Total synthesis, NMR solution structure, and binding model of the potent histone deacetylase inhibitor FR235222

Article abstract of DOI:10.1002/anie.200501995

An alternative route: The fungal metabolite FR235222, a potent inhibitor of mammalian histone deacetylase (HDAC), has been synthesized. Key steps are the preparation of unusual amino acids Ahoda and (2R,4S)-MePro. A 3D model for cyclopeptide inhibitor int

Full text of DOI:10.1002/anie.200501995

Angewandte
Chemie
Total Synthesis
DOI: 10.1002/anie.200501995
Total Synthesis, NMR Solution Structure, and
Binding Model of the Potent Histone Deacetylase
Inhibitor FR235222**
Manuela Rodriquez, Stefania Terracciano, Elena Cini,
Giulia Settembrini, Ines Bruno, Giuseppe Bifulco,
Maurizio Taddei,* and Luigi Gomez-Paloma*
The organization of DNA into chromatin plays a major role in
gene regulation. Changes in chromatin architecture (chroma-
tin remodeling) can be modulated by acetylation, methyl-
ation, phosphorylation, and ubiquitination of N-terminal
histone tails.[1] Histone acetylation is the best-understood
mechanism of chromatin remodeling, and is essentially
governed by the antagonistic activity of histone acetyltrans-
ferases (HATs) and histone deacetylases (HDACs).[2] As
acetylation is a key component of gene expression, HAT and
HDAC inhibitors or activators have been studied as potential
therapeutic agents.[3] In the frame of the various histone
acetylase inhibitors, natural cyclopeptides play an important
role as reversible or irreversible zinc chelators with effective
anticancer and antiprotozoal activities. The conformational
features of selected members of this class of compounds (for
example, apicidin and congeners) have been established by X-
ray and NMR spectroscopic studies and by comparison of
their biological activity with that of simplified synthetic
analogues.[4]
Cyclopeptide FR235222 (1; see Scheme 2), which was
recently isolated from the fermentation broth of Acremonium
sp.,[5] showed a potent and selective inhibition of T cell
[*] Dr. M. Rodriquez, E. Cini, G. Settembrini, Prof. M. Taddei
Dipartimento Farmaco Chimico Tecnologico
Università di Siena
via A. Moro 2, 53100 Siena (Italy)
Fax: (+39)0577-234-333
E-mail: taddei.m@unisi.it
Dr. M. Rodriquez, Dr. S. Terracciano, Prof. I. Bruno, Prof. G. Bifulco,
Prof. L. Gomez-Paloma
Dipartimento di Scienze Farmaceutiche
Università di Salerno
via Ponte don Melillo, 84084 Fisciano, Salerno (Italy)
Fax: (+39)089-962-828
E-mail: gomez@unisa.it
[**] Financial support by the University of Salerno, the University of
Siena, and MIUR (Rome) is gratefully acknowledged. L.G.P., I.B.,
and G.B. also wish to acknowledge the use of the instrumental
(NMR spectroscopy and mass spectrometry) facilities of the Center
of Competence in Diagnostics and Molecular Pharmaceutics
supported by Regione Campania (Italy) through POR funds. L.G.P.
thanks Professor R. Abagyan for the use of ICM 3.2 (The Scripps
Research Institute and Molsoft LLC, La Jolla, USA). We thank
undergraduates P. De Nicola, M. De Vito, and F. Cristofano for
technical assistance.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 423 –427
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
423

Communications
proliferation and lymphokine production, and also exhibited
a potent inhibition of HDAC.[6] This immunosuppressant
activity highlights 1 as an attractive target for chemical
synthesis and, in fact, Ma and co-workers recently reported its
total synthesis.[7] The availability of a synthetic sample of 1
was also deemed important to examine its HDAC binding
features, taking into account that details of the determinants
for the recognition between cyclopeptides and HDAC
enzymes are only partly and indirectly known. Finally, besides
the inherent interest evoked by its biological properties, we
envisaged that a convergent synthetic approach, combined
with key information on its bioactive conformation and
HDAC binding interaction, would pave the way to the
assembly of focused libraries of simplified agents and to the
rational design of more powerful and selective analogues.
Hence, in the context of our continuous efforts in
synthesizing peptide-related molecules,[8] and animated by
the opportunities that would result in terms of chemical
biology and medicinal chemistry studies,[9] we looked for an
effective and possibly flexible route to 1. Even at first glance it
is apparent that the most difficult task, from a synthetic
viewpoint, resides in the stereoselective access to the two key
noncoded amino acids present in 1, namely (2S,9R)-2-amino-
9-hydroxy-8-oxodecanoic acid (Ahoda) and trans-4-methyl-d-
proline, in the form of suitably protected building blocks for
their subsequent incorporation into standard solid-phase
peptide chemistry protocols.
Ahoda, the long-chain a-amino acid residue, is a typical
structural element of class I and II HDAC peptide inhibi-
tors,[3b] all of which invariably present one of such substrate-
analogue moieties,[10] with the function to chelate the ZnII ion
in the active site. Although Ahoda is present in many other
cyclopeptide HDAC inhibitors,[4b,11] only one synthesis has
been reported to date.[7,12] We decided to follow a retrosyn-
thetic approach, in which the configuration at C2 was derived
from l-glutamic acid and the configuration at C9 was derived
from (R)-lactic acid. Thus, compound 2, prepared as pre-
viously reported,[13] was homologated to the six-carbon-chain
aldehyde 4 via the enol ether 3 (Scheme 1). Wittig–Horner–
Emmons reaction with dimethyl (R)-[3-(tert-butyldimethylsi-
lyloxy)-2-oxobutyl]phosphonate[14] gave compound 5 in good
yield as a single diastereoisomer. Reduction and contempo-
raneous protecting-group exchange was carried out by
catalytic hydrogenation (in a Parr apparatus) to give N-Boc-
O-TBDMS-Ahoda (6), ready for use in peptide synthesis.
Scheme 1. Reagents and conditions: a) MeOCH2PPh3Cl, KHDMS,
THF, 08C; 75%; b) HCl/AcOEt; 94%; c) (R)-(MeO)2P(O)CH2COCH(O-
TBDMS)CH3, LiCl, DIEA, MeCN; 87%; d) Pd(OH)2, H2 (6 atm), Boc2O,
MeOH; 75%; e) BnBr, NaOH, Na2CO3, H2O followed by SOCl2 in
MeOH; 76%; f) KHMDS, THF, MeI, À788C; 92%; g) Pd(OH)2, H2
(6 atm), MeOH, then Boc2O, Et3N, DMAP, MeCN followed by column
chromatography; 60%; h) BH3·Me2S in THF, 408C, 72 h; 60%;
i) LiOH, THF, H2O; j) TFA/CH2Cl2, TIS, followed by FmocOSu,
Na2CO3, H2O; 77%. KHMDS=potassium 1,1,1,3,3,3-hexamethyldisila-
zane, TBDMS=tert-butyldimethylsilyl, DIEA=ethyldiisopropylamine,
Boc=tert-butyloxycarbonyl, Bn=benzyl, DMAP=4-dimethylaminopyri-
dine, TFA=trifluoroacetic acid, TIS=triisopropylsilane, FmocOSu=9-
fluorenylmethyloxycarbonyl-N-hydroxysuccinimide.
1
HPLC and 600-MHz H NMR analyses of 6 established that
racemization did not occur along the full synthetic pathway.
Substituted prolines have been extensively investigated as
rigid elements for conformationally restricted peptidomimet-
ics, and many synthetic procedures have been described to
implement a wide variety of functional groups on pyrrolidine-
2-carboxy derivatives.[15] In particular, 4-methylproline (4-
MePro), which is also present in other natural products,[16] has
been recently prepared by stereocontrolled hydrogenation of
the corresponding methylene derivative obtained from 4-
hydroxyproline.[17] Appealing as it is, with its use of a naturally
occurring starting material, this method becomes econom-
ically demanding when access to 4-MePro of the d series is
desired, as in our case. In an attempt to rationalize our needs
in terms of the chiral pool,[18] we sought an efficient route by
starting again from glutamic acid. Clearly, a key step was the
stereoselective preparation of (2R,4S)-4-MeGlu.
Special care was taken in the selection of the protection
scheme, as it was known that the NHBoc group strongly
favors the attack of MeI on the opposite side of the Glu g-
enolate, as a consequence of a chelat-ion-controlled transition
state.[19] Assuming that the formation of another anionic site
at the nitrogen atom was therefore detrimental for our
purposes, we found that the N,N-dibenzyl dimethyl ester 7
could be stereoselectively methylated in position 4 to give 8 as
a 4:1 mixture of diastereomers.[20] Upon catalytic hydro-
424
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 423 –427

Angewandte
Chemie
genation, 8 directly produced the ring-closure lactam, from
which, after N-Boc protection, the major isomer 9 was
isolated by column chromatography (60% yield). The con-
figuration of 9 was determined by comparison with previously
reported NMR data.[15,21] Chemoselective reduction of the
lactam carbonyl group, followed by ester hydrolysis and Boc/
Fmoc protecting-group exchange, finally gave the required
Fmoc-amino acid 12 in three steps (46% yield).
was cyclized by exposure to HATU (2 equiv) and DIEA
(2 equiv) in a very dilute solution to avoid side formation of
the cyclodimeric sequence of 1 as by-product. The cyclized
product was obtained in 68% yield after purification by
HPLC. The analytical data ([a]D, electrospray mass spec-
1
trometry, H and 13C NMR spectroscopy) measured on the
synthetic sample were virtually indistinguishable from those
reported for the natural product.[24]
With compounds 6 and 12 in hand, the synthesis of 1 was
attempted by following a typical solid-phase peptide chemis-
try protocol (Scheme 2), using a 2-chlorotrityl resin.[22] Fmoc-
d-4-MePro was loaded on the resin and followed by HATU/
HOBT promoted couplings with l-Phe and l-Iva. Product 6
Next we determined the solution conformation of 1, a
logical prerequisite for the subsequent analysis of the nature
of its binding interactions with HDAC enzymes. Although
cyclopeptides represent an important class of HDAC inhib-
itors with potential activity as antitumor and antiprotozoal
compounds, no relevant information on the type and nature of
the molecular determinants for their interactions with HDAC
isozymes has yet been reported. The crystal structures of
HDAC–ligand complexes are, in fact, limited to small-
molecule inhibitors, such as trichostatin A (TSA) and sub-
eroylanilide hydroxamic acid (SAHA), with no examples of
inhibitors with a large head (cap) group.[25] A conformational
analysis of 1 was carried out by restrained molecular
mechanics (MM) and dynamics (MD) calculations (1 ns, see
Supporting Information).
The solution structure depicted in Figure 1 was obtained
by using 21 distance constraints, previously collected in a
Figure 1. NMR solution conformation of FR235222 (1) obtained by
restrained MD calculations by using ROESY (tmix =100 ms) data
collected at 600 MHz in CDCl3 and [D6]DMSO.
series of ROESY spectra (CDCl3 and [D6]DMSO, 600 MHz,
t
mix = 50, 100, and 150 ms). Visual inspection of this confor-
mation shows that the molecule consists of a central core,
which comprises the tetrapeptide backbone, the methylpro-
line (MePro) ring, and the whole isovaline moiety, from which
phenylalanine and Ahoda side-chain appendages project
facing in opposite directions (Figure 1). The MePro residue
is characterized by a trans geometry of its peptide linkage, as
witnessed by an intense ROESY cross-peak between Ha of
Phe and Hd2 of MePro. The tetrapeptide backbone describes a
g turn and a g inverse turn, as indicated by the f/y torsion
angle values (4-MePro: + 76/À83; Ahoda: + 78/À60; Iva:
À89/ + 93; Phe: À108/ + 102). Three hydrogen-bonding inter-
actions, all between residues i and i + 2, stabilize the two
g turns. Thus, the Phe-CO and NH moieties are H-bonded to
Ahoda-NH and CO (2.09 and 1.92 Š, respectively), whereas
4-MePro-CO is weakly H-bonded to Iva-NH (2.7 Š).
Scheme 2. Final stages of the total synthesis of 1: a) DIEA, CH2Cl2;
b) 20% piperidine, DMF; c) Fmoc-AA or Ahoda derivative 6, HOBT,
HBTU, NMM, DMF; d) AcOH/TFE/CH2Cl2 (2:2:6); e) TFA/H2O/TIS
(94:4:1); f) HATU, CH2Cl2, c=7.7”10À5 m. DIEA=N,N-diisopropyl-
ethylamine, DMF=N,N-dimethylformamide, AA=amino acid,
HOBT=N-hydroxybenzotriazole, HBTU=O-(benzotriazol-1-yl)-1,1,3,3-
tetramethyluronium hexafluorophosphate, NMM=N-methylmorpho-
line, TFE=2,2,2-trifluoroethanol, PS-2-ClTrt-Cl resin=polystyrene/2-
chlorotrityl chloride resin.
was introduced as the last amino acid before cleavage from
the resin with AcOH/TFE/TIS, which gave the linear
precursor (15) of FR235222 in 70% yield. After removal of
TBDMS and Boc groups from Ahoda by treatment with TFA/
H2O/TIS (95% yield),[23] the unprotected linear tetrapeptide
The next step was a docking study between FR235222 (1)
and HDAC with the software ICM 3.2 (MolSoft LLC,
Angew. Chem. Int. Ed. 2006, 45, 423 –427
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
425

Communications
La Jolla, CA), which takes advantage of a particularly
Also remarkable is the difference of positioning between
the TSA and FR235222 head groups (Figure 2b). On the basis
of such a model, our hypothesis is that cyclic tetrapeptide
inhibitors of this kind are able to bind with high affinity to
HDAC because there is a large, shallow cavity that can
accommodate the globe-shaped tetrapeptide core, thus allow-
ing the substrate-mimetic chain to extend and reach the zinc
ion. In more detail, the phenyl group of TSA is bent toward
Y91 (E98 in HDAC1) and P22 (P29), whereas the FR235222
macrolactam domain is projected on the other side, and fills a
cavity delimited by F200 and Y91 (Figure 2a). The role of the
methyl group of 4-MePro raises the speculation that more
bulky groups at this position could potentially allow addi-
tional van der Waals and hydrophobic interactions with the
protein surface. A final consideration relates to the backbone
features of the cyclopeptide core. The consecutive g turn and
g inverse turn are certainly essential for the correct global
shape of the core and for the positioning of the Ahoda side
chain. This should be taken into account for the design of
simplified active analogues.
efficient proprietary docking algorithm. For ICM docking
the receptor is represented by a series of grid potentials, and
both ligand and receptor are flexible and explicit molecules.
Moves of ligand and receptor (side chain and loops) are
computed on the basis of pseudo-Brownian dynamics.[26] The
docking procedure was first validated by reproducing TSA
positioning in the histone deacetylase-like protein (HDLP)
binding pocket, with the TSA–HDLP complex (3D coordi-
nates: PDB archive code 1C3R) as reference.[25,27] As the final
root-mean-square deviation (RMSD) computed on TSA
heavy atoms (docked versus X-ray) was less than 0.1 Š, we
proceeded further on docking FR235222. Figure 2 shows an
On the basis of this model it appears that, besides the
macrolactam scaffold, three quite voluminous appendages
may be housed within the cavity between F200 and Y91. In
other words, there should be quite a wide range of allowed
decorations that may conveniently replace the Phe and Iva
residues of FR235222, a finding partly confirmed by struc-
ture–activity relationship data on cyclopeptide HDAC inhib-
itors of this type.[28] In a rational design of new peptidomi-
metic HDAC inhibitors, assuming that the zinc-binding
element can be easily implemented by a variety of different
solutions,[10] novel scaffolds should be selected from among
rigid mono- or bicyclic elements bearing two or three
decorating appendages. These should preferably be aromatic
side chains to take advantage of the location of F200 and Y91,
two sites for favorable p–p stacking interactions.
In conclusion, we have synthesized FR235222 (1) by using
a convergent approach, classical solid-phase peptide synthesis
(SPPS) protocols and macrolactamization in dilute solution.
Two non-proteinogenic residues, Ahoda and 4-MePro, were
stereoselectively prepared in suitably protected forms for
their subsequent employment in SPPS. In addition, we have
studied the NMR solution structure of 1 and, more impor-
tantly, carried out docking studies with HDAC-like protein.
Our integrated study allowed us to derive a 3D model for
cyclopeptide interaction with the active site of HDAC, thus
highlighting critical differences between the binding modes of
small-molecule and cyclopeptide inhibitors.
Figure 2. a) 3D model of the interaction between FR235222 (1) and
HDLP. For a more convenient visual inspection of the positioning of 1
in the active site, TSA is also displayed in cyan. The protein is
represented by ribbons (blue to red, N to C terminus) with only a
selection of side chains shown. Gold: side chains externally delimiting
the binding site; white: those involved in H-bonding interactions
(black dashed lines with distances in Š). Only heavy atoms are
II
displayed, exceptfor polar hydrogen atoms (gray). The Zn ion is
represented by a CPK sphere in light blue. b) The different kinds of
surface complementarity exhibited by TSA (carbon atoms: white) and
FR235222 (carbon atoms: yellow) cap groups toward HDLP are
shown. The HDLP molecular surface is depicted as green wire frames
(see also text).
expanded view of the model of FR235222 bound to the active
site of HDLP. To allow a meaningful comparison, TSA is also
displayed in the same original positioning as that found in the
crystallographic structure of the TSA–HDLP complex. Nota-
bly, this 3D model highlights the fact that, although substan-
tially different in cap groups, both TSA and FR235222 project
their zinc-chelating elements in a very similar fashion.
Among the principal polar interactions that stabilize the
FR235222 molecule within the HDAC binding pocket, the
two oxygen atoms of the a-hydroxyketone terminal function
of Ahoda embrace the metal ion, with the hydroxy group also
involved in two key ligand–protein hydrogen bonds with
H131 and D168, two of the main residues taking an active role
in the enzymatic catalysis (see Figure 2a). Van der Waals
interactions are also of great importance in the FR235222–
HDLP recognition process, especially for the accommodation
of the cap group. Indeed, the tetrapeptide cap group of 1 and
the outward side of the enzyme exhibit complementary
surfaces that correspond with the rim of the tubular pocket
that leads to the zinc ion, the latter being filled by the Ahoda
long chain.
Received: June 9, 2005
Published online: November 28, 2005
Keywords: inhibitors · natural products · peptides ·
.
solid-phase synthesis · total synthesis
[1] a) M. Biel, V. Wascholowski, A. Giannis, Angew. Chem. 2005,
117, 3248 – 3280; Angew. Chem. Int. Ed. 2005, 44, 3186 – 3216;
b) E. J. Richards, S. C. Elgin, Cell 2002, 108, 489 – 500; c) B. M.
Turner, Cell 2002, 111, 285 – 291.
426
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 423 –427

Angewandte
Chemie
[2] a) B. D. Strahl, C. D. Allis, Nature 2000. 403, 41 – 45; b) M. Izuka,
M. M. Smith, Curr. Opin. Genet. Dev. 2003, 13, 154 – 160; c) D.
Chung, Trends Mol. Med. 2002, 8, S10 – S11; d) H. Hess-Stummp,
Eur. J. Cell Biol. 2005, 84, 109 – 121.
Chem. 1986, 51, 3140 – 3143. However, in our hands this route
was less convenient.
[19] S. Hanessian, R. Margarita, Tetrahedron Lett. 1998, 39, 5887 –
5890.
[20] Unfortunately, the tetrabenzyl Glu derivative, an intermediate
also used in Ahoda synthesis, showed too little reactivity under
the same conditions, likely as a result of steric hindrance from
the benzyl groups.
[21] M. C. Moody, D. W. Young, J. Chem. Soc. Perkin Trans. 1 1997,
3519 – 3530.
[3] For HAT inhibitors, see: a) M. Biel, A. Krestovali, E. Karatzali,
J. Papamatheakis, A. Giannis, Angew. Chem. 2004, 116, 4065 –
4067; Angew. Chem. Int. Ed. 2004, 43, 3974 – 3976; for HDAC
inhibitors, see: b) P. A. Marks, T. Miller, V. M. Richon, Curr.
Opin. Pharmacol. 2003, 3, 344 – 351; c) T. A. Miller, D. J. Witter,
S. Belvedere, J. Med. Chem. 2003, 46, 5097 – 5116.
[22] a) S. Terracciano, I. Bruno, G. Bifulco, J. E. Copper, C. D. Smith,
L. Gomez-Paloma, R. Riccio, J. Nat. Prod. 2004, 67, 1325 – 1331;
b) A. Napolitano, M. Rodriquez, I. Bruno, S. Marzocco, G.
Autore, R. Riccio, L. Gomez-Paloma, Tetrahedron 2003, 59,
10203 – 10211.
[23] The amount of TIS used as carbocation scavenger had to be
carefully adjusted, because of its ability to promote a reductive
amination (as side reaction) between the free amino terminus
and the Ahoda carbonyl group in position 8, which gives rise to
an eight-membered ring.
[24] H. Mori, Y. Urano, T. Kinoshita, S. Yoshimura, S. Takase, M.
Hino, J. Antibiot. 2003, 56, 181 – 185.
[25] M. S. Finnin, J. R. Donigian, A. Cohen, V. M. Richon, R. A.
Rifkind, P. A. Marks, R. Breslow, N. P. Pavletich, Nature 1999,
401, 188 – 193.
[4] a) P. T. Meinke, P. Liberator, Curr. Med. Chem. 2001, 8, 211 –
235; b) S. B. Singh, D. L. Zink, J. M. Liesch, R. T. Mosley, A. W.
Dombrowski, G. F. Bills, S. J. Darkin-Rattray, D. M. Schmatz,
M. A. Goetz, J. Org. Chem. 2002, 67, 815 – 825.
[5] H. Mori, Y. Urano, F. Abe, S. Furukawa, S. Furukawa, Y.
Tsurumi, K. Sakamoto, M. Hashimoto, S. Takase, M. Hino, T.
Fujii, J. Antibiot. 2003, 56, 76 – 79.
[6] H. Mori, Y. Urano, F. Abe, S. Furukawa, S. Furukawa, F. Sakai,
M. Hino, T. Fujii, J. Antibiot. 2003, 56, 80 – 86.
[7] W. Xie, B. Zou, D. Pei, D. Ma, Org. Lett. 2005, 7, 2775 – 2777.
This publication appeared in print while our article was being
reviewed.
[8] L. R. Lampariello, D. Piras, M. Rodriquez, M. Taddei, J. Org.
Chem. 2003, 68, 7893 – 7895, and references therein.
[9] For a review on combinatorial libraries in medicinal chemistry,
see: D. L. Boger, J. Desharnais, K. Capps, Angew. Chem. 2003,
115, 4270 – 4309; Angew. Chem. Int. Ed. 2003, 42, 4138 – 4176.
[10] The solutions sought by Nature and medicinal chemists to
engineer substrate (acetyllysine) analogues for HDAC inhibition
encompass various chains of appropriate length, all ending with
a functional group with good metal-ion binding properties, such
as hydroxamate, hydroxycarbonyl, or diamine groups. Accord-
ingly, a relatively unsubstituted C6 chain with a suitable end
group can be considered the minimal structural requirement for
HDAC inhibition, even if, for optimal activity, a so-called cap
(head) group is also necessary. The latter is the most variable
portion among different HDAC inhibitors, and ranges from a
simple phenyl group to large molecular moieties, as in
FR235222.
[26] a) R. A. Abagyan, M. Totrov, D. Kuznetsov, J. Comput. Chem.
1994, 15, 488 – 506; for more recent applications of ICM docking,
see: b) M. Shapira, B. M. Raaka, F. L. Sharmistha, M. Totrov, Z.
Zhou, S. R. Wilson, R. A. Abagyan, H. G. Samuels, Proc. Natl.
Acad. Sci. USA 2003, 100, 7354 – 7359; c) M. Shapira, R. A.
Abagyan, M. Totrov, J. Med. Chem. 2003, 46, 3045 – 3059;
d) C. N. Cavasotto, R. A. Abagyan, J. Mol. Biol. 2004, 337, 209 –
225.
[27] D. F. Wang, O. Wiest, P. Helquist, H. Y. Lan-Hargest, N. Wiech,
J. Med. Chem. 2004, 47, 3409 – 3417.
[28] a) S. L. Colletti, R. W. Myers, S. J. Darkin-Rattray, A. M.
Gurnett, P. M. Dulski, S. Galuska, J. J. Allocco, M. B. Ayer, C.
Li, J. Lim, T. M. Crumley, C. Cannova, D. M. Schmatz, M. J.
Wyvratt, M. H. Fisher, P. T. Meinke, Bioorg. Med. Chem. Lett.
2001, 11, 107 – 111; b) S. L. Colletti, R. W. Myers, S. J. Darkin-
Rattray, A. M. Gurnett, P. M. Dulski, S. Galuska, J. J. Allocco,
M. B. Ayer, C. Li, J. Lim, T. M. Crumley, C. Cannova, D. M.
Schmatz, M. J. Wyvratt, M. H. Fisher, P. T. Meinke, Bioorg. Med.
Chem. Lett. 2001, 11, 113 – 117; c) R. Furumai, Y. Komatsu, N.
Nishino, S. Khochbin, M. Yoshida, S. Horinouchi, Proc. Natl.
Acad. Sci. USA 2001, 98, 87 – 92.
[11] For examples, see: a) Y. Masuoka, K. Shin-Ya, K. Furihata, K.
Nagai, Y. Suzuki, H. Seto, J. Antibiot. 2001, 54, 187 – 190; b) K.
Yoskimura, S. Tsuboya, K. Okazaky, Chem. Abstr. 1995, 123,
548.
[12] For synthesis of the simple 2-amino-8-oxodecanoic acid, see:
a) J. A. Bajgrowicz, A. El Hallaoui, R. Jacquier, C. Pigiere, P.
Viallefont, Tetrahedron 1985, 41, 1833 – 1841; b) W. Kuriyama, T.
Kitahara, Heterocycles 2001, 55, 1 – 6; c) L. Mou, L. Singh,
Tetrahedron Lett. 2001, 42, 6603 – 6606; d) S. Kim, E.-Y. Kim, H.
Ko, Y. H. Jung, Synthesis 2003, 2194 – 2198; e) T. S. Cooper, P.
Laurent, C. J. Moody, A. K. Takle, Org. Biomol. Chem. 2004, 2,
265 – 276; f) M. Rodriquez, I. Bruno, E. Cini, M. Marchetti, M.
Taddei, L. Gomez-Paloma, J. Org. Chem. 2005, in press.
[13] M. Rodriquez, M. Taddei, Synthesis 2005, 3, 493 – 495.
[14] A. Chen, I. Savage, E. J. Thomas, P. D. Wilson, Tetrahedron Lett.
1993, 34, 6769 – 6772.
[15] For examples, see: a) A. M. P. Koskinen, H. Rapoport, J. Org.
Chem. 1989, 54, 1859 – 1866; b) Q. Wang, N. A. Sasaki, P. Potier,
Tetrahedron 1998, 54, 15759 – 15780; c) J. Zhang, W. Wang, C.
Xiong, V. J. Hruby, Tetrahedron Lett. 2003, 44, 1413.
[16] T. Golakoti, W. Y. Yoshida, S. Chaganty, R. E. Moore, Tetrahe-
dron 2000, 56, 9093 – 9102.
[17] J. L. Del Valle, M. Goodman, J. Org. Chem. 2003, 68, 3923 –
3931.
[18] An alternative route based on stereoselective taken into
consideration, see: a) N. K. Yee, Y. Dong, S. R. Kapadia, J. J.
Song, J. Org. Chem. 2002, 67, 8688 – 8691; b) J. K. Thottathil, J. L.
Moniot, R. H. Mueller, M. K. Y. Wong, T. P. Kissick, J. Org.
Angew. Chem. Int. Ed. 2006, 45, 423 –427
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
427