Two-colour screening in combinatorial chemistry: prospecting for enantioselectivity in a library of steroid-based receptors

Article abstract of DOI:10.1016/j.tet.2009.06.018

The screening of resin-bound combinatorial libraries with pairs of dye-tagged substrates is a powerful strategy for discovering selective receptors. However, implementation has been hampered by a lack of complementary but chemically similar dyes. We now s

Full text of DOI:10.1016/j.tet.2009.06.018

Tetrahedron 65 (2009) 6370–6381
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Two-colour screening in combinatorial chemistry: prospecting for
enantioselectivity in a library of steroid-based receptors
*
´
Vicente del Amo, Adam P. McGlone, Jose M. Soriano, Anthony P. Davis
School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The screening of resin-bound combinatorial libraries with pairs of dye-tagged substrates is a powerful
strategy for discovering selective receptors. However, implementation has been hampered by a lack of
complementary but chemically similar dyes. We now show that the well-established Disperse Red 1 and
the recently-introduced Bristol Blue 1 can be used in parallel to synthesise ‘pseudoenantiomeric’ ana-
Received 8 January 2009
Received in revised form 19 May 2009
Accepted 4 June 2009
Available online 10 June 2009
logues of N-acetyl-
library has been prepared and screened with these substrates. Preliminary results suggest that some
members may be highly enantioselective receptors for N-acetyl- -amino acids.
Ó 2009 Elsevier Ltd. All rights reserved.
a-amino acids and of the anti-inflammatory drug Naproxen. A steroid-based receptor
Keywords:
a
Combinatorial chemistry
Enantioselective recognition
Amino acids
Steroids
1. Introduction
are labelled with complementary chromophores and the library
exposed to the mixture. Beads are now selected according to the
purity (not intensity) of the colour they accumulate. Although this
protocol has been successfully demonstrated, difficulties with the
chromophores have hampered further development. Of the two
dyes employed by Still, Disperse Red 1 (1) (Fig.1) has proved widely
useful but Disperse Blue 3 (2) is problematic. Dye 2 is hard to obtain
pure and lacks chemical stability, being vulnerable to oxidation,
reduction and deprotonation. Furthermore, it is not a good struc-
tural match for 1, increasing the risk of false hits due to dye-based
selectivity.
The design of selective receptors for complex molecules is still
a difficult challenge for supramolecular chemistry. Ideally one
might aspire to a fully rational approach, employing computer
based design guided by the quantitative modelling of recognition
processes. However barriers are imposed by the flexibility of most
target molecules, the difficulty of accounting for solvation and the
need to consider entropy as well as enthalpy in binding calcula-
tions. Where these problems are insurmountable, the ‘semi-ratio-
nal’ approach of combinatorial chemistry provides
a useful
alternative. As shown by Still and others,1 it is possible to prepare
very large libraries of receptor-like molecules using split-and-mix
solid phase synthesis. The library members are mixed in a single
vessel, but are spatially segregated in that each resin bead carries
only one member. Visual screening for binding, employing dye-
tagged substrates, allows the whole library to be tested in a single
operation. By exploring many structures at once, the method can
reveal design solutions which may not have been predicted.
The common variant of Still’s procedure employs a single dye-
tagged substrate, and is suitable for identifying strong receptors for
one particular guest. In many cases, however, it is selectivity rather
than binding strength which is most important. For these situations
Still developed a two-colour assay, originally employed to optimise
selectivity between two peptide substrates.1g The two substrates
O
O
NHMe
O2N
N
N
N
HN
1
2
OH
OH
NO2
N
N
O2N
S
N
3
OH
* Corresponding author. Tel.: þ44 (0)117 954 6334; fax: þ44 (0)117 929 8611.
E-mail address: anthony.davis@bristol.ac.uk (A.P. Davis).
Figure 1. Dyes used for substrate tagging in combinatorial chemistry: Disperse Red 1
(1), Disperse Blue 3 (2) and Bristol Blue 1 (3).
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.06.018

V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
6371
In response to this problem, we have reported the synthesis
and preliminary assessment of ‘Bristol Blue 1’ (3), an alternative
partner for 1 in the two-colour screening system.2 3 is chemically
and structurally similar to 1 while being complementary in colour.
Our long term goal is to discover enantioselective receptors3
through the screening of libraries with substrates tagged with
these chromophores. We now show that 1 and 3 can be employed
to prepare ‘pseudoenantiomeric’ pairs of carboxylate substrates,
related to N-acetyl amino acids 4 and to the anti-inflammatory
drug S-Naproxen (S-5) (Fig. 2). We also describe the synthesis of
a library of steroid-based receptors, and the use of the coloured
substrates in preliminary screens of the library. The results sug-
gest that the protocol does indeed have potential for the discovery
of enantioselective receptors.
NH2
HN
O
O
O
a
O
O
96 %
OH
OH
L-10
L-9
80%
b,c
HN
(over 2 steps)
O
HN
O
O
O
b,c
O
O
80%
O
R
OH
(over 2 steps)
O
OH
N
H
O
O
OH
OH
MeO
O
D-10
D- or L-7
4
e
S-5
58 %
Figure 2. Substrates for enantioselective recognition: N-acetyl amino acids 4 and
S-Naproxen (S-5).
NH2
HN
O
HO
HO
2. Results and discussion
d
O
O
2.1. Design of dye labelled substrates
Prepared as
in Ref. 18
OH
OH
The dye-labelled substrates were intended to be representative
of N-acetylamino acids 4, and of Naproxen 5, such that receptors
discovered through two-colour screening would also be effective
for the unlabelled prototypes. It was therefore desirable to avoid
disturbance of the key recognition features of 4 and 5, placing the
labels remotely from the chiral centers. In the case of 4, this could
D-6
D-11
Scheme 1. Synthesis of N-acetyl amino acid derivatives 7 for coupling to chromo-
phores. Reagents and conditions: (a) AcCl, Et3N, DCM, 0 ꢀC to rt; (b) bromobenzyl
acetate, Cs2CO3, DMF, rt; (c) Pd(C)/H2 (1 atm), THF, rt; (d) Ac2O, H2O, 80 ꢀC; (e) tert-
butyl bromide, K2CO3, Bu4NþBrꢁ, DMA, 55 ꢀC.
be achieved by starting from D- and L-tyrosine 6 and linking the
a slightly different route, from
lated using acetic anhydride in water to give acetamide
good yield and high purity.4 tert-Butyl ester
-10 was then
formed by treatment of
-11 with tert-butyl bromide (Scheme 1).5
-10 was then converted to -7 using the same procedure as for
the enantiomeric series.
D
-tyrosine (
D
-6).
D
-6 was acety-
chromophores to the phenolic hydroxyl groups. In the case of
the Naproxen enantiomers 5, the methyl ethers could potentially be
cleaved to give naphtholic hydroxyl groups which could be
exploited similarly. To make the connections between dyes and
substrates, we decided to convert the Ar–OH groups to Ar–OCH2-
CO2H, and then to esterify with 1 or 3. Our initial targets were
therefore the carboxylic acids 7 (from 4) and 8 (from 5) (Fig. 3). The
tert-butyl esters in 7 and 8 would be compatible with the esterifi-
cation conditions, and then removable with acid to give the dye-
tagged carboxylate substrates.
D
-11 in
D
D
D
D
2.3. Synthesis of Naproxen derivatives (R)- and (S)-8
Both enantiomers of the Naproxen-derived intermediate 8 were
prepared as shown in Scheme 2. Either (R)- or (S)-Naproxen 5 was
heated under reflux in concentrated hydrobromic acid to give the
corresponding hydroxy derivative 12,6 which was then protected as
tert-butyl ester by treatment with TFAA followed by tert-butanol
then ammonium hydroxide.7 The resulting ester 13 was then
O-alkylated with bromobenzyl acetate/K2CO3 in MeCN. Finally,
hydrogenolysis of the benzyl ester provided carboxylic acid 8.
t
CO2 Bu
NHAc
NH2
tBuO2C
HO2C
O
O
O
OH
O
OH
OH
6
7
8
2.4. Synthesis of dye tagged pseudoenantiomeric
carboxylates
Figure 3. Tyrosine (6) and intermediates 7 and 8.
Intermediates D-7 and L-7 were attached to chromophores 1 and
2.2. Synthesis of N-acetyltyrosine derivatives
L
-7 and
D
-7
3 respectively, via ester linkages formed using 1,10-carbonyl-
diimidazole-mediated couplings (Scheme 3). The coupled products
14 and 17 were deprotected with TFA to give acids 15 and 18.
Treatment of the acids with tetrabutylammonium hydroxide gave
the lipophilic salts 16 and 19, suitable for the screening experi-
ments. A similar sequence applied to Naproxen-derived acids (R)-8
and (S)-8 gave salts 22 and 25 (Scheme 4).
Intermediate
-tyrosine tert-butyl ester (
of -9 with acetyl choride, to give acetamide
by alkylation with bromobenzyl acetate and debenzylation with
H2/Pd(C) to give -7. Its enantiomer -7 was prepared using
L
-7 was prepared from commercially available
-9), as shown in Scheme 1. Treatment
-10, was followed
L
L
L
L
L
D

6372
V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
O
O
electron poor aromatic urea at the C7 position was chosen as the
primary binding motif, ureas being well established as carboxylate
binding units.12 Both the C3 and C12 positions of the steroid were to
be appended with peptidic moieties, enclosing the substrate and
creating diversity. It was decided to attach the steroid to the resin
via a robust amide linkage, with a view to analysing single beads by
Edman degradation after selection experiments. Although this
method would not give a single definitive anwer in most cases,
because the two chains would be cleaved simultaneously to reveal
two amino acids at each stage of the process, any analysis would
imply a maximum of four different structures. These could be
resynthesised for confirmatory testing using parallel solid phase
synthesis.13
OH
OH
a
96%
OMe
OH
(S)- or (R)-12
(S)- or (R)-5
88%
b
The starting point for library synthesis was the differentially
protected triaminosteroid 27.9a As shown in Scheme 5, this was
converted to urea 28 which was then hydrolysed to carboxylic acid
29. Attachment of 29 to high loading TentagelÔ aminomethyl resin
gave polymer-bound intermediate 30. The NF31 test14 was
employed to ensure that the amido coupling had proceeded to high
conversion. The test indicated nearly complete coupling between
carboxylic acid 29 and the resin, only a few free amine function-
alities being detected on the polymer. These few amino groups
were capped with acetyl functions by treatment with a large excess
of acetic anhydride in CH2Cl2/pyridine. The C12 o-Ns group was
removed, to afford the corresponding free amine derivative 31.
‘Split-and-mix’ synthesis15 was then used to introduce dipeptide
O
O
O
O
c,d
74%
(over 2 steps)
O
O
OH
OH
(S)- or (R)-8
(S)- or (R)-13
Scheme 2. Synthesis of Naproxen derivative 8 for coupling to chromophores. Reagents
and conditions: (a) 48% (w/w) HBr, reflux, 2 h; (b) (i) (CF3CO)2O, dry THF, 0 ꢀC, 4 h, (ii)
tert-butanol, 0 ꢀC to rt, 16 h, (iii) NH4OH, 0 ꢀC to rt, 30 min; (c) bromobenzyl acetate,
K2CO3, dry MeCN, rt, 40 h; (d) H2/Pd(C), CHCl3, rt, 2 h.
NHAc
NHAc
NHAc
Bu4N+
O
RO
O
O
O
a
O
c
O
82%
97%
O
O
N
O
N
O
O
O
O
OH
N
N
N
D-7
N
14 R = t-Bu
15 R = H
16
b
NO2
NO2
99%
NHAc
O
NHAc
O
NHAc
O
Bu4N+
O
RO
O
O
d
O
c
O
O
62%
97%
O
O
N
N
O
O
OH
N
S
N
S
L-7
N
O2N
N
O2N
17 R = t-Bu
18 R = H
NO2
19
NO2
b
74%
Scheme 3. Synthesis of tyrosine-based pseudoenantiomeric substrates 16 and 19. Reagents and conditions: (a) 1,10-carbonyldiimidazole, dry DCM, ꢁ10 ꢀC, 90 min, then 1, dry DCM,
0 ꢀC to rt, 16 h; (b) TFA/DCM, 0 ꢀC to rt, 6 h; (c) Bu4NOH (1.0 M in MeOH), CHCl3/H2O (2:1), rt, 30 min; (d) 1,10-carbonyldiimidazole, dry DCM, ꢁ10 ꢀC 90 min, then 3, dry DCM, 0 ꢀC to
rt, 3 days.
2.5. Design and synthesis of receptor library
‘legs’ at position 12. Twelve common N-Fmoc protected amino
acids were employed, with acid-removable side-chain protection
(side-chain protecting groups are given in brackets); Ala, Arg (Pbf),
Asp (O-t-Bu), Cys (Trt), Gly, His (Trt), Lys (Boc), Phe, Pro, Ser (O-t-
Bu), Trp (Boc) and Val. Resin-bound intermediate 31 was divided
into 12 portions of approximately equal weight. Each portion was
coupled to one of the above amino acids using HATU, HOBt and
DIPEA in dry DMF.16 At this point, the loading of the steroid onto the
resin was evaluated by applying the ‘Fmoc test’17 to a sample of
resin from the Gly coupling (glycine being the most likely to react
quantitatively). An approximate loading of 0.29 mmol of steroid per
gram of resin was inferred. The resin portions were recombined
The general structure chosen for the library is shown in Figure 4.
The design was based on our previous experience with cholic acid
26 as a scaffold for supramolecular8 and combinatorial9 chemistry.
In particular, we have shown that ‘cholapod’ receptors derived from
26 can bind carboxylates with quite good enantioselectivities (er up
to 10:1).8c,10 The tripodal11 architecture can surround a substrate,
maximising the number of interactions. The steroidal skeleton
preorganises the binding functionality imposing a specific orien-
tation on the interactions. The C24 of cholic acid provides a con-
venient anchor point, allowing attachment to a solid support. An

V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
6373
O
O
O
O
O
Bu4N+
OR
O
O
O
a
c
65%
O
97%
N
N
O
O
O
O
OH
N
N
N
N
20 R = t-Bu
21 R = H
22
(R)-8
b
NO2
NO2
Quant.
O
O
O
Bu4N+
OR
O
O
d
c
77 %
O
O
93%
O
O
N
O
N
O
O
O
OH
N
N
S
S
N
O2N
N
O2N
23 R = t-Bu
24 R = H
NO2
25
NO2
(S)-8
b
71%
Scheme 4. Synthesis of Naproxen-based pseudoenantiomeric substrates 22 and 25. Reagents and conditions: (a) 1,10-carbonyldiimidazole, dry DCM, 0 ꢀC, 1 h, then 1, dry DCM, 24 h,
0
0
ꢀC to rt; (b) TFA/DCM, 0 ꢀC to rt, 3 h (red derivative) or 10 h (blue derivative); (c) Bu4NOH (1.0 M in MeOH), CHCl3/H2O (2:1), 0 ꢀC, 10 min; (d) 1,10-carbonyldiimidazole, dry DCM,
ꢀC, 1 h, then 3, dry DCM, 0 ꢀC to rt, 3 days.
mixture 22þ25. Solvent choice is an important parameter for this
24
O
12
type of experiment. Binding must be strong enough that substrates
accumulate in the beads, but not so strong that all beads become
coloured. It is also important to ensure that complexation is re-
versible, so that substrate molecules can seek out the most effective
receptors. Library 33 was designed to operate principally through
hydrogen bonding, so that non-polar solvent systems were in-
dicated. However, the use of chloroform, or less polar solvent sys-
tems, seemed to give non-specific irreversible binding. The
problem was solved by the addition of small amounts of alcohols,
which moderated the binding strength and promoted exchange
between bound and unbound substrates.
A typical experiment involved (a) exposure of library to a pseu-
doenantiomeric substrate mixture in CHCl3/t-BuOH (12:1) for 1 h, (b)
separation of the beads by filtration, (c) washing with CHCl3 to
remove unbound substrate, and (d) suspension of the library in
hexane for microscopic examination and bead picking. Studies with
22þ25 were disappointing. Binding was weak, even with low levels
of alcohol in the solvent, and there was little variation between
beads. However, the results for the tyrosine-derived pseudoe-
nantiomers were more encouraging. When a representative portion
of library 33 was equilibrated with a 1:1 mixture of 16 and 19 in
CHCl3/t-BuOH (12:1), many of the beads acquired strong colours
(Fig. 5). As shown in Figure 5 more blue beads were observed than
red, implying that the library was generally biassed towards 19. This
could result from a preference for the blue chromophore, but it is
reasonable to suppose that, at least in some cases, the blue beads are
OH
7
OH
OH
26
3
OH
24
O
12
HN
7
3
N
O
H
peptide leg
N
H
Ar
urea as carboxylate
binding motif
peptide leg
Figure 4. Design of library, and its relationship with cholic acid 6.
after coupling and the Fmoc group was cleaved by treating the
beads with 20% piperidine in DMF. Re-splitting, a second amino
acid coupling (HOBt, HBTU, DIPEA in dry DMF), recombination,
Fmoc cleavage and Boc protection finished the synthesis of the first
peptide chain at C12 to afford 32. The azide group at C3 was then
reduced with a large excess of PMe3 in THF, followed by addition of
water, to give the corresponding primary amine. Linkage of the first
amino acid to the C3 amine (HOBt, HBTU, DIPEA in dry DMF), Fmoc
cleavage and linkage of a second amino acid completed the syn-
thesis of library 33 (Scheme 5). Library 33 possesses 124¼20,736
members and is, to our knowledge, the first example of a combi-
natorial library prepared from an orthogonally protected triamino
cholanoate derivative.18
L
-selective. This sense of selectivity is consistent with earlier results
on steroid-based carboxylates receptors, where a range of guanidi-
nium- and urea-substituted systems have all shown -selectivity
with N-acetyl-
-aminocarboxylates.10 Washing of the beads with
L
a
methanol rendered them almost colourless, proving that binding
was due to non-covalent interactions.
The aim of library screening is to identify the best (in this case the
most selective) receptors among the many present. The above ex-
periment does not serve this purpose well, as most of the beads are
perceived as ‘hits’. However, discrimination can be improved by the
simple modification of biassing the substrate mixture. In the present
2.6. Preliminary screening of steroid-based library 33
As a preliminary test of our strategy, the steroid-based receptor
library 33 was screened against the tyrosine-derived pseudo-
enantiomeric mixture 16þ19, and against the Naproxen-derived

6374
V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
O
O
O
OR
HN
a
NHoNs
NHoNs
c
OMe
NHoNs
N
H
N
H
O
O
NH2
75%
N3
N3
N3
27
N
N
30
H
H
R = Me
28
oNs = o-nitrosulfonyl
b
R = H
88%
CF3
CF3
d
29
O
O
O
HN
HN
HN
NH2
HN
HN
N
N
AA1
AA1
N
O
O
i,g,f,g
O
e,f g,f,h
H
N3
H
H
NH
AA1
AA2NHFmoc
N3
AA2NHBoc
AA2NHBoc
N
N
N
31
H
H
H
33
32
CF3
CF3
CF3
Scheme 5. Synthesis of library of receptors 33. Reagents and conditions: (a) p-(trifluoromethyl)phenyl isocyanate, NEt3, DMAP, dry THF, reflux, 6 h; (b) NaOH, MeOH/H2O, rt, 3 days;
(c) NovaSynÒ TG HL aminomethyl resin, TBTU, DMAP, DMF, rt, overnight, then Ac2O, pyridine, DCM, rt, overnight; (d) PhSH, Cs2CO3, dry DMF, 55 ꢀC, 4 days; (e) N-Fmoc-
-amino acid
-amino acid derivative, HBTU, HOBt, DIPEA, dry DMF, rt, overnight; (h) (Boc)2O,
a
derivative, HATU, HOBt, DIPEA, dry DMF, rt, overnight; (f) piperidine, DMF, rt, 30 min; (g) N-Fmoc-
a
DCM, rt, overnight; (i) PMe3 (1.0 M in THF), THF, rt, 3 h, then H2O, rt, 2 h.
Figure 5. Micrograph of a portion of library 33 after equilibration with a 1:1 mixture of
red carboxylate 16þblue carboxylate 19 for 1 h in CHCl3/t-BuOH (12:1). Before viewing
the beads were filtered, washed briefly with chloroform and suspended in pure hex-
ane. Most beads appear pale (implying weak binding) or deep blue.
case, the addition of extra red substrate challenges the library such
that only the most selective members appear blue. The results of two
such experiments, employing 2:1 and 5:1 ratios of red/blue sub-
strates, are shown in Figure 6. Even in the latter case, with a five-fold
excess of red substrate, a few beads appeared clearly blue among
a majority of red and greenish (unselective) beads. These blue beads
were separated from the rest, and set aside for future studies.
3. Conclusions
Figure 6. Micrographs of portions of library 33 after equilibration with unbalanced
mixtures of 16 (red)þ19 (blue). Experimental procedure as for Figure 5. (a) Red/blue¼2:1.
Beads are mainly pale or deep blue, as for Figure 5. (b) Red/blue¼5:1. Most beads appear
pale or shades of green/brown, but a few (e.g., arrow) are clearly perceived as blue.
Combinatorial chemistry with two-colour screening offers hope
for the discovery of selective receptors, even in cases where rational

V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
6375
design is especially difficult. However there are experimental issues
which need to be addressed, including the design and synthesis of
substrates suitable for screening. In this paper, we have shown that
the similar but complementary dyes 1 and 3 can make a valuable
contribution. Using parallel reaction sequences they have been
attached to substrates of real interest, without disturbing the
substrates’ key recognition features. The resulting pseudo-
enantiomers have been subjected to preliminary testing against
a novel steroid-based receptor library. Our observations suggest
that library members can vary greatly in their binding properties,
and that some distinguish effectively between the substrates. For
an individual member, we cannot say for certain that selectivity is
based on substrate configuration, rather than the nature of the
dye units. However, by picking and studying a number of selective
beads, we give ourselves several chances of finding enantiose-
lectivity. Moreover, in principle the issue can be resolved by
re-screening the picked beads with substrates in which dyes have
been swapped between chiral centers (in other words, using
the enantiomers of 16 and 19). Beads which now select for the
opposite colour (red, in this case) should be enantioselective and
worthy of detailed study. In future work we will aim to realise this
programme and emerge with highly selective receptors.
d, J¼8.3 Hz, ArH), 6.98 (2H, d, J¼8.3 Hz, ArH), 7.64 (1H, br s, phenolic
OH); 13C NMR (100 MHz, CDCl3):
(CH2), 53.7 (CH), 82.6 ((CH3)3C), 115.4 (ArCH), 127.4 (ArC), 130.4
(ArCH), 155.4 (ArC),170.1 (CONH),171.1 (CO2C(CH3)3); MS (ESIþ): m/z¼
302 [MþNa]þ; HRMS (ESIþ): m/z calcd for [C15H21NO4þNa]þ
302.1363, found 302.1367.
d¼23.1 (CH3), 27.9 ((CH3)3C), 37.4
4.3. N-Acetyl-D-tyrosine tert-butyl ester
D-10
N-Acetyl- -tyrosine
D
D
-1119 (1.00 g; 4.48 mmol), K2CO3 (16.0 g,
116.5 mmol) and Bu4NþClꢁ (1.02 g, 4.48 mmol) were suspended in
dimethylacetamide (50 mL) with vigorous stirring. tert-Butyl bro-
mide (33 mL, 215 mmol) was added to the stirred suspension and
the reaction heated to 55 ꢀC and left to stir overnight. The reaction
mixture was then poured onto 1.0 M KHSO4 in water (150 mL) and
extracted with ethyl acetate (3ꢂ100 mL). The combined organic
layers were evaporated under reduced pressure leaving a yellow oil.
Residual dimethylacetamide was removed by dissolving the oily
residue in DCM (100 mL) and washing with ice-water (3ꢂca.
100 mL). The DCM layer was dried (MgSO4), filtered and evaporated
under reduced pressure to give a yellow oil. Purification by flash
column chromatography (hexanes/EtOAc, 3:1) gave
58%) as a clear pale yellow oil. [
D þ0.9 (c 0.09, DCM). Other data as
for -10.
D-10 (1.25 g,
a
]
4. Experimental
L
4.1. General methods
4.4. N-Acetyl-O-carboxymethyl-L-tyrosine tert-butyl ester L-7
All reagents and solvents were obtained from commercial sup-
pliers and used without further purification unless otherwise
stated. Methanol was distilled over calcium chloride, magnesium
and iodine. DMF was obtained dry from Aldrich. THF and DCM were
obtained dry from an Anhydrous Engineering Solvent Purification
System (AESPS). Analytical TLC was carried out on DC-Alufolien
Kieselgel 60F254 0.2 mm plates (Merck) and compounds were
visualised by UV fluorescence, 5% phosphomolybdic acid in etha-
nol, ninhydrin solution or by charring over a Bunsen burner flame.
Flash chromatography of reaction products was carried out using
Note: The
tert-Butyl ester
D
enantiomer was prepared in the same fashion.
-10 (2.00 g, 7.17 mmol) and Cs2CO3 (7.01 g,
L
21.5 mmol) were dissolved in DMF (75 mL) and stirred vigorously
for 15 min. 2-Bromobenzyl acetate (4.5 mL, 28.7 mmol) was added
to the pale yellow solution and the mixture was left to stir over-
night at room temperature. DMF was removed by evaporation
under reduced pressure. The oily residue was dissolved in EtOAc
(100 mL) and washed with ice-water (3ꢂca. 100 mL). The organic
layer was dried (MgSO4), filtered and evaporated under reduced
pressure giving the crude product as an oil. Purification by flash
column chromatography on silica gel using hexanes/ethyl acetate
Silica 60A, particle size 35–70 mm (Fischer Scientific). IR spectra
were recorded on a Perkin–Elmer Spectrum One spectrometer.
Melting points were obtained using Gallenkamp melting point
blocks and are quoted as uncorrected values. 1H NMR and 13C NMR
spectra were recorded on Jeol Delta/GX270 or Jeol Delta/GX400
spectrometers, using deuterated solvents and referenced internally
to the residual solvent peak or TMS (dH¼0.00 ppm, dC¼0.0 ppm)
signal. Coupling constants (J-values) are given in Hz. The DEPT 135ꢀ
technique was used to assign (CH2) signals. Chemical shifts are
reported as follows: value (number of protons, description of ab-
sorption, coupling constant(s) where applicable, assignment).
(1:2) as eluent gave N-acetyl-O-benzyloxycarbonylmethyl-
sine tert-butyl ester as an oily foam (5.95 g, 82%). [
D þ0.8 (c 0.04,
MeOH); 1H NMR (400 MHz, CDCl3, 298 K, TMS):
¼1.41 (9H, s,
L-tyro-
a
]
d
C(CH3)3), 2.00 (3H, s, CH3), 3.03 (2H, m, CH2), 4.58 (2H, s, CH2), 4.70
(1H, m, H), 5.28 (2H, s, CH2), 6.09 (1H, d, J¼6.4 Hz, NH), 6.81 (2H, d,
J¼8.3 Hz, ArH), 7.06 (2H, d, J¼8.3 Hz, ArH), 7.35 (4H, m, ArH), 8.00
(1H, s, ArH); 13C NMR (100 MHz, CDCl3, 298 K, TMS):
d¼20.7 (CH3),
27.3 (C(CH3)3), 37.3 (CH2), 53.8 (CH), 65.1 (CH2), 66.4 (CH2), 79.1
(C(CH3)3), 114.2 (ArC), 121.1 (ArC), 128.0 (ArC), 128.3 (ArC), 130.0
(ArC), 134.1 (ArC), 135.3 (ArC), 149.2 (ArC), 168.4 (C]O), 170.6
(C]O).
4.2. N-Acetyl-
L
-tyrosine tert-butyl ester
L
-10
To the foregoing diester (1.80 g, 4.21 mmol) dissolved in THF
(20 mL) was added activated (heated in vacuo) Pd/C (Degussa,
10% w/w, 180 mg). The mixture was stirred under an atmosphere
of H2 (1 atm), until TLC indicated that the reaction had pro-
ceeded to completion. Residual catalyst was removed by filtra-
tion through Celite and washing with THF (3ꢂ50 mL). The filtrate
was then evaporated to dryness under reduced pressure giving
L-Tyrosine tert-butyl ester (
L-9) (4.81 g, 20.3 mmol) and tri-
ethylamine (2.85 g, 3.95 mL, 28.4 mmol) were dissolved in DCM
(100 mL). The solution was vigorously stirred and cooled in an ice–
salt bath to 0 ꢀC. Acetyl chloride (1.04 mL, 22.3 mmol) was added
dropwise over 15 min. The mixture was then stirred for 20 min at
0 ꢀC, allowed to warm to room temperature slowly and left stirring
overnight. DCM was removed by evaporation under reduced
pressure, and the oily residue was extracted with EtOAc (3ꢂ50 mL).
The combined organic layers were washed with water (50 mL) then
dried (MgSO4), filtered and evaporated under reduced pressure to
give the crude product as a clear oil. Purification by flash column
acid
MeOH); 1H NMR (400 MHz, CDCl3, 298 K, TMS):
C(CH3)3), 2.00 (3H, s, CH3), 2.90–2.97 (2H, m, CH2), 4.53 (2H, s,
L
-7 (1.40 g, 98%) as a colourless oil. [
a
]
þ9.42 (c 0.05,
D
d
¼1.42 (9H, s,
CH2), 4.74 (1H, m,
a
-H), 6.09 (1H, d, J¼7.8 Hz, NH), 6.85 (2H, d,
J¼8.3 Hz, ArH), 7.07 (2H, d, J¼8.3 Hz, ArH); 13C NMR (100 MHz,
CDCl3, 298 K, TMS):
d
¼23.2 (CH3), 28.0 (C(CH3)3), 37.2 (CH2), 53.7
chromatography (hexanes/EtOAc, 3:1) gave
L-10 (5.43 g, 96%) as
(CH), 67.0 (OCH2CO2H), 82.4 (C(CH3)3), 114.7 (ArCH), 128.6
(ArCH), 128.7 (ArC), 130.6 (ArC), 169.5 (C]O), 169.2 (C]O),
(C]O); MS (CIþ): m/z (%)¼338 [M]þ (7), 278 (59), 228 (5); HRMS
(ESþ): m/z calcd for [C17H23NO6þNa]þ 360.1418, found 360.1413.
a colourless oil. [
298 K, TMS):
a
]
D
ꢁ0.9 (c 0.09, DCM); 1H NMR (400 MHz, D2O,
d
¼1.44 (9H, s, C(CH3)3), 1.98 (3H, s, CH3), 2.96–3.03
(2H, m, CH2), 4.74 (1H, m, CH), 6.20 (1H, d, J¼8.3 Hz, NH), 6.74 (2H,

6376
V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
The enantiomeric purity of
L
-7 and
D
-7 was confirmed by HPLC
4.7. D-Tyrosine-derived red carboxylate salt 16
analysis using a Daicel Chiralcel OD-H column (250ꢂ4.6 mm)
eluting with 0.1% TFA in hexane/isopropanol (9:1) at a flow rate
of 1.0 mL minꢁ1 with the UV detector set at 254 nm. Retention
To red acid 15 (55 mg, 0.095 mmol) dissolved in CHCl3 (2 mL) was
added a solution of NaHCO3 (64 mg, 0.76 mmol) and Bu4Nþ$HSO4ꢁ
(130 mg, 0.38 mmol) in deionised water (1.8 mL). After 5 min of
vigorous stirring, the bi-phasic solution was poured into a separating
funnel containing CHCl3 (15 mL) and deionised water (15 mL). The
organic layer was separated, dried (Na2SO4), filtered and evaporated
to dryness under reduced pressure giving the Bu4Nþ salt 16 (76 mg,
97%) as a red amorphous solid. 1H NMR (400 MHz, CDCl3, 298 K,
times:
L-7, 13.97 min; D-7, 16.97 min.
4.5. -Tyrosine-derived red tert-butyl ester 14
D
1,10-Carbonyldiimidazole (442 mg, 2.73 mmol) dissolved in
dry DCM (5 mL) under nitrogen was cooled to 0 ꢀC in an ice–salt
bath. To this solution was added acid
D
-7 (478 mg, 1.42 mmol) in
TMS):
d
¼0.81 (3H, m, NCH2CH3), 0.87 (12H, m, Bu4NþCH3), 1.17 (8H,
dry DCM (7 mL) over ca. 5 min. The resultant pale yellowish so-
lution was then stirred for 1 h at 0 ꢀC. Disperse Red 1 1 (391 mg,
1.09 mmol) was added as a solid to the mixture, while main-
taining an outward flow of nitrogen. The mixture was then stir-
red at 0 ꢀC for a further 30 min and at room temperature for 36 h.
The residue was diluted with DCM (50 mL) and washed with
water (50 mL). The water layer was extracted with DCM
(3ꢂ30 mL) and the combined organic layers were dried (MgSO4),
filtered and evaporated giving the crude product as a red solid.
Purification by flash chromatography on silica gel using hexanes/
ethyl acetate (1:1) as eluent, followed by precipitation from
CHCl3/hexanes gave 14 as a red amorphous solid (565 mg, 82%).
m, Bu4NþCH2), 1.35 (8H, m, Bu4NþCH2), 1.83 (3H, s, CH3), 3.03–3.25
(10H, m, Bu4NþCH2N and benzylic CH2), 3.41 (2H, q, J¼7.3 Hz,
NCH2CH3), 3.62 (2H, t, J¼6.4 Hz, NCH2CH2O), 4.33 (3H, m, NCH2CH2O
and
a
-H), 4.48 (2H, s, OCH2CO), 6.65 (2H, d, J¼8.3 Hz, ArH), 6.71 (2H,
d, J¼8.8 Hz, ArH), 6.81 (1H, d, J¼8.3 Hz, NH), 7.08 (2H, d, J¼8.3 Hz,
ArH), 7.83 (4H, m, ArH), 8.23 (2H, d, J¼8.3 Hz, ArH); MS (ESþ): m/z
(%)¼578 (5) [Mꢁ(NC16H36)þ2H]þ, 242 (41) [NC16H36]þ; HRMS (ESþ):
Mass calcd for [C45H66N6O6ꢁ(C16H36N)þ2H]þ¼578.2245, found
578.2256.
4.8. L-Tyrosine-derived blue tert-butyl ester 17
1H NMR (400 MHz, CDCl3, 298 K, TMS):
d
¼1.25 (3H, t, J¼6.8 Hz,
1,10-Carbonyldiimidazole (196 mg, 1.0 mmol) was dissolved in
DCM (5 mL) under nitrogen and cooled to 0 ꢀC in an ice–salt bath. A
solution of acid 7 (219 mg, 0.65 mmol) in DCM (3 mL) was then
added with vigorous stirring at 0 ꢀC. The solution was stirred for 2 h
at 0 ꢀC, after which blue azo dye 3 (183 mg, 0.50 mmol) was added
as a solution in DCM (7 mL). The solution was stirred for a further
30 min at 0 ꢀC, then for 6 h at room temperature, before partioning
between DCM (30 mL) and water (30 mL). The water layer was
separated and extracted with DCM (3ꢂ30 mL) the combined or-
ganic layers were then dried (Na2SO4), filtered and evaporated
under reduced pressure to leave a blue residue. Purification by flash
column chromatography on silica gel using hexane/ethyl acetate
(1:2) gave 17 (212 mg, 62%) as a blue amorphous solid. 1H NMR
NCH2CH3), 1.40 (9H, s, C(CH3)3), 2.00 (3H, s, CH3), 3.04 (2H, m,
CH2), 3.51 (2H, q, J¼6.8 Hz, NCH2CH3), 3.73 (3H, t, J¼6.4 Hz,
NCH2CH2O), 4.44 (3H, t, J¼6.4 Hz, NCH2CH2O), 4.60 (2H, s,
OCH2CO2R), 4.72 (1H, m,
a
-H), 5.91 (1H, d, J¼7.6 Hz, NH), 6.80
(4H, m, ArH), 7.08 (2H, m, ArH), 7.92 (4H, m, ArH), 8.33 (2H, m,
ArH); 13C NMR (100 MHz, CDCl3, 298 K, TMS):
d
¼12.4 (CH3), 23.4
(CH3), 28.0 (C(CH3)3), 37.3 (CH2), 46.3 (CH2), 49.0 (CH2), 61.9
(CH2), 65.4 (CH2), 82.4 (C(CH3)3), 112.8 (ArC), 114.6 (ArC), 124.9
(ArC), 125.0 (ArC), 129.9 (ArC), 130.6 (ArC), 137.1 (ArC), 144.8
(ArC), 145.3 (ArC), 154.3 (ArC), 156.8 (ArC), 168.9 (C]O), 169.4
(C]O), 170.9 (C]O); MS (ESþ): m/z (%)¼633 (5). The material
was contaminated with a small amount of 1, which was removed
in the next step.
(400 MHz, CDCl3, 298 K, TMS):
d
¼1.28 (3H, t, J¼7.08 Hz, CH3CH2N),
1.39 (9H, s, C(CH3)3), 1.96 (3H, s, NHCOCH3), 3.04 (2H, m, CH2Ph),
3.55 (2H, q, J¼7.08 Hz, CH3CH2N), 3.78 (2H, t, J¼6.0 Hz, OCH2CH2N),
4.47 (2H, t, J¼6.0 Hz, OCH2CH2N), 4.58 (2H, s, OCH2CO), 4.69 (1H, m,
NHCHCO), 5.90 (1H, d, J¼7.6 Hz, NH), 6.75 (2H, d, J¼8.6 Hz, ArH),
6.82 (2H, d, J¼8.6 Hz, ArH), 7.04 (2H, d, J¼8.8 Hz, ArH), 7.94 (2H, d,
4.6.
D-Tyrosine-derived red acid 15
Ester 14 (831 mg, 1.31 mmol) was dissolved in DCM (2 mL) and
cooled in an ice–salt bath. TFA (13 mL) was added dropwise with
stirring to the red solution. After 15 min the mixture was allowed to
warm to room temperature. After stirring for a further 3 h, the
solvent was removed by evaporation. The resultant purple residue
was dissolved in DCM (100 mL), the solution was washed with
brine (100 mL), and the aqueous layer was extracted with DCM
(3ꢂ100 mL). The combined organic layers were dried (MgSO4),
filtered and evaporated under reduced pressure to give a red resi-
due. Purification by flash column chromatography on silica gel,
eluting with ethyl acetate followed by 2% acetic acid in ethyl ace-
tate, gave acid 15 (745 mg; 1.29 mmol; 99%) as a red solid. Melting
point (CHCl3/Hexane)¼160.2–163.1 ꢀC; IR: nmax (solid)¼3300, 3450,
2971,1738, 1600,1509,1371 cmꢁ1; 1H NMR (400 MHz, CDCl3, 298 K,
J¼8.8 Hz, ArH), 8.35 (1H, s, ArH); 13C NMR (100 MHz, CDCl3):
¼12.4
d
(CH3), 23.3 (CH3), 28.0 ((CH3)3C), 37.2 (CH2), 46.2 (CH2), 48.9 (CH),
53.6 (CH2), 61.8 (CH2), 65.3 (CH2), 82.4 ((CH3)3C), 114.4 (ArCH), 124.9
(ArCH), 129.8 (ArC), 130.7 (ArCH), 137.11 (ArC), 144.0 (ArC), 154.2
(ArC), 156.6 (ArC), 162.2 (ArC), 168.9 (CO2), 169.4 (CONH), 171.0
(CO2); MS (ESIþ): m/z¼685 [MþH]þ, 707 [MþNa]þ; HRMS (ESIþ):
m/z calcd for [C31H37N6O10SþH]þ 685.2286, found 685.2269 and
[C31H37N6O10SþNa]þ 707.2106, found 707.2084.
4.9. L-Tyrosine-derived blue acid 18
tert-Butyl ester 22 (250 mg, 0.365 mmol) was dissolved in DCM
(1 mL) and the blue solution was cooled to 0 ꢀC in an ice–salt bath
with vigorous stirring. Dropwise addition of TFA (7 mL) at 0 ꢀC gave
a purple solution which was left to stir in the ice–salt bath for
15 min. Stirring was continued for a further 8 h at room tempera-
ture. Evaporation gave a metallic-purple coloured residue which
was dissolved in DCM (50 mL) and washed with brine (3ꢂ50 mL).
The blue organic layer was dried (Na2SO4) and evaporated to give
a blue foam. This material was dry-loaded onto silica gel and sub-
jected to gravity column chromatography on silica gel using DCM/
MeOH/acetic acid (92:5:3) as eluent. The blue acid 18 (169 mg, 74%)
TMS):
d
¼1.24 (3H, t, J¼6.8 Hz, NCH2CH3), 1.99 (3H, s, acetyl CH3),
3.04 (2H, m, CH2Ar), 3.50 (2H, q, J¼6.8 Hz, NCH2CH3), 3.71 (3H, t,
J¼5.1 Hz, NCH2CH2O), 4.44 (3H, t, J¼5.1 Hz, NCH2CH2O), 4.56 (2H, s,
OCH2CO2R), 4.77 (1H, m,
a-H), 6.72 (2H, d, J¼8.4 Hz, ArH), 6.80 (2H,
d, J¼9.2 Hz, ArH), 7.10 (2H, m, ArH), 7.23 (1H, d, J¼7.9 Hz, NH), 7.91
(4H, m, ArH), 8.31 (2H, d, ArH); 13C NMR (100 MHz, CDCl3, 298 K,
TMS):
d
¼12.7 (CH3), 23.0 (CH3), 38.1 (CH2), 46.6 (CH2), 49.3 (CH2),
61.7 (CH2), 66.1 (CH2), 113.8 (ArCH), 115.0 (ArCH), 124.0 (ArCH),
126.5 (ArCH), 128.7 (ArC), 129.6 (ArCH), 130.4 (ArC), 136.3 (ArC),
146.3 (ArC), 147.6 (ArC), 155.2 (ArC), 156.5 (ArC), 168.7 (C]O), 169.2
(C]O), 172.0 (C]O); elemental analysis: calcd for C29H31N5O8: C
59.38, H 5.50, N 11.94; found C 58.87, H 5.23, N 12.31.
was obtained as
a blue amorphous solid. IR: nmax (solid
state)¼2938, 1737, 1598, 1543, 1511, 1388 cmꢁ1; 1H NMR (400 MHz,

V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
6377
CDCl3, 298 K, TMS):
d
¼1.25 (3H, t, J¼8.5 Hz, CH3CH2N), 1.98 (3H, s,
stirred overnight. The reaction was cooled again to ice-bath tem-
perature and NH4OH (35% in water, 6 mL) was added dropwise.
When the addition was finished the mixture was allowed to rise to
room temperature again and was finally stirred for a further 30 min
before the volatiles were evaporated under vacuum. The crude
residue was triturated with boiling DCM and the crystalline solid
formed was removed by filtration. The filtrate was washed with
saturated aqueous NaHCO3 and dried over MgSO4, then the solvent
was evaporated under reduced pressure to give the tert-butyl ester
(S)-13 (2.66 g, 88%) as a white solid. An analytical sample was
NHCOCH3), 3.11 (2H, dd, J¼14.1, 5.9 Hz, CH2), 3.56 (2H, q, J¼7.1 Hz,
CH3CH2N), 3.80 (2H, t, J¼6.0 Hz, OCH2CH2N), 4.48 (2H, m,
OCH2CH2N), 4.59 (2H, s, OCH2CO), 4.78 (1H, m, NHCHCO), 6.01 (1H,
d, J¼7.6 Hz, NH), 6.72 (2H, d, J¼8.6 Hz, ArH), 6.84 (2H, d, J¼8.6 Hz,
ArH), 7.04 (2H, d, J¼9.3 Hz, ArH), 7.93 (2H, d, J¼9.3 Hz, ArCH), 8.34
(1H, s, ArCH); 13C NMR (100 MHz, CDCl3):
d
¼12.3 (CH3), 23.0 (CH3),
36.3 (CH2), 46.0 (CH2), 48.8 (CH), 53.6 (CH2), 61.9 (CH2), 65.1 (CH2),
114.6 (ArC), 125.0 (ArCH), 129.1 (ArC), 130.5 (ArC), 137.1 (ArC), 144.0
(ArC), 154.5 (ArC), 156.7 (ArC), 162.6 (ArC), 168.9 (C]O), 170.7
(C]O); MS (ESIþ): m/z¼629 [MþH]þ, 651 [MþNa]þ; HRMS (ESIþ):
m/z calcd for [C27H28N6O10SþH]þ 629.1660, found 629.1649 and
[C27H28N6O10SþNa]þ 651.1480, found 651.1463.
prepared by crystallising (S)-13 from hot benzene/hexane. Melting
20
point (benzene/hexane)¼140–141 ꢀC; [
a
]
þ24.9 (c 0.5, CHCl3); IR:
D
nmax (solid state)¼3373, 2985, 1696 cmꢁ1
;
1H NMR (400 MHz,
CDCl3):
d
¼1.41 (9H, s, C(CH3)3), 1.52 (3H, d, J¼7.2 Hz, CHCH3), 3.74
4.10.
L-Tyrosine-derived blue carboxylate salt 19
(1H, q, J¼7.1 Hz, CHCH3), 7.04–7.09 (2H, m, ArH), 7.38 (1H, dd, J¼8.5,
1.6 Hz, ArH), 7.59 (1H, d, J¼8.6 Hz, ArH), 7.63 (1H, s, ArH), 7.66 (1H,
The procedure described above for 16 was employed to convert
acid 18 into salt 19. IR: nmax (solid state)¼2936,1737,1598,1543,1511,
d, J¼8.8 Hz, ArH); 13C NMR (100 MHz, CDCl3):
¼18.8 (CH3), 28.4
d
(C(CH3)3), 46.8 (CH), 81.0 (C(CH3)3), 109.7 (ArCH), 118.2 (ArCH),
126.2 (ArCH), 126.9 (ArCH), 126.9 (ArCH), 129.3 (ArC), 134.0 (ArCH),
136.6 (ArC), 153.7 (ArC), 175.3 (CO2); MS (EIþ): m/z (%)¼272 (25)
[M]þ, 171 (100) [MꢁCO2C(CH3)3]þ; HRMS (EIþ): m/z calcd for
[C17H20O3]þ 272.1412, found 272.1409.
1388 cmꢁ1; 1H NMR (400 MHz, CDCl3, 298 K, TMS):
d¼0.99 (12H, t,
J¼8.5 Hz, Bu4NþCH3), 1.25 (3H, t, J¼8.5 Hz, CH3CH2N), 1.41 (8H, m,
Bu4NþCH2), 1.65 (8H, m, Bu4NþCH2), 1.98 (3H, s, CH3), 3.13 (2H, dd, J
14.2, 5.9, CH2Ph), 3.24 (8H, t, J¼8.5 Hz, Bu4NþCH2), 3.55 (2H, q,
J¼7.1 Hz, CH3CH2N), 3.80 (2H, t, J¼6.1 Hz, OCH2CH2N), 4.44 (2H, t,
J¼6.1 Hz, OCH2CH2N), 4.59 (3H, m), 6.73 (2H, d, J¼8.6 Hz, ArH), 6.84
(2H, d, J¼8.6 Hz, ArH), 7.02 (1H, d, J¼7.80 Hz, NH), 7.12 (2H, d,
J¼8.5 Hz, ArH), 7.96 (2H, d, J¼8.5 Hz, ArCH), 8.37 (1H, s, ArCH); 13C
4.13. (S)-2-(6-Benzyloxycarbonylmethoxy-naphthalen-
2-yl)-propionic acid tert-butyl ester
NMR (100 MHz, CDCl3):
d¼12.4 (CH3), 13.6 (CH3), 21.2 (CH2), 23.1
Note: The (R)-enantiomer was prepared in the same fashion.
The ester (S)-13 (2.50 g, 9.18 mmol) and K2CO3 (1.29 g,
9.19 mmol) were suspended in dry acetonitrile (90 mL) under
a nitrogen atmosphere. Benzyl 2-bromoacetate (2.74 g, 1.89 mL,
11.96 mmol) was added and the reaction mixture was stirred at
room temperature for 40 h. The solvent was then evaporated under
reduced pressure. The residue was dissolved in DCM, washed with
water, dried (MgSO4) and the organic solvent was evaporated. Flash
(CH3), 23.9 (CH2), 31.9 (CH2), 46.4 (CH2), 48.8 (CH), 58.9 (CH2), 61.8
(CH2), 62.5 (CH2), 65.1 (CH2), 114.6 (ArCH), 124.9 (ArCH), 129.1 (ArC),
130.5 (ArCH), 137.1 (ArC), 144.0 (ArC), 154.5 (ArC), 156.7 (ArC), 162.6
(ArC), 168.9 (CO2), 170.7 (CONH); MS (ESIþ): m/z¼629 [MþH]þ, 651
[MþNa]þ from the tyrosine carboxylate and 242.28 [M]þ counterion
salt; HRMS (ESIþ): m/z calcd for [C27H28N6O10SþH]þ 629.1660, found
629.1649 and [C27H28N6O10SþNa]þ 651.1480, found 651.1463 and
[C16H36N]þ 242.2842, found 242.2854.
chromatography (hexane/EtOAc, 8:1) afforded the title product
20
(2.87 g, 74%) as a colourless oil. [
a
]
D
þ9.2 (c 0.5, CHCl3); IR: nmax
4.11. (S)-2-(6-Hydroxy-naphthalen-2-yl)-propionic
acid (S)-126
(oil)¼2978, 2934, 1754, 1720, 1605 cmꢁ1
;
1H NMR (400 MHz,
CDCl3):
d
¼1.39 (9H, s, C(CH3)3), 1.52 (3H, d, J¼7.1 Hz, CHCH3), 3.74
(1H, q, J¼7.1 Hz, CHCH3), 4.77 (2H, s, CH2), 5.26 (2H, s, CH2), 7.01
(1H, d, J¼2.4 Hz, ArH), 7.21 (1H, dd, J¼8.9, 2.5 Hz, ArH), 7.26–7.42
(6H, m, ArH), 7.62 (1H, d, J¼8.5 Hz, ArH), 7.66 (1H, s, ArH), 7.73 (1H,
Note: The (R)-enantiomer was prepared in the same fashion.
(S)-Naproxen 5 (3.0 g, 13.0 mmol) was suspended in HBr (48%
w/w in water, 120 mL) and heated at reflux for 2 h. The mixture was
allowed to warm to room temperature and filtered. The solid was
d, J¼9.0 Hz, ArH); 13C NMR (100 MHz, CDCl3):
¼18.9 (CH3), 28.3
d
(C(CH3)3), 46.8 (CH), 65.9 (CH2), 67.4 (CH2), 80.9 (C(CH3)3), 107.5
(ArCH), 119.0 (ArCH), 126.1 (ArCH), 126.9 (ArCH), 127.4 (ArCH), 128.8
(ArC), 129.0 (ArCH), 130.0 (ArCH), 133.6 (ArC), 135.6 (ArC), 137.3
(ArC), 156.1 (ArC), 169.0 (CO2), 174.2 (CO2); MS (EIþ) m/z (%)¼420
(45) [M]þ, 364 (18) [MꢁC(CH3)3þH]þ, 319 (100) [MꢁCO2C(CH3)3]þ;
HRMS (EIþ): m/z calcd for [C26H28O5]þ 420.1937, found 420.1937.
washed with chilled water to afford the title product (S)-12 (2.71 g,
20
96%) as a white solid. [
a
]
þ56.1 (c 0.5, acetone); IR (solid state):
D
nmax¼3215, 1698, 1128 cmꢁ1; 1H NMR (400 MHz, DMSO-d6):
¼1.46
d
(3H, d, J¼7.1 Hz, CH3), 3.80 (1H, q, J¼7.1 Hz, CH), 7.08–7.13 (2H, m,
ArH), 7.36 (1H, dd, J¼8.4, 1.5 Hz, ArH), 7.65–7.69 (2H, m, ArH), 7.76
(1H, d, J¼8.4 Hz, ArH), 9.72 (1H, s, ArOH), 12.32 (1H, br s, CO2H); 13
C
NMR (100 MHz, DMSO-d6):
d¼19.3 (CH3), 45.4 (CH), 109.3 (ArC),
4.14. (S)-2-(6-Carboxymethoxy-naphthalen-2-yl)-
propionic acid tert-butyl ester (S)-8
119.6 (ArC), 126.4 (ArC), 126.9 (ArC), 127.0 (ArC), 127.0 (ArC), 128.5
(ArC), 129.9 (ArC), 134.4 (ArC), 156.0 (ArC), 176.3 (CO2H) (lit.20
d
¼18.4, 44.54, 108.4, 118.7, 125.5, 126.09, 126.13, 127.6, 129.1, 133.5,
Note: The (R)-enantiomer was prepared in the same fashion.
The above benzylic ester (1.08 g, 2.57 mmol) and Pd/C (Degussa
type, 10% w/w Pd content, 50% of water content) (220 mg) were
suspended in chloroform (25 mL) and vigorously stirred under
1 atm of hydrogen for 2 h. The mixture was filtered through a plug
of Celite, washing with chloroform. The filtrate was evaporated
under reduced pressure to obtain the carboxylic acid (S)-9 (846 mg,
quantitative yield) as an off-white solid. For analytical purposes
a sample was crystallised from dichloromethane/hexane obtaining
135.4, 155.1, 175.5).
4.12. (S)-2-(6-Hydroxy-naphthalen-2-yl)-propionic
acid tert-butyl ester (S)-13
Note: The (R)-enantiomer was prepared in the same fashion.
The carboxylic acid (S)-12 (2.39 g, 11.05 mmol) was dissolved in
dry THF (110 mL) under a nitrogen atmosphere. The solution was
cooled to 0 ꢀC, then trifluoroacetic anhydride (13.93 g, 9.42 mL,
66.32 mmol) was added dropwise and the mixture was further
stirred for 4 h maintaining the temperature below 5 ꢀC. tert-Buta-
nol (32 mL) was added dropwise and the resulting mixture was
then allowed to rise to room temperature and was vigorously
(S)-8 as white plates. Melting point (DCM/hexane)¼117–118 ꢀC;
20
[a
]
17.5 (c 0.5, CHCl3); IR: nmax (solid state)¼3055, 2978, 2904,
D
1748, 1720, 1607, 1505 cmꢁ1
;
1H NMR (400 MHz, CDCl3):
d¼1.39
(9H, s, C(CH3)3), 1.52 (3H, d, J¼7.1 Hz, CHCH3), 3.75 (1H, q, J¼7.1 Hz,
CHCH3), 4.78 (2H, s, CH2), 7.08 (1H, d, J¼2.4 Hz, ArH), 7.21 (1H, dd,

6378
V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
J¼8.9, 2.5 Hz, ArH), 7.42 (1H, dd, J¼8.4, 1.4 Hz, ArH), 7.68 (2H, dþs,
J¼9.4 Hz, ArH), 7.74 (1H, d, J¼9.0 Hz, ArH), 9.95 (s, 1H, CO2H); 13C
(2H, d, J¼8.9 Hz, ArH), 7.91 (2H, d, J¼8.9 Hz, ArH), 8.32 (2H, d,
J¼8.7 Hz, ArH); 13C NMR (100 MHz, CDCl3):
d
¼12.5 (CH3), 18.5
NMR (100 MHz, CDCl3):
d
¼18.8 (CH3), 28.3 (C(CH3)3), 46.8 (CH),
(CH3), 45.5 (CH), 45.9 (CH2), 48.9 (CH2), 62.6 (CH2), 65.7 (CH2), 107.4
(ArCH), 111.9 (ArCH), 119.1 (ArCH), 123.0 (ArCH), 125.0 (ArCH), 126.5
(ArCH), 126.6 (ArCH), 126.8 (ArCH), 127.7 (ArCH), 129.1 (ArC), 129.8
(ArC), 130.1 (ArCH), 133.8 (ArC), 136.0 (ArC), 144.3 (ArC), 151.5 (ArC),
156.1 (ArC), 157.0 (ArC), 169.2 (CO2), 179.9 (CO2H); MS (ESþ) m/z
(%)¼593 (60) [MþNa]þ, 571 (100) [MþH]þ; HRMS (ESIþ): m/z calcd
for [C31H30N4O7þH]þ 571.2187 found, 571.2190. Anal. found: C,
60.99; H, 5.22; N, 8.77%. C31H30N4O7$2H2O requires C, 61.38; H,
5.65; N, 9.24%. The enantiopurity of the carboxylic acid 21 was
analysed by chiral HPLC on an (S,S)-Whelk O1 analytical column,
eluting with EtOH/CHCl3/CF3CO2H (80:20:0.1) (flow rate¼1 mL/
65.3 (CH2), 81.1 (C(CH3)3), 107.5 (ArCH), 118.8 (ArCH), 126.2 (ArCH),
127.0 (ArCH), 127.4 (ArCH), 130.0 (ArC), 130.1 (ArCH), 133.6 (ArC),
137.5 (ArC), 155.7 (ArC), 173.9 (CO2), 174.4 (CO2); MS (EIþ) m/z
(%)¼330 (23) [M]þ, 274 (8) [MꢁC(CH3)3þH]þ, 229 (100)
[MꢁCO2C(CH3)3]þ; HRMS (EIþ): m/z calcd for [C19H22O5]þ 330.1467,
found 330.1467. Anal. found: C, 69.00; H, 6.79%. C19H22O5 requires
C, 69.07; H, 6.71%.
4.15. (R)-Naproxen-derived red tert-butyl ester 20
Acid derivative (R)-8 (492 mg, 1.49 mmol) and 1,10-carbon-
yldiimidazole (241 mg, 1.49 mmol) were dissolved in dry DCM
(15 mL) at 0 ꢀC and the mixture was stirred for 1 h under a nitrogen
atmosphere. Disperse Red 1 1 (347 mg, 1.10 mmol) dissolved in dry
DCM (1 mL) was added. The mixture was allowed to rise to room
temperature then further stirred for 24 h. The mixture was then
diluted with DCM (15 mL) and washed with saturated aqueous
NaHCO3. The organic phase was dried (MgSO4) and the solvent
evaporated under reduced pressure. Flash chromatography (hex-
ane/EtOAc, 4:1) afforded the title compound 20 (447 mg, 65%) as an
amorphous red solid. An analytical sample was crystallised from
dichloromethane/hexanes as red platelets. Melting point (DCM/
hexane) >96–97 ꢀC (decomposed); IR: nmax (solid state)¼2971,
min, UV detection at
l
¼254 nm). A small sample of the corre-
sponding S enantiomer was prepared and analysed to confirm that
the separation was possible. The enantiomeric excess of 21 was
found to be 99%. Retention times: 21, 8.8 min; (S)-21, 12.5 min.
4.17. (R)-Naproxen-derived red carboxylate salt 22
The carboxylic acid 21 (785 mg, 1.39 mmol) was dissolved in
chloroform (15 mL) and cooled to 0 ꢀC. Tetrabutylammonium hy-
droxide (1.0 M in methanol, 1.44 mL, 1.44 mmol) was diluted in
15 mL of chloroform and was added dropwise with vigorous stir-
ring. The mixture was stirred for a further 5 min before the solvent
was evaporated under reduced pressure. The residue was redis-
solved in ethyl acetate (50 mL) and washed with water (10 mL). The
organic phase was dried (MgSO4) and the solvent evaporated to
give 22 (1.09 g, 97%) as a brownish-red amorphous solid. 1H NMR
1757, 1728, 1603, 1519, 1509 cmꢁ1
;
1H NMR (400 MHz, CDCl3):
d
¼1.20 (3H, t, J¼6.9 Hz, CH2CH3), 1.39 (9H, s, C(CH3)3), 1.51 (3H, d,
J¼7.1 Hz, CHCH3), 3.44 (2H, q, J¼7.1 Hz, CH2CH3), 3.69–3.77 (3H, m,
CHCH3þCH2), 4.45 (2H, t, J¼6.1 Hz, CH2), 4.73 (2H, s, CH2), 6.77 (2H,
d, J¼9.2 Hz, ArH), 7.01 (1H, d, J¼2.4 Hz, ArH), 7.18 (1H, dd, J¼8.9,
2.5 Hz, ArH), 7.40 (1H, dd, J¼8.5, 1.4 Hz, ArH), 7.64 (1H, d, J¼8.9 Hz,
ArH), 7.65 (1H, s, ArH), 7.73 (1H, d, J¼9.0 Hz, ArH), 7.87–7.93 (4H, m,
ArH), 8.32 (2H, d, J¼9.0 Hz, ArH); 13C NMR (100 MHz, CDCl3):
(400 MHz, CDCl3):
d¼0.90 (12H, t, J¼7.3 Hz, CH3), 1.20 (3H, t,
J¼7.1 Hz, CH2CH3), 1.25–1.34 (8H, m, CH2), 1.38–1.48 (8H, m, CH2),
1.54 (3H, d, J¼7.0 Hz, CHCH3), 3.05–3.09 (8H, m, CH2), 3.51 (2H, q,
J¼7.1 Hz, CH2CH3), 3.57 (2H, t, J¼6.0 Hz, CH2), 3.73 (1H, q, J¼7.1 Hz,
CHCH3), 4.70 (4H, sþm, 2ꢂCH2), 6.79 (2H, d, J¼9.3 Hz, ArH), 6.99
(1H, d, J¼2.4 Hz, ArH), 7.13 (1H, dd, J¼9.0, 2.6 Hz, ArH), 7.57 (1H, d,
J¼8.5 Hz, ArH), 7.67 (1H, d, J¼8.9 Hz, ArH), 7.68 (1H, dd, J¼8.5,
1.6 Hz, ArH), 7.78 (1H, s, ArH), 7.85 (2H, d, J¼9.2 Hz, ArH), 7.89 (2H,
d, J¼8.9 Hz, ArH), 8.30 (2H, d, J¼9.1 Hz, ArH); MS (ESIþ): m/z
(%)¼571 (100) [MþH]þ, 242 (47) [NBu4]þ.
d¼12.6 (CH3), 18.9 (CH3), 28.3 (C(CH3)3), 42.0 (CH2), 46.0 (CH2), 46.8
(CH), 62.6 (CH2), 65.7 (CH2), 84.8 (C(CH3)3), 108.2 (ArCH), 111.9
(ArCH),118.9 (ArCH),123.1 (ArCH),125.0 (ArCH),126.2 (ArCH),126.6
(ArCH), 127.0 (ArCH), 127.4 (ArCH), 130.1 (ArCH), 133.6 (ArC), 142.1
(ArC), 176, 157.1 (ArC), 157.6 (ArC), 164.8 (CO2), 176.1 (CO2); MS
(ESIþ) m/z (%)¼649 (11) [MþNa]þ, 627 (44) [MþH]þ, 571 (65)
[MꢁC(CH3)3þH]þ; HRMS (ESIþ): m/z calcd for [C38H38N4O7þH]þ
627.2813, found 627.2819. Anal. found: C, 67.11; H, 6.37; N, 8.92%.
C35H38N4O7 requires C, 67.08; H, 6.11; N, 8.94%.
4.18. (S)-Naproxen-derived blue tert-butyl ester 23
The carboxylic acid derivative 8 (600 mg, 1.82 mmol) and car-
bonyl diimidazole (300 mg, 1.86 mmol) were dissolved in dry DCM
(20 mL) at 0 ꢀC and the mixture was stirred for 1 h under nitrogen
atmosphere. Bristol Blue 1 3 (347 mg, 1.14 mmol) dissolved in dry
DCM (2 mL) was added dropwise. The resulting mixture was
allowed to rise to room temperature and was further stirred for 3
days. The volatiles were evaporated and the crude solid was dis-
solved in EtOAc (50 mL). The solution was washed with a saturated
aqueous NaHCO3 solution (50 mL), the organic phase was dried
(MgSO4) and the solvent evaporated under reduced pressure. Flash
chromatography (hexane/EtOAc, 3:2) afforded the title compound
23 (596 mg, 77%) as a deep-blue amorphous solid. IR: nmax (solid
4.16. (R)-Naproxen-derived red acid 21
The tert-butyl ester 20 (870 mg, 1.39 mmol) was dissolved in
DCM (4 mL) and the solution was cooled to 0 ꢀC before TFA (4 mL)
was added dropwise with vigorous stirring. The mixture was stir-
red for 1 h at 0 ꢀC and then allowed to warm to room temperature
before further stirring for 3 h. The volatiles were evaporated, and
the crude residue was dissolved in chloroform and washed with
water. The aqueous layer from the washings was extracted several
times with chloroform until no red colouration remained in the
aqueous phase. The combined organic layers were dried (MgSO4)
and the solvent evaporated under reduced pressure to give the
carboxylic acid 21 (785 mg, quantitative yield) as a red solid. An
analytical sample was prepared by crystallisation from dichloro-
methane/hexane. Melting point (DCM/hexane)¼152–153 ꢀC; IR:
nmax (solid state)¼3054, 2908,1737, 1704,1600,1587, 1508 cmꢁ1; 1H
state)¼2978, 1762, 1724, 1599, 1545 cmꢁ1
;
1H NMR (400 MHz,
¼1.23 (3H, t, J¼7.1 Hz, CH2CH3), 1.39 (9H, s, C(CH3)3), 1.52
CDCl3):
d
(3H, d, J¼7.1 Hz, CHCH3), 3.49 (2H, q, J¼7.1 Hz, CH2CH3), 3.74 (1H, q,
J¼7.1 Hz, CHCH3), 3.78 (2H, t, J¼5.9 Hz, CH2), 4.48 (2H, t, J¼5.9 Hz,
CH2), 4.74 (2H, s, CH2), 6.78 (2H, d, J¼9.3 Hz, ArH), 6.99 (1H, d,
J¼2.4 Hz, ArH), 7.16 (1H, dd, J¼8.9, 2.5 Hz, ArH), 7.41 (1H, dd, J¼8.5,
1.4 Hz, ArH), 7.63 (1H, d, J¼8.5 Hz, ArH), 7.66 (1H, s, ArH), 7.73 (1H,
d, J¼8.9 Hz, ArH), 7.92 (2H, d, J¼8.8 Hz, ArH), 8.35 (1H, s, ArH); 13C
NMR (400 MHz, CDCl3):
d
¼1.20 (3H, t, J¼7.0 Hz, CH2CH3), 1.58 (3H,
d, J¼6.8 Hz, CHCH3), 3.43 (2H, q, J¼7.0 Hz, CH2CH3), 3.70 (2H, t,
J¼6.0 Hz, CH2), 3.87 (1H, m, CHCH3), 4.44 (2H, t. J¼6.0 Hz, CH2), 4.72
(2H, s, CH2), 6.74 (2H, d, J¼8.9 Hz, ArH), 6.99 (1H, d, J¼2.0 Hz, ArH),
7.18 (1H, dd, J¼8.8,1.8 Hz, ArH), 7.40 (1H, d, J¼8.4 Hz, ArH), 7.63 (1H,
d, J¼8.3 Hz, ArH), 7.68 (1H, s, ArH), 7.71 (1H, d, J¼8.9 Hz, ArH), 7.86
NMR (100 MHz, CDCl3):
d
¼12.6 (CH3), 18.8 (CH3), 28.3 (C(CH3)3),
46.6 (CH2), 46.7 (CH), 49.3 (CH2), 62.3 (CH2), 65.7 (CH2), 80.9
(C(CH3)3), 107.1 (ArCH), 113.0 (ArCH), 118.8 (ArCH), 125.3 (ArCH),
126.1 (ArCH), 127.0 (ArCH), 127.3 (ArCH), 129.9 (ArC), 130.1 (ArCH),

V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
6379
133.5 (ArC), 137.5 (ArC), 144.4 (ArC), 154.5 (ArC), 155.8 (ArC), 169.2
(CO2), 174.1 (CO2); MS (EIþ) m/z (%)¼677 (68) [M]þ; HRMS (ESIþ):
m/z calcd for [C33H35N5O9SþNa]þ 700.2053, found 700.2048.
evaporated under reduced pressure. The residue was redissolved
in dichloromethane, washed with aqueous HCl (1 M) and dried
(MgSO4). The organic solvent was evaporated under reduced
pressure. Flash chromatography (hexane/EtOAc, 2:1) afforded 28
(2.29 g, 75%) as a pale yellowish amorphous solid; IR: nmax (solid
4.19. (S)-Naproxen-derived blue acid 24
state)¼3422, 3286, 2956, 2870, 2090, 1697, 1535, 1159 cmꢁ1
;
1H
The tert-butyl ester 23 (560 mg, 0.83 mmol) was dissolved in
DCM (8 mL) and the solution was cooled to 0 ꢀC before TFA (6 mL)
was added dropwise with vigorous stirring. The mixture was stir-
red for 1 h at 0 ꢀC and then allowed to warm to room temperature
before further stirring for 10 h. The volatiles were evaporated and
the crude residue was redissolved in chloroform and washed with
water. The aqueous layer from the washings was extracted several
times with chloroform until no blue colouration remained in the
aqueous phase. The combined organic layers were dried (MgSO4)
and the solvent evaporated under reduced pressure. Flash chro-
matography (DCM/EtOAc, 6:1 to DCM/EtOAc, 2:1) afforded the
carboxylic acid 24 (364 mg, 71%) as a deep-blue amorphous solid.
IR: nmax (solid state)¼3110, 3058, 2976, 2932,1758,1736,1704,1596,
NMR (400 MHz, CDCl3):
(3H, s, 18-CH3), 0.86 (3H, s, 19-CH3), 3.09 (1H, m, 3
d
¼0.82 (3H, d, J¼5.6 Hz, 21-CH3), 0.83
b
-H), 3.62 (3H,
-H),
s, CO2CH3), 3.82 (1H, d, J¼9.6 Hz, 7
b
-H), 4.02 (1H, br s, 12b
5.17 (1H, d, J¼7.8 Hz, NHSO2), 6.47 (1H, br s, CH–NHC(O)), 7.39
(1H, s, NHC(O)NHAr), 7.53 (2H, d, J¼8.7 Hz, ArH), 7.62 (2H, d,
J¼8.7 Hz, ArH), 7.78–7.86 (2H, m, ArH), 7.93–7.95 (1H, m, ArH),
8.16–8.19 (1H, m, ArH); 13C NMR (100 MHz, CDCl3):
d
¼14.6 (CH3),
23.1 (CH3), 23.5 (CH3), 26.2 (CH2), 26.9 (CH2), 29.6 (CH), 32.0
(CH2), 32.5 (CH2), 35.2 (CH2), 35.4 (CH2), 35.6 (CH2), 37.0 (CH),
42.0 (CH), 44.4 (CH), 46.9 (CH), 49.3 (CH), 51.7 (CO2CH3), 57.7
(CH), 61.2 (CH), 118.4 (ArCH), 125.8 (ArCH), 126.5 (ArCH), 126.5
(ArC), 131.0 (ArCH), 133.4 (ArCH), 134.5 (ArCH), 143.4 (ArC), 154.4
(NHC(O)NH), 174.7 (CO2CH3); MS (ESIþ): m/z (%)¼840 (49)
[MþNa]þ, 818 (87) [MþH]þ; HRMS (ESIþ): m/z calcd for
[C39H50F3N7O7SþNH4]þ 835.3783, found 835.3787.
1542 cmꢁ1
;
1H NMR (400 MHz, CDCl3):
d
¼1.21 (3H, t, J¼7.1 Hz,
CH2CH3), 1.57 (3H, d, J¼7.1 Hz, CHCH3), 3.44 (2H, q, J¼7.1 Hz,
CH2CH3), 3.75 (2H, t, J¼5.8 Hz, CH2), 3.86 (1H, q, J¼7.1 Hz, CHCH3),
4.46 (2H, t, J¼5.8 Hz, CH2), 4.72 (2H, s, CH2), 6.71 (2H, d, J¼9.3 Hz,
ArH), 6.95 (1H, d, J¼2.4 Hz, ArH), 7.13 (1H, dd, J¼8.9, 2.5 Hz, ArH),
7.34 (1H, dd, J¼8.6, 1.5 Hz, ArH), 7.59 (1H, d, J¼8.5 Hz, ArH), 7.65
(1H, br s, ArH), 7.68 (1H, d, J¼9.0 Hz, ArH), 7.85 (2H, d, J¼8.9 Hz,
4.22. 3
aminocarbonylamino]-12
amino]-5 -cholan-24-oic acid 29
a
-Azido-7
a-[(4-trifluoromethylphenyl)-
a-[(2-nitrobenzenesulfonyl)-
b
ArH), 8.28 (1H, s, ArH); 13C NMR (100 MHz, CDCl3):
d¼12.6 (CH3),
Steroid 28 (2.29 g, 2.80 mmol) was dissolved in methanol
(23 mL). A solution of NaOH (448 mg, 11.20 mmol) in deionised
water (7 mL) was added and the mixture was stirred at room
temperature for three days. The reaction was quenched with
aqueous HCl (2 M, 10 mL). The mixture was evaporated under re-
duced pressure, redissolved in ethyl acetate, washed with water
and dried (MgSO4). The solvent was removed under vacuum. Flash
chromatography (DCM/MeOH, 98:2) gave the carboxylic acid 29
(1.73 g, 88%) as a white solid. IR: nmax (solid state)¼3293, 2939,
18.5 (CH3), 45.4 (CH2), 46.5 (CH), 49.2 (CH2), 62.4 (CH2), 65.6 (CH2),
107.1 (ArCH), 113.0 (ArCH), 119.0 (ArCH), 125.3 (ArCH), 126.9 (ArCH),
127.0 (ArCH), 127.7 (ArCH), 130.2 (ArCH), 130.6 (ArC), 133.7 (ArC),
143.6 (ArC), 154.5 (ArC), 169.8 (CO2), 175.3 (CO2); MS (ESIþ) m/z
(%)¼644 (76) [MþNa]þ, 622 (65) [MþH]þ; HRMS (ESIþ): m/z calcd
for [C29H27N5O9SþH]þ 622.1602, found 622.1605.
4.20. (S)-Naproxen-derived blue carboxylate salt 25
2869, 2091, 1684, 1537, 1159 cmꢁ1 1H NMR (400 MHz, CDCl3):
;
The carboxylic acid 24 (374 mg, 0.60 mmol) was dissolved in
chloroform (7 mL) and cooled to 0 ꢀC. Tetrabutylammonium hy-
droxide (1.0 M in methanol, 0.63 mL, 0.63 mmol) was diluted in
chloroform (7 mL) and was added dropwise, with vigorous stirring.
The mixture was stirred for a further 5 min before the solvent was
evaporated under reduced pressure. The residue was dissolved in
EtOAc (30 mL) and washed with water (8 mL). The organic layer
was dried (MgSO4) and the solvent was evaporated to give the title
compound 25 (480 mg, 93%) as a dark blue amorphous solid. 1H
d
¼0.76 (d, J¼6.5 Hz, 3H, 21-H3), 0.79 (s, 3H, 18-H3), 0.86 (s, 3H, 19-
H3), 3.08 (br s, 1H, 3b-H), 3.83 (m, 1H, 7b-H), 3.96 (br s, 1H, 12b-H),
5.39 (b s,1H, NHSO2R), 6.38 (br s,1H, CH–NHC(O)), 7.52 (d, J¼8.6 Hz,
2H, ArH), 7.59 (d, J¼8.6 Hz, 2H, ArH), 7.74–7.92 (m, 3H, ArH), 8.15 (d,
J¼7.8 Hz, 1H, ArH); 13C NMR (100 MHz, CDCl3/CD3OD 5:1):
¼13.4
d
(CH3), 17.2 (CH3), 22.7 (CH3), 23.1 (CH2), 26.0 (CH2), 26.4 (CH2), 27.1
(CH2), 28.4 (CH), 30.5 (CH2), 31.0 (CH2), 31.8 (CH2), 34.6 (C), 34.9
(CH2), 35.1 (CH2), 35.2 (CH), 36.7 (CH), 41.4 (CH), 44.1 (CH), 45.8 (C),
46.1 (CH), 47.4 (CH), 57.9 (CH), 61.0 (CH), 117.5 (ArCH), 123.4 (q,
J¼32.3 Hz, ArC–CF3), 124.6 (q, J¼216.0 Hz, CF3), 125.1 (ArCH), 126.0
(q, J¼3.8 Hz, F3C–C]CH), 130.7 (ArCH), 133.0 (ArCH), 133.7 (ArCH),
135.0 (ArC), 143.0 (ArC), 147.9 (ArC), 155.1 (NHCO), 177.0 (CO2H); MS
(ESIþ): m/z (%)¼826 (100) [MþNa]þ, 804 (26) [MþH]þ, 615 (27)
NMR (400 MHz, CDCl3):
d
¼0.91 (12H, t, J¼7.3 Hz, CH3), 1.21 (3H, t,
J¼7.1 Hz, CH2CH3), 1.25–1.34 (8H, m, CH2), 1.38–1.48 (8H, m, CH2),
1.55 (3H, d, J¼7.2 Hz, CHCH3), 3.05–3.09 (8H, m, CH2), 3.46 (2H, q,
J¼7.1 Hz, CH2CH3), 3.73 (2H, t, J¼5.8 Hz, CH2), 3.84 (1H, q, J¼7.1 Hz,
CHCH3), 4.46 (2H, t, J¼5.8 Hz, CH2), 4.72 (2H, s, CH2), 6.71 (2H, d,
J¼9.3 Hz, ArH), 6.95 (1H, d, J¼2.4 Hz, ArH), 7.13 (1H, dd, J¼8.9,
2.5 Hz, ArH), 7.34 (1H, dd, J¼8.6, 1.5 Hz, ArH), 7.59 (1H, d, J¼8.5 Hz,
ArH), 7.65 (1H, br s, ArH), 7.68 (1H, d, J¼9.0 Hz, ArH), 7.85 (2H, d,
J¼8.9 Hz, ArH), 8.28 (1H, s, ArH); MS (ESIþ): m/z (%)¼622 (71)
[MþH]þ, 242 (49) [NBu4]þ.
[MꢁCF3–(C6H4)–NHCO]þ;
HRMS
(ESIþ):
m/z
calcd
for
[C38H48F3N7O7SþNH4]þ 821.3626, found 821.3627. Anal. found: C,
54.12; H, 6.26; N, 11.35%. C38H48F3N7O7S$2H2O requires C, 54.34; H,
6.24; N, 11.67%.
4.23. Protocols in solid phase
4.21. Methyl 3
aminocarbonylamino]-12
amino]-5 -cholan-24-oate 28
a
-azido-7
a
-[(4-trifluoromethylphenyl)-
a
-[(2-nitrobenzenesulfonyl)-
Reactions on solid phase were carried out in polypropylene 1, 2,
5 or 25 mL solid phase extraction tubes (SPET) purchased from
Supelco. SPETs (therefore resins) were agitated by a Stuart SB2 solid
phase rotator. After reaction, beads were dried on a Supelco Visi-
prep solid phase extraction vacuum manifold. For observation after
screening experiments, beads were placed on Petri dishes and
viewed through an optical Olympus ST-PZ microscope equipped
with an Olympus U-PMTVC and C3040-ADL camera adaptors. Mi-
crographs such as those shown in Figures 5 and 6 were taken with
b
To
a
solution of the steroidal derivative 279a (2.37 g,
3.76 mmol) and DMAP (229 mg, 1.88 mmol), in dry THF (40 mL),
were added triethylamine (378 mg, 0.52 mL, 3.74 mmol) and
p-trifluoromethylphenyl
isocyanate
(838 mg,
0.64 mL,
4.48 mmol) at room temperature with vigorous stirring. The re-
action mixture was heated at reflux for 6 h before the solvent was

6380
V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
an Olympus C-5050 ZOOM digital camera and processed with
standard computer software.
4.28. Standard procedure for estimation of first
amino acid attachment
Completely dry Fmoc-containing resin (w3 mg, weighed accu-
rately) was placed in a 10 mm quartz UV cuvette. Freshly prepared
20% piperidine in DMF (3 mL, measured accurately) was added to
the cuvette and to another cuvette as reference. The resin was
carefully agitated for 5 min with the aid of a Pasteur pipette. The
cuvettes were placed in a spectrophotometer and the absorbance at
301 nm read. The level of loading on the resin can be estimated
from the following equation:
4.24. Typical NF31 test for amino groups on solid phase
A solution of the NF31 dye13 in acetonitrile (ca. 1 M, 0.5 mL) was
added to a collection of approximately 20 beads in a small sample
vial and heated to reflux and left to cool. The heating/cooling cycle
was repeated, then the beads were left to settle and the solvent
carefully removed with the aid of a pipette. The beads were washed
with acetonitrile (3ꢂ1 mL) and DCM (3ꢂ1 mL), the solvent being
carefully removed by pipette each time, before being analysed
under the microscope. Intense red beads indicate the presence of
amino groups (positive result).
Fmoc loading (mmol gꢁ1)¼(AbssampleꢁAbsref)/(1.65ꢂweight of
resin in mg). In the present case the loading was calculated as
0.29 mmol of steroid per gram of resin.
4.29. Standard procedure for removal of Fmoc
protecting groups
4.25. 3
aminocarbonylamino]-12
amino]-5 -cholan-24-oic acid loaded onto Tentagel (30)
a
-Azido-7
a
-[(4-trifluoromethylphenyl)-
a
-[(2-nitrobenzenesulfonyl)-
b
The twelve resin portions employed for peptide synthesis were
combined and washed with MeOH (3ꢂ10 mL) and DMF (3ꢂ10 mL)
and agitated with 20% piperidine in DMF (10 mL) for 30 min. This
protocol was repeated twice on the same beads to ensure complete
removal of the Fmoc group. Care was taken to vent the sample to
avoid pressure build-up of CO2 in the tube. The beads were then
filtered and washed with DMF (3ꢂ10 mL), MeOH (3ꢂ10 mL) and
DCM (3ꢂ10 mL). The NF31 test produced intense red beads,
confirming deprotection.
Steroidal derivative 29 (1.60 g, 1.99 mmol), TBTU (636 mg,
1.99 mmol), DMAP (725 mg, 5.94 mmol) were dissolved in DMF
(11 mL). This mixture was then added to high loading NovaSynÒ TG
HL (0.44 mmol gꢁ1 21
amino resin (3.00 g, 1.32 mmol) in a solid
)
phase extraction tube (SPET) and agitated overnight. The resin was
then drained, washed with DMF (3ꢂ10 mL), MeOH (3ꢂ10 mL) and
DCM (3ꢂ10 mL) and dried. Testing with NF31 (see above) gave
a negative result. The resin was then agitated overnight with acetic
anhydride (1 mL), pyridine (1 mL) and DCM (10 mL) to cap any
remaining amine groups, giving resin-bound steroid 30. The char-
acteristic signal for the C3 azide at nmax¼2092 cmꢁ1 in the IR
spectrum confirmed the presence of the steroid.
4.30. Standard procedure for Boc-protecting
the peptide podand at C12
Following Fmoc removal of the second amino acid on the C12
position, the resin was agitated overnight with di-tert-butyl dicar-
bonate (4 equiv) in DCM (12 mL). The resin was then drained and
washed with DCM (3ꢂ10 mL), MeOH (3ꢂ10 mL) and DCM again
(3ꢂ10 mL). NF31 test on a few beads indicated the reaction was
complete.
4.26. C12 o-Ns deprotection on solid phase (30/31)
Resin 30 (3.50 g, ꢃ1.32 mmol), Cs2CO3 (3.87 g, 11.88 mmol) and
thiophenol (2.62 g, 2.44 mL, 23.76 mmol) were suspended in dry
DMF (25 mL) into a two-necked flask. Nitrogen gas was bubbled
through the mixture (via syringe needle) for agitating purposes
while it was heated at 60 ꢀC for 3 days. DMF lost by evaporation was
replaced as necessary. The resin was removed by filtration, washed
with water (3ꢂ10 mL), DMF (3ꢂ10 mL), MeOH (3ꢂ10 mL) and
finally DCM (3ꢂ10 mL) and dried. This resin was subjected again to
the same reaction conditions (Cs2CO3 and thiophenol in hot DMF)
to ensure complete cleavage of the C12 o-Ns group. NF31 test
showed preponderance of brightly red beads as expected from the
presence of the free amino function.
4.31. Standard procedure for reducing the azide at C3 position
A solution of 1.0 M of trimethylphosphine in THF (5 mL) was
diluted with THF (5 mL) and added to the resin, which was left open
to the atmosphere to avoid build-up of nitrogen gas. This was
gently agitated every 20 min until no more nitrogen was seen to be
evolving (w3 h). Deionised water (4 mL) was added and the tube
agitated for 2 h. The resin was then washed with THF (3ꢂ10 mL),
MeOH (3ꢂ10 mL) and finally DCM (3ꢂ10 mL) before being air-
dried. NF31 test on a few beads gave an intense red colour in-
dicating that the reaction was successful. IR of the resin showed the
4.27. Standard procedure for peptide synthesis
on solid phase (as carried out for library 33)
absence of the azide peak at
n
¼2091 cmꢁ1
.
Acknowledgements
The bulk resin was divided into twelve equal portions in twelve
SPET tubes. Each N-Fmoc-a-amino acid derivative (3 equiv) was
Financial support from the EU (RTN contract HPRN-CT-2001-
00182) and the EPSRC (GR/R42757/01) is gratefully acknowledged.
We thank Prof. Christian Roussel and Nicolas Vanthuyne, University
of Marseille, for performing the chiral HPLC analyses.
placed in a sample vial to which was added 1 mL of solution A and
1 mL of solution B. Solution A consisted of HATU (12ꢂ3 equiv) and
HOBt (12ꢂ2.9 equiv) dissolved in DMF (12 mL). Solution B con-
sisted of DIPEA (12ꢂ6 equiv) in DMF (12 mL). Each tube was gently
shaken to ensure complete dissolution. After being left to stand for
1 min, each amino acid coupling solution was added to a portion of
resin (1 equi loading) with DMF (4 mL) and agitated overnight.
Each portion of resin was then filtered, washed with DMF
(3ꢂ10 mL), MeOH (3ꢂ10 mL) and DCM (3ꢂ10 mL) and a few beads
of each portion subjected to NF31 test as shown above. A negative
test indicated that the reaction was complete. Any red colouration
indicated that the reaction needed to be repeated.
References and notes
1. (a) Burger, M. T.; Still, W. C. J. Org. Chem. 1997, 62, 4785; (b) Wennemers, H.;
Conza, M.; Nold, M.; Krattinger, P. Chem.dEur. J. 2001, 7, 3342; (c) Wennemers,
H.; Conza, M. J. Org. Chem. 2002, 67, 2696; (d) Davies, M.; Bonnat, M.; Guiller, F.;
Kilburn, J. D.; Bradley, M. J. Org. Chem. 2002, 63, 8696; (e) Nitz, M.; Franz, K. J.;
Maglathin, R. L.; Imperiali, B. ChemBioChem 2003, 4, 272; (f) Xu, R.; Greiveldinger,
G.; Marenus, L. E.; Cooper, A.; Ellman, J. A. J. Am. Chem. Soc. 1999, 121, 4898; (g)
Boyce, R.; Li, G.; Nestler, H. P.; Suenaga, T.; Still, W. C. J. Am. Chem. Soc. 1994, 116,

V. del Amo et al. / Tetrahedron 65 (2009) 6370–6381
6381
7955; (h) Borchardt, A.; Still, W. C. J. Am. Chem. Soc. 1994, 116, 373; (i) Ohlmeyer,
M. H. J.; Swanson, R. N.; Dillard, L. W.; Reader, J. C.; Ausoline, G.; Kobayashi, R.;
Wigler, M.; Still, W. C. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 10922; (j) Still, W. C. Acc.
Chem. Res. 1996, 29, 155.
2. del Amo, V.; McGlone, A. P.; Foster, C.; Davis, A. P. J. Comb. Chem. 2005, 7, 1.
3. For the use of two-colour screening to discover enantioselective reagents
(specifically, chiral amines which select for one enantiomer of an activated
ester), see: Weingarten, M. D.; Sekanina, K.; Still, W. C. J. Am. Chem. Soc. 1998,
120, 9112 As far as we know, the method has not yet been used to find enan-
tioselective receptors (binding through non-covalent interactions).
4. Easton, C.; Harper, J. B. Tetrahedron Lett. 1998, 39, 5269.
3499; (c) Kocis, P.; Issakova, O.; Sepetov, N. F.; Lebl, M. Tetrahedron Lett. 1995, 36,
6623; (d) Virta, P.; Leppa¨nen, M.; Lo¨nnberg, H. J. Org. Chem. 2004, 69, 2008; (e)
Farcy, N.; de Muynck, H.; Madder, A.; Hosten, N.; De Clercq, P. J. Org. Lett. 2001, 3,
4299; (f) Gea, A.; Farcy, N.; Hosten, N.; Martins, J.; De Clercq, P. J.; Madder, A. Eur. J.
Org. Chem. 2006, 18, 4135; (g) Van der Plas, S. E.; Gea, A.; Figaroli, S.; De Clercq, P. J.;
Madder, A. Eur. J. Org. Chem. 2008, 1582.
12. (a) Wilcox, C. S.; Kim, E.-I.; Romano, D.; Kao, L.-H.; Burt, A. L.; Curran, D. P.
Tetrahedron 1995, 51, 621; (b) Kyne, G. M.; Light, M. E.; Hursthouse, M. B.; de
Mendoza, J.; Kilburn, J. D. J. Chem. Soc., Perkin Trans. 1 2001, 1258; (c) Pelizzi, N.;
Cosnati, A.; Friggeri, A.; Ungaro, R. J. Chem. Soc., Perkin Trans. 2 1998, 6, 1307; (d)
Laurent, P.; Miyaji, H.; Collinson, S. R.; Prokes, I.; Moody, C. J.; Tucker, J. H. R.;
Slawin, A. M. Z. Org. Lett. 2002, 4, 4037.
13. Although Edman degradation should be applicable to the library members, the
method was not demonstrated during the present work. If it proved difficult to
implement, a photocleavable linker could be incorporated in the design to
allow analysis by MS (see Ref. 9c).
5. (a) Chevallet, P.; Garouste, P.; Malawska, B.; Martinez, J. Tetrahedron Lett. 1993,
34, 7409; (b) Stazzaolimi, P.; Scuccato, M.; Giumonini, A. G. Tetrahedron 2000,
56, 3625.
6. Franssen, E. J. F.; Koiter, J.; Kuipers, C. A. M.; Bruins, A. P.; Moolenaar, F.; Dezeeuw,
D.; Kruizinga, W. H.; Kellogg, R. M.; Meijer, D. K. F. J. Med. Chem. 1992, 35, 1246.
7. (a) Parish, R. C.; Stock, L. M. J. Org. Chem. 1965, 30, 927; (b) Bonar-Law, R. P.;
Davis, A. P.; Sanders, J. K. M. J. Chem. Soc., Perkin Trans. 1 1990, 8, 2245.
8. (a) Davis, A. P. Chem. Soc. Rev. 1993, 22, 243; (b) Davis, A. P.; Bonar-Law, R. P.;
Sanders, J. K. M. In Supramolecular Reactivity and Transport: Bioorganic Systems;
Murakami, Y., Ed.; Comprehensive Supramolecular Chemistry; Pergamon: Oxford,
1996; Vol. 4, p 257; (c) Davis, A. P.; Joos, J.-B. Coord. Chem. Rev. 2003, 240, 143; (d)
Davis, A. P. Coord. Chem. Rev. 2006, 250, 2939; (e) Davis, A. P. Molecules 2007, 12,
2106.
14. Madder, A.; Farcy, N.; Hosten, N. G. C.; De Muynck, H.; De Clerq, P. J.; Barry, J.;
Davis, A. P. Eur. J. Org. Chem. 1999, 7, 2787.
´
15. For a review on ‘‘split-and-mix’’ methodology see: Lam, K. S.; Lebl, M.; Krchna,
V. Chem. Rev. 1997, 97, 411; For remarkable examples see: (a) Lam, K. S.; Salmon,
S. E.; Hersh, E. M.; Hruby, V. J.; Kazmierski, W. M.; Knapp, R. J. Nature 1991, 354,
82; (b) Houghten, R. A.; Pinilla, C.; Blondelle, S. E.; Appel, J. R.; Dooley, C. T.;
Cuervo, J. H. Nature 1991, 354, 84.
16. (a) Abdelmoty, I.; Albericio, F.; Carpino, L. A.; Foxman, B. M.; Kates, S. A. Lett.
Pept. Sci. 1994, 1, 52; (b) Carpino, L. A.; El-Fahan, A.; Albericio, F.; Minor, C. A.
J. Chem. Soc., Chem. Commun. 1994, 201.
9. (a) del Amo, V.; Siracusa, L.; Markidis, T.; Baragan˜a, B.; Bhattarai, K. M.; Galobardes,
´
´
M.; Naredo, G.; Perez-Payan, M. N.; Davis, A. P. Org. Biomol. Chem. 2004, 2, 3320; (b)
De Muynck, H.; Madder, A.; Farcy, N.; De Clercq, P. J.; Pe´rez-Paya´n, M. N.;
Ohberg, L. M.; Davis, A. P. Angew. Chem., Int. Ed. 2000, 39,145; (c) Madder, A.; Li,
L.; De Muynck, H.; Farcy, N.; Van Haver, D.; Fant, F.; Vanhoenacker, G.; Sandra,
17. The Fmoc groups are removed by treatment with piperidine, and the soluble
UV-active side-product is assayed by spectroscopy. See: Fields, G. B.; Noble, R. L.
Int. J. Pept. Protein Res. 1990, 35, 161.
¨
P.; Davis, A. P.; De Clercq, P. J. J. Comb. Chem. 2002, 4, 552.
18. For a review on steroid-based combinatorial libraries see: Maltais, R.; Tremblay,
M. R.; Tiobanu, L. C.; Poirier, D. J. Comb. Chem. 2004, 6, 445.
19. Behr, C. J. Am. Chem. Soc. 1932, 54, 1630; Easton, C.; Harper, J. B. Tetrahedron Lett.
1998, 39, 5269.
´
´
10. (a) Siracusa, L.; Hurley, F. M.; Dresen, S.; Lawless, L. J.; Perez-Payan, M. N.; Davis,
A. P. Org. Lett. 2002, 4, 4639; (b) Baragan˜a, B.; Blackburn, A. G.; Breccia, P.; Davis,
´
A. P.; de Mendoza, J.; Padron-Carrillo, J. M.; Prados, P.; Riedner, J.; de Vries, J. G.
Chem.dEur. J. 2002, 8, 2931; (c) Lawless, L. J.; Blackburn, A. G.; Ayling, A. J.;
Pe´rez-Paya´n, M. N.; Davis, A. P. J. Chem. Soc., Perkin Trans. 1 2001, 1329; (d) Davis,
A. P.; Lawless, L. J. Chem. Commun. 1999, 9.
20. Gannett, P. M.; Miller, L.; Daft, J.; Locuson, C.; Tracy, T. S. J. Labelled Compd.
Radiopharm. 2007, 50, 1272.
21. Purchased from NovaBiochem (catalogue number 01-64-0144). The batch
employed for the synthesis of library 13 presented a loading equal to 0.
11. For other tripodal scaffolds see: (a) Sonntag, L. S.; Ivan, S.; Langer, M.; Conza, M. A.;
Wennemers, H. Synlett 2004,1270; (b) Opatz, T.; Liskamp, R. M. J. Org. Lett. 2001, 3,
44 g molꢁ1
.