifluoroacetic acid
(TFA) before the Cbz group was installed by means of N-(ben-
zyloxycarbonyloxy)succinimide (CbzOSuc), to afford acid 17.
The Cbz analogue of the aldehyde may also be used directly,
but in this case the aldol reaction proceeded with considerably
lower yields and diastereoselectivities.
Chemistry
As shown in Figure 10, the proposed hybrid analogues can be
derived retrosynthetically from three main fragments: a western
fragment (corresponding to fragment 1), a middle fragment
(fragment 2), and an eastern fragment (fragment 3). These are
linked by amide functionalities to enable rapid access to ana-
logues for SAR studies.
Synthesis of the western fragment
As shown in Scheme 1, synthesis of the western fragment
started from racemic ethyl 3-piperidinylacetate (23). Chiral res-
olution was achieved by fractional crystallization using the dia-
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
7
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
good yields as a single isomer after careful chromato-
graphic separation (Scheme 5).[40,41]
After O-methylation, the auxiliary was replaced
with N,O-dimethylhydroxylamine to give a Weinreb
amide (not shown) in good yield following a proce-
dure that was likewise developed in our research
group.[42] Careful control of reaction conditions was
crucial to avoid elimination of the methoxy group
under the strongly basic reaction environment to
give an enone byproduct, which proved inseparable
Scheme 2. Synthesis of compound 17. Reagents and conditions: a) Boc2O, Al2O3, neat, RT,
99%; b) IBX, EtOAc, reflux, 91%; c) Masamune’s auxiliary, cHex2BOTf, Et3N, CH2Cl2, ꢀ788C,
90%, d.r.=96:4; d) NaOH, H2O2, tBuOH/MeOH, 08C, 99%; e) TFA, CH2Cl2, RT, 96%;
f) CbzOSuc, NaHCO3, acetone/H2O, RT, 97%. IBX=2-iodoxybenzoic acid, OTf=trifluoro-
methanesulfonate, Mes=mesityl.
Scheme 3. Synthesis of compound 18. Reagents and conditions: a) NaN3,
DMF, reflux, 85%; b) KOH, EtOH/H2O, 97%; c) H2, Pd/C, MeOH, RT, 96%;
d) CbzOSuc, NaHCO3, acetone/H2O, RT, 96%.
Synthesis of amino acid 18 involved preparation of azide 28
from chloride 27 according to a sequence developed by Haya-
shi and co-workers (Scheme 3).[38] Azide introduction proceed-
ed smoothly with sodium azide in N,N-dimethylformamide
(DMF). Saponification of the ethyl ester gave 28, which was
further elaborated to the amino acid by hydrogenolysis in the
presence of palladium on activated charcoal. For amine protec-
tion we relied again on a Cbz group, which was introduced in
excellent yield under the conditions described above to give
18.
Scheme 5. Synthesis of compound 21. Reagents and conditions: a) Masa-
mune’s auxiliary, cHex2BOTf, Et3N CH2Cl2, ꢀ788C, 89%; b) MeI, Ag2O, MS
(3 ꢁ), Et2O, RT, 92%; c) CH3NHOCH3·HCl, iPrMgCl, THF, ꢀ158C!08C, 85%;
d) (MeO)2P(O)Me, nBuLi, THF, ꢀ788C, 93%; e) NaHMDS, THF, ꢀ788C!RT,
97%; f) DDQ, CH2Cl2/buffer (pH 7), 08C!RT, 97%; g) MeI, Ag2O, MS (3 ꢁ),
Et2O, RT, 96%; h) CSA, MeOH, RT, 97%; i) PPh3, DIAD, DPPA, THF, 08C!RT,
86%; j) H2, Pd/C, HCl/MeOH, RT. TBS=tert-butyldimethylsilyl, MS=molecular
sieves, NaHMDS=sodium bis(trimethylsilyl)amide, DDQ=2,3-dichloro-5,6-di-
cyano-p-benzoquinone, PMB=para-methoxybenzyl, CSA=camphorsulfonic
acid, DIAD=diisopropyl azodicarboxylate, DPPA=diphenyl phosphoryl
azide.
Synthesis of the middle fragments 19 and 20 started from
Cbz-protected amino acid 29 (Scheme 4). For O-methylation io-
domethane and 2.0 equivalents of silver(I) oxide were used as
a mild base to prevent racemization of the labile a-hydroxy
moiety. Higher amounts of base also gave the N-methylated
building block 31, which can be separated by careful column
chromatography. Finally, saponification of the methyl esters
gave the desired compounds 19 and 20.[39]
by column chromatography. To set stage for the Horner–Wads-
worth–Emmons (HWE) reaction, the phosphonate moiety was
introduced with dimethyl methylphosphonate to give com-
pound 34. The HWE reaction with para-methoxybenzyl (PMB)-
protected Roche aldehyde 35[43] gave 36 in excellent yields
and exclusively as the E isomer, as determined from the
1H NMR spectra. After PMB deprotection of 36 with 2,3-di-
chloro-5,6-dicyano-p-benzoquinone (DDQ), the liberated hy-
droxy function was methylated with a combination of iodome-
thane and silver(I) oxide. Alternatively, the O-methylated Roche
aldehyde may also be used, but this resulted in con-
siderably lower yields; therefore, this longer se-
quence was chosen. Finally, the primary tert-butyldi-
methylsilyl (TBS) group was removed under mildly
acidic conditions. For introduction of the nitrogen,
displacement of the terminal alcohol with an azide,
bis-Boc-protected ammonia, or phthalimide were
evaluated under Mitsunobu conditions.[44,45] From
Synthesis of the eastern fragments
Efforts were next directed toward synthesis of the eastern frag-
ment 21, starting with a sequence previously developed in our
research group during the course of the total synthesis of rhi-
zopodin (11).[22c] We relied on a Masamune–Abiko aldol ap-
proach starting from aldehyde 32 (derived in two steps from
1,4-butanediol), to give an intermediate alcohol (not shown) in
Scheme 4. Synthesis of compounds 19 and 20. Reagents and conditions: a) MeI, Ag2O,
DMF, 58C, 78% (30/31=1:3); b) NaOH, acetone/H2O, RT, 97% (for 19) and 99% (for 20).
these alternatives, introduction of the azide worked
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
8
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
best, using diphenyl phosphoryl azide (DPPA) to give the de-
sired azide (not shown).
nate and sodium bis(trimethylsilyl)amide (NaHMDS) as base to
give the desired a,b-unsaturated nitrile 40 in good yields with
an E/Z ratio of 5:2.[51] The obtained mixture was then directly
subjected to a double reduction sequence, first of the double
bond to its saturated congener using hydrogen and palladium
on activated charcoal, then of the nitrile and amide functionali-
ties to the amines using lithium aluminum hydride, to give
building block 22 in good yields over these two steps.[52]
At this stage, however, subsequent efforts to further ad-
vance this azide to 21 were thwarted by a seemingly impossi-
ble reduction of the azide moiety to the amine. Among the
methods tested were the use of H2, Pd/C, Staudinger condi-
tions,[46] and Zn, NH4Cl.[47] However, all of these methods led to
decomposition or formation of a variety of undefined prod-
ucts. At this point we anticipated that the formed amine and
the carbonyl group could undergo side reactions to prevent
clean formation of 21. Indeed, if the reaction was carried out
under carefully controlled acidic conditions, leading to proto-
nation of the amine, 21a was formed in good yields, and the
obtained crude hydrochloric acid salt was immediately subject-
ed to amide couplings. Notably, the b-methoxy functionality
again proved prone to elimination under these conditions, so
that to some degree the eliminated product was formed.
For synthesis of the simplified eastern fragment 22, (R)-(ꢀ)-
3-piperidinecarboxylic acid (38) was identified as a suitable
starting material, already containing the required stereogenic
center (Scheme 6). Reduction[48] of the carboxylic acid yielded
a primary alcohol (not shown), which was further elaborated
Fragment union and completion of synthesis of the side chain
analogues
As shown in Scheme 7, for fragment union, our strategy relied
on DEPBT-mediated amide formation, which proceeded in a re-
liable manner. In detail, coupling of amine 22 with Cbz-pro-
tected amino acids 17–20 gave amides 41–44 in good yields
(76–93%). For carbamate removal, palladium-catalyzed hydro-
genation proved most efficient. Finally, for coupling of the li-
berated free amines with building block 16a, addition of
sodium carbonate as additional base was critical. Otherwise,
considerably lower yields (35–53%) were obtained. In summa-
ry, following this sequence, the side chain analogues 12, 13,
45, and 46 were prepared in a highly convergent manner over
nine linear steps.
The synthesis of the side chain analogues 14, 15, 51, and 52
was initiated by DEPBT-mediated amide formation of 21a with
17, 18, 19, and 20 to give 47–50 in pure form, after removal
of elimination products by HPLC if required (Scheme 8). Cleav-
age of the protecting group by hydrogenolysis followed by
DEPBT-mediated amide coupling with 16a gave the desired
products in good yields.
Biological results
Scheme 6. Synthesis of compound 22. Reagents and conditions: a) LiAlH4,
THF, reflux, 97%; b) isovaleric acid, DEPBT, Et3N, THF, RT, 85%; c) IBX, EtOAc,
reflux, 96%; d) (EtO)2P(O)CH2CN, NaHMDS, THF, ꢀ788C; e) H2, Pd/C, MeOH,
RT, 84% over two steps; f) LiAlH4, THF, reflux, 84%. DEPBT=3-(diethoxyphos-
phoryloxy)-1,2,3-benzotriazin-4(3H)-one.
For biological evaluation of the foregoing analogues, we first
evaluated the inhibitory efficacy on the growth of the mamma-
lian cervical carcinoma cell line KB-31 and the human mamma-
ry carcinoma cell line MCF-7. Both the final analogues (12–15,
by 3-(diethoxyphosphoryloxy)-
1,2,3-benzotriazin-4(3H)-one
(DEPBT)-mediated amide forma-
tion with isovaleric acid, which
gave amide 39 exclusively with-
out the competing formation of
an ester.[49] In the next steps ho-
mologation and introduction of
the nitrogen atom by HWE reac-
tion was envisaged. Therefore,
the primary alcohol had to be
oxidized into an intermediate al-
dehyde, which was best per-
formed with 2-iodoxybenzoic
acid (IBX) in ethyl acetate at
Scheme 7. Fragment union and formation of the side chain analogues 12, 13, 45, and 46. Reagents and condi-
tions: a) DEPBT, Et3N, THF, RT, 76–93%; b) H2, Pd/C, MeOH, RT, 93–99%; c) 16a, DEPBT, Et3N, Na2CO3, THF, RT, 70–
reflux.[50] The HWE reaction used
diethyl cyanomethylphospho- 86%.
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
9
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
nounced effects on the treated cells at these concentrations
(e.g., 54: Figure 11c). Only for one of the compounds (i.e., 50:
Figure 11d), a very weak effect at the border area of the cells
may be visible, resulting in small knots in the cortex of the
cells.[57]
Conclusions
In summary, we have developed an initial set of simplified
hybrid analogues of the highly potent actin binding agents rhi-
zopodin and bistramide, based on an interdisciplinary ap-
proach at the interface of molecular modeling, organic synthe-
sis, and chemical biology. Computational analyses of noncova-
lent actin–inhibitor interactions were performed with Auto-
Dock and the DrugScore scoring function, and were validated
by re-docking of known actin binding ligands. This guided the
design of a novel, highly modular and simplified class of
hybrid analogues with a rationale to address both the bistra-
mide and rhizopodin binding sites. Synthesis of these com-
pounds was carried out by a highly modular and convergent
three-fragment coupling strategy, enabling rapid access to dra-
matically simplified rhizopodin–bistramide hybrids. This new
analogue class was analyzed for its antiproliferative and actin
binding properties. Most analogues demonstrated only moder-
ate or no antiproliferative or actin binding properties. However,
a certain degree of antiproliferative activity was retained for
analogues resembling the original side chain of rhizopodin.
This may suggest that more elaborate and structurally related
analogues may indeed retain the activity of the original natural
products, in agreement with previous results from the Mar-
riott[12f,26] and Nicolaou[24] research groups. These results also
suggest that the in silico model may need further refinement.
Furthermore, a broader set of SAR data will be helpful for de-
lineation and simplification of the pharmacophore. In general,
a more detailed theoretical understanding and predictive anal-
ysis of intermolecular noncovalent interactions in target–inhibi-
tor studies for functional design is warranted.
Scheme 8. Fragment union and formation of the side chain analogues 14,
15, 51, and 52. Reagents and conditions: a) DEPBT, Et3N, Na2CO3, THF, RT, 57–
62%; b) H2, Pd/C, HCl, MeOH, RT, 91–95%; c) 16a, DEPBT, Et3N, Na2CO3, THF,
RT, 70–78%.
45, 46, 51, and 52) as well as the synthetic intermediates (41–
44 and 47–50) were evaluated. As shown in Table 3, among all
the novel ligands, three compounds (49, 50, and 54[56]
)
showed moderate activities, with IC50 values in the mid-micro-
molar range, and two further ligands (48 and 53[55]) showing
effects at high micromolar concentrations. Notably, all the
active compounds contained the original rhizopodin side
chain, while the corresponding compounds bearing the simpli-
fied piperidine segments (48 vs. 41, 49 vs. 43, and 50 vs. 44)
did not show any activity. The configuration at C10 had only
a minor influence on the activity (49 and 54). None of the
compounds that incorporated the piperidine substitute for the
bistramide segment (i.e., fragment 1, see Figure 8) showed ac-
tivity at the concentrations studied, suggesting the piperidine
to be an unsuitable substitute. It is possible that the proposed
fragment 1 may be too voluminous (compare 14, 15, 51, and
52 with 48–50, 53, and 54).
We then analyzed the effect of these compounds (i.e., 48–
50, 53, and 54) on the morphology of PtK2 epithelial kidney
cells derived from a male potoroo by fluorescence microscopy.
As previously described,[17b] F-actin filaments were made visible
with fluorescein isothiocyanate (FITC)-labeled phalloidin which
binds specifically to polymerized actin. The cells were first
treated with these compounds at concentrations of 25 mgmLꢀ1
(~50 mm), in direct comparison with rhizopodin. As shown in
Figure 11b, treatment of the cells with rhizopodin leads to
a gradual decay of the stress fibers within minutes, relative to
control cells (Figure 11a), resulting in a complete restitution of
F-actin.[17b] In contrast, the analogues did not cause any pro-
Experimental Section
Computational details
To model the ‘barbed end’ of actin, the X-ray structures of actin
with bound ‘barbed end’ inhibitors such as kabiramide C (1QZ5),
jaspisamide A (1QZ6), ulapualide A (1S22), aplyronine A (1WUA),
swinholide A (1YXQ), reidispongiolide A (2ASM), reidispongiolide C
(2ASP), sphinxolide B (2ASO), bistramide A (2FXU), rhizopodin
(2VYP), and lobophorolide (3M6G) were considered. All structures
were aligned with the crystal structure of actin with bound aplyro-
nine A (1WUA) in order to adjust the binding pockets. Crystallo-
graphic water molecules, metal ions, and cofactors were removed,
as were alternative protein residues (Table S1, Supporting Informa-
tion (SI)). Missing atoms located close to the ligand binding site
were added manually, followed by a brief minimization with the
molecular modeling package Macromodel 9.7,[58] constraining the
rest of the structure to the X-ray coordinates. Processing the li-
gands was carried out by adding absent atoms followed by brief
minimization (100 steps, OPLS2005 force field,[59] aqueous solution)
of the added ligand part within the corresponding protein binding
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
10
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 3. Inhibitory effects of ligands on the growth of the mammalian cervical carcinoma cell line KB-31 and human mammary carcinoma cell line MCF-
7.[a,c]
Ligand[b]
MCF-7[a]
KB-31[b]
IC50 [mm]
IC90 [mm]
>370
IC50 [mm]
IC90 [mm]
>370
103.0ꢁ29.0
27.5ꢁ19.4
55.0ꢁ39.0
370ꢁ0.0
16.1ꢁ16.2
>370
41.2ꢁ0.0
350ꢁ0.0
9.2ꢁ0.0
>370
>370
123.5ꢁ0.0
>370
27.5ꢁ0.0
61.9ꢁ29.2
>370
11.5ꢁ3.2
>370
[a] Cell lines were obtained from the German collection of Microorganisms and Cell Cultures (DSMZ). Growth inhibition was measured as previously de-
scribed, and metabolic activity was determined after five days using the WST-1 assay. Data are the meanꢁSD from experiments performed in duplicate.[53]
[b] All other compounds showed no activity at concentrations up to 370 mm. [c] Rhizopodin shows activities in the low-nanomolar range toward similar
cell lines.[54]
pocket to generate a sensible bond length (Table S2, SI). All other
atoms were kept at the positions of the crystal structure. Process-
ing the side chain hybrids was performed by drawing the struc-
tures in Maestro (including polar hydrogen atoms), followed by
a brief minimization within the protein of reidispongiolide A
(2ASM). All co-crystal structures of actin exhibit similar to nearly
identical positions of the amino acid backbone for the binding site
of the ligand macrocycle as well as in the binding pocket of the
ligand side chain (Figure S1, SI). Thus, the conformation of actin in
the crystal structures is considered as the low-energy binding con-
formation for the ‘barbed end’ inhibitors and was kept fixed to X-
ray coordinates during docking. Docking was conducted with
a combination of rigid protein structure and partly flexible ligand
structure using AutoDock 3.0[30] with the DrugScore objective func-
tion.[31] This combination has proven reliable for binding mode pre-
diction in a “re-docking” evaluation.[60] The number of flexible
bonds in the ligand was determined automatically and checked for
plausibility (Tables S3 and S4, SI). Default parameters were used,
except for the number of GA runs, which was set to 100, the popu-
lation size, which was set to 200, the number of generations,
which was set to 50000, and the number of evaluations, which
was set to 3ꢂ107. The length of the ligands in the crystal structure
varies from 20.5 to 37.5 ꢁ. However, the search area and the elec-
trostatic map size was chosen to be a cube, centered on the
ligand, with a space diagonal of ~58.5 ꢁ (edge length 33.75 ꢁ, 90
grid points, 0.375 ꢁ grid point distance) to cover the size of all li-
gands (results are depicted in Table 1 and SI Figure S2). Further-
more, smaller cubes with a space diagonal of 39.3 ꢁ (edge length
22.7 ꢁ, 60 grid points, 0.375 ꢁ grid point distance) as well as 48.7 ꢁ
(edge length 28.1 ꢁ, 75 grid points, 0.375 ꢁ grid point distance)
were applied to evaluate the influence of the search area (results
are depicted in Figures S3 and S4, SI). For the re-docking process,
each ligand was docked 100 times to the actin structure of the cor-
Figure 11. Fluorescence microscopy of PtK2 cells after staining for F-actin
(green); nuclei are blue: a) MeOH control 0.5%, b) rhizopodin 500 ngmLꢀ1
,
c) 54 25 mgmLꢀ1, d) 50 25 mgmLꢀ1
.
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
11
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
responding co-crystal structure. Thus, 100 solutions were obtained
for each ligand docking. The structures were clustered at a ligand
RMSD of 1.0 ꢁ. The highest populated clusters for each ligand
were compared with the mean docked energy as well as the bind-
ing efficiency calculated with respect to the number of heavy
atoms. The structure with lowest docking energy was regarded as
cluster representative and was compared with the crystal structure
(Tables S5 and S6, SI). To evaluate the docking results for various
actin conformations, each ligand was docked onto each protein
crystal structure in a cross-docking process. The results are sum-
marized in Tables S7–S9, SI.
13C NMR (125 MHz, CD3OD): d=23.8, 29.2, 32.6, 38.5, 53.6, 57.5,
62.1, 130.5, 130.6, 131.4, 132.6, 174.7 ppm; HRMS-ESI m/z [M]+
calcd for C14H20NO2+: 234.1489, found: 234.1489.
Synthesis of the middle fragments
Masamune aldol reaction to form ester 26. tert-Butyl protection
of 25 was previously described.[36] Following this procedure gave
2-[N-(tert-butyloxycarbonyl)-N-methylamino]ethanol (99%) as a col-
1
orless oil: Rf =0.25 (PE/EtOAc, 6:4); H NMR (300 MHz, CDCl3): d=
1.47 (s, 9H), 2.93 (s, 3H), 3.40 (t, J=5.3 Hz, 2H), 3.76 ppm (t, J=
5.3 Hz, 2H); 13C NMR (75 MHz, CDCl3): d=28.3, 35.3, 51.2, 61.0,
79.7, 155.8, 157.0* ppm;[61a] HRMS-ESI m/z [M+Na]+ calcd for
C8H17NNaO3+: 198.1101, found: 198.1096.
Synthesis details
Experimental details for the syntheses are given below. Please refer
to the Supporting Information for the syntheses of compounds
32–34, the Weinreb amide en route to 34, as well as for the gener-
al methods.
2-[N-(tert-Butyloxycarbonyl)-N-methylamino]ethanol
(166 mg,
0.948 mmol, 1.0 equiv) was dissolved in EtOAc (7 mL) before IBX
(796 mg, 2.84 mmol, 3.0 equiv) was added. The suspension was
held at reflux for 4 h before it was filtered over a pad of silica gel.
The solvent was removed in vacuo to yield tert-butyl methyl(2-ox-
oethyl)carbamate (149 mg, 91%) as a colorless oil: Rf =0.48 (PE/
EtOAc, 6:4); 1H NMR (300 MHz, CDCl3): d=1.46 (s, 4.5H), 1.41 (s,
4.5H), 2.92 (s, 1.5H), 2.94 (s, 1.5H), 3.90 (s, 1H), 4.00 (s, 1H),
9.59 ppm (s, 1H); 13C NMR (75 MHz, CDCl3): d=28.2, 35.8, 58.7,
59.1*, 80.4, 80.6*, 155.3, 156.0*, 198.5 ppm;[61a] HRMS-ESI m/z [M+
Na]+ calcd for C8H15NNaO3+: 196.0944, found: 196.0940.
Synthesis of the western fragment
Chiral resolution of amine 23. Synthesis was previously de-
scribed.[34] This procedure gave 24 (36%) as a white solid; mp:
1
117 8C, [a]2D0 = +38.78 (c=1.0 in CHCl3); H NMR (500 MHz, CDCl3):
d=1.03 (qd, J=12.4, 4.7 Hz, 1H), 1.28 (t, J=7.1 Hz, 3H), 1.51–1.65
(m, 2H), 1.74 (d, J=12.4 Hz, 1H), 1.91–2.01 (m, 1H), 2.09–2.16 (m,
2H), 2.21 (td, J=12.4, 4.4 Hz, 1H), 2.24 (t, J=12.0 Hz, 1H), 2.97 (d,
J=12.4 Hz, 1H), 3.07 (d, J=12.0 Hz, 1H), 4.14 (q, J=7.1 Hz, 2H),
4.88 (s, 1H), 7.22 (t, J=7.4 Hz, 1H), 7.29 (t, J=7.4 Hz, 2H),
7.46 ppm (d, J=7.4 Hz, 2H); 13C NMR (125 MHz, CDCl3): d=14.2,
Masamune aldol reaction to form ester 26 was previously de-
scribed.[37] Following this procedure gave 26 (90%, d.r.=96:4) as
a colorless oil: Rf =0.27 (PE/EtOAc, 7:3); [a]2D0 = +8.38 (c=0.8 in
1
CHCl3); H NMR (300 MHz, CDCl3): d=1.18 (d, J=7.4 Hz, 3H), 1.31
(d, J=6.9 Hz, 3H), 1.45 (s, 9H), 2.30 (s, 3H), 2.43 (s, 6H), 2.56–2.69
(m, 1H), 2.78 (s, 3H), 2.91 (s, 3H), 3.15–3.47 (m, 2H), 3.82–3.96 (m,
1H), 3.96–4.16 (m, 1H), 5.77 (d, J=5.5 Hz, 1H), 6.87 (s, 2H), 6.97–
7.11 (m, 2H), 7.11–7.25 ppm (m, 3H); 13C NMR (75 MHz, CDCl3): d=
12.3, 13.3, 20.9, 22.6, 28.3, 28.4, 36.2, 43.9, 52.9, 55.5, 72.8, 78.1,
80.1, 126.1, 128.0, 128.4, 131.9, 132.2, 138.0, 140.4, 142.3, 157.6,
173.8 ppm; HRMS-ESI m/z [M+H]+ calcd for C30H45N2O7S+:
577.2942, found: 577.2938.
21.7, 28.5, 30.1, 38.0, 43.3, 47.6, 60.7, 74.4, 126.6, 127.1, 128.1,
142.5, 170.9, 178.9 ppm; HRMS-FAB m/z [M]+ calcd for C9H18NO2
172.1332, found: 172.1341; d.r.ꢂ99:1%.[34b]
:
+
(3R)-1-Benzyl-3-(carboxymethyl)piperidin-1-ium chloride (16a):
The salt 24 (208 mg, 0.642 mmol) was dissolved in 2m Na2CO3
(5 mL). This solution was extracted with EtOAc (5ꢂ5 mL). The com-
bined organic layers were dried over MgSO4, filtered and concen-
trated in vacuo. The residue was dissolved in acetone (8 mL), and
K2CO3 (488 mg, 3.53 mmol, 5.5 equiv) was added followed by
benzyl bromide (103 mL, 0.706 mmol, 1.1 equiv). The mixture was
stirred overnight at RT. Afterward the mixture was filtered and con-
centrated in vacuo. Purification by column chromatography (SiO2,
PE/iPrNH2, 200:3) gave (R)-ethyl 2-(1-benzylpiperidin-3-yl)acetate as
(2S,3R)-4-(((Benzyloxy)carbonyl)(methyl)amino)-3-hydroxy-2-
methylbutanoic acid (17). Saponification of 26 was previously de-
scribed.[37] Following this procedure gave (2S,3R)-4-((tert-butoxycar-
bonyl)(methyl)amino)-3-hydroxy-2-methylbutanoic acid (99%) as
a colorless oil: Rf =0.29 (CH2Cl2/MeOH, 7:3); [a]2D0 = +2.38 (c=1.0 in
1
CHCl3); H NMR (300 MHz, CDCl3): d=1.26 (d, J=7.0 Hz, 3H), 1.45
a colorless oil (146 mg, 87%): Rf =0.30 (PE/iPrNH2, 200:3); [a]D20
=
(s, 9H), 2.58 (quint., J=7.0 Hz, 1H), 2.93 (s, 3H), 3.14–3.38 (m, 1H),
3.38–3.57 (m, 1H), 3.85–3.97 (m, 1H), 5.92 ppm (brs, 2H); 13C NMR
1
ꢀ2.28 (c=1.0 in CHCl3); H NMR (300 MHz, CDCl3): d=0.89–1.11 (m,
1H), 1.22 (t, J=7.2 Hz, 3H), 1.54–1.70 (m, 2H), 1.70–1.83 (m, 2H),
2.00 (td, J=10.6, 3.9 Hz, 1H), 2.04–2.17 (m, 1H), 2.19 (dd, J=14.6,
6.5 Hz, 1H), 2.25 (dd, J=14.6, 7.6 Hz, 1H), 2.68–2.82 (m, 2H), 3.49
(s, 2H), 4.10 (q, J=7.1 Hz, 2H), 7.18–7.27 (m, 1H), 7.28–7.36 ppm
(m, 4H); 13C NMR (75 MHz, CDCl3): d=14.2, 24.8, 30.6, 33.2, 39.2,
53.9, 59.4, 60.2, 63.4, 126.8, 128.1, 129.0, 138.6, 172.7 ppm; HRMS-
EI m/z [MꢀH]+ calcd for C16H22NO2+: 260.1645, found: 260.1660.
(75 MHz, CDCl3): d=13.8, 28.3, 36.1, 43.2, 53.1, 72.5, 80.7, 157.7,
178.3 ppm; HRMS-ESI m/z [M+K]+ calcd for C11H21KNO5
:
+
286.1051, found: 286.1052.
(2S,3R)-4-((tert-Butoxycarbonyl)(methyl)amino)-3-hydroxy-2-methyl-
butanoic acid (153 mg, 0.617 mmol, 1.0 equiv) was dissolved in
CH2Cl2 (16 mL) at RT. TFA (0.95 mL, 12.3 mmol, 20 equiv) was
added and the mixture was stirred for 90 min. Afterward the vola-
tiles were removed in vacuo and the residue was co-evaporated
with toluene (2ꢂ5 mL) to give crude (2R,3S)-3-carboxy-2-hydroxy-
(R)-Ethyl 2-(1-benzylpiperidin-3-yl)acetate (138 mg, 0.528 mmol)
was dissolved in 6m HCl (9.5 mL). The reaction mixture was stirred
at RT for 4 h. Afterward the reaction mixture was concentrated in
vacuo, co-evaporated with toluene (2ꢂ4 mL) to remove residual
H2O and finally dried in vacuo to give hydrochloride 16a (141 mg,
99%) as a white crystalline solid; mp: 2058C (dec.); [a]2D0 = +1.48
(c=1.0 in MeOH); 1H NMR (500 MHz, CD3OD): d=1.18–1.38 (m,
1H), 1.78–2.01 (m, 3H), 2.22–2.41 (m, 3H), 2.80 (t, J=11.5 Hz, 1H),
2.95 (t, J=11.6 Hz, 1H), 3.44 (d, J=11.6 Hz, 1H), 3.51 (d, J=11.5 Hz,
1H), 4.34 (s, 2H), 7.44–7.52 (m, 3H), 7.52–7.61 ppm (m, 2H);
N-methylbutan-1-aminium trifluoroacetate as
a colorless gum
(154 mg, 96%), which was used directly without further purifica-
tion.
Crude (2R,3S)-3-carboxy-2-hydroxy-N-methylbutan-1-aminium tri-
fluoroacetate (154 mg, 0.590 mmol, 1.0 equiv) was dissolved in ace-
tone/H2O (1:1 v/v, 2 mL) at RT before NaHCO3 (198 mg, 2.36 mmol,
4.0 equiv) was added followed by CbzOSuc (166 mg, 0.666 mmol
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
12
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
1.1 equiv). The reaction mixture was stirred for two days before it
was diluted with H2O (4 mL) and acetone was removed in vacuo.
The aqueous layer was washed with CH2Cl2 (2ꢂ3 mL) before it was
acidified with 1m HCl to pH 3. Afterward it was extracted with
EtOAc (4ꢂ5 mL). The combined organic layers were washed with
H2O (1ꢂ4 mL) and brine (1ꢂ4 mL), dried (MgSO4) and concentrat-
ed to yield 17 (161 mg, 97%) as a colorless oil: Rf =0.73 (CH2Cl2/
MeOH, 9:1); [a]2D0 =ꢀ2.68 (c=1.0 in CHCl3); 1H NMR (300 MHz,
CD3OD): d=1.10 (d, J=7.0 Hz, 1.5H), 1.19 (d, J=7.0 Hz, 1.5H),
2.44–2.59 (m, 1H), 3.00 (s, 1.5H), 3.02 (s, 1.5H), 3.15–3.30 (m, 1H),
3.47–3.62 (m, 1H), 3.90–4.01 (m, 1H), 5.11 (s, 2H), 7.25–7.44 ppm
(m, 5H); 13C NMR (75 MHz, CD3OD): d=13.9, 14.0*, 36.5, 36.9*, 45.5,
53.8, 54.3*, 68.5, 72.9, 73.1*, 128.9, 129.1*, 129.2, 129.6, 138.3,
158.3, 158.6*, 178.2 ppm;[61a] HRMS-ESI m/z [MꢀH]ꢀ calcd for
C14H18NO5ꢀ: 280.1190, found: 280.1182.
(29) (1.00 g, 3.95 mmol, 1.0 equiv) was dissolved in dry DMF
(10 mL) and cooled to 58C. Ag2O (3.66 g, 15.8 mmol, 4.0 equiv) and
iodomethane (2.5 mL, 39.5 mmol, 10 equiv) were added and the
mixture was stirred overnight at 58C in the dark. Afterward, the re-
action mixture was diluted with EtOAc and filtered over a plug of
Celite. Next, all volatiles were removed in vacuo. Purification by
column chromatography (SiO2, PE/EtOAc, 3:1) gave 30 (201 mg,
18%) and 31 (698 mg, 60%) as colorless oils. Data for 30: Rf =0.16
1
(PE/EtOAc, 3:1); [a]2D0 =ꢀ15.08 (c=0.5 in CHCl3); H NMR (300 MHz,
CDCl3): d=1.87–2.08 (m, 2H), 3.25–3.39 (m, 2H), 3.40 (s, 3H), 3.75
(s, 3H), 3.86 (dd, J=7.7, 4.4 Hz, 1H), 5.10 (s, 3H), 7.28–7.41 ppm
(m, 5H); 13C NMR (75 MHz, CDCl3): d=32.6, 37.7, 52.0, 58.3, 66.6,
78.9, 128.1, 128.1, 128.5, 136.6, 156.3, 172.7 ppm; HRMS-ESI m/z
[M+H]+ calcd for C14H20NO5+: 282.1336, found: 282.1339. Data for
31: Rf =0.22 (PE/EtOAc, 3:1); [a]2D0 =ꢀ16.28 (c=1.0 in CHCl3);
1H NMR (400 MHz, CDCl3): d=1.85–1.96 (m, 1H), 1.97–2.09 (m, 1H),
2.93 (s, 3H), 3.29–3.42 (m, 4H), 3.44–3.59 (m, 1H), 3.68–3.82 (m,
4H), 5.14 (s, 2H), 7.28–7.43 ppm (m, 5H); 13C NMR (100 MHz,
CDCl3): d=30.8, 31.2*, 34.4, 35.0*, 45.0, 45.8*, 51.9, 58.1, 67.0, 67.1*,
77.9, 78.1*, 127.8, 127.9, 128.4, 136.8, 156.2, 172.7, 172.8* ppm;[61a]
HRMS-ESI m/z [M+Na]+ calcd for C15H21NNaO5+: 318.1312, found:
318.1312.
(R)-4-Azido-3-hydroxybutanoic acid (28). Synthesis was previously
described.[38] Following this procedure gave (R)-ethyl 4-azido-hy-
droxybutanoate (85%) as a yellow oil: Rf =0.24 (PE/EtOAc, 3:1);
1
[a]2D0 = +15.68 (c=1.0 in CHCl3); H NMR (400 MHz, CDCl3): d=1.29
(t, J=7.2 Hz, 3H), 2.52 (dd, J=16.6, 4.8 Hz, 1H), 2.57 (dd, J=16.6,
7.6 Hz, 1H), 2.96 (brs, 1H), 3.33 (dd, J=12.6, 5.9 Hz, 1H), 3.38 (dd,
J=12.6, 4.4 Hz, 1H), 4.19 (q, J=7.2 Hz, 2H) 4.21 ppm (dddd, J=
7.6, 5.9, 4.8, 4.4 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=14.1, 38.4,
55.5, 61.0, 67.3, 172.1 ppm; HRMS-ESI m/z [M+Na]+ calcd for
C6H11N3NaO3+: 196.0693, found: 196.0697.
(S)-4-(((Benzyloxy)carbonyl)(methyl)amino)-2-methoxybutanoic
acid (20). Methyl ester 31 (544 mg, 1.84 mmol) was dissolved in
a mixture of acetone and 1m NaOH solution (6:1 v/v, 19 mL). The
mixture was stirred at RT for 3.5 h before it was adjusted to pH~7
by adding 1m HCl and then acetone was removed in vacuo. The
mixture was basified to pH 12 by adding 1m NaOH solution before
it was washed with Et2O (1ꢂ3 mL) to remove traces of unreacted
starting material. The basic aqueous phase was acidified to pH 3
by addition of 1m HCl solution which caused formation of a white
precipitate. The aqueous phase was extracted with EtOAc (3ꢂ
5 mL). The combined organic layers were dried over MgSO4, fil-
tered and concentrated in vacuo to give 20 (512 mg, 99%) as a col-
orless oil: Rf =0.36 (CH2Cl2/MeOH, 19:1); [a]2D0 =ꢀ8.88 (c=1.0 in
Saponification of (R)-ethyl 4-azido-hydroxybutanoate was previous-
ly described.[38] Following this procedure gave 28 (97%) as a color-
less oil: Rf =0.53 (CH2Cl2/MeOH, 9:1); [a]2D0 = +22.48 (c=0.5 in
CH2Cl2); 1H NMR (400 MHz, CDCl3): d=2.60 (dd, J=16.8, 4.8 Hz,
1H), 2.65 (dd, J=16.8, 7.8 Hz, 1H), 3.38 (dd, J=12.6, 6.2 Hz, 1H),
3.43 (dd, J=12.6, 4.4 Hz, 1H), 4.24 (dddd, J=7.8, 6.2, 4.8, 4.4 Hz,
1H), 6.53 ppm (brs, 2H); 13C NMR (75 MHz, CDCl3): d=38.3, 55.ꢀ5,
67.1, 176.8 ppm; HRMS-ESI m/z [MꢀH]ꢀ calcd for C4H6N3O3
:
144.0415, found: 144.0414.
1
CHCl3); H NMR (400 MHz, CDCl3): d=1.89–2.01 (m, 1H), 2.02–2.14
(R)-4-(((Benzyloxy)carbonyl)amino)-3-hydroxybutanoic acid (18).
To a stirred solution of azide 28 (1.26 g, 8.69 mmol) in MeOH
(225 mL) Pd/C (10%, 900 mg) was added. The resulting reaction
mixture was stirred for 2 h under an atmosphere of H2. Afterward,
the reaction mixture was filtered through a pad of Celite which
was rinsed thoroughly with MeOH (600 mL) and finally with H2O
(600 mL). The resulting filtrate was concentrated in vacuo to yield
(R)-4-amino-3-hydroxybutanoic acid (993 mg, 96%) as a colorless
solid: Rf =0.27 (n-butanol/acetic acid/H2O, 4:1:1); mp: 2138C (dec.);
(m, 1H), 2.94 (s, 3H), 3.32–3.44 (m, 4H), 3.49–3.59 (m, 1H), 3.75–
3.84 (m, 1H), 5.15 (s, 2H), 7.28–7.40 (m, 5H), 8.52 ppm (brs, 1H);
13C NMR (100 MHz, CDCl3): d=30.2, 30.8*, 34.3, 35.0*, 45.0, 45.5*,
58.2, 67.3, 77.4, 78.1*, 127.9, 128.0, 128.5, 136.6, 156.4, 156.8ꢀ*,
175.5, 176.1* ppm;[61a] HRMS-ESI m/z [MꢀH]ꢀ calcd for C14H18NO5
:
280.1190, found: 280.1181.
(S)-4-(((Benzyloxy)carbonyl)amino)-2-methoxybutanoic acid (19).
Synthesis previously described.[39] Following the procedure for 20
gave 19 (97%) as a colorless oil: Rf =0.28 (CH2Cl2/MeOH, 19:1);
1
[a]2D0 =ꢀ17.48 (c=1.0 in H2O); H NMR (400 MHz, D2O): d=2.41 (d,
J=6.5 Hz, 2H), 2.93 (dd, J=13.0, 9.5 Hz, 1H), 3.14 (dd, J=13.0,
3.2 Hz, 1H), 4.18 ppm (dtd, J=9.5, 6.5, 3.2 Hz, 1H); 13C NMR
(75 MHz, D2O): d=44.3, 46.2, 68.0, 180.7 ppm; HRMS-ESI m/z [M+
Na]+ calcd for C4H9NNaO3+: 142.0475, found: 142.0478.
1
[a]2D0 =ꢀ21.38 (c=1.0 in CHCl3); H NMR (400 MHz, CDCl3): d=1.92–
2.11 (m, 2H), 3.28–3.43 (m, 2H), 3.44 (s, 3H), 3.84–3.91 (m, 1H),
5.11 (s, 2H), 5.17 (brs, 1H), 7.30–7.34 (m, 1H), 7.35–7.40 ppm (m,
4H); 13C NMR (75 MHz, CDCl3): d=32.4, 37.6, 58.4, 66.8, 78.5, 128.1,
128.1, 128.5, 136.4, 156.6, 175.9 ppm; HRMS-ESI m/z [M+Na]+
calcd for C13H17NNaO5+: 290.0999, found: 290.0993.
Cbz protection of (R)-4-amino-3-hydroxybutanoic acid was per-
formed following the procedure for compound 17. Cbz-protected
amino acid 18 (96%) was obtained as a colorless oil: Rf =0.58
(CH2Cl2/MeOH, 9:1); [a]2D0 = +2.38 (c=1.0 in CHCl3); 1H NMR
(400 MHz, CD3OD): d=2.35 (dd, J=15.6, 8.5 Hz, 1H), 2.49 (dd, J=
15.6, 4.3 Hz, 1H), 3.17 (dd, J=13.8, 6.6 Hz, 1H), 3.22 (dd, J=13.8,
5.3 Hz, 1H), 4.04–4.12 (m, 1H), 5.08 (s, 2H), 7.26–7.40 ppm (m, 5H);
13C NMR (75 MHz, CD3OD): d=40.6, 47.5, 67.7, 68.7, 129.0, 129.1,
129.6, 138.4, 159.2, 175.3 ppm; HRMS-ESI m/z [MꢀH]ꢀ calcd for
C14H18NO5ꢀ: 252.0877, found: 252.0869.
Synthesis of the eastern fragments
Dimethyl-((3R,4R)-7-(tert-butyldimethylsilyloxy)-4-methoxy-3-
methyl-2-oxoheptyl) phosphonate (34). To a cold (ꢀ788C) solu-
tion of dimethyl methylphosphonate (0.60 mL, 5.54 mmol,
2.3 equiv) in dry THF (4.3 mL) nBuLi (1.6m in hexanes, 3.3 mL,
5.30 mmol, 2.2 equiv) was added dropwise over a period of 20 min
and stirring was continued for 1.5 h at ꢀ788C. Afterward (2R,3R)-6-
(tert-butyldimethylsilyloxy)-N,3-dimethoxy-N,2-dimethylhexanamide
(803 mg, 2.41 mmol, 1.0 equiv) in THF (3 mL) was added dropwise
over a period of 30 min. The reaction mixture was stirred for 2 h at
(S)-Methyl 4-(((benzyloxy)carbonyl)amino)-2-methoxybutanoate
(30) and (S)-methyl 4-(((benzyloxy)carbonyl)(methyl)amino)-2-
methoxybutanoate (31). (S)-(+)-Z-4-Amino-2-hydroxybutyric acid
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
13
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
ꢀ788C before it was quenched by addition of saturated NH4Cl so-
lution (4 mL). The mixture was allowed to warm to RT and then
Et2O (20 mL) and H2O (5 mL) were added. The organic layer was
separated and the aqueous layer was extracted with Et2O (2ꢂ
20 mL) and CH2Cl2 (3ꢂ20 mL). The combined organic layers were
dried (MgSO4) and the solvent was removed in vacuo. Purification
by column chromatography (SiO2, PE/EtOAc, 4:6) yielded 34
15.8, 7.6 Hz, 1H); 13C NMR (75 MHz, CDCl3): d=ꢀ5.3, 12.7, 15.6,
18.3, 25.9, 26.4, 27.7, 39.5, 47.2, 57.7, 63.1, 66.6, 82.4, 130.0, 148.6,
202.8 ppm; HRMS-ESI m/z [M+H]+ calcd for C19H39O4Si+: 359.2612,
found: 359.2612.
Molecular sieves (3 ꢁ), Ag2O (2.13 g, 9.20 mmol, 20 equiv), and io-
domethane (2.2 mL, 34.5 mmol, 75 equiv) were added to a solution
of alcohol (2S,6R,7R,E)-10-((tert-butyldimethylsilyl)oxy)-1-hydroxy-7-
(889 mg, 93%) as a pale-yellow oil: Rf =0.13 (PE/EtOAc, 3:7); [a]D20
=
methoxy-2,6-dimethyldec-3-en-5-one
(165 mg,
0.460 mmol,
1
ꢀ61.48 (c=1.0 in CHCl3); H NMR (300 MHz, CDCl3): d=0.04 (s, 6H),
0.88 (s, 9H), 1.01 (d, J=6.8 Hz, 3H), 1.39–1.71 (m, 4H), 2.97–3.08
(m, 1H), 3.10 (dd, J=22.3, 14.2 Hz, 1H), 3.25 (s, 3H), 3.32 (dd, J=
22.3, 14.2 Hz, 1H), 3.33–3.41 (m, 1H), 3.61 (t, J=6.0 Hz, 2H), 3.76
(d, J=1.1 Hz, 3H), 3.80 ppm (d, J=1.1 Hz, 3H); 13C NMR (75 MHz,
CDCl3): d=ꢀ5.2, 12.6, 18.5, 26.1, 26.4, 27.3, 42.3 (d, J=129.3 Hz),
50.0 (d, J=1.9 Hz), 53.1 (t, J=6.2 Hz, 2C), 57.6, 63.1, 83.3,
205.9 ppm (d, J=6.2 Hz); HRMS-ESI m/z [M+Na]+ calcd for
C17H37NaO6PSi+: 419.1989, found: 419.1993.
1.0 equiv) in dry Et2O (10 mL). The reaction mixture was stirred at
RT and exclusion of light overnight. Afterward, the mixture was fil-
tered through a pad of Celite and it was washed thoroughly with
Et2O (100 mL). The solution was concentrated in vacuo to yield
(2S,6R,7R,E)-10-((tert-butyldimethylsilyl)oxy)-1,7-dimethoxy-2,6-di-
methyl-dec-3-en-5-one (165 mg, 96%), as a colorless oil: Rf =0.37
1
(PE/EtOAc, 3:1); [a]2D0 =ꢀ33.18 (c=1.0 in CHCl3); H NMR (300 MHz,
CDCl3): d=0.05 (s, 6H), 0.90 (s, 9H), 1.01 (d, J=7.0 Hz, 3H), 1.09 (d,
J=6.9 Hz, 3H), 1.41–1.68 (m, 4H), 2.57–2.70 (m, 1H), 3.02 (dq, J=
8.0, 7.0 Hz, 1H), 3.29 (s, 3H), 3.31–3.38 (m, 2H), 3.34 (s, 3H), 3.45–
3.53 (m, 1H), 3.59–3.65 (m, 2H), 6.22 (dd, J=15.8, 1.4 Hz, 1H),
6.83 ppm (dd, J=15.8, 7.1 Hz, 1H); 13C NMR (75 MHz, CDCl3): d=
ꢀ5.3, 12.6, 16.1, 18.3, 25.9, 26.6, 27.8, 36.9, 47.2, 57.7, 58.9, 63.1,
76.6, 82.3, 129.1, 149.0, 202.8 ppm; HRMS-ESI m/z [M+Na]+ calcd
for C20H40NaO4Si+: 395.2588, found: 395.2589.
(2S,6R,7R,E)-10-(tert-Butyldimethylsilyloxy)-7-methoxy-1-(4-me-
thoxybenzyloxy)-2,6-dimethyldec-3-en-5-one (36). Phosphonate
34 (341 mg, 0.861 mmol, 1.1 equiv) was dissolved in dry THF
(32 mL). After cooling to ꢀ788C, NaHMDS (1.0m in THF, 0.78 mL,
0.780 mmol, 1.0 equiv) was added dropwise over a period of
15 min. Stirring was continued for 1 h at ꢀ788C. Afterward, alde-
hyde 35[43] (163 mg, 0.783 mmol, 1.0 equiv) was added over
a period of 20 min. The reaction mixture was stirred for 2 h at
ꢀ788C and was then slowly allowed to warm to RT. The mixture
was stirred overnight before it was quenched by addition of buffer
(pH 7, 55 mL). Et2O (75 mL) was added and the organic layer was
separated. The aqueous layer was extracted with Et2O (3ꢂ50 mL)
and the combined organic extracts were dried over MgSO4. The
solvent was removed in vacuo. After purification by column chro-
matography (SiO2, PE/EtOAc, 10:1) 36 was yielded (363 mg, 97%)
as a colorless oil: Rf =0.25 (PE/EtOAc, 8:1); [a]2D0 =ꢀ29.28 (c=1.0 in
Camphorsulfonic acid (43.9 mg, 0.189 mmol, 0.3 equiv) was added
to a solution of (2S,6R,7R,E)-10-((tert-butyldimethylsilyl)oxy)-1,7-di-
methoxy-2,6-dimethyl-dec-3-en-5-one
(235 mg,
0.630 mmol,
1.0 equiv) in MeOH (6.5 mL). The reaction mixture was diluted with
EtOAc (65 mL) after 15 min. The mixture was washed with saturat-
ed NaHCO3 solution (2ꢂ15 mL) and brine (15 mL). The organic
layer was separated and the combined aqueous layers were ex-
tracted with EtOAc (2ꢂ30 mL). The combined organic layers were
dried over MgSO4, filtered and concentrated in vacuo. Purification
by column chromatography (SiO2, PE/EtOAc, 1:1) yielded 37
(158 mg, 97%) as a colorless oil: Rf =0.34 (EtOAc); [a]2D0 =ꢀ36.68
1
CHCl3); H NMR (300 MHz, CDCl3): d=0.04 (s, 6H), 0.89 (s, 9H), 1.00
1
(d, J=7.1 Hz, 3H), 1.09 (d, J=6.9 Hz, 3H), 1.38–1.66 (m, 4H), 2.64
(virt. sept., J=6.6 Hz, 1H), 3.01 (virt. quint., J=7.3 Hz, 1H), 3.26 (s,
3H), 3.34–3.43 (m, 2H), 3.46–3.51 (m, 1H), 3.61 (t, J=6.0 Hz, 2H),
3.80 (s, 3H), 4.44 (s, 2H), 6.20 (dd, J=15.9, 1.1 Hz, 1H), 6.80 (dd, J=
15.9, 7.1 Hz, 1H), 6.86 (d, J=8.5 Hz, 2H), 7.24 ppm (d, J=8.5 Hz,
2H); 13C NMR (75 MHz, CDCl3): d=ꢀ5.1, 12.8, 16.3, 18.5, 26.1, 26.7,
27.9, 37.1, 47.2, 55.4, 57.8, 63.3, 72.9, 73.9, 82.4, 113.9, 129.2, 129.3,
130.4, 149.4, 159.3, 203.0 ppm; HRMS-ESI m/z [M+Na]+ calcd for
C27H46NaO5Si: 501.3007, found: 501.3007.
(c=1.0 in CHCl3); H NMR (300 MHz, CDCl3): d=1.03 (d, J=7.1 Hz,
3H), 1.09 (d, J=6.7 Hz, 3H), 1.48–1.59 (m, 1H), 1.60–1.73 (m, 3H),
2.17 (brs, 1H), 2.57–2.71 (m, 1H), 3.11 (quint., J=7.1 Hz, 1H), 3.31–
3.36 (m, 2H), 3.32 (s, 3H), 3.34 (s, 3H), 3.55 (td, J=7.1, 2.5 Hz, 1H),
3.60–3.71 (m, 2H), 6.22 (dd, J=15.9, 1.2 Hz, 1H), 6.84 ppm (dd, J=
15.9, 7.1 Hz, 1H); 13C NMR (75 MHz, CDCl3): d=12.2, 16.1, 26.9,
28.0, 36.9, 46.5, 57.6, 58.9, 62.9, 76.6, 82.0, 128.9, 149.4, 202.6 ppm;
HRMS-ESI m/z [M+Na]+ calcd for C14H26NaO4+: 281.1723, found:
281.1726.
(2S,6R,7R,E)-10-Hydroxy-1,7-dimethoxy-2,6-dimethyldec-3-en-5-
one (37). PMB ether 36 (227 mg, 0.474 mmol, 1.0 equiv) was dis-
solved in a mixture of CH2Cl2 and buffer (pH 7, 12 mL, 10:1 v/v).
The mixture was cooled to 08C and DDQ (161 mg, 0.711 mmol,
1.5 equiv) was added. The reaction mixture was stirred for 30 min
at 08C before it was allowed to warm to RT and stirred for addi-
tional 90 min. Afterward the mixture was quenched with saturated
NaHCO3 solution (15 mL). The organic layer was separated and the
aqueous layer was extracted with CH2Cl2 (3ꢂ10 mL). The combined
organic layers were washed with brine (15 mL), dried over MgSO4,
filtered and concentrated in vacuo. Purification by column chroma-
tography (SiO2, PE/EtOAc, 3:1) yielded (2S,6R,7R,E)-10-((tert-butyldi-
methylsilyl)oxy)-1-hydroxy-7-methoxy-2,6-dimethyldec-3-en-5-one
(4R,5R,9S,E)-4,10-dimethoxy-5,9-dimethyl-6-oxodec-7-en-1-ami-
nium chloride (21a). Triphenylphosphine (70.5 mg, 0.269 mmol,
1.3 equiv) was dissolved in dry THF (0.8 mL) at 08C and DIAD
(53 mL, 0.269 mmol, 1.3 equiv) was added. After stirring the reac-
tion mixture for 15 min, alcohol 37 (53.0 mg, 0.205 mmol,
1.0 equiv) dissolved in dry THF (1.5 mL) was added followed by
slow addition of DPPA (58 mL, 0.269 mmol, 1.3 equiv). The reaction
mixture was stirred for 3 h in the cold (08C) and for an additional
1 h at RT, before the reaction mixture was adsorbed onto Celite
and purified directly by flash column chromatography (SiO2, PE/
EtOAc, 8:1) to afford (2S,6R,7R,E)-10-azido-1,7-dimethoxy-2,6-dime-
thyldec-3-en-5-one as a colorless oil (50.1 mg, 86%): Rf =0.49 (PE/
EtOAc, 2:1); [a]2D0 =ꢀ53.48 (c=1.0 in CH2Cl2); 1H NMR (300 MHz,
CDCl3): d=1.05 (d, J=7.1 Hz, 3H), 1.12 (d, J=6.8 Hz, 3H), 1.41–1.57
(m, 1H), 1.58–1.79 (m, 3H), 2.59–2.75 (m, 1H), 3.08 (quint., J=
7.1 Hz, 1H), 3.26–3.40 (m, 4H), 3.33 (s, 3H), 3.37 (s, 3H), 3.53 (td,
J=7.1, 3.2 Hz, 1H), 6.25 (dd, J=15.9, 1.1 Hz, 1H), 6.87 ppm (dd, J=
15.9, 7.1 Hz, 1H); 13C NMR (75 MHz, CDCl3): d=12.2, 16.1, 24.3,
27.5, 36.9, 46.8, 51.6, 57.8, 58.9, 76.6, 81.7, 128.8, 149.5, 202.3 ppm;
(166 mg, 97%) as a colorless oil: Rf =0.25 (PE/EtOAc, 2:1); [a]D20
=
1
ꢀ41.18 (c=1.0 in CHCl3); H NMR (300 MHz, CDCl3): d=0.05 (s, 6H),
0.89 (s, 9H), 1.02 (d, J=7.0 Hz, 3H), 1.10 (d, J=6.9 Hz, 3H), 1.38–
1.53 (m, 1H), 1.54–1.68 (m, 3H), 1.70 (brs, 1H), 2.56 (virt. sept., J=
6.8 Hz, 1H), 2.96–3.08 (m, 1H), 3.28 (s, 3H), 3.44–3.54 (m, 1H),
3.55–3.66 (m, 4H), 6.26 (dd, J=15.8, 1.0 Hz, 1H), 6.78 ppm (dd, J=
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
14
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
HRMS-ESI m/z [M+Na]+ calcd for C14H25N3NaO3+: 306.1788, found:
306.1790.
consumption of starting material 39, the mixture was allowed to
cool to RT, filtered and concentrated in vacuo. The resulting resi-
due was quickly purified by column chromatography (SiO2, PE/
EtOAc, 1:1) to yield (R)-1-(3-methylbutanoyl)piperidine-3-carbalde-
hyde as a colorless oil (388 mg, 96%): Rf =0.27 (PE/EtOAc, 1:1);
Pd/C (10%, 30.0 mg) was added to
a stirred solution of
(2S,6R,7R,E)-10-azido-1,7-dimethoxy-2,6-dimethyldec-3-en-5-one
(57.9 mg, 0.204 mmol) in methanolic HCl (0.25m, 12 mL). The re-
sulting reaction mixture was stirred for 35 min under an atmos-
phere of H2. Afterward, the reaction mixture was filtered through
a PTFE syringe filter (Chromafil Xtra, PTFE-45/25 0.45 mm) that was
washed with MeOH (3ꢂ5 mL). The resulting filtrate was quickly
concentrated in vacuo and the residue was co-evaporated with
MeOH (5 mL) and with MeCN (5 mL) to finally yield the crude
amine as hydrochloride 21a (60.0 mg, max. 99%) which was used
directly without further purification.
1
[a]2D0 =ꢀ40.28 (c=1.0 in CHCl3); H NMR (500 MHz, CDCl3): d=0.92–
1.00 (m, 6H), 1.48–1.60 (m, 1H), 1.66–1.82 (m, 2H), 1.91–2.00 (m,
0.6H), 2.05–2.17 (m, 1.4H) 2.17–2.23 (m, 1H), 2.24–2.29 (m, 1H),
2.39–2.49 (m, 1H), 3.12 (ddd, J=12.8, 9.8, 2.6 Hz, 0.4H), 3.26 (ddd,
J=13.0, 9.6, 3.2 Hz, 0.6H), 3.39–3.50 (m, 1H), 3.58 (dt, J=13.0,
4.7 Hz, 0.6H), 3.81 (dd, J=13.7, 2.7 Hz, 0.4H), 4.01 (dt, J=12.8,
4.1 Hz, 0.4H), 4.23 (dd, J=13.4, 3.6 Hz, 0.6H), 9.68 (s, 0.4H),
9.71 ppm (s, 0.6H); 13C NMR (125 MHz, CDCl3): d=22.7, 24.0*, 24.0,
24.5, 24.8*, 25.6*, 25.7, 41.5, 41.8*, 42.0*, 42.1, 45.2*, 46.3, 48.0,
48.4*, 171.2*, 171.2, 202.1*, 202.2 ppm;[61b] HRMS-EI m/z [MꢀH]+
calcd for C11H18NO2+: 196.1343, found: 196.1351.
(R)-1-(3-Hydroxymethyl)piperidin-1-yl)-3-methylbutan-1-one
(39). LiAlH4 (992 mg, 26.1 mmol, 2.5 equiv) was suspended in dry
THF (45 mL) at RT. The resulting mixture was held at reflux and (R)-
piperidine-3-carboxylic acid (38) (1.35 g, 10.4 mmol, 1.0 equiv) was
added in small portions to the hot suspension (NOTE: violent reac-
tion!). The resulting reaction mixture was held at reflux for 5 h. Af-
terward it was cooled to 08C and H2O (1.0 mL), aqueous NaOH so-
lution (3m, 1.0 mL) and further H2O (1.0 mL) were added in this
order with an interval of 5 min each to give a white precipitate.
The mixture was filtered and the precipitate was washed with
warm Et2O (300 mL). The organic phase was dried over Na2SO4, fil-
tered and concentrated in vacuo to yield (R)-piperidin-3-ylmethanol
as a colorless liquid (1.16 g, 97%): Rf =0.28 (CH2Cl2/MeOH/iPrNH2,
70:30:3); [a]2D0 = +13.88 (c=0.8 in CHCl3); 1H NMR (300 MHz,
CDCl3): d=1.09–1.21 (m, 1H), 1.40–1.55 (m, 1H), 1.59–1.83 (m, 3H),
2.31 (brs, 2H) 2.41 (dd, J=11.9, 9.7 Hz, 1H), 2.58 (td, J=11.4,
3.0 Hz, 1H), 2.97 (dt, J=11.7, 3.9 Hz, 1H), 3.14 (dd, J=11.9, 3.6 Hz,
1H), 3.44 (dd, J=10.6, 6.7 Hz, 1H), 3.54 ppm (dd, J=10.6, 5.5 Hz,
1H); 13C NMR (75 MHz, CDCl3): d=25.7, 27.8, 39.3, 46.8, 49.9,
66.1 ppm; HRMS-ESI m/z [M+H]+ calcd for C6H14NO+: 116.1070,
found: 116.1069.
To a cold (ꢀ788C) stirred solution of diethyl cyanomethylphospho-
nate (106 mL, 0.654 mmol, 1.6 equiv) in dry THF (15 mL), a solution
of NaHMDS (1.0m solution in THF, 0.61 mL, 0.610 mmol, 1.5 equiv)
was slowly added over a period of 20 min at ꢀ788C. The resulting
solution was stirred for further 90 min at ꢀ788C. Afterward, a solu-
tion of (R)-1-(3-methylbutanoyl)piperidine-3-carbaldehyde (80.6 mg,
0.409 mmol, 1.0 equiv) in dry THF (3 mL) was added over a period
of 15 min via syringe pump. The reaction mixture was stirred for
3 h at ꢀ788C and afterward allowed to slowly warm to RT over-
night. The reaction mixture was quenched by the addition of
buffer (pH 7, 5 mL) and Et2O (15 mL). The organic layer was sepa-
rated and the aqueous layer was extracted with EtOAc (3ꢂ10 mL).
The combined organic layers were dried (MgSO4), filtered and con-
centrated in vacuo. The resulting product 40 was obtained as
a mixture of E and Z isomers (E/Z=5:2) and hydrogenated without
further purification: Rf Z isomer=0.37 (PE/EtOAc, 3:7), E isomer=
0.46 (PE/EtOAc, 3:7).
(R)-3-(1-Isopentylpiperidin-3-yl)propan-1-amine (22). To a stirred
solution of crude 40 (max. 0.409 mmol) in MeOH (15 mL) Pd/C
(10%, 30 mg) was added. The resulting reaction mixture was
stirred for 2 h under an atmosphere of H2. Afterward the reaction
mixture was filtered through a pad of Celite, washed with MeOH
(15 mL) and finally with EtOAc (30 mL). The resulting filtrate was
concentrated in vacuo and purified by column chromatography
(SiO2, PE/EtOAc, 1:1) to yield (S)-3-(1-(3-methylbutanoyl)piperidin-3-
yl)propanenitrile (76.4 mg, 84% over two steps) as a colorless oil:
Rf =0.39 (PE/EtOAc, 3:7); [a]2D0 =ꢀ15.98 (c=0.8 in CHCl3); 1H NMR
(500 MHz, CDCl3): d=0.93–0.98 (m, 6H), 1.12–1.28 (m, 1H), 1.39–
1.57 (m, 2H), 1.58–1.75 (m, 3H), 1.86–1.94 (m, 1H), 2.05–2.12 (m,
1H), 2.17–2.23 (m, 2H), 2.37–2.43 (m, 2H) 2.60–2.66 (m, 1H), 2.74
(dd, J=13.2, 9.9 Hz, 0.35H), 3.10 (ddd, J=13.5, 10.4, 3.0 Hz, 0.65H),
3.67 (dt, J=13.5, 4.1 Hz, 0.65H), 3.75–3.82 (m, 0.35H), 4.20–4.23 (m,
0.65H), 4.45 ppm (dt, J=12.8, 3.6 Hz, 0.35H); 13C NMR (125 MHz,
CDCl3): d=14.7*, 14.7, 22.6, 22.7, 22.7*, 24.4*, 25.0, 25.7, 28.7, 28.9*,
30.1, 30.2*, 34.7, 35.8*, 42.0*, 42.1, 42.2*, 46.1, 46.5, 51.2*, 119.1*,
119.5, 170.8*, 171.0 ppm;[61c] HRMS-ESI m/z [M+H]+ calcd for
C13H23N2O+: 223.1805, found: 223.1082.
To
a stirred solution of (R)-piperidin-3-ylmethanol (786 mg,
6.82 mmol, 1.0 equiv) in dry THF (60 mL) isovaleric acid (0.82 mL,
7.50 mmol, 1.1 equiv), Et3N (4.7 mL, 34.1 mmol, 5.0 equiv) and
DEPBT (3.06 g, 10.2 mmol, 1.5 equiv) were added. The resulting re-
action mixture was stirred overnight, before Et2O (35 mL) and a sa-
turated solution of NaHCO3 (30 mL) were added. After stirring for
15 min, the organic layer was separated and the aqueous layer was
extracted with EtOAc (3ꢂ50 mL). The combined organic layers
were dried (MgSO4), filtered and concentrated in vacuo. The result-
ing residue was purified by column chromatography (SiO2, PE/
EtOAc/iPrNH2, 50:50: 3) to yield 39 (1.15 g, 85%) as a colorless oil:
Rf =0.26 (PE/EtOAc/iPrNH2, 50:50:3); [a]2D0 =ꢀ25.08 (c=1.0 in
1
CHCl3); H NMR (500 MHz, CDCl3): d=0.92–0.97 (m, 6H), 1.18–1.25
(m, 0.5H), 1.34–1.50 (m, 1.5H), 1.62–1.73 (m, 1.5H), 1.74–1.84 (m,
1.5H), 2.04–2.14 (m, 1H), 2.17–2.27 (m, 2H), 2.69–2.75 (m, 0.5H),
2.86 (dd, J=13.4 Hz, 10.2 Hz, 0.5H), 3.20 (dd, J=13.2 Hz, 7.4 Hz,
0.5H), 3.29 (ddd, J=13.2, 8.4, 3.6 Hz, 0.5H), 3.40–3.49 (m, 1.5H),
3.50–3.59 (m, 1H), 3.87 (dd, J=13.2, 3.3 Hz, 0.5H), 3.89–3.93 (m,
0.5H), 4.33 (dt, J=13.2, 3.7 Hz, 0.5H), 5.49 ppm (brs, 1H); 13C NMR
(125 MHz, CDCl3): d=22.4*, 22.7, 22.8, 22.8*, 24.3*, 24.4, 25.8*, 25.8,
26.5, 27.3*, 37.4, 39.4*, 42.0, 42.4*, 44.4*, 47.2, 49.1*, 63.4, 65.0*,
171.2*, 171.8 ppm;[61a] HRMS-ESI m/z [M+Na]+ calcd for
C11H21NNaO2+: 222.1465, found: 222.1465.
To a stirred suspension of LiAlH4 (190 mg, 5.00 mmol, 4.0 equiv) in
dry THF (4 mL) a solution of (S)-3-(1-(3-methylbutanoyl)piperidin-3-
yl)propanenitrile (277 mg, 1.25 mmol, 1.0 equiv) in dry THF (8 mL)
was added carefully at RT. The resulting reaction mixture was held
at reflux for 3.5 h. After TLC indicated the consumption of starting
material, Et2O (12 mL) was added and the reaction mixture was
cooled to 08C. H2O (0.23 mL), aqueous NaOH solution (3m,
0.23 mL) and further H2O (0.70 mL) were added in this order with
an interval of 5 min each to give a white precipitate. Finally, the
(R)-3-(1-(3-Methylbutanoyl)piperidin-3-yl)acrylonitrile (40). To
a stirred solution of 39 (409 mg, 2.05 mmol, 1.0 equiv) in EtOAc
(20 mL) IBX (1.73 g, 6.16 mmol, 3.0 equiv) was added and the re-
sulting mixture was heated at reflux for 2 h. After TLC indicated
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
15
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
mixture was dried (Na2SO4), filtered, washed with Et2O (50 mL) and
concentrated in vacuo to yield 22 (223 mg, 84%) after purification
by column chromatography (SiO2, Et2O/iPrNH2, 100:5) as a colorless
liquid: Rf =0.21 (PE/EtOAc/iPrNH2, 50:50:5); [a]2D0 = +3.38 (c=1.0 in
CHCl3); 1H NMR (500 MHz, CDCl3): d=0.80–0.86 (m, 1H), 0.89 (d, J=
6.6 Hz, 6H), 1.17–1.34 (m, 4H), 1.36–1.41 (m, 2H), 1.42–1.49 (m,
2H), 1.50–1.59 (m, 4H), 1.62–1.67 (m, 1H), 1.74–1.78 (m, 1H), 1.79–
1.82 (m, 1H), 2.27–2.31 (m, 2H), 2.66 (t, J=7.1 Hz, 2H), 2.84–
2.87 ppm (m, 2H); 13C NMR (125 MHz, CDCl3): d=22.7, 22.8, 25.5,
26.9, 31.1, 31.2, 31.9, 36.0, 36.1, 42.5, 54.5, 57.7, 60.8 ppm; HRMS-
ESI m/z [M+H]+ calcd for C13H29N2+: 213.2325, found: 213.2324.
(89.6 mg, 92%) as a yellow oil: Rf =0.22 (PE/EtOAc/iPrNH2, 80:20:3);
[a]2D0 =ꢀ16.18 (c=1.0 in CH2Cl2); 1H NMR (400 MHz, CDCl3): d=
0.79–0.88 (m, 1H), 0.90 (d, J=6.6 Hz, 6H), 1.14–1.27 (m, 2H), 1.37–
1.47 (m, 2H), 1.47–1.71 (m, 7H), 1.72–1.80 (m, 1H), 1.82–2.02 (m,
3H), 2.29–2.41 (m, 2H), 2.84–2.96 (m, 2H), 3.17–3.32 (m, 3H), 3.33–
3.44 (m, 1H), 3.39 (s, 3H), 3.68 (dd, J=6.4, 5.1 Hz, 1H), 5.09 (s, 2H),
5.12–5.20 (m, 1H), 6.60 (t, J=5.5 Hz, 1H), 7.28–7.39 ppm (m, 5H);
13C NMR (125 MHz, CDCl3): d=22.7, 22.7, 25.1, 26.8, 27.0, 30.9, 31.8,
32.3, 35.5, 35.7, 37.5, 39.1, 54.2, 57.5, 58.4, 60.4, 66.6, 80.8, 128.0,
128.1, 128.5, 136.6, 156.3, 171.8 ppm; HRMS-ESI m/z [M+H]+ calcd
for C26H44N3O4+: 462.3326, found: 462.3326.
Benzyl ((S)-4-((3-((R)-1-isopentylpiperidin-3-yl)propyl)amino)-3-
methoxy-4-oxobutyl)(methyl)carbamate (44). The reaction was
performed according to general procedure 1 with 22 (46.9 mg,
0.220 mmol, 1.0 equiv), and 20 (77.3 mg, 0.275 mmol, 1.25 equiv)
to give 44 (85.0 mg, 81%) as a colorless oil: Rf =0.30 (PE/EtOAc/
iPrNH2, 80:20:3); [a]2D0 =ꢀ16.48 (c=1.0 in CH2Cl2); 1H NMR
(400 MHz, CDCl3): d=0.79–0.87 (m, 1H), 0.89 (d, J=6.6 Hz, 6H),
1.11–1.28 (m, 2H), 1.36–1.45 (m, 2H), 1.45–1.70 (m, 7H), 1.71–1.79
(m, 1H), 1.80–1.93 (m, 2H), 1.93–2.10 (m, 1H), 2.31–2.39 (m, 2H),
2.82- 2.98(m, 2H), 2.91 (s, 3H), 3.13–3.32 (m, 4.5H), 3.37 (s, 1.5H),
3.41–3.68 (m, 2H), 5.11 (s, 2H), 6.43–4.56 (m, 0.5H), 6.60–6.73 (m,
0.5H), 7.28–7.42 ppm (m, 5H); 13C NMR (100 MHz, CDCl3): d=22.6,
22.7, 25.0, 26.8, 27.0, 30.3, 30.6*, 30.8, 31.7, 33.9, 34.6*, 35.4, 35.6,
39.1, 44.8, 45.1*, 54.1, 57.4, 58.1, 58.2*, 60.3, 66.9, 80.1, 127.7, 127.8,
128.4, 136.8, 156.1*, 156.2, 171.6*, 171.9 ppm;[61a] HRMS-ESI m/z
[M+H]+ calcd for C27H46N3O4+: 476.3483, found: 476.3482.
Fragment union
General procedure 1: coupling of fragment 22 with middle frag-
ments 17–20 to give amides 41–44. The Cbz-protected g-amino
acid (1.15–1.25 equiv) and amine 22 were dissolved in dry THF
(0.1m) at RT. Et3N was added (5 equiv with respect to the amino
acid) followed by DEPBT (1.5 equiv with respect to the amino acid)
and the mixture was stirred at RT overnight. Half-saturated NaHCO3
solution was added, and it was stirred for 30 min before it was ex-
tracted with EtOAc (five times). The combined organic extracts
were dried (MgSO4) and concentrated in vacuo. The crude product
was purified by column chromatography (SiO2, 3% iPrNH2 in PE/
EtOAc) giving the desired amides.
Benzyl
((2R,3S)-2-hydroxy-4-((3-((R)-1-isopentylpiperidin-3-yl)-
propyl)amino)-3-methyl-4-oxobutyl)(methyl)carbamate (41). The
reaction was performed according to general procedure 1 with 22
(42.0 mg, 0.198 mmol, 1.0 equiv), and 17 (67.1 mg, 0.239 mmol,
1.2 equiv) to give 41 (72.7 mg, 77%) as a yellow oil: Rf =0.37 (PE/
EtOAc/iPrNH2, 50:50:3); [a]2D0 = +6.28 (c=1.0 in CH2Cl2); 1H NMR
(400 MHz, CDCl3): d=0.81–0.89 (m, 1H), 0.90 (d, J=6.6 Hz, 6H),
1.20–1.36 (m, 5H), 1.36–1.49 (m, 3H), 1.49–1.71 (m, 6H), 1.71–1.80
(m, 1H), 1.82–1.96 (m, 1H), 2.17–2.42 (m, 3H), 2.82–2.95 (m, 2H),
3.01 (s, 3H), 3.07–3.35 (m, 3H), 3.41–3.61 (m, 1H), 3.68–3.83 (m,
1H), 4.31 (brs, 1H), 5.13 (s, 2H), 5.68 (brs, 0.2H), 6.54 (brs, 0.8H),
7.29–7.44 ppm (m, 5H); 13C NMR (150 MHz, CDCl3): d=15.9, 16.0*,
22.7, 22.7, 25.1, 26.8, 26.8, 30.9, 31.8, 35.5, 35.7, 35.9, 36.4*, 39.6,
42.7, 43.0*, 53.5*, 54.2, 54.3, 57.5, 60.4, 67.2*, 67.3, 73.0, 127.8,
128.1, 128.5, 136.5, 136.6*, 156.2*, 157.4, 175.2, 175.5* ppm;[61d]
HRMS-ESI m/z [M+H]+ calcd for C27H46N3O4+: 476.3483, found:
476.3483.
General procedure 2: Cbz deprotection of amides 41–44. Pd/C
(10%, 0.5 mg per 1.0 mg amide) was added to a stirred solution of
Cbz-protected amides 41–44 in MeOH (7.0 mgmLꢀ1). The resulting
reaction mixture was stirred for 90 min under an atmosphere of H2.
Afterward the reaction mixture was filtered through a PTFE syringe
filter (Chromafil Xtra, PTFE-45/25 0.45 mm). The resulting filtrate
was concentrated in vacuo to yield the crude amines (93–99%),
which were used directly without further purification.
General procedure 3: coupling of amines with western fragment
16a. To a stirred suspension of Na2CO3 (5.0 equiv), carboxylic acid
(1.2 to 1.7 equiv) and crude amine (1.0 equiv) in dry THF (0.1m)
Et3N (5.0 equiv) and DEPBT (2.0 equiv) were added. The resulting
reaction mixture was stirred at RT overnight. After TLC indicated
consumption of the amine, half-saturated NaHCO3 solution was
added and the mixture was stirred for 30 min before it was extract-
ed with EtOAc (five times). The combined organic layers were
dried over MgSO4, filtered and concentrated in vacuo. The crude
product was purified by column chromatography (SiO2, 3% NEtMe2
or iPrNH2 in PE/EtOAc) to give the desired side chain analogues
12–15, 45, 46, 51 and 52.
Benzyl ((R)-2-hydroxy-4-((3-((R)-1-isopentylpiperidin-3-yl)propyl)-
amino)-4-oxobutyl)carbamate (42). The reaction was performed
according to general procedure 1 with 22 (47.3 mg, 0.223 mmol,
1.0 equiv), and 18 (69.8 mg, 0.276 mmol, 1.24 equiv) to give 42
(75.9 mg, 76%) as
a colorless oil: Rf =0.17 (PE/EtOAc/iPrNH2,
50:50:3); [a]2D0 = +3.38 (c=1.0 in CH2Cl2); H NMR (400 MHz, CDCl3):
d=0.77–0.86 (m, 1H), 0.88 (d, J=6.6 Hz, 6H), 1.14–1.26 (m, 2H),
1.35–1.44 (m, 2H), 1.44–1.55 (m, 3H), 1.56–1.68 (m, 4H), 1.69–1.77
(m, 1H), 1.88 (td, J=11.3, 2.3 Hz, 1H), 2.25–2.40 (m, 4H), 2.82–2.95
(m, 2H), 3.10–3.26 (m, 3H), 3.26–3.36 (m, 1H), 4.00–4.08 (m, 1H),
4.88 (brs, 1H), 5.07 (s, 2H), 5.63 (t, J=5.7 Hz, 1H), 6.50 (t, J=
5.0 Hz, 1H), 7.28–7.36 ppm (m, 5H); 13C NMR (100 MHz, CDCl3): d=
22.6, 22.6, 24.9, 26.5, 26.8, 30.7, 31.7, 35.2, 35.4, 39.6, 39.9, 46.2,
1
(2S,3R)-4-(2-((R)-1-Benzylpiperidin-3-yl)-N-methylacetamido)-3-
hydroxy-N-(3-((R)-1-isopentylpiperidin-3-yl)propyl)-2-methylbu-
tanamide (12). The reaction was performed according to general
procedure 3 with acid 16a (27.1 mg, 100 mmol, 1.12 equiv), and
crude amine (derived from amide 41, max. 89.0 mmol, 1.0 equiv) to
give 12 (37.6 mg, 76%) as a colorless oil: Rf =0.27 (EtOAc/iPrNH2,
1
100:5); [a]2D0 = +7.78 (c=1.0 in CH2Cl2); H NMR (400 MHz, CDCl3):
d=0.79–0.87 (m, 1H), 0.89 (d, J=6.6 Hz, 6H), 0.97–1.09 (m, 1H),
1.12–1.25 (m, 2H), 1.25 (d, J=7.1 Hz, 2.7H), 1.29 (d, J=7.2 Hz,
0.3H), 1.36–1.44 (m, 2H), 1.48–1.69 (m, 9H), 1.71–1.79 (m, 2H),
1.79–1.88 (m, 2H), 2.03 (m, 1H), 2.10–2.25 (m, 2H), 2.25–2.36 (m,
4H), 2.64–2.72 (m, 1H), 2.76 (d, J=9.5 Hz, 1H), 2.81–2.91 (m, 2H),
2.92 (s, 0.3H), 3.06 (s, 2.7H), 3.17–3.28 (m, 3H), 3.40–3.54 (m, 2H),
3.58–3.65 (m, 1H), 3.66–3.72 (m, 0.9H), 3.75–3.82 (m, 0.1H), 4.58
54.0, 57.4, 60.2, 66.7, 67.9, 127.9, 128.0, 128.4, 136.4, 157.1,
171.8 ppm; HRMS-ESI m/z [M+H]+ calcd for C25H42N3O4
:
+
448.3170, found: 448.3169.
Benzyl ((S)-4-((3-((R)-1-isopentylpiperidin-3-yl)propyl)amino)-3-
methoxy-4-oxobutyl) carbamate (43). The reaction was performed
according to general procedure 1 with 22 (44.3 mg, 0.209 mmol,
1.0 equiv), and 19 (64.1 mg, 0.240 mmol, 1.15 equiv) to give 43
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
16
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
(brs, 1H), 6.26 (brs, 0.1H). 6.74 (t, J=4.3 Hz, 0.9H), 7.19–7.27 (m,
1H), 7.28–7.35 ppm (m, 4H); 13C NMR (100 MHz, CDCl3): d=15.7,
22.7, 22.7, 24.7, 25.3, 26.9, 26.9, 30.8, 31.0, 31.9, 33.0, 34.3*, 35.7,
35.8, 37.6, 37.9, 39.6, 41.2*, 43.5, 53.9, 53.9, 54.3, 57.6, 59.7, 60.5,
63.4, 71.6*, 73.2, 126.9, 128.1, 129.0, 129.1*, 138.5, 172.6*, 173.8,
175.0, 175.2* ppm;[61e] IR (film): n˜ =3310, 2930, 2362, 1635, 1542,
(m, 0.6H), 6.51 (t, J=5.7 Hz, 0.4H), 6.73 (t, J=5.8 Hz, 0.6H), 7.17–
7.24 (m, 1H), 7.25–7.31 ppm (m, 4H), 13C NMR (125 MHz, CDCl3):
d=22.7, 22.7, 24.8, 25.3, 26.8, 27.0, 27.0*, 29.9, 30.8*, 30.8, 31.0,
31.1*, 31.8*, 31.8, 33.0, 33.2*, 33.3*, 35.5, 35.8, 35.9, 37.3*, 38.0,
39.1, 39.1*, 43.6, 45.8*, 53.7*, 53.8, 54.2*, 54.3, 57.6, 58.1, 58.3*,
59.7, 59.8*, 60.6, 63.3*, 63.3, 79.9*, 80.0, 126.7, 128.0, 129.0, 138.6,
171.1*, 171.8*, 171.8, 171.9 ppm;[61b] IR (film): n˜ =3309, 2929, 2362,
1646, 1532, 1456, 1111, 699 cmꢀ1; HRMS-ESI m/z [M+H]+ calcd for
C33H57N4O3+: 557.4425, found: 557.4423.
1456, 732, 699 cmꢀ1
;
HRMS-ESI m/z [M+Na]+ calcd for
C33H56N4NaO3+: 579.4245, found: 579.4242.
(R)-4-(2-((R)-1-Benzylpiperidin-3-yl)acetamido)-3-hydroxy-N-(3-
((R)-1-isopentylpiperidin-3-yl)propyl)butanamide (45). The reac-
tion was performed according to general procedure 3 with acid
16a (29.6 mg, 110 mmol, 1.10 equiv), and crude amine (derived
from amide 42, max. 94.4 mmol, 1.0 equiv) to give 45 (35.0 mg,
70%) as a colorless oil: Rf =0.15 (EtOAc/iPrNH2, 100:5); [a]2D0 = +
4.48 (c=1.0 in CH2Cl2); (1H NMR 400 MHz, CDCl3): d=0.80–0.89 (m,
1H), 0.90 (d, J=6.6 Hz, 6H), 0.99–1.10 (m, 1H), 1.15–1.29 (m, 2H),
1.36–1.45 (m, 2H), 1.49–1.70 (m, 9H), 1.70–1.78 (m, 2H), 1.79–1.92
(m, 2H), 2.02–2.14 (m, 3H), 2.14–2.23 (m, 1H), 2.24–2.30 (m, 2H),
2.30–2.39 (m, 2H), 2.64–2.75 (m, 2H), 2.84–2.94 (m, 2H), 3.16–3.29
(m, 3H), 3.31–3.40 (m, 1H), 3.44 (d, J=13.2 Hz, 1H), 3.50 (d, J=
13.2 Hz, 1H), 3.98 (tt, J=6.7, 4.1 Hz, 1H), 4.86 (brs, 1H), 6.17 (t, J=
5.6 Hz, 1H), 6.40 (brs, 1H), 7.20–7.26 (m, 1H), 7.28–7.33 ppm (m,
4H); 13C NMR (100 MHz, CDCl3): d=22.7, 22.7, 24.6, 25.1, 26.7, 26.8,
30.5, 30.9, 31.9, 33.5, 35.5, 35.7, 39.7, 39.9, 41.3, 44.7, 54.0, 54.2,
57.6, 59.2, 60.4, 63.3, 68.2, 126.9, 128.1, 129.0, 138.5, 171.8,
General procedure 4: coupling of fragment 21a with middle
fragments 17–20 to give amides 47–50. The Cbz-protected g-
amino acid (1.3–2.0 equiv) was dissolved in dry THF (0.1m) at RT.
Et3N was added (5 equiv with respect to the amino acid) followed
by DEPBT (2 equiv with respect to the amino acid) and the mixture
was stirred at RT for 1–2 h. The resulting bright-green solution was
cannulated to a mixture of the amine 21a (1.0 equiv) and solid
Na2CO3 (5 equiv with respect to 21a) and the reaction mixture was
stirred at RT overnight. Half-saturated NaHCO3 solution was added,
and after being stirred for 30 min it was extracted with EtOAc (five
times). The combined organic extracts were dried (MgSO4) and
concentrated in vacuo. The crude product was purified by column
chromatography (SiO2, 3% NEtMe2 in PE/EtOAc) followed by re-
versed-phase HPLC purification. (KNAUER Eurospher II100 RP-C18;
5 mm, 250ꢂ16 mm; eluent: MeCN/H2O)
Benzyl ((2R,3S)-4-(((4R,5R,9S)-4,10-dimethoxy-5,9-dimethyl-6-ox-
odecyl)amino)-2-hydroxy-3-methyl-4-oxobutyl)(methyl)carba-
mate (47). The reaction was performed according to general pro-
cedure 4 with 17 (39.0 mg, 0.139 mmol, 1.6 equiv), and crude
amine 21a (25.5 mg, max. 86.7 mmol, 1.0 equiv) to give 47
(28.1 mg, 62% over two steps) as a colorless oil; Rf =0.37 (PE/
EtOAc/NEtMe2, 30:70:3); [a]2D0 =ꢀ14.38 (c=0.5 in CH2Cl2); 1H NMR
(500 MHz, CD2Cl2): d=0.88 (d, J=6.7 Hz, 3H), 0.94 (d, J=7.0 Hz,
3H), 1.23 (d, J=6.9 Hz, 3H), 1.31–1.45 (m, 2H), 1.46–1.59 (m, 3H),
1.60–1.72 (m, 2H), 2.16–2.24 (m, 0.3H), 2.27–2.34 (m, 0.7H), 2.45
(ddd, J=17.5, 8.9, 6.3 Hz, 1H), 2.51 (ddd, J=17.5, 9.0, 6.3 Hz, 1H),
2.76 (virt. quint., J=7.2 Hz, 1H), 2.99 (s, 3H), 3.05–3.23 (m, 3H),
3.15 (dd, J=9.2, 6.1 Hz, 1H), 3.19 (dd, J=9.2, 6.1 Hz, 1H), 3.24 (s,
3H), 3.28 (s, 3H), 3.37–3.44 (m, 1.3H), 3.49 (dd, J=14.2, 5.6 Hz,
0.7H), 3.68–3.77 (m, 1H), 4.04 (brs, 1H), 5.11 (s, 2H), 5.92 (brs,
0.3H), 6.59 (brs, 0.7H), 7.28–7.34 (m, 1H), 7.35–7.40 ppm (m,
4H);13C NMR (125 MHz, CD2Cl2): d=12.5*, 12.7, 16.2, 16.4*, 17.3,
24.9, 25.0*, 27.8, 27.8, 33.4, 36.2, 36.6*, 39.7*, 39.8, 40.9, 43.3, 43.5*,
49.4, 54.0*, 54.7, 57.8, 59.0, 67.5*, 67.7, 73.3, 73.5*, 78.7, 82.4*, 82.6,
128.2, 128.5, 129.0, 137.5, 157.9, 175.7, 176.0*, 213.8 ppm;[61f]
HRMS-ESI m/z [M+H]+ calcd for C28H47N2O7+: 523.3378, found:
523.3379.
173.2 ppm; IR (film): n˜ =3293, 2929, 2362, 1640, 1557, 1453, 1100,
734, 699 cmꢀ1; HRMS-ESI m/z [M+Na]+ calcd for C31H52N4NaO3
:
+
551.3932, found: 551.3933.
(S)-4-(2-((R)-1-Benzylpiperidin-3-yl)acetamido)-N-(3-((R)-1-isopen-
tylpiperidin-3-yl)propyl)-2-methoxybutanamide (13). The reac-
tion was performed according to general procedure 3 with acid
16a (33.5 mg, 124 mmol, 1.19 equiv), and amine (derived from
amide 43, max. 104 mmol, 1.0 equiv) to give 13 (44.0 mg, 78%) as
a colorless oil: Rf =0.29 (PE/EtOAc/iPrNH2, 50:50:3); [a]2D0 =ꢀ13.78
1
(c=1.0 in CH2Cl2); H NMR (500 MHz, CDCl3): d=0.81–0.89 (m, 1H),
0.90 (d, J=6.6 Hz, 6H), 0.97–1.06 (m, 1H), 1.15–1.28 (m, 2H), 1.38–
1.45 (m, 2H), 1.50–1.69 (m, 9H), 1.71–1.84 (m, 3H), 1.84–1.91 (m,
3H), 1.99–2.05 (m, 2H), 2.06–2.15 (m, 2H), 2.30–2.38 (m, 2H), 2.64–
2.70 (m, 1H), 2.73 (d, J=10.0 Hz, 1H), 2.85–2.94 (m, 2H), 3.17–3.32
(m, 3H), 3.36–3.42 (m, 1H), 3.38 (s, 3H), 3.44 (d, J=13.3 Hz, 1H),
3.49 (d, J=13.3 Hz, 1H), 3.63 (t, J=5.8 Hz, 1H), 5.98 (t, J=5.4 Hz,
1H), 6.62–6.68 (m, 1H), 7.21–7.26 (m, 1H), 7.28–7.32 ppm (m, 4H);
13C NMR (125 MHz, CDCl3): d=22.7, 22.7, 24.6, 25.2, 26.8, 27.0, 30.5,
30.9, 31.8, 32.0, 33.6, 35.6, 35.7, 35.9, 39.2, 41.6, 53.9, 54.2, 57.5,
58.3, 59.4, 60.4, 63.3, 80.9, 126.9, 128.1, 129.0, 138.6, 171.9,
171.9 ppm; IR (film): n˜ =3274, 2931, 2362, 1648, 1541, 1456, 1115,
700 cmꢀ1
;
HRMS-ESI m/z [M+Na]+ calcd for C32H54N4NaO3
:
+
Benzyl ((R)-4-(((4R,5R,9S)-4,10-dimethoxy-5,9-dimethyl-6-oxode-
cyl)amino)-2-hydroxy-4-oxobutyl)carbamate (48). The reaction
was performed according to general procedure 4 with 18 (57.8 mg,
0.228 mmol, 2.0 equiv), and crude amine 21a (33.2 mg, max.
0.112 mmol, 1.0 equiv) to give 48 (34.3 mg, 62% over two steps) as
a colorless oil: Rf =0.22 (EtOAc/NEtMe2, 100:3); [a]2D0 =ꢀ17.78 (c=
1.0 in CH2Cl2); 1H NMR (400 MHz, CD2Cl2): d=0.88 (d, J=6.7 Hz,
3H), 0.95 (d, J=7.1 Hz, 3H), 1.29–1.44 (m, 2H), 1.47–1.59 (m, 3H),
1.60–1.73 (m, 2H), 2.23–2.35 (m, 2H), 2.45 (ddd, J=17.5, 8.6,
5.9 Hz, 1H), 2.52 (ddd, J=17.5, 8.9, 6.3 Hz, 1H), 2.77 (quint., J=
7.1 Hz, 1H), 3.15 (dd, J=9.3, 6.2 Hz, 1H), 3.17–3.24 (m, 3H), 3.19
(dd, J=9.3, 6.0 Hz, 1H), 3.25 (s, 3H), 3.28 (s, 3H), 3.29–3.35 (m, 1H),
3.38–3.43 (m, 1H), 4.02 (dddd, J=7.9, 6.5, 4.0, 3.9 Hz, 1H), 4.41
(brs, 1H), 5.08 (s, 2H), 5.42 (t, J=6.1 Hz, 1H), 6.18 (t, J=6.1 Hz,
1H), 7.28–7.33 (m, 1H), 7.33–7.37 ppm (m, 4H); 13C NMR (100 MHz,
CD2Cl2): d=12.5, 17.3, 25.0, 27.8, 27.8, 33.4, 39.9, 40.1, 40.8, 46.7,
565.4088, found: 565.4086.
(S)-4-(2-((R)-1-Benzylpiperidin-3-yl)-N-methylacetamido)-N-(3-
((R)-1-isopentylpiperidin-3-yl)propyl)-2-methoxybutanamide
(46). The reaction was performed according to general procedure 3
with acid 16a (35.9 mg, 133 mmol, 1.09 equiv), and crude amine
(derived from amide 44, max. 122 mmol, 1.0 equiv) to give 46
(58.7 mg, 86%) as
a colorless oil: Rf =0.40 (PE/EtOAc/iPrNH2,
50:50:3); [a]2D0 =ꢀ12.88 (c=1.0 in CH2Cl2); 1H NMR (500 MHz,
CDCl3): d=0.78–0.86 (m, 1H), 0.88 (d, J=6.6 Hz, 6H), 0.95–1.05 (m,
1H), 1.14–1.26 (m, 2H), 1.34–1.41 (m, 2H), 1.48–1.67 (m, 9H), 1.70–
1.91 (m, 5H), 1.91–2.04 (m, 2H), 2.10–2.20 (m, 2H), 2.20–2.33 (m,
3H), 2.62–2.70 (m, 1H), 2.74–2.81 (m, 1H), 2.82–2.87 (m, 2H), 2.87
(s, 1.2H), 2.95 (s, 1.8H), 3.12–3.31 (m, 3H), 3.33–3.39 (m, 0.4H), 3.35
(s, 1.8H), 3.38 (s, 1.2H), 3.40–3.45 (m, 1H), 3.46–3.53 (m, 1H), 3.55
(dd, J=7.0, 4.4 Hz, 0.6H), 3.60 (dd, J=7.1, 4.1 Hz, 0.4H), 3.65–3.72
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
17
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
49.3, 57.9, 59.0, 67.2, 68.6, 78.7, 82.5, 128.4, 128.6, 129.0, 137.4,
157.5, 172.4, 213.9 ppm; HRMS-ESI m/z [M+Na]+ calcd for
C26H42N2NaO7+: 517.2884, found: 517.2886.
1H), 1.36–1.46 (m, 1H), 1.49–1.59 (m, 4H), 1.59–1.71 (m, 3H), 1.72–
1.83 (m, 2H), 1.94–2.02 (m, 1H), 2.04–2.14 (m, 1H), 2.16–2.22 (m,
1H), 2.22–2.30 (m, 2H), 2.45 (ddd, J=17.5, 8.9, 5.8 Hz, 1H), 2.51
(ddd, J=17.5, 9.1, 6.2 Hz, 1H), 2.62–2.70 (m, 1H), 2.72–2.80 (m,
2H), 2.88 (s, 0.3H), 3.03 (s, 2.7H), 3.12–3.19 (m, 2H), 3.19–3.23 (m,
3H), 3.24 (s, 2.7H), 3.25 (s, 0.3H), 3.28 (s, 3H), 3.35–3.52 (m, 3H),
3.57 (dd, J=13.9, 5.2 Hz, 1H), 3.61–3.67 (m, 0.9H), 3.73–3.78 (m,
0.1H), 4.40 (d, J=5.6 Hz, 1H), 6.17 (t, 6.0 Hz, 0.1H), 6.75 (t, J=
5.3 Hz, 0.9H), 7.19–7.26 (m, 1H), 7.26–7.34 ppm (m, 4H); 13C NMR
(125 MHz, CD2Cl2): d=12.5*, 12.8, 16.0, 16.2*, 17.3, 24.9, 25.1*, 25.5,
27.7, 27.8, 31.4, 33.4, 33.6, 34.5*, 37.9, 38.4, 39.8, 40.8*, 41.0, 43.9*,
44.1, 49.2*, 49.4, 54.3, 54.5, 54.8*, 57.8, 59.0, 60.4, 63.9, 72.5*, 73.6,
78.6, 82.4*, 82.6, 127.3, 128.6, 129.5, 139.6, 174.3, 175.4,
213.8 ppm;[61e] IR (film): n˜ =3309, 2931, 2361, 1711, 1646, 1558,
1456, 1100, 700 cmꢀ1; HRMS-ESI m/z [M+H]+ calcd for C34H57-
N3O6+: 604.4320, found: 604.4324.
Benzyl ((S)-4-(((4R,5R,9S)-4,10-dimethoxy-5,9-dimethyl-6-oxode-
cyl)amino)-3-methoxy-4-oxobutyl)carbamate (49). The reaction
was performed according to general procedure 4 with 19 (51.5 mg,
0.193 mmol, 1.5 equiv), and crude amine 21a (37.5 mg, max.
0.127 mmol, 1.0 equiv) to give 49 (36.7 mg, 57% over two steps)
as a colorless oil: Rf =0.26 (PE/EtOAc/NEtMe2, 50:50:3); [a]D20
=
ꢀ33.98 (c=1.0 in CH2Cl2); 1H NMR (400 MHz, CD2Cl2): d=0.88 (d,
J=6.6 Hz, 3H), 0.94 (d, J=7.0 Hz, 3H), 1.29–1.44 (m, 2H), 1.50–1.61
(m, 3H), 1.62–1.71 (m, 2H), 1.81–1.97 (m, 2H), 2.45 (ddd, J=17.5,
8.7, 5.9 Hz, 1H), 2.51 (ddd, J=17.5, 8.9, 6.3 Hz, 1H), 2.77 (dq, J=
7.8, 7.0 Hz, 1H), 3.15 (dd, J=9.3, 6.2 Hz, 1H), 3.19 (dd, J=9.3,
6.0 Hz, 1H), 3.21–3.29 (m, 3H), 3.24 (s, 3H), 3.28 (s, 3H), 3.30–3.35
(m, 1H), 3.38 (s, 3H), 3.39–3.44 (m, 1H), 3.64 (dd, J=6.7, 5.0 Hz,
1H), 5.06 (s, 2H), 5.19–5.27 (m, 1H), 6.68 (t, J=5.2 Hz, 1H), 7.27–
7.34 (m, 1H), 7.34–7.41 ppm (m, 4H); 13C NMR (75 MHz, CD2Cl2):
d=12.6, 17.3, 25.1, 27.7, 27.8, 33.1, 33.4, 38.2, 39.4, 40.9, 49.3, 57.8,
58.9, 59.0, 66.9, 78.7, 81.4, 82.5, 128.5, 128.4, 129.0, 137.7, 156.8,
(R)-4-(2-((R)-1-Benzylpiperidin-3-yl)acetamido)-N-((4R,5R,9S)-4,10-
dimethoxy-5,9-dimethyl-6-oxodecyl)-3-hydroxybutanamide (51).
The reaction was performed according to general procedure 3 with
16a (14.8 mg, 54.9 mmol, 2.1 equiv), and crude amine (derived
from amide 48, max. 26.5 mmol, 1.0 equiv) to give 51 (10.6 mg,
70% over two steps) as a colorless oil: Rf =0.19 (EtOAc/NEtMe2,
172.2, 213.7 ppm; HRMS-ESI m/z [M+Na]+ calcd for C27H44N2NaO7
+
: 531.3041, found: 531.3040.
1
100:6); [a]2D0 =ꢀ14.78 (c=0.5 in CH2Cl2); H NMR (500 MHz, CD2Cl2):
Benzyl ((S)-4-(((4R,5R,9S)-4,10-dimethoxy-5,9-dimethyl-6-oxode-
cyl)amino)-3-methoxy-4-oxobutyl)(methyl)carbamate (50). The
reaction was performed according to general procedure 4 with 20
(48.6 mg, 0.178 mmol, 1.5 equiv), and amine 21a (33.2 mg, max.
0.112 mmol, 1.0 equiv) to give 50 (35.7 mg, 61% over two steps) as
a colorless oil: Rf =0.30 (PE/EtOAc/NEtMe2, 50:50:3); [a]2D0 =ꢀ41.38
d=0.88 (d, J=6.7 Hz, 3H), 0.94 (d, J=7.0 Hz, 3H), 0.97–1.07 (m,
1H), 1.30–1.45 (m, 2H), 1.48–1.59 (m, 4H), 1.59–1.69 (m, 3H), 1.69–
1.75 (m, 1H), 1.79 (t, J=9.9 Hz, 1H), 1.96–2.09 (m, 3H), 2.09–2.18
(m, 1H), 2.18–2.29 (m, 2H), 2.46 (ddd, J=17.5, 8.8, 5.7 Hz, 1H), 2.51
(ddd, J=17.5, 9.0, 6.2 Hz, 1H), 2.61–2.69 (m, 1H), 2.63–2.68 (m,
1H), 2.73–2.80 (m, 1H), 3.12–3.24 (m, 5H), 3.24 (s, 3H), 3.28 (s, 3H),
3.34 (ddd, J=13.9, 6.3, 3.6 Hz, 1H), 3.38–3.43 (m, 1H), 3.41 (d, J=
13.1 Hz, 1H), 3.47 (d, J=13.1 Hz, 1H), 3.94 (ddt, J=7.9, 6.3, 4.0 Hz,
1H), 4.66 (brs, 1H), 6.11 (t, J=5.5 Hz, 1H), 6.34 (t, J=5.3 Hz, 1H),
7.20–7.25 (m, 1H), 7.27–7.31 ppm (m, 4H); 13C NMR (125 MHz,
CD2Cl2): d=12.6, 17.3, 25.0, 25.4, 27.8, 27.8, 31.1, 33.4, 34.2, 39.9,
40.4, 40.9, 41.7, 45.3, 49.3, 54.6, 57.9, 59.0, 60.0, 63.8, 68.8, 78.7,
82.5, 127.4, 128.6, 129.5, 139.6, 172.3, 173.5, 213.8 ppm; IR (film):
1
(c=1.0 in CH2Cl2); H NMR (400 MHz, CD2Cl2): d=0.88 (d, J=6.7 Hz,
3H), 0.94 (d, J=7.0 Hz, 3H), 1.31–1.44 (m, 2H), 1.49–1.60 (m, 3H),
1.60–1.72 (m, 2H), 1.75–1.88 (m, 1H), 1.99 (dddd, J=14.0, 8.5, 6.6,
4.2 Hz, 1H), 2.45 (ddd, J=17.5, 8.6, 5.9 Hz, 1H), 2.51 (ddd, J=17.5,
8.9, 6.2 Hz, 1H), 2.76 (virt. quint., J=7.3 Hz, 1H), 2.90 (s, 3H), 3.15
(dd, J=9.1, 6.0 Hz, 1H), 3.19 (dd, J=9.1, 6.0 Hz, 1H), 3.20–3.27 (m,
2H), 3.24 (s, 3H), 3.27–3.31 (m, 2H), 3.28 (s, 3H), 3.33–3.50 (m, 4H),
3.57 (dd, J=7.6, 4.2 Hz, 1H), 5.10 (s, 2H), 6.71 (brs, 0.5H), 6.59 (brs,
0.5H), 7.27–7.33 (m, 1H), 7.34–7.38 ppm (m, 4H); 13C NMR
(100 MHz, CD2Cl2): d=12.6, 17.3, 25.1, 27.8, 27.8, 31.1, 31.4*, 33.4,
34.4, 35.0*, 39.4, 40.9, 45.5, 45.9*, 49.3, 57.8, 58.7, 58.8*, 59.0, 67.3,
78.7, 80.9, 82.6, 128.2, 128.4, 128.9, 137.9, 156.5, 156.7*, 172.0,
172.2*, 213.7 ppm;[61a] HRMS-ESI m/z [M+Na]+ calcd for
C28H46N2NaO7+: 545.3197, found: 545.3197.
n˜ =3295, 2931, 2360, 1711, 1646, 1542, 1456, 1100, 700 cmꢀ1
;
HRMS-ESI m/z [M+H]+ calcd for C32H54N3O6+: 576.4007, found:
576.4005.
(S)-4-(2-((R)-1-Benzylpiperidin-3-yl)acetamido)-N-((4R,5R,9S)-4,10-
dimethoxy-5,9-dimethyl-6-oxodecyl)-2-methoxybutanamide (15).
The reaction was performed according to general procedure 3 with
16a (18.6 mg, 68.9 mmol, 2.1 equiv), and crude amine (derived
from amide 49, max. 32.2 mmol, 1.0 equiv) to give 15 (14.7 mg,
77% over two steps) as a colorless oil: Rf =0.18 (PE/EtOAc/NEtMe2,
40:60:3); [a]2D0 =ꢀ30.88 (c=0.5 in CH2Cl2); 1H NMR (500 MHz,
CD2Cl2): d=0.88 (d, J=6.7 Hz, 3H), 0.94 (d, J=7.1 Hz, 3H), 0.97–
1.04 (m, 1H), 1.30–1.43 (m, 2H), 1.50–1.58 (m, 3H), 1.58–1.67 (m,
3H), 1.68–1.87 (m, 5H), 1.95–2.08 (m, 4H), 2.45 (ddd, J=17.5, 9.0,
6.1 Hz, 1H), 2.51 (ddd, J=17.5, 9.0, 6.2 Hz, 1H), 2.63–2.69 (m, 1H),
2.69–2.74 (m, 1H), 2.77 (dq, J=7.8, 7.1 Hz, 1H), 3.15 (dd, J=9.2,
6.1 Hz, 1H), 3.18–3.24 (m, 2H), 3.19 (dd, J=9.2, 6.1 Hz, 1H), 3.24 (s,
3H), 3.25–3.35 (m, 2H), 3.28 (s, 3H), 3.37 (s, 3H), 3.38–3.43 (m, 1H),
3.40 (d, J=13.2 Hz, 1H), 3.46 (d, J=13.2 Hz, 1H), 3.59 (dd, J=6.5,
5.3 Hz, 1H), 5.95 (t, J=4.8 Hz, 1H), 6.72 (t, J=5.4 Hz, 1H), 7.20–7.25
(m, 1H), 7.27–7.32 ppm (m, 4H); 13C NMR (125 MHz, CD2Cl2): d=
12.6, 17.3, 25.1, 25.5, 27.7, 27.8, 31.2, 32.8, 33.4, 34.2, 36.4, 39.4,
40.9, 42.0, 49.3, 54.5, 57.8, 58.8, 59.0, 60.1, 63.9, 78.6, 81.5, 82.5,
127.3, 128.6, 129.5, 139.7, 172.1, 172.3, 213.7 ppm; IR (film): n˜ =
3296, 2930, 2361, 1711, 1648, 1539, 1452, 1104, 699 cmꢀ1; HRMS-
ESI m/z [M+Na]+ calcd for C33H55N3NaO6+: 612.3983, found:
612.3982.
General procedure 5: Cbz deprotection of amides 47–50. Pd/C
(10%, 2.0 mg per 1.0 mg amide) was added to a stirred solution of
Cbz protected amide 47–50 in 0.075m methanolic HCl
(7.0 mgmLꢀ1). The resulting reaction mixture was stirred for 25 min
under an atmosphere of H2. Afterward, the reaction mixture was fil-
tered through a PTFE syringe filter (Chromafil Xtra, PTFE-45/25
0.45 mm). The resulting filtrate was concentrated in vacuo to yield
the crude amines (91–95%) as hydrochlorides which were used di-
rectly without further purification.
(2S,3R)-4-(2-((R)-1-Benzylpiperidin-3-yl)-N-methylacetamido)-N-
((4R,5R,9S)-4,10-dimethoxy-5,9-dimethyl-6-oxodecyl)-3-hydroxy-
2-methylbutanamide (14). The reaction was performed according
to general procedure 3 with 16a (18.6 mg, 68.9 mmol, 2.1 equiv),
and crude amine (derived from amide 47, max. 32.2 mmol,
1.0 equiv) to give 14 (14.7 mg, 76% over two steps) as a colorless
oil: Rf =0.13 (PE/EtOAc/NEtMe2, 25:75:3); [a]2D0 =ꢀ9.88 (c=0.5 in
1
CH2Cl2); H NMR (500 MHz, CD2Cl2): d=0.88 (d, J=6.7 Hz, 3H), 0.94
(d, J=7.0 Hz, 2.7H), 0.96 (d, J=7.3 Hz, 0.3H), 0.97–1.08 (m, 1H),
1.20 (d, J=7.1 Hz, 2.7H), 1.27 (d, J=7.1 Hz, 0.3H), 1.29–1.36 (m,
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
18
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[6] I. Spector, F. Braet, N. R. Shochet, M. Bubb, Microsc. Res. Tech. 1999, 47,
18–37.
(S)-4-(2-((R)-1-Benzylpiperidin-3-yl)-N-methylacetamido)-N-
((4R,5R,9S)-4,10-dimethoxy-5,9-dimethyl-6-oxodecyl)-2-methoxy-
butanamide (52). The reaction was performed according to gener-
al procedure 3 with 16a (15.3 mg, 56.7 mmol, 1.8 equiv), and crude
amine (derived from amide 50, max. 31.4 mmol, 1.0 equiv) to give
52 (14.8 mg, 78% over two steps) as a colorless oil: Rf =0.19 (PE/
EtOAc/NEtMe2, 60:40:3); [a]2D0 =ꢀ32.78 (c=0.5 in CH2Cl2); 1H NMR
(500 MHz, CD2Cl2): d=0.88 (d, J=6.6 Hz, 3H), 0.94 (d, J=7.1 Hz,
1.8H), 0.95 (d, J=7.1 Hz, 1.2H), 0.98–1.05 (m, 1H), 1.30–1.43 (m,
2H), 1.50–1.70 (m, 7H), 1.70–1.80 (m, 2H), 1.81–2.02 (m, 3H), 2.04–
2.11 (m, 1H), 2.11–2.27 (m, 2H), 2.42–2.54 (m, 2H), 2.62–2.70 (m,
1H), 2.73–2.80 (m, 2H), 2.83 (s, 1.2H), 2.94 (s, 1.8H), 3.13–3.21 (m,
2H), 3.21–3.30 (m, 2H), 3.24 (s, 1.8H), 3.25 (s, 1.2H), 3.28 (s, 3H),
3.34 (s, 1.8H), 3.34–3.43 (m, 3H), 3.38 (s, 1.2H), 3.45–3.59 (m, 3H),
6.63 (t, J=5.8 Hz, 0.4H), 6.78 (t, J=5.6 Hz, 0.6H), 7.19–7.25 (m, 1H),
7.26–7.32 ppm (m, 4H); 13C NMR (125 MHz, CD2Cl2): d=12.4*, 12.5,
17.1, 24.9, 25.0*, 25.4, 27.6 (2C), 30.7, 31.3*, 31.4, 31.6*, 33.2, 33.3*,
33.5*, 33.7, 35.8, 37.7*, 38.3, 39.2, 39.3*, 40.7*, 40.8, 44.2, 46.2*,
49.1*, 49.2, 54.3, 54.3*, 57.7, 57.7*, 58.6, 58.7*, 58.8, 60.3, 60.4*,
63.7, 78.5, 80.4*, 80.8, 82.3*, 82.4, 127.1, 128.4, 129.3, 139.5, 171.5,
171.8*, 172.0*, 172.1, 213.4*, 213.5 ppm;[61b] IR (film): n˜ =3310,
2930, 2363, 1712, 1646, 1526, 1405, 1106, 700 cmꢀ1; HRMS-ESI m/z
[M+H]+ calcd for C34H57N3O6+: 604.4320, found: 604.4323.
[7] L. R. Otterbein, P. Graceffa, R. Dominguez, Science 2001, 293, 708–711.
[8] For reviews, see: a) K.-S. Yeung, I. Paterson, Angew. Chem. Int. Ed. 2002,
41, 4632–4653; Angew. Chem. 2002, 114, 4826–4847; b) J. S. Allingham,
V. A. Klenchin, I. Rayment, Cell. Mol. Life Sci. 2006, 63, 2119–2134.
[9] Other natural products, such as the latrunculins, selectively address the
ATP binding pocket between domains 2 and 4, leading to significant
conformational changes within both these domains. As a result, the
actin monomers can no longer interact with the existing filament or
other monomers, and therefore, polymerization is effectively sup-
pressed: a) I. Spector, N. R. Shochet, Y. Kashman, A. Groweiss, Science
1983, 219, 493–495; b) M. Couꢃ, S. L. Brenner, I. Spector, E. D. Korn,
FEBS Lett. 1987, 213, 316–318.
[10] a) J. Tanaka, Y. Yan, J. Choi, J. Bai, V. A. Klenchin, I. Rayment, G. Marriott,
Proc. Natl. Acad. Sci. USA 2003, 100, 13851–13856; b) V. A. Klenchin, J.
Allingham, R. King, J. Tanaka, G. Marriott, I. Rayment, Nat. Struct. Mol.
Biol. 2003, 10, 1058–1063.
[11] a) J. A. Roesener, P. J. Scheuer, J. Am. Chem. Soc. 1986, 108, 846–847;
b) S. Matsunaga, N. Fusetani, K. Hashimoto, J. Am. Chem. Soc. 1986, 108,
847–849; c) M. R. Kernan, T. F. Molinski, D. J. Faulkner, J. Org. Chem.
1988, 53, 5014–5020; d) S. Matsunaga, N. Fusetani, K. Hashimoto, K.
Koseki, M. Noma, H. Noguchi, U. Sankawa, J. Org. Chem. 1989, 54,
1360–1363.
[12] a) G. Guella, I. Mancini, G. Chiasera, F. Pietra, Helv. Chim. Acta 1989, 72,
237–246; b) M. V. D’Auria, L. Gomez-Paloma, L. Minale, A. Zampella, J. F.
Verbist, C. Roussakis, C. Debitus, Tetrahedron 1993, 49, 8657–8664;
c) M. V. D’Auria, L. Gomez-Paloma, L. Minale, A. Zampella, J. F. Verbist, C.
Roussakis, C. Debitus, J. Patissou, Tetrahedron 1994, 50, 4829–4834;
d) C. Bassarello, G. Bifulco, A. Zampella, M. V. D’Auria, R. Riccio, L.
Gomez-Paloma, Eur. J. Org. Chem. 2001, 39–44; e) J. S. Allingham, A.
Zampella, M. V. D’Auria, I. Rayment, Proc. Natl. Acad. Sci. USA 2005, 102,
14527–14532; f) R. D. Perrins, G. Cecere, I. Paterson, G. Marriott, Chem.
Biol. 2008, 15, 287–294.
[13] a) S. Carmely, Y. Kashman, Tetrahedron Lett. 1985, 26, 511–514; b) Y.
Kato, N. Fusetani, S. Matsunaga, K. Hashimoto, R. Sakai, T. Higa, Y. Kash-
man, Tetrahedron Lett. 1987, 28, 6225–6228; c) M. Kobayashi, J. Tanaka,
T. Katori, M. Matsuura, M. Yamashita, I. Kitagawa, Chem. Pharm. Bull.
1990, 38, 2409–2418; d) M. Doi, T. Ishida, M. Kobayashi, I. Kitagawa, J.
Org. Chem. 1991, 56, 3629–3632; e) S. Tsukamoto, M. Ishibashi, T.
Sasaki, J. Kobayashi, J. Chem. Soc. Perkin Trans. 1 1991, 3185–3188;
f) J. S. Todd, K. A. Alvi, P. Crews, Tetrahedron Lett. 1992, 33, 441–442;
g) M. Kobayashi, K. Kawazoe, T. Okamoto, T. Sasaki, I. Kitagawa, Chem.
Pharm. Bull. 1994, 42, 19–26.
Biological evaluations
Cell culture and growth inhibition assays. Mammalian cervical carci-
noma cell line KB-31 and human mammary carcinoma cell line
MCF-7 were obtained from the German collection of Microorgan-
isms and Cell Cultures (DSMZ). Growth inhibition was measured as
previously described, and metabolic activity was determined after
five days by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT) assay.[53]
Fluorescence microscopy. Potoroo (Potorous tridactylis) PtK2 (NCL-5)
kidney cells (ATCC CCL56) from the German collection of Microor-
ganisms and Cell Cultures (DSMZ) were grown on ME medium sup-
plemented with nonessential amino acids and 10% fetal bovine
serum. Growth conditions and details on the fluorescence micros-
copy studies were described previously.[17b] The novel ligands were
evaluated at concentrations of 25 mgmLꢀ1
.
[14] a) K. Yamada, M. Ojika, T. Ishigaki, Y. Yoshida, H. Ekimoto, M. Arakawa, J.
Am. Chem. Soc. 1993, 115, 11020–11021; b) M. Ojika, H. Kigoshi, T. Ishi-
gaki, T. Tsuboi, T. Ogawa, K. Yamada, J. Am. Chem. Soc. 1994, 116, 7441–
7442.
Acknowledgements
[15] a) D. Gouiffꢄs, S. Moreau, N. Helbecque, J. L. Bernier, J. P. Hꢃnichart, Y.
Barbin, D. Laurent, J. F. Verbist, Tetrahedron 1988, 44, 451–459; b) C.
Roussakis, N. Robillard, D. Riou, J. F. Biard, G. Pradal, P. Piloquet, C. Deb-
itus, J. F. Verbist, Cancer Chemother. Pharmacol. 1991, 28, 283–292;
c) J. F. Biard, C. Roussakis, J. M. Kornprobst, D. Gouffes-Barbin, J. F. Verb-
ist, J. Nat. Prod. 1994, 57, 1336–1345; d) D. Riou, C. Roussakis, J. F.
Biard, J. F. Verbist, Anticancer Res. 1993, 13, 2331–2334; e) A. V. Statsuk,
R. Bai, J. L. Baryza, V. A. Verma, E. Hamel, P. A. Wender, S. A. Kozmin, Nat.
Chem. Biol. 2005, 1, 383–388; f) S. A. Rizvi, V. Tereshko, A. A. Kossiakoff,
S. A. Kozmin, J. Am. Chem. Soc. 2006, 128, 3882–3883.
We thank the Deutsche Forschungsgemeinschaft (DFG; Me 2756/
4-1 and Forschergruppe 1406) and the DFG Graduiertenkol-
leg 850 (scholarship to S.D.) for generous funding, Julian Horn
and Raiko Hahn for technical assistance, and Andreas Schneider
for excellent assistance with HPLC separations.
[16] J. C. Blain, Y.-F. Mok, J. Kubanek, J. S. Allingham, Chem. Biol. 2010, 17,
802–807.
Keywords: actin · analogues · bistramide · polyketides ·
rhizopodin
[17] a) F. Sasse, H. Steinmetz, G. Hçfle, H. Reichenbach, J. Antibiot. 1993, 46,
741–748; b) T. M. A. Gronewold, F. Sasse, H. Lꢅnsdorf, H. Reichenbach,
Cell Tissue Res. 1999, 295, 121–129; c) N. Horstmann, D. Menche, Chem.
Commun. 2008, 5173–5175; d) N. Oku, A. Matsumoto, T. Matsunaga, Y.
Asano, H. Kasai, S. Matoba, Y. Igarashi, Chem. Pharm. Bull. 2014, 62,
294–300.
[1] T. D. Pollard, L. Blanchoin, R. D. Mullins, Annu. Rev. Biophys. Biomol.
Struct. 2000, 29, 545–576.
[2] T. P. Stossel, J. Biol. Chem. 1989, 264, 18261–18264.
[3] M. A. Jordan, L. Wilson, Curr. Opin. Cell Biol. 1998, 10, 123–130.
[4] A. Giganti, E. Friederich, Prog. Cell Cycle Res. 2003, 5, 511–525.
[5] a) L. A. Cameron, P. A. Giardini, F. S. Soo, J. A. Theriot, Nat. Rev. Mol. Cell
Biol. 2000, 1, 110–119; b) A. Ploubidou, M. Way, Curr. Opin. Cell Biol.
2001, 13, 97–105.
[18] For a review, see Ref. [10b].
[19] Rhizopodin: G. Hagelueken, G. Albrecht, R. Jansen, H. Steinmetz, D.
Heinz, M. Kalesse, F. Sasse, W.-D. Schubert, Angew. Chem. Int. Ed. 2009,
48, 595–598; Angew. Chem. 2009, 121, 603–606.
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
19
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[20] For a leading review on the total synthesis of actin binding macrolides,
see Ref. [8].
[37] L. A. Marcaurelle, E. Comer, S. Dandapani, J. R. Duvall, B. Gerard, S. Kesa-
van, M. D. Lee IV, H. Liu, J. T. Lowe, J. Marie, C. A. Mulrooney, B. A.
Pandya, A. Rowley, T. D. Ryba, B.-C. Suh, J. Wei, D. W. Young, L. B. Akella,
N. T. Ross, Y.-L. Zhang, D. M. Fass, S. A. Reis, W.-N. Zhao, S. J. Haggarty,
M. Palmer, M. A. Foley, J. Am. Chem. Soc. 2010, 132, 16962–16976.
[38] S. Nishiguchi, M. O. Sydnes, A. Taguchi, T. Regnier, T. Kajimoto, M. Node,
Y. Yamazaki, F. Yakushiji, Y. Kiso, Y. Hayashi, Tetrahedron 2010, 66, 314–
320.
[39] Y. Kobayashi, T. Tsuchiya, S. Umezawa, T. Yoneta, S. Fukatsu, H. Umeza-
wa, Bull. Chem. Soc. Jpn. 1987, 60, 713–720.
[40] For a comprehensive review, see: J. Li, D. Menche, Synthesis 2009,
2293–2315.
[41] a) A. Abiko, J.-F. Liu, S. Masamune, J. Am. Chem. Soc. 1997, 119, 2586–
2587; b) T. Inoue, J.-F. Liu, D. C. Buske, A. Abiko, J. Org. Chem. 2002, 67,
5250–5256; c) A. Abiko, J.-F. Liu, J. Org. Chem. 1996, 61, 2590–2591.
[42] J. Li, P. Li, D. Menche, Synlett 2009, 2417–2420.
[43] M. G. Organ, J. Wang, J. Org. Chem. 2003, 68, 5568–5574.
[44] B. Lal, B. N. Pramanik, M. S. Manhas, A. K. Bose, Tetrahedron Lett. 1977,
18, 1977–1980.
[45] a) O. Mitsunobu, Synthesis 1981, 1–28; b) K. C. Kumara Swamy, N. N.
Bhuvan Kumar, E. Balaraman, K. V. P. Pavan Kumar, Chem. Rev. 2009, 109,
2551–2651.
[46] Y. G. Gololobov, I. N. Zhmurova, L. F. Kasukhin, Tetrahedron 1981, 37,
437–472.
[21] Total syntheses of bistramides: a) P. Wipf, T. D. Hopkins, Chem. Commun.
2005, 3421–3423; b) M. T. Crimmins, A. C. DeBaillie, J. Am. Chem. Soc.
2006, 128, 4936–4937; c) J. T. Lowe, I. E. Wrona, J. S. Panek, Org. Lett.
2007, 9, 327–330; d) J. S. Yadav, L. Chetia, Org. Lett. 2007, 9, 4587–
4589; e) L. Tomas, G. B. Gennas, M. A. Hiebel, P. Hampson, D. Gueyrard,
B. Pelotier, J. Yli-Kauhaluoma, O. Piva, J. M. Lord, P. G. Goekjian, Chem.
Eur. J. 2012, 18, 7452–7466.
[22] Total syntheses of rhizopodin: a) M. Dieckmann, M. Kretschmer, P. Li, S.
Rudolph, D. Herkommer, D. Menche, Angew. Chem. Int. Ed. 2012, 51,
5667–5670; Angew. Chem. 2012, 124, 5765–5768; b) S. M. Dalby, J. M.
Goodwin-Tindall, I. Paterson, Angew. Chem. Int. Ed. 2013, 52, 6517–
6521; Angew. Chem. 2013, 125, 6645–6649; c) M. Kretschmer, M. Die-
ckmann, P. Li, S. Rudolph, D. Herkommer, J. Troendlin, D. Menche,
Chem. Eur. J. 2013, 19, 15993–16018; formal total synthesis: d) K. K. Pu-
lukuri, T. K. Chakraborty, Org. Lett. 2014, 16, 2284–2287.
[23] For synthetic studies toward rhizopodin, see: a) Z. Cheng, L. Song, Z.
Xu, T. Ye, Org. Lett. 2010, 12, 2036–2039; b) T. Chakraborty, K. K. Pulu-
kuri, M. Sreekanth, Tetrahedron Lett. 2010, 51, 6444–6446; c) T. K. Chak-
raborty, M. Sreekanth, K. K. Pulukuri, Tetrahedron Lett. 2011, 52, 59–61;
d) K. K. Pulukuri, T. K. Chakraborty, Org. Lett. 2012, 14, 2858–2861; e) M.
Kretschmer, D. Menche, Org. Lett. 2012, 14, 382–385; f) M. Dieckmann,
S. Rudolph, S. Dreisigacker, D. Menche, J. Org. Chem. 2012, 77, 10782–
10788; g) M. Dieckmann, S. Rudolph, C. Lang, W. Ahlbrecht, D. Menche,
Synthesis 2013, 45, 2305–2315; h) M. Dieckmann, D. Menche, Org. Lett.
2013, 15, 228–231; i) T. Bender, D. Loits, J. M. White, M. A. Rizzacasa,
Org. Lett. 2014, 16, 1450–1453.
[47] W. Lin, X. Zhang, Z. He, Y. Jin, L. Gong, A. Mi, Synth. Commun. 2002, 32,
3279–3284.
[48] D. Enders, C. Thiebes, Synthesis 2000, 510–512.
[49] H. Li, X. Jiang, Y. Ye, C. Fan, T. Romoff, M. Goodman, Org. Lett. 1999, 1,
91–93.
[50] J. D. More, N. S. Finney, Org. Lett. 2002, 4, 3001–3003.
[51] W. S. Wadsworth, W. D. Emmons, J. Am. Chem. Soc. 1961, 83, 1733–
1738.
[24] For a previous SAR study of a monomeric rhizopodin analogue, see:
K. C. Nicolaou, X. Jiang, P. J. Lindsay-Scott, A. Corbu, S. Yamashiro, A.
Bacconi, V. M. Fowler, Angew. Chem. Int. Ed. 2011, 50, 1139–1144;
Angew. Chem. 2011, 123, 1171–1176.
[25] For previous SAR studies of bistramide analogues, see: I. E. Wrona, J. T.
Lowe, T. J. Turbyville, T. R. Johnson, J. Beignet, J. A. Beutler, J. S. Panek, J.
Org. Chem. 2009, 74, 1897–1916.
[26] During preparation of this manuscript, Marriott and co-workers report-
ed the evaluation of side chain analogues based on these vinylamide
side chains: J. H. Pereira, C. Petchprayoon, A. C. Hoepker, N. Moriarty,
S. J. Fink, G. Cecere, I. Paterson, P. D. Adams, G. Marriott, ChemMedChem
2014, 9, 2286–2293.
[52] a) T. Yasuhara, K. Nishimura, M. Yamashita, N. Fukuyama, K.-I. Yamada, O.
Muraoka, K. Tomioka, Org. Lett. 2003, 5, 1123–1126; b) M. E. Kuehne,
S. D. Cowen, F. Xu, L. S. Borman, J. Org. Chem. 2001, 66, 5303–5316.
[53] a) M. Ishiyama, H. Tominaga, M. Shiga, K. Sasamoto, Y. Ohkura, K. Ueno,
Biol. Pharm. Bull. 1996, 19, 1518–1520; b) F. Sasse, H. Steinmetz, T.
Schupp, F. Petersen, K. Memmert, H. Hofmann, C. Heusser, V. Brinkmann,
P. Von Matt, G. Hçfle, H. Reichenbach, J. Antibiot. 2002, 55, 543–551.
[54] Exemplarily data for comparison, Ref. [17]: chronic myelogenous leuke-
mia cell line K-562: 15 ngmLꢀ1; BHK (GBF, hamster): 30 ngmLꢀ1; L929
[27] H. Kigoshi, K. Suenaga, M. Takagi, A. Akao, K. Kanematsu, N. Kamei, Y.
Okugawa, K. Yamada, Tetrahedron 2002, 58, 1075–1102.
(mouse): 15 ngmLꢀ1
.
[28] A detailed analysis of the X-ray-derived binding interactions of aplyro-
nine with actin has been reported: K. Hirata, S. Muraoka, K. Suenaga, T.
Kuroda, K. Kato, H. Tanaka, M. Yamamoto, M. Takata, K. Yamada, H. Ki-
goshi, J. Mol. Biol. 2006, 356, 945–954.
[29] All figures represent the actin protein from the co-crystal with reidis-
pongiolide A (PDB ID: 2ASM). Gelsolin was aligned from the corre-
sponding co-crystal with actin (PDB ID: 1H1V). The position of bistra-
mide in panels a) and b) was obtained by alignment from co-crystal
PDB ID: 1FXU, and the position of kabiramide C was obtained by align-
ment from structure PDB ID: 1QZ5.
[55] Compound 53 was prepared by coupling a diastereomeric mixture of
21a (derived from reaction of 34 with an older, partially racemized
batch of Roche aldehyde 35) with 1.5 equiv (2S,3R)-4-((tert-butoxycarbo-
nyl)(methyl)amino)-3-hydroxy-2-methylbutanoic acid (see synthesis of
17) using DEPBT (3.0 equiv), Et3N (7.5 equiv), and Na2CO3 (5.0 equiv) in
51% yield.
[56] Compound 54 was prepared by coupling a diastereomeric mixture of
21a (derived from reaction of 34 with an older, partially racemized
batch of Roche aldehyde 35) with 19 (1.4 equiv) using DEPBT
(3.0 equiv), Et3N (7.5 equiv), and Na2CO3 (5.0 equiv) in 68% yield.
[57] In agreement with these data, very little or no effects on the actin poly-
merization were visible in in vitro assays (actin polymerization biochem
kit muscle actin, with rhizopodin as control) at concentrations up to
1000 mm; see the Supporting Information for details.
[30] G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey, W. E. Hart, R. K.
Belew, A. J. Olson, J. Comput. Chem. 1998, 19, 1639–1662.
[31] H. Gohlke, M. Hendlich, G. Klebe, J. Mol. Biol. 2000, 295, 337–356.
[32] The results for the smaller search areas (75 and 60 grid points in each
dimension) are depicted in the Supporting Information. The docking
convergence (number of structures in the highest populated cluster)
are also summarized there.
[58] Macromodel ver. 9.7, Schrçdinger LLC, New York, NY (USA), 2009.
[59] W. L. Jorgensen, D. S. Maxwell, J. Tirado-Rivers, J. Am. Chem. Soc. 1996,
118, 11225–11236.
[33] Residue F375 is located at the end of the protein sequence and not re-
solved in all crystal structures.
[60] C. A. Sotriffer, H. Gohlke, G. Klebe, J. Med. Chem. 2002, 45, 1967–1970.
[61] Extra signals due to amide conformers marked with an asterisk: a) con-
former ratio=1:1; b) conformer ratio=6:4; c) conformer ratio=65:35;
d) conformer ratio=8:2; e) conformer ratio=9:1; f) conformer ratio=
75:25.
[34] a) Y.-G. Zhu, H.-Z. Kan, L.-Q. Jiang, W.-H. Hu, Synth. Commun. 2012, 42,
1137–1145; b) Enantiomeric excess was determined by HPLC analysis
of a corresponding urea derivative according to: L. Merritt, J. S. Ward
(Eli Lilly Co.), Int. PCT Pub. No. WO1998054179, 1998; Chem. Abstr.
1998, 130, 38385; absolute configuration was confirmed by X-ray crys-
tallography.
Received: November 25, 2014
Revised: December 12, 2014
Published online on && &&, 0000
[35] P. Abraham, J. B. Pitner, A. H. Lewin, J. W. Boja, M. J. Kuhar, F. I. Carrol, J.
Med. Chem. 1992, 35, 141–144.
[36] V. K. Yadav, K. G. Babu, J. Org. Chem. 2004, 69, 577–580.
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
20
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
Less talk, more actin: X-ray-structure-
based in silico studies of the noncova-
lent interactions between complex poly-
ketides and their natural target actin re-
sulted in the design, modular synthesis,
and biological evaluation of a novel and
structurally simplified analogue class
based on a hybrid structure of bistra-
mide A and rhizopodin.
D. Herkommer, S. Dreisigacker, G. Sergeev,
F. Sasse, H. Gohlke, D. Menche*
&& – &&
Design, Synthesis, and Biological
Evaluation of Simplified Side Chain
Hybrids of the Potent Actin Binding
Polyketides Rhizopodin and
Bistramide
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
21
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ