Synthesis of the protein phosphatase 2A inhibitor (4S,5S,6S,10S,11S,12S)-cytostatin

Article abstract of DOI:10.1002/1521-3773(20020517)41:10<1748::AID-ANIE1748>3.0.CO;2-X

Reagent-controlled reactions are key to the introduction of the stereocenters in the total synthesis of the cytostatin isomer 1. The stereochemical flexibility of the synthesis allows efficient and convenient access to other isomers.

Full text of DOI:10.1002/1521-3773(20020517)41:10<1748::AID-ANIE1748>3.0.CO;2-X

COMMUNICATIONS
O
Synthesis of the Protein Phosphatase 2A
O
-
Na+
O
P
Inhibitor (4S,5S,6S,10S,11S,12S)-Cytostatin**
1
4
O
HO
O
OH
12
6
11
5
10
Laurent Bialy and Herbert Waldmann*
Dedicated to Professor Lutz F. Tietze
on the occasion of his 60th birthday
(4S,5S,6S,10S,11S,12S) –1
O
O
-
Na+
O
P
OH
The reversible phosphorylation of proteins is employed by
living organisms for the regulation of innumerable cellular
processes, and aberrant protein phosphorylation contributes
to the development of many human diseases, including cancer
and diabetes.[1] Protein kinases (PKs) catalyze protein phos-
phorylation, whereas protein phosphatases (PPs) are respon-
sible for dephosphorylation. PKs are established targets for
drug discovery.[2] However, the development of small-mole-
cule inhibitors of PPs is emerging only very recently[3] as a
very rapidly growing area of investigation in clinical biology
and medicinal chemistry.[4] Naturally occurring PP inhibi-
tors have been used widely to antagonize PP action in
biology experiments.[5] Thus, natural products with PP-inhib-
itory activity can serve as invaluable starting points in
structural space for the development of potent and selec-
tive PP inhibitors. This was recently demonstrated by the
successful solid-phase synthesis and biological investiga-
tion of analogues of the Cdc25 phosphatase inhibitor dysi-
diolide.[6]
The naturally occurring PP2A inhibitor cytostatin 1 (IC50 ˆ
210 nm[7a]), which was isolated from a Streptomyces strain by
Ishizuka and co-workers,[7b] inhibits the adhesion of B16
melanoma cells to laminin and collagen, displays antimeta-
static and cytotoxic activity,[7c] and induces apoptosis of B16
melanoma cells[7d] at submicromolar concentrations. Herein
we disclose the synthesis of the 4S,5S,6S,10S,11S,12S isomer of
cytostatin.
O
HO
O
OH
OH
OH
2
3
O
O
P
-
Na+
O
O
HO
O
OH
Et
H2N
R
products in both enantiomeric forms[10]), the asymmetric
Evans alkylation, and the enantioselective reduction of an
acetylenic ketone (for which efficient reagent-controlled
processes that give rise to both possible stereoisomers are
known).[11]
Retrosynthetic analysis of cytostatin (1) led to vinyl iodide 4
and dienylstannane 5 (retro Stille coupling). In this strategy,
the labile triene unit is generated in a late step of the synthesis,
which allows convenient and efficient access to more stable
analogues (Scheme 1).[12] The a,b-unsaturated lactone incor-
porated in 4 was traced back to b-hydroxyaldehyde 6 from
which it should be accessible by a Still Gennari olefination[13]
and subsequent lactonization. It was planned to generate both
syn diols (corresponding to C5/C6 and C10/C11 in the natural
product) by means of an asymmetric aldol reaction with N-
Ishizuka et al. determined the constitution of cytostatin, but
not its relative and absolute configuration.[7e] Therefore, in
planning the synthesis, we drew from the structurally related
natural products fostriecin 2[8] and the phoslactomycins 3.[9]
Given that the biosynthesis of these natural products might
follow similar pathways and/or use identical starting materials
and biocatalysts, the configurations of stereocenters 4, 5, 10,
and 12 were chosen by analogy. The configuration of stereo-
center 6 was assigned arbitrarily. The unknown absolute
configuration of cytostatin prompted us to design the syn-
thesis with a high degree of flexibility to allow rapid and
reliable variation of absolute and relative configuration if
desired. This was ensured by employing the asymmetric Evans
aldol condensation (which gives access to syn and anti aldol
O
O
-
Na+
O
P
1
4
O
HO
O
OH
12
6
11
5
10
1
O
O
P
RO
RO
O
O
OH
I
+
Bu3Sn
4
5
O
OR'
OR'' OR'''
12
6
[*] Prof. Dr. H. Waldmann, Dipl.-Chem. L. Bialy
Max-Planck-Institut f¸r molekulare Physiologie
Abteilung Chemische Biologie
6
Otto-Hahn-Strasse 11, 44227Dortmund (Germany)
Fax : (‡49)231-133-2499
O
O
and
O
Fachbereich 3, Organische Chemie, Universit‰t Dortmund
44221 Dortmund (Germany)
O
N
SiMe3
Ph
Fax : (‡49)231-133-2499
7
8
9
E-mail: herbert.waldmann@mpi-dortmund.mpg.de
[**] This research was supported by the Fonds der Chemischen Industrie.
Scheme 1. Retrosynthetic analysis of (4S,5S,6S,10S,11S,12S)-cytostatin.
1748
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4110-1748 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 10

COMMUNICATIONS
propionyloxazolidinone 7 as nucleophile.[14] The chiral alde-
hyde 8, which is accessible by means of an asymmetric
alkylation of the same acyloxazolidinone 7,[15] should serve as
electrophile. After conversion of the double bond into an
aldehyde group, a second aldol reaction would be carried out.
Elaboration of the resulting N-acyl group into an alkynyl
ketone by using trimethylsilylacetylene (9) as nucleophile and
subsequent reduction would yield 6, the precursor to vinyl
iodide 4.
Chiral alcohol 10 was synthesized as described by Evans
et al.[15] and subjected to Swern oxidation to yield the
corresponding aldehyde. The volatile aldehyde was employed
in an asymmetric aldol reaction with the dibutylboron enolate
derived from acyloxazolidinone 7 (Scheme 2). Crystalline
aldol adduct 11 was obtained in high yield and with very high
stereoselectivity. Only traces of undesired diastereomers were
detected and were readily separated by crystallization.
Protection of the hydroxy group as the methoxymethyl
(MOM) ether, reductive removal of the chiral auxiliary with
™LiBH3OH∫[16] (generated in situ), and masking of the
primary alcohol as the tert-butyldimethylsilyl (TBS) ether
yielded olefin 12 with high overall efficiency. Hydroboration
of the double bond by using 9-BBN, oxidative workup, and
subsequent Swern oxidation of the resulting primary alcohol
delivered the desired aldehyde intermediate. As expected, a
second asymmetric syn aldol reaction with the boron enolate
formed from acyloxazolidinone 7 gave b-hydroxyamide 13
with high overall yield and high stereoselectivity (only one
diastereomer could be detected by 1H NMR spectroscopic
analysis). The absolute configuration of aldol adduct 13 was
assigned based on the assumption that the aldol reaction
proceeds in a reagent-controlled fashion with syn selectivity,
as described for various other cases.[14] In the course of the
aldol reaction the TBS group was cleaved, but the primary
alcohol was readily resilylated.
To generate the last stereocenter, an alkynyl ketone was
required. This ketone was obtained by conversion of acylox-
azolidinone 13 into the fully protected Weinreb amide 14,
which was treated with lithium trimethylsilylacetylide. After
removal of the TMS group by treatment with catalytic
amounts of borax in methanol, alkynyl ketone 15 was
obtained in high overall yield. The stereoselective reduction
of the keto group was attempted with different reagents,
including Alpine borane[11a] and borane in the presence of a
phenylglycine-derived oxazaborolidine.[11e] The highest yields
were obtained with borane in the presence of oxazaborolidine
(R)-16 (2 equiv).[11c,d] The desired secondary alcohol was
formed in nearly quantitative yield, and a second diaster-
O
O
OH
a, b
c–e
HO
O
N
Ph
10
11
1
eomer could not be detected by means of H NMR spectro-
TBSO
OPG
O
TBSO
OPG
OH O
scopic analysis. The protection of the propargylic alcohol
required a blocking group that is stable under the conditions
of the planned Still Gennari olefination and the subsequent
acid-mediated lactonization. Furthermore, it had to survive
the removal of the TBS ether. After substantial experimenta-
tion, it was found that the tert-butyldiphenylsilyl (TBDPS)
group fulfils these criteria, and the protecting group pattern
was arranged accordingly. Primary alcohol 17 was selectively
formed by removal of the TBS group with HF/pyridine.
Alcohol 17 was oxidized to the corresponding aldehyde
with DMP,[17] and the carbonyl compound was converted into
Z-olefin 18 under Still Gennari conditions (Scheme 3).
Subsequently, the MOM protecting groups were removed by
treatment with CBr4 in 2-propanol. Under these conditions
(HBr is probably formed in situ),[18] lactone 19 was formed
simultaneously.
i, j
f–h
N
O
Ph
12
13
O
OPG
TBSO
OPG
OPG O
TBSO
OPG
k, l
N
OMe
14
15
Ph
H
N
Ph
O
B
(R)-16
m–o
HO
OPG
OPG OTBDPS
17
To develop a reaction sequence that would give flexible
access to different phosphorylated analogues of the natural
product, it was planned to introduce the phosphate group at
this stage of the synthesis. The correct choice of phosphate-
protecting group proved to be crucial. Initial experiments with
the methoxybenzyl group, which was successfully applied in
the synthesis of fostriecin,[8b] were not successful. Similarly,
the cyanoethyl ester was not suitable because only one
cyanoethyl group could be removed in the final deprotection
step without destroying the entire molecule. Finally, the use of
the fluorenylmethyl ester[19] allowed the successful cleavage of
the protecting group.
Scheme 2. Synthesis of intermediate 17, which incorporates all stereogenic
centers. a) (COCl)2, DMSO, NEt3, CH2Cl2, À788C !RT; b) 7, Bu2BOTf,
DIPEA, CH2Cl2, À788C !RT, then H2O2, pH 7 , 08C !RT, 68% over two
steps; c) MOMCl, DIPEA, CH2Cl2, room temperature, 97% ; d) LiBH4,
H2O, Et2O, 0 8C !RT, 78%; e) TBSCl, imidazole, DMF, room temperature,
97%; f) 9-BBN, THF, room temperature, then H2O2, NaOH, 83%;
g) (COCl)2, DMSO, NEt3, CH2Cl2, À788C !RT; h) 7, Bu2BOTf, DIPEA,
CH2Cl2, À788C !RT, then H2O2, pH 7 , 08C !RT; TBSCl, NEt3, DMAP,
CH2Cl2, room temperature, 86% over two steps; i) Me3Al,
‡
(H2NMeOMe) ClÀ, THF, À108C !08C, 84%; j) MOMCl, DIPEA,
CH2Cl2, room temperature, 92%; k) 9, BuLi, À788C; THF, À788C !
À 108C; l) borax, methanol, H2O, À108C !RT, 95% over two steps;
m) (R)-16, BH3 ¥ Me2S, THF, À308C, 96%; n) TBDPSCl, imidazole, DMF,
room temperature; o) HF/pyridine, THF, room temperature, 40 min, 88%
over two steps. DMSO ˆ dimethyl sulfoxide, Tf ˆ trifluoromethanesulfon-
yl, DIPEA ˆ diisopropylethylamine, DMF ˆ N,N-dimethylformamide,
9-BBN ˆ borabicyclo[3.3.1]nonane, DMAP ˆ 4-dimethylaminopyridine,
PG ˆ protecting group ˆ MOM ˆ methoxymethyl.
The phosphate was formed by phosphitylation and subse-
quent oxidation. After selective cleavage of the TBDPS ether,
the desired alkyne 20 was obtained in high yield. Conversion
Angew. Chem. Int. Ed. 2002, 41, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4110-1749 $ 20.00+.50/0
1749

COMMUNICATIONS
4S,5S,6S,10S,11S,12S isomer of cytostatin 1 was isolated in
85% yield.
OPG
OPG OTBDPS
OH OTBDPS
OH
OH OPG
OPG OTBDPS
a, b
Synthetic 1 displays a specific rotation of [a]2D0 ˆ ‡50 (c ˆ
0.114 in [D4]MeOH). Unfortunately, this value can not be
employed to ascertain the absolute configuration of the
natural product because Ishizuka and co-workers did not
report the specific rotation of the isolated sample.[7e]
The phosphatase-inhibiting activity of synthetic 1 was
investigated with p-nitrophenyl phosphate as substrate.[7d]
Isomer 1 inhibited PP2A (IC50 ˆ 33 nm). This value is one
order of magnitude lower than that of the natural product.
This finding suggests that our assumption about the absolute
configuration of the natural product may be correct, at least
for most of the stereogenic centers.
MeO
O
17
18
c
O
O
O
FmO
P
O
O
OH
d, e
FmO
O
19
20
f, g
O
O
O
O
FmO
FmO
P
In conclusion we have developed an asymmetric synthesis
of the 4S,5S,6S,10S,11S,12S of the PP2A inhibitor cytostatin.
The synthesis is stereochemically flexible and employs only
reagent-controlled transformations, thus allowing access to
each desired isomer of cytostatin in a reliable and efficient
manner. This successful synthesis now opens up new oppor-
tunities for the development of new tools for biological
studies and of new cancer drugs.
P
O
O
OH
I
O
O
h
FmO
FmO
21
22
i
O
-
O
Na+
O
O
P
O
OH
HO
Received: January 31, 2002 [Z18620]
Fm
1
[1] T. Hunter, Cell 2000, 100, 113 127.
[2] For a review, see: A. J. Bridges, Chem. Rev. 2001, 101, 2541 2572.
[3] Z.-Y. Zhang, Curr. Opin. Chem. Biol. 2001, 5, 416 423.
[4] For example, see: J. S. Lazo, D. C. Aslan, E. C. Southwick, K. A.
Cooley, A. P. Ducruet, B. Joo, A. Vogt, P. Wipf, J. Med. Chem. 2001, 44,
4042 4049; R. A. Urbanek, S. J. Suchard, G. B. Steelman, K. S.
Knappenberger, L. A. Sygowski, C. A. Veale, M. J. Chapdelaine, J.
Med. Chem. 2001, 44, 1777 1793.
[5] a) K. Hinterding, D. Alonzo-DÌaz, H. Waldmann, Angew. Chem. 1998,
110, 716 780; Angew. Chem. Int. Ed. 1998, 37, 668 749; b) J. E.
Sheppeck II, C.-M. Gauss, A. R. Chamberlin, Bioorg. Med. Chem.
1997, 5, 1739 1750.
Scheme 3. Synthesis of 1. a) DMP, CH2Cl2, NaHCO3, 93%;
b) (CF3CH2O)2P(O)CH2CO2Me, [18]crown-6, KHMDS, THF, À788C,
92%; c) CBr4, 2-propanol, 828C, 83%; d) (FmO)2PNiPr2, tetrazole,
CH2Cl2/CH3CN, 08C !RT, then I2, pyridine, H2O, THF; 95%; e) HF/
pyridine, THF, room temperature, 28 h, 82%; f) NIS, AgNO3, DMF, room
‡À
À
‡
À ˆ À
temperature, quant.; g) K OOC N N COO K , HOAc, 2-propanol/
dioxane, 63%; h) 5, [PdCl2(CH3CN)2] (cat.), DMF/THF, 62%;
‡
i) NEt3:CH3CN 1:5, room temperature, then Na -Dowex resin, 85%.
DMP ˆ Dess–Martin periodinane, HMDS ˆ hexamethyldisilazane, NIS
N-iodosuccinimide; PG ˆ MOM.
[6] a) D. Brohm, S. Metzger, A. Bhargava, O. M¸ller, F. Lieb, H.
Waldmann, Angew. Chem. 2002, 114, 319 323; Angew. Chem. Int. Ed.
2002, 41, 307 311; b) J. L. Blanchard, D. M. Epstein, M. D. Boisclair,
J. Rudolph, K. Pal, Biorg. Med. Chem. Lett. 1999, 9, 2537-2538; c) M.
Takahashi, K. Dodo, Y. Sugimoto, Y. Aoyagi, Y. Yamada, Y.
Hashimoto, R. Shirai, Biorg. Med. Chem. Lett. 2000, 10, 2571-2574.
[7] a) M. Kawada, M. Amemiya, M. Ishizuka, T. Takeuchi, Biochim.
Biophys. Acta 1999, 1452, 209 217; b) M. Amemiya, M. Ueno, M.
Osono, T. Masuda, N. Kinoshita, C. Nishida, M. Hamada, M. Ishizuka,
T. Takeuchi, J. Antibiot. 1994, 47, 536 540; c) T. Masuda, S.-I.
Watanabe, M. Amemiya, M. Ishizuka, T. Takeuchi, J. Antibiot. 1995,
48, 528 529; d) M. Kawada, M. Amemiya, M. Ishizuka, T. Takeuchi,
Jpn. J. Cancer Res. 1999, 90, 219 225; e) M. Amemiya, T. Someno, R.
Sawa, H. Naganawa, M. Ishizuka, T. Takeuchi, J. Antibiot. 1994, 47,
541 544.
[8] a) Structure determination: G. C. Hokanson, J. C. French, J. Org.
Chem. 1985, 50, 462 466; D. L. Boger, M. Hikota, B. M. Lewis, J. Org.
Chem. 1997, 62, 1748 1753; b) Synthesis: D. L. Boger, S. Ichikawa, W.
Zhong, J. Am. Chem. Soc. 2001, 123, 4161 4167; D. E. Chavez, E. N.
Jacobsen, Angew. Chem. 2001, 113, 3779 3782; Angew. Chem. Int. Ed.
2001, 40, 3667 3670.
[9] Structure determination: T. Kohama, T. Nakamura, T. Kinoshita, I.
Kaneko, A. Shiraishi, J. Antibiot. 1993, 46, 1512 1519; T. Shibata, S.
Kurihara, K. Yoda, H. Haruyama, Tetrahedron 1995, 51, 11999
12012.
of 20 into vinyl iodide (Z)-21 proved to be problematic. After
substantial experimentation, 21 was obtained by conversion of
the alkyne into the corresponding alkynyl iodide in the
presence of silver nitrate followed by reduction of the triple
bond with diimide.[20] All attempts to reduce the triple bond
with different reagents gave undesired side reactions, in
particular, the reduction of the a,b-unsaturated lactone and
ring opening. However, when diimide was generated in situ in
2-propanol, these side reactions were largely suppressed and
vinyl iodide (Z)-21 was obtained in 63% yield (Scheme 3).
This advanced intermediate was then subjected to Stille
coupling with Z,E-dienylstannane 5 with [PdCl2(CH3CN)2] as
catalyst and without additional phosphane.[12] Under these
conditions, sensitive (Z,Z,E)-22 was formed in a satisfactory
yield (62%). Dienylstannane 5 was synthesized from croto-
naldehyde: the latter was converted into the dibromoolefin
with CBr4 and PPh3 treated with nBuLi followed by Bu3SnCl
to give an enyne stannane, and subsequently subjected to
hydrozirconation (not shown). Finally, the phosphate was
unmasked. Upon treatment of phosphoric acid triester 22
with excess triethylamine, both fluorenylmethyl groups
[10] M. Braun, Methoden Org. Chem. (Houben-Weyl) 4th Ed. 1952
Vol. E21b, 1995, pp. 1612 1712.
,
[11] a) M. M. Midland, D. C. McDowell, R. L. Hatch, A. Tramontano, J.
Am. Chem. Soc. 1980, 102, 867 869; b) P. V. Ramachandran, A. V.
were cleaved in
a
b-elimination reaction, and the
1750
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4110-1750 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 10

COMMUNICATIONS
Teodorovic, M. V. Rangaishenvi, H. C. Brown, J. Org. Chem. 1992, 57,
2379 2386; c) K. A. Parker, M. W. Ledeboer, J. Org. Chem. 1996, 61,
3214 3217; d) E. J. Corey, R. K. Bakshi, S. Shibata, J. Am. Chem. Soc.
1987, 109, 5551 5553; e) J. Bach, R. Berenguer, J. Garcia, T.
Loscertales, J. Vilarrasa, J. Org. Chem. 1996, 61, 9021 9025; f) K.
Matsumura, S. Hashiguchi, T. Ikariya, R. Noyori, J. Am. Chem. Soc.
1997, 119, 8738 8739.
recognition. Is it possible to create a set of modules, each of
which recognizes a certain amino acid in a peptidic environ-
ment or, even better, a short peptide sequence?
We recently reported the stabilization of small peptides in
their b-sheet conformation by external ligands.[7] N-acylated
3-aminopyrazoles were shown to interact with every hydro-
gen-bond-donor (D) and acceptor (A) available at the top
face of a dipeptide. Two consecutive amino acids can be
clamped together with one of these heterocycles, which causes
the formation of three almost-linear hydrogen bonds in a
DAD sequence (Figure 1). This recognition process simulta-
neously fixes the peptide in the thermodynamically favorable
b-sheet conformation. Thus, small soluble models of this
important secondary structure have been prepared.
[12] J. K. Stille, B. L. Groh, J. Am. Chem. Soc. 1987, 109, 813 817.
[13] W. C. Still, C. Gennari, Tetrahedron Lett. 1983, 24, 4405 4408.
[14] D. A. Evans, J. A. Bartroli, T. L. Shih, J. Am. Chem. Soc. 1981, 103,
2127 2129.
[15] D. A. Evans, S. L. Bender, J. Morris, J. Am. Chem. Soc. 1988, 110,
2506 2526.
[16] T. D. Penning, S. D. Djuric, R. A. Haack, V. J. Kalish, J. M. Myashiro,
B. W. Rowell, S. S. Yu, Synth. Commun. 1990, 20, 307 312.
[17] D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277 7287.
[18] A. S.-Y. Lee, Y.-J. Hu, S.-F. Chu, Tetrahedron 2001, 57, 2121 2126.
[19] Y. Watanabe, T. Nakamura, H. Mitsumoto, Tetrahedron Lett. 1997, 38,
7407 7410.
[20] For example, see: J.-M. Duffault, J. Einhorn, A. Alexakis, Tetrahedron
Lett. 1991, 32, 3701 3704.
Modular Building Blocks for Amino Acid
Recognition in Peptides
Figure 1. b-Sheet stabilization with aminopyrazoles–three-point binding
of the top face of a dipeptide by the DAD binding sites of the amino-
pyrazole; left: Lewis structure; right: results of molecular-mechanics
calculation (Cerius2, Molecular Simulations, Dreiding 2.21).
Mark Wehner and Thomas Schrader*
The selective recognition of short peptide sequences is a
key concept in many natural regulatory processes. Thus, the
natural antibiotic vancomycin binds tightly to the C-terminal
d-Ala-d-Ala fragment in peptides, which is used for the
construction for bacterial cell walls.[1] Numerous cell cell
recognition events rely on the recognition of the specific Arg-
Gly-Asp sequence.[2]
À
ˆ
In a b sheet, the N H and C O bonds of the backbone
point up and down, whereas the amino acid residues (R) are
extended horizontally away from the peptide in a well-defined
geometry. This preorganized arrangement was our starting
point for a new modular concept of peptide receptors: If it
were possible to attach to the aminopyrazole a rigid U-shaped
substituent with a properly placed binding site at its tip, this
could reach down to the side chain of the respective amino
acid and lead to an additional, specific noncovalent inter-
action. With an interchangeable tip, various binding sites
could be introduced into the basic peptide receptor, which
would lead to a modular set of building blocks, selective for
the typical classes of amino acids. To our knowledge, no
example of such a rationally designed set of peptide receptors
exists to date.
The first attempts to mimic Nature×s potent peptide hosts
with artificial structures, confined the conformational free-
dom of a host molecule by creating a cleft (Rebek and co-
workers), a macrocycle (Still and co-workers), or even a cavity
(Still and co-workers); the specific binding sites were often
taken from (non)natural amino acids.[3] Several groups have
created receptor molecules for important secondary struc-
tures found in peptides and proteins. Thus, a-helical[4] and b-
sheet[5] portions of polypeptides can be recognized by
synthetic ligands with a complementary hydrogen-bond-
donor and acceptor pattern. In recent years, considerable
progress has been achieved with a combinatorial approach.[6]
Various site-specific proteases, such as thrombin, trypsin,
and many others provide a shallow groove for the efficient
recognition of the backbone of the peptide to be cleaved, in its
extended conformation. This array is combined with a specific
binding pocket for the side chains of the target amino acid,
and thus defines the cleavage site. We asked ourselves if such
a rational design could also be used for artificial peptide
For the U-shaped spacer, we chose the framework of
Kemp×s triacid, because it is exceptionally rigid and its
chemistry is well developed.[8] Two acid groups were desig-
nated to carry the monoacylated 3,5-diaminopyrazole unit by
way of an imide functionality, while the third could be coupled
to an aniline derivative with the correct binding site in the m-
position. Scheme 1 shows the general structure of the building
blocks: both the imide and the neighboring aminopyrazole
unit are in the same plane, locked together by an intra-
molecular hydrogen bond.
During the synthesis, care must be taken to chemoselec-
tively address the three amino functionalities of 3,5-diamino-
pyrazole, without an extensive use of protecting groups. We
begin with imide formation between mono(trifluoroacetylat-
ed) diaminopyrazole 1 and Kemp×s acid anhydride
(Scheme 2). The imide intermediate 2 must be N-Boc pro-
[*] Prof. Dr. T. Schrader, M. Wehner
Fachbereich Chemie der Universit‰t
Hans Meerwein Strasse
35032 Marburg (Germany)
Fax : (‡49)6421-28-28917
E-mail: schradet@mailer.uni-marburg.de
Angew. Chem. Int. Ed. 2002, 41, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4110-1751 $ 20.00+.50/0
1751