A sequential high-yielding large-scale solution-method for synthesis of philanthotoxin analogues

Article abstract of DOI:10.1016/S0223-5234(02)00003-X

A general, improved procedure for rapid synthesis of philanthotoxin analogues, a pharmacologically important class of polyamine conjugates, is described. The solution-phase procedure is illustrated by gram-scale synthesis of philanthotoxins PhTX-343 and P

Full text of DOI:10.1016/S0223-5234(02)00003-X

European Journal of Medicinal Chemistry 38 (2003) 117ꢀ122
/
www.elsevier.com/locate/ejmech
Short communication
A sequential high-yielding large-scale solution-method for synthesis
of philanthotoxin analogues
Petrine Wellendorph, Jerzy W. Jaroszewski, Steen Honore´ Hansen, Henrik Franzyk *
Department of Medicinal Chemistry and Department of Analytical Chemistry, Royal Danish School of Pharmacy, Universitetsparken 2, DK-2100
Copenhagen, Denmark
Received 22 July 2002; received in revised form 29 October 2002; accepted 4 November 2002
Abstract
A general, improved procedure for rapid synthesis of philanthotoxin analogues, a pharmacologically important class of
polyamine conjugates, is described. The solution-phase procedure is illustrated by gram-scale synthesis of philanthotoxins PhTX-
343 and PhTX-12. Selectively protected polyamines are coupled to Na-Fmoc-protected amino acid pentafluorophenyl esters. After
removal of the Na-Fmoc group, the amine is coupled with carboxylic acid pentafluorophenyl esters. Deprotection followed by a
rapid and efficient purification by vacuum liquid chromatography on octadecylsilyl silica (RP-18 phase) gave the philanthotoxin
analogues in 74ꢀ78% overall yield.
/
´
# 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
Keywords: Polyamines; Philanthotoxins; Solution-phase synthesis; Purification
1. Introduction
However, this route is not sequential as the N-acyl
group is introduced at an early stage. A similar
Venoms of spiders and wasps contain polyamine
toxins used to paralyse prey [1,2]. Their analogues
comprise a pharmacologically important class of non-
competitive inhibitors of ionotropic receptors. Recently,
potency at a nanomolar level and selectivity towards
various classes of receptors have been demonstrated [1]
with certain analogues of philanthotoxin-433, a venom
constituent of the Egyptian digger wasp. This empha-
approach involving monoacylation of unprotected sper-
mine (1, Fig. 1) with N-butyryl-L-tyrosine via a dicy-
clohexylcarbodiimide coupling has been described [4].
Recently, a number of methods for solid-phase
synthesis of polyamine toxins have been developed [5ꢀ
/
7]. Solid-phase methods are quite convenient for pre-
paration of small amounts of compounds for high-
throughput screening. When a particular compound is
required in larger amounts for more detailed studies,
solid-phase methods become quite expensive due to the
cost of the resin and the need of using a large excess of
reagents, often repeatedly, to ensure a full conversion.
For example, for preparation of ca. 100 mg of a toxin on
solid phase, several grams of reagents are typically used.
Moreover, if the polyamine moiety is symmetrical as in
the case of spermine (1) or other a,v-diamines, a
solution-phase strategy seems more appropriate. The
majority of philanthotoxin analogues contain spermine
(1) as the polyamine moiety [3,8], and the interest in
analogues having an a,v-diamine part (e.g. 1,12-di-
aminododecane) has emerged recently due to their
pronounced selectivity for the nicotinic acetylcholine
receptors [7].
sises the interest in further studies of structureꢀactivity
/
relationships for which a rapid, efficient and scalable
synthetic method is necessary.
The modular structure of philanthotoxins, which
comprises an amino acid moiety, an N-acyl group and
a polyamine, linked via amide bonds, enables easy
construction of combinatorial libraries of philantho-
toxin analogues. Classical methods of polyamine toxin
synthesis, involving coupling of p-nitrophenyl esters of
N-acylamino acids with appropriately protected poly-
amines, have been developed by Nakanishi et al. [3].
* Corresponding author.
E-mail address: hf@dfh.dk (H. Franzyk).
´
0223-5234/02/$ - see front matter # 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
PII: S 0 2 2 3 - 5 2 3 4 ( 0 2 ) 0 0 0 0 3 - X

118
P. Wellendorph et al. / European Journal of Medicinal Chemistry 38 (2003) 117ꢀ122
/
The present work is concerned with the development
of improved methods of synthesis as well as purification
of polyamine toxins. Such methods should allow con-
venient parallel synthesis of small combinatorial li-
braries from readily available and stable building
blocks. A sequential strategy similar to that used in
solid-phase peptide synthesis was chosen. Purification of
the highly polar philanthotoxins is troublesome, and
preparative reverse-phase HPLC, used thus far [6,7], is
not easily applied for gram-scale purifications.
The pharmacologically important toxins, PhTX-343
(2, Fig. 1) (typically employed as a pharmacological
reference compound) and PhTX-12 (3, Fig. 2) each
contain a symmetrical, commercially available polya-
mines, spermine (1) or 1,12-diaminododecane (4), re-
spectively. The amino acid unit (Tyr) was selected to be
incorporated via the corresponding Na-Fmoc-protected
pentafluorophenyl (Pfp) ester, as a vast number of this
type of amino acid derivatives are commercially avail-
able or can readily be prepared [9,10]. The N-acylation
Fig. 2. Synthesis of PhTX-12 (3).
was likewise envisaged to be effected using an active
Pfp-ester of a carboxylic acid.
2. Chemistry
A prerequisite for the chosen strategy is an easy access
to large amounts of tri-Boc-protected spermine (7). A
one-pot literature procedure [11] involving a selective
monoprotection of 1 in methanol at low temperature
(ꢁ78 8C) with a trifluoroacetyl (Tfa) group, followed by
/
Boc-protection of the remaining amino functionalities
and subsequent ammonolysis, posed two obstacles.
Firstly, a distinct precipitation of 1 was observed upon
cooling of the methanolic solution below ꢁ50 8C.
/
Secondly, removal of the N-trifluoroacetyl groups in
6, formed as a substantial by-product, was incomplete
using methanolic ammonia under the conditions de-
scribed [11]. This resulted in an inseparable mixture of 7
and 8 (the latter could be detected in the crude reaction
product by the presence of 13C-NMR signals at d 116
and 157, corresponding to trifluoroacetamide, as op-
posed to trifluoroacetate anion, which appears at d 118
and 163 (in CDCl3)].
These problems were solved by performing the initial
Tfa-protection at ꢁ
/
50 8C (92 8C, manual control)
/
instead of at ꢁ78 8C, and by using aqueous sodium
/
hydroxide to remove the Tfa-protecting groups, to give
pure 7 in 43% yield. This procedure also avoids the
Fig. 1. Synthesis of PhTX-343 (2).

P. Wellendorph et al. / European Journal of Medicinal Chemistry 38 (2003) 117ꢀ
/122
119
formation of trifluoroacetamide as a by-product, which
turned out to be surprisingly difficult to remove.
Coupling of 7 with Fmoc-protected L-tyrosine penta-
enantiomerically pure as shown by capillary electro-
phoresis [6] using heptakis-(2,6-di-O-methyl)-b-cyclo-
dextrin as a chiral additive (Fig. 3). Thus, the chirality
of the amino acid moiety has been preserved during all
synthetic steps. The overall yield of 2 from 7 was 78%.
To demonstrate the efficiency of this protocol further,
a large-scale synthesis of PhTX-12 (3) starting from
mono(Boc)-protected 1,12-diaminododecane (11) [15]
was performed. A modified procedure for mono-Boc-
protection of the diamine 4 using 0.4 equiv. of di-tert-
fluorophenyl ester (with the side chain protected as a
tert-butyl ether) in the presence of one equivalent of
triethylamine and subsequent Fmoc-removal using
octanethiolꢀDBU in THF [12] proceeded in high yield.
/
However, the solubility properties of 9 did not allow a
facile removal of impurities by a trituration procedure,
often feasible for protected peptides [12,13]. Conse-
quently, a purification step was necessary, readily
achieved by vacuum liquid chromatography (VLC) on
silica gel. Acylation of 9 with pentafluorophenyl buty-
rate (obtained from the corresponding acid chloride by
treatment with pentafluorophenol in the presence of
pyridine [14]) and removal of the Boc-protecting groups
gave the final product 2.
butyl pyrocarbonate (Boc2O) in methanolꢀdichloro-
/
methane (3:2) was employed. VLC purification gave 11
in 78% yield. The conversion of 11 via 12 and 13 to 3
proceeded in 74% overall yield.
VLC using C18-coated silica gel proved to be a very
convenient method for purification of gram quantities of
2 in a short time. The final product was, as expected,
3. Conclusion
In conclusion, the present work represents an un-
precedently high-yielding and cost-efficient route to
philanthotoxin analogues; e.g. the overall yield of
PhTX-12 (3) from diamine 4 was 58%. Although the
selective protection of spermine (1) only afforded 43% of
the desired 7, this is reasonably close to the expected ca.
50%. This lower yield is due to the simultaneous
formation of bis-Tfa-spermine protected at both pri-
mary amino groups, which ultimately results in the
formation of di-Boc-spermine as a substantial by-
product (18%). This by-product is, however, useful for
solid-phase synthesis of philanthotoxins [6], or it may be
converted to 7. The method is very atom-economical, as
most reactions are performed using ca. equimolar
amounts of reactants. Also, the present procedure is
based on readily available protected amino acid building
blocks, and we have already prepared several other
analogues with equally satisfactory results. An extensive
biological testing of these analogues are currently being
pursued, and these results will be published in due
course. The present method can be expanded to
analogues containing other diamine or polyamine moi-
eties, the only requirement being the access to suitably
protected precursors, possibly by extension of the
procedures described for 7 or 11. We believe that due
to its sequential nature, the present method will be very
convenient for parallel synthesis of larger amounts of
members of small, focused combinatorial libraries,
exhibiting variations in the amino acid and acyl
moieties. The use of reverse-phase VLC is a major
improvement with respect to the resources spent on
purification. By using a manually packed VLC column
of an appropriate size, it has proved possible to transfer
elution results from analytical reverse-phase TLC (RP-
Fig. 3. Capillary electrophoresis of PhTX-343 (2) prepared as shown
in Fig. 1 (64.5 cmꢂ50 mm I.D. fused silica capillary (56 cm to
/
detector), 30 kV polarity, 30 mM heptakis-(2,6-di-O-methyl)-b-cyclo-
dextrin, 50 mM 6-aminocaproic acid, pH 4.0, 15 8C, detection at 214
nm). Top, (S)-PhTX-343 obtained from
(S)- and (R)-PhTX-343. (i) 1 equiv. F3CCOOEt, then 4 equiv. Boc2O;
(ii) NaOH, H2OꢀMeOH; (iii) Fmoc-NHCH(R)COOPfp, NEt3; (iv)
DBU, octanethiolꢀTHF; (v) R?COOPfp; (vi) TFAꢀCH2Cl2.
L-Tyr. Bottom, Mixture of
/
/
/
18, Rf ca. 0.15ꢀ0.2) directly to the VLC column, thus
/

120
P. Wellendorph et al. / European Journal of Medicinal Chemistry 38 (2003) 117ꢀ122
/
enabling gram-scale purification within a few hours with
a gradient elution whenever necessary.
2.57 (2H, t, Jꢃ
CDCl3): d 12.3 (C-4), 17.4 (C-3), 34.1 (C-2), 124.2 (C-
1?), 135.4ꢀ C-6?), 168.5 (C-1). Anal.
C10H7O2F5 requires C, 47.26; H, 2.78, Found: C,
47.54; H, 2.75%.
/
7.3 Hz, H-2). 13C-NMR (75 MHz,
/
141.7 (C-2?Ã
/
4. Experimental
4.1. Tri-Boc-spermine (7)
4.3. Mono-Boc-1,12-diaminododecane (11)
Spermine (1, 6.03 g, 30 mmol) was dissolved in MeOH
1,12-Diaminododecane (3.00 g, 15 mmol) was dis-
(350 mL) under N2. The mixture was cooled to ꢁ
/
50 8C
solved in CH2Cl2ꢀ
Boc2O (1.31 g, 0.4ꢂ
added. The mixture was stirred for 2 days. Concentra-
tion yielded a residue, which was suspended in CH2Cl2ꢀ
MeOH (12.5:1, 54 mL) and loaded onto a VLC column
(6ꢂ7 cm; Merck 60H silica gel). Gradient elution with
hexane, CH2Cl2, CH2Cl2ꢀMeOH (100:5), CH2Cl2ꢀ
MeOHꢀ32% aqueous NH3 (200:10:1 to 100:10:1)
/
MeOH (3:2, 150 mL), and then
(92 8C, manually temperature-controlled dry iceꢀ
/
/
ace-
/
15 mmol) in MeOH (5 mL) was
tone bath), when ethyl trifluoroacetate (4.1 mL, 35
mmol) was added dropwise during 0.5 h. The tempera-
ture was then allowed to rise to 0 8C, and the mixture
was stirred for an additional 0.5 h at 0 8C. A solution of
di-tert-butyl dicarbonate (30.5 g, 410 mmol) in MeOH
(30 mL) was then added during 10 min. The reaction
mixture was stirred at room temperature (r.t.) for 16 h
(TLC showed a fully acylated product at Rf 0.8 in
EtOAc), at which point H2O (40 mL) and concentrated
aqueous NaOH (30.6 mL) were added. The mixture was
stirred at r.t. for 6 h followed by 16 h at 4 8C. Most of
the MeOH was removed in vacuo. To the residual
suspension was added CH2Cl2 (250 mL) and H2O (100
mL). The aqueous layer was extracted repeatedly with
/
/
/
/
/
yielded 11 (1.41 g, 78%).
4.4. PhTX-343 (2)
Tri-Boc-spermine (7, 2.0 g, 3.98 mmol) and Fmoc-
Tyr(But)-OPfp (2.74 g, 1.1ꢂ
3.98 mmol) were dissolved
in dry CH2Cl2 (10 mL), and then Et3N (0.55 mL, 3.97
mmol) was added. After the mixture was stirred at r.t.
/
CH2Cl2 (7ꢂ
/
100 mL). The combined organic phases
for 20 h (TLC with petroleum etherꢀ
product at Rf 0.3), the reaction mixture was diluted with
CH2Cl2 (100 mL) and washed with water (2ꢂ50 mL).
/
EtOAc 1:1 showed
were dried (Na2SO4) and concentrated. The crude 7
(14.74 g) was dissolved in CH2Cl2 and loaded onto a
/
VLC column (7ꢂ
was eluted successively with petroleum ether, CH2Cl2,
CH2Cl2ꢀMeOH (10:1), and CH2Cl2ꢀMeOHꢀ32% aqu-
eous NH3 (100:5:1 to 40:10:1). This yielded fraction A
followed by 7 (4.49 g, Rf 0.54 in CH2Cl2ꢀMeOHꢀ32%
aqueous NH3; 30%) and di-Boc-spermine (2.21 g, Rf
0.11 in CH2Cl2ꢀMeOHꢀ32% aqueous NH3; 18%).
Repeated chromatography of fraction A afforded tet-
ra-Boc-spermine (5.36 g, Rf 0.86 in CH2Cl2ꢀMeOHꢀ
/
7 cm; Merck 60H silica gel), which
The organic layer was dried (Na2SO4) and concentrated.
The residue was dissolved in dry THF (15 mL) and then
/
/
/
octanethiol (39.8 mL, 10ꢂ
mL, 0.2ꢂ3.98 mmol) were added. The mixture was kept
at r.t. for 24 h (after 3 h, TLC with CH2Cl2ꢀMeOHꢀ
/3.98 mmol) and DBU (0.79
/
/
/
/
/
32% aqueous NH3 200:10:1 showed almost full conver-
sion; product at Rf 0.25). The reaction mixture was
concentrated (removal of THF) and then loaded onto a
/
/
/
/
VLC column (6ꢂ
was eluted successively with CH2Cl2 (300 mL) and
CH2Cl2ꢀMeOHꢀ32% aqueous NH3 200:10:0.5 (1500
/6 cm; Merck 60H silica gel), which
32% aqueous NH3; 35%) and an additional amount of
7 (1.96 g, 13%).
/
/
mL) to give 9 (2.60 g, 91%).
A portion of 9 (2.09 g, 2.89 mmol) was treated with
4.2. Pentafluorophenyl butyrate
C3H7COOPfp (1.48 g, 2.0ꢂ/2.89 mmol) in the presence
Pentafluorophenol (3.94 g, 21.4 mmol) and butyryl
chloride (2.3 mL, 22.5 mmol) were dissolved in CH2Cl2
(100 mL). The mixture was cooled to 0 8C, and then
pyridine (2.7 mL, 34.2 mmol) was added slowly. After
the mixture was stirred at r.t. for 2 h, it was diluted with
CH2Cl2 (250 mL), and subsequently washed with 0.1 M
of Et3N (0.55 mL, 1.37ꢂ2.89 mmol) for 2 h, when the
mixture was concentrated. The residue was redissolved
in CH2Cl2 (5 mL) and treated with an additional
/
amount of C3H7COOPfp (0.27 g, 0.37ꢂ
for an additional 1 h. The reaction was followed by TLC
with petroleum etherꢀacetone 2:1, Rf 0.21 and 0.32
/2.89 mmol)
/
aqueous HCl (2ꢂ
/
200 mL) and brine (200 mL). The
correspond to the reactant and product, respectively.
The reaction mixture was concentrated, and the residue
was dissolved in CH2Cl2 (15 mL) and loaded onto a
organic phase was dried (Na2SO4), filtered, and con-
centrated in vacuo (bath temperature below 30 8C) to
give the pure product as a liquid (5.31 g, 98%, purity at
least 99%), which can be stored as a crystalline solid in a
VLC column (6.5ꢂ
was eluted successively with hexane (300 mL), hexaneꢀ
EtOAc (1:1, 500 mL), hexaneꢀEtOAc (1:1.5, 1000 mL),
and hexaneꢀEtOAc (1:2, 2300 mL) to give 10 (1.95 g,
86%).
/6 cm; Merck 60H silica gel), which
/
freezer for several months; Rf 0.85 (petroleum etherꢀ
/
/
1
EtOAc 2:1). H-NMR (300 MHz, CDCl3): d 0.98 (3H,
t, Jꢃ7.3 Hz, H-4), 1.74 (2H, sextet, Jꢃ7.3 Hz, H-3),
/
/
/

P. Wellendorph et al. / European Journal of Medicinal Chemistry 38 (2003) 117ꢀ
/122
121
Deprotection of 10 (1.66 g, 2.10 mmol) was performed
in CH2Cl2ꢀTFA (9:1, 160 mL) at r.t. during 2.5 h. The
TFA (10 mL) and loaded onto a VLC column packed
with Merck Lichroprep RP-18 (40ꢀ63 mm; 10ꢂ5 cm
/
/
/
mixture was concentrated to dryness, the residue was
dissolved in 0.1% aqueous TFA (10 mL), and the
solution was loaded onto a VLC column packed with
column), which was eluted with 5% MeCN (in 0.1%
aqueous TFA; 200 mL), 10% MeCN (in 0.1% aqueous
TFA; 500 mL), 20% MeCN (in 0.1% aqueous TFA; 500
mL), and 30% MeCN (in 0.1% aqueous TFA; 1000 mL).
RP-18 TLC with 40% MeCN (in 0.1% aqueous TFA)
showed 3 at Rf 0.30 in fractions eluted with 30% MeCN
(in 0.1% aqueous TFA). Concentration of these frac-
tions yielded 3 (0.93 g, 88%; 74% from 11), 1H- and 13C-
NMR spectra of which were identical with those of an
authentic sample [6].
Merck Lichroprep RP-18 (40ꢀ
/
63 mm; 10ꢂ
/
7 cm col-
umn), which was eluted with 0.1% aqueous TFA (500
mL), 1% MeCN (in 0.1% aqueous TFA; 1000 mL), 5%
MeCN (in 0.1% aqueous TFA; 1000 mL), and 10%
MeCN (in 0.1% aqueous TFA; 1500 mL). RP-18 TLC
with 20% MeCN (in 0.1% aqueous TFA) showed 2 at Rf
0.38 in fractions eluted with 5ꢀ10% MeCN (in 0.1%
/
aqueous TFA). Concentration of these fractions yielded
2 (1.63 g, 100%; 78% from 7), 1H- and 13C-NMR spectra
of which were identical with those of an authentic
sample [6].
Acknowledgements
We thank Ms. Uraiwan Ngamrabiab for technical
assistance during the preparative scale-up work.
4.5. PhTX-12 (3)
Mono-Boc-1,12-diaminododecane (11, 0.89 g, 2.96
mmol) and Fmoc-Tyr(But)-OPfp (2.74 g, 1.1ꢂ
2.96
mmol) were dissolved in dry CH2Cl2 (15 mL), and
then Et3N (0.41 mL, 2.96 mmol) was added. After the
mixture was stirred at r.t. for 1 h, a distinct precipitation
occurred, and more of dry CH2Cl2 (7 mL) was added.
After an additional 1 h, when TLC with petroleum
/
References
[1] K. Strømgaard, K. Andersen, P. Krogsgaard-Larsen, J.W.
Jaroszewski, Mini Rev. Med. Chem. 1 (2000) 317ꢀ
[2] (a) P.N.R. Usherwood, Farmaco 55 (2000) 202ꢀ205;
/338.
/
(b) M.G. Bixel, M. Krauss, C. Weise, M.L. Bolognesi, M. Rosini,
P.N.R. Usherwood, C. Melchiorre, F. Hucho, Farmaco 56 (2001)
etherꢀ
reaction mixture was filtered extracting the insolubles
with CH2Cl2ꢀhexane (1:1, 50 mL). The filtrate was
/EtOAc 2:1, showed a single spot with Rf 0.4, the
133ꢀ135;
/
(c) M.G. Bixel, M. Krauss, M.L. Bolognesi, M. Rosini, I.S.
Mellor, P.N.R. Usherwood, C. Melchiorre, K. Nakanishi, F.
/
concentrated to give a crude product (1.99 g, 92%), a
portion of which (1.89 g, 2.59 mmol) was dissolved in
Hucho, Eur. J. Biochem. 267 (2000) 110ꢀ120.
[3] (a) R. Goodnow, K. Konno, Jr., M. Niwa, T. Kallimopoulos, R.
Bukownik, D. Lenares, K. Nakanishi, Tetrahedron 46 (1990)
/
dry THF (20 mL), and then octanethiol (25.9 mL, 10ꢂ
2.59 mmol) and DBU (0.52 mL, 0.2ꢂ2.59 mmol) were
added. After 3 h at r.t. (TLC with CH2Cl2ꢀMeOHꢀ32%
/
3267ꢀ3286;
/
/
(b) K. Nakanishi, X. Huang, H. Jiang, Y. Liu, K. Fang, D.
Huang, S.-K. Choi, E. Katz, M. Eldefrawi, Bioorg. Med. Chem. 5
/
/
aqueous NH3 200:10:1 showed the product at Rf 0.75)
the reaction mixture was concentrated, and then loaded
(1997) 1969ꢀ
[4] E. Moya, I.S. Blagbrough, Tetrahedron 36 (1995) 9401ꢀ
[5] (a) I.A. Nash, B.W. Bycroft, W.C. Chan, Tetrahedron Lett. 37
(1996) 2625ꢀ2628;
(b) Y. Hidai, T. Kan, T. Fukuyama, Tetrahedron Lett. 40 (1999)
4711ꢀ4714;
(c) N.D. Hone, L.J. Payne, Tetrahedron Lett. 41 (2000) 6149ꢀ
6152;
/
1988.
/
9404.
onto a VLC column (6ꢂ
which was eluted successively with petroleum ether (400
mL), petroleum etherꢀEtOAc (1:1, 800 mL; and 1:2, 750
mL), and CH2Cl2ꢀMeOHꢀ32% aqueous NH3
/6 cm; Merck 60H silica gel),
/
/
/
/
/
/
100:10:0.5 (1000 mL) to give 12 (1.23 g, 91%).
Acylation of 12 (1.21 g, 2.33 mmol) with
(d) F. Wang, S. Manku, D.G. Hall, Organic Lett. 2 (2000) 1581ꢀ
/
1583;
(e) K. Strømgaard, K. Andersen, T. Ruhland, P. Krogsgaard-
C3H7COOPfp (0.71 g, 1.2ꢂ
(10 mL) was performed in the presence of Et3N (0.32
mL, 2.33 mmol) for 20 h (TLC with petroleum etherꢀ
/
2.33 mmol) in dry CH2Cl2
Larsen, J.W. Jaroszewski, Synthesis (2001) 877ꢀ884.
/
/
[6] K. Strømgaard, I. Bjørnsdottir, K. Andersen, M.J. Brierley, S.
Rizoli, N. Eldursi, I.R. Mellor, P.N.R. Usherwood, S.H. Hansen,
ME2CO 2:1 showed the product at Rf 0.32), when the
mixture was concentrated. The residue was redissolved
P. Krogsgaard-Larsen, J.W. Jaroszewski, Chirality 12 (2000) 93ꢀ
/
102.
[7] (a) K. Strømgaard, T.J. Brier, K. Andersen, I.R. Mellor, A.
Saghyan, D. Tikhonov, P.N.R. Usherwood, P. Krogsgaard-
in CH2Cl2 (10 mL) and loaded onto a VLC column (6ꢂ
6 cm; Merck 60H silica gel), which was eluted succes-
sively with hexane (500 mL) and hexaneꢀEtOAc (1:1,
800 mL), which afforded 13 (1.32 g, 96%). Deprotection
of 13 (1.17 g, 1.98 mmol) was performed in CH2Cl2ꢀ
TFA (9:1, 10 mL) at r.t. for 3 h (RP-18 TLC with 60%
MeCNꢀ0.1% aqueous TFA showed the product at Rf
0.55). The mixture was concentrated to dryness, and the
residue was dissolved in 50% MeCNꢀ0.1% aqueous
/
/
Larsen, J.W. Jaroszewski, J. Med. Chem. 43 (2000) 4526ꢀ4533;
/
(b) K. Strømgaard, I.R. Mellor, K. Andersen, I. Neagoe, F.
Pluteanu, P.N.R. Usherwood, P. Krogsgaard-Larsen, J.W. Jar-
/
oszewski, Bioorg. Med. Chem. Lett. 12 (2002) 1159ꢀ
[8] S.-K. Choi, R.A. Goodnow, A. Kalivretenos, G.W. Chiles, S.
Fushiya, K. Nakanishi, Tetrahedron 48 (1992) 4793ꢀ4822.
[9] (a) L. Kisfaludy, I. Schon, Synthesis (1983) 325ꢀ327;
(b) M. Green, J. Berman, Tetrahedron Lett. 31 (1990) 5851ꢀ5852.
/
1162.
/
/
/
¨
/
/

122
P. Wellendorph et al. / European Journal of Medicinal Chemistry 38 (2003) 117ꢀ122
/
[10] L. Carpino, G.Y. Han, J. Org. Chem. 37 (1972) 3404ꢀ
[11] A.J. Geall, I.S. Blagbrough, Tetrahedron 56 (2000) 2449ꢀ
[12] J.E. Sheppeck, II, H. Kar, H. Hong, Tetrahedron Lett. 41 (2000)
5329ꢀ5333.
[13] Y.H. Wang, J.C. Xu, Synthesis (1990) 845ꢀ
/
3409.
[14] T. Sammakia, T.B. Hurley, J. Org. Chem. 64 (1999) 4652ꢀ
[15] (a) B. Gallot, A. Douy, Mol. Cryst. Liq. Cryst. 153 (1988) 367ꢀ
/
4664.
/
2460.
/
374;
/
(b) F. Jourdain, M. Caron, J.C. Pommelet, Synth. Commun. 29
/846.
(1999) 1785ꢀ1799.
/