Synthesis of a family of cyclic peptide-based anion receptors

Article abstract of DOI:10.1039/c0ob01072c

We report here the design and synthesis of a family of novel backbone modified cyclic peptides, bearing dipicolylamine side chains for metal complexation and subsequent anion binding studies. Two approaches to the cyclic peptides were investigated. Initia

Full text of DOI:10.1039/c0ob01072c

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 3471
www.rsc.org/obc
PAPER
Synthesis of a family of cyclic peptide-based anion receptors†
Stephen J. Butler and Katrina A. Jolliffe*
Received 24th November 2010, Accepted 16th February 2011
DOI: 10.1039/c0ob01072c
We report here the design and synthesis of a family of novel backbone modified cyclic peptides, bearing
dipicolylamine side chains for metal complexation and subsequent anion binding studies. Two
approaches to the cyclic peptides were investigated. Initially, a stepwise approach was employed,
involving solid-phase assembly of oxazole-based building blocks, followed by solution-phase
macrolactamisation of the resulting linear precursor. The alternative strategy involved the formation of
linear bisoxazole fragments in solution-phase, followed by a cyclodimerisation reaction. The zinc(II)
complexes of these receptors bind selectively to di- and tri-phosphate ions over hydrogenphosphate.
Introduction
In recent years, the Lissoclinum family of natural heterocycle-
containing cyclic peptides has been identified as suitable scaffolds
for molecular recognition.1–6 The cyclic hexa- and octapeptides
are biologically modified to contain aromatic oxazole and thi-
azole heterocycles derived from serine, threonine and cysteine
moieties, which alternate with standard amino acid residues. The
heterocycles are linked by trans-amide bonds to form a rigid
macrocyclic structure which is almost planar, a virtue which
results from a network of bifurcated hydrogen bonds between
the heterocyclic nitrogen atoms and the amide hydrogen atoms
which line the interior of the macrocycle. Furthermore, if all amino
acid side chains are of the same stereochemical configuration,
they are presented from the same face of the cyclic peptide. The
side chains can be readily elaborated to a range of functional
groups. These attractive features have recently been utilised
in the development of molecular receptors, chiral ligands and
combinatorial libraries.4,6–11
We are particularly interested in using these scaffolds for
anion recognition. Anions such as pyrophosphate (P2O74-) and
adenosine triphosphate (ATP) are of importance in numerous
biological processes (e.g. as enzyme substrates or the products of
enzymatic reactions). Therefore, sensors capable of the selective
detection of these anions under physiological conditions have
the potential for real-time monitoring of such reactions. We
recently reported the first example of an anion receptor based
on a Lissoclinum-type cyclic octapeptide scaffold, bearing amino
acid side chains functionalised with two Zn(II)-dipicolylamine
ligands (1·Zn2, Fig. 1).9 We showed that receptor 1·Zn2 binds
to pyrophosphate (P2O74-) with high affinity (log Ka = 8.0 0.1)
and selectivity of two orders of magnitude over ATP and ADP
Fig. 1 Structures of cyclic peptide-based anion receptors 1·Zn2–6·Zn2.
under mimicked physiological conditions (aqueous solution, pH
7.4, high salt concentrations), as demonstrated by a fluorescent
indicator displacement assay. Such high binding affinity for
pyrophosphate can be ascribed to the large size of the receptor
cavity and the large distance between the binding sites, which
complement the size and shape of the large pyrophosphate anion.
This was confirmed by the synthesis of a comparatively smaller
receptor 2·Zn2,12 based on a diketopiperazine scaffold, which
displayed lower affinity (log Ka = 6.0 0.2) and selectivity between
School of Chemistry, The University of Sydney, 2006, NSW, Australia.
E-mail: kate.jolliffe@sydney.edu.au; Fax: +61 2 9351 3329; Tel: +61 2
9351 2297
† Electronic supplementary information (ESI) available: 1H and 13C NMR
spectroscopic data for new compounds. See DOI: 10.1039/c0ob01072c
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3471–3483 | 3471
©

View Article Online
Scheme 1 Synthesis of oxazole-based amino acids 16–18. Conditions: (i) Deoxo-Fluor, CH2Cl2, -20 ◦C then NaHCO3, H2O; (ii) DBU, BrCCl3, CH2Cl2,
0
◦C; (iii) H2 (1 atm.), Pd/C, MeOH, THF; (iv) Fmoc-OSu, NaHCO3, H2O, 1,4-dioxane; (v) NaOH, MeOH–H2O (7 : 3 v/v); (vi) HCl, 1,4-dioxane.
di- and triphosphate oxoanions compared to the larger receptor
1·Zn2. Given that these receptors differ only in the size of the cyclic
peptide scaffold, it is clear that a larger cyclic peptide is required
to effectively accommodate pyrophosphate. Even so, a higher
degree of selectivity is required if receptor 1·Zn2 is to be used as a
bioanalytical sensing tool. For example, an effective sensor for the
real-time monitoring of the enzymatic action of DNA polymerase
must detect low concentrations of pyrophosphate byproduct, in
the presence of a large excess of the structurally similar enzyme
substrate ATP.13,14
To this end, we have now synthesised a second generation of
cyclic peptide-based anion receptors 3·Zn2–6·Zn2 (Fig. 1). The
structures of these new compounds vary in the spacing between the
two Zn(II)-Dpa binding sites and the steric bulk and functionality
of the ‘non-binding’ side chains. It was anticipated that these
structural modifications would have significant effects on the
binding affinity and selectivity for di- and triphosphate oxoanions,
allowing us to determine the optimal structure of a pyrophosphate
selective receptor.
We required sufficient quantities of the cyclic-peptide based
receptors to allow a comprehensive investigation of their anion
binding capabilities. Therefore, several synthetic approaches to
these compounds were explored. The first route involves the
macrolactamisation of a linear precursor, which is prepared in
a traditional stepwise manner by coupling together oxazole-
based building blocks. The second approach expands on the
success of cyclooligomerisation procedures,10,15–17 and involves
the cyclodimerisation of a linear peptide, which is prepared by
coupling together two oxazole-based amino acids. We report
here the synthesis of receptor family 1 and 3–6 via these two
approaches.
thereby defining the sequence and position of the binding sites
relative to the ‘non-binding’ amino-acid side chains. This stepwise
approach was also utilised for the synthesis of the C2-symmetric
receptor 3; however, this resulted in an unacceptable overall yield.
Therefore, we investigated alternative routes for the preparation
of C2-symmetric cyclic peptides 1, 3 and 4, which involved fewer
synthetic steps (see below).
We chose to prepare receptors 5 and 6 using solid-phase peptide
synthesis (SPPS), due to its operational simplicity. Kelly and
coworkers have previously utilised the solid-phase strategy in their
total syntheses of several naturally occurring azole-modified cyclic
peptides.18,19 We proposed that similar methodology would be
applicable in the solid-phase assembly of the linear precursors
to receptors 5 and 6.
We first prepared dipeptides 7 and 8 (Scheme 1) in good
yields by coupling Cbz-leucine or Cbz-phenylalanine with ser-
ine benzyl ester utilising 1-hydroxybenzotriazole (HOBt) and
O-(benzotriazol-1-yl)-N,N,N¢,N¢-tetramethyluronium hexafluo-
rophosphate (HBTU) as the coupling reagents. Likewise, dipeptide
9 was synthesised by coupling the orthogonally protected Boc-
Orn(Cbz)-OH with serine methyl ester under standard peptide
coupling conditions. The dipeptides were then converted to
the corresponding oxazoles 10–12 in high yields using a well-
established two-step procedure,20 involving treatment with Deoxo-
Fluor reagent followed by oxidation of the intermediate oxazolines
using bromotrichloromethane and 1,8-diazabicyclol[5.4.0]undec-
7-ene (DBU).
With oxazoles 10–12 in hand, we turned our attention to
installing the appropriate functionalisation for use in Fmoc-
strategy SPPS. Thus, the N- and C-terminal protecting groups
of 10 and 11 were removed in a single synthetic step by catalytic
hydrogenolysis, to provide the fully-deprotected oxazole 13 and
the known oxazole 1421 in quantitative yield. Saponification of the
C-terminal methyl ester of 12 using sodium hydroxide, followed
by removal of the N-terminal Boc protecting group using 4 M
hydrochloric acid in dioxane, gave the hydrochloride salt 15 in
excellent yield. Subsequent treatment of amines 13–15 with Fmoc-
succinimide afforded the desired Fmoc-protected oxazoles 16–18
in good yields, ready for incorporation into SPPS.
Results and discussion
Macrolactamisation approach
The unsymmetrical cyclic peptide scaffolds of receptors 5 and 6
were synthesised by the macrolactamisation route, as this enabled
the side chain functional groups to be selectively addressed,
3472 | Org. Biomol. Chem., 2011, 9, 3471–3483
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Conditions for the macrolactamisation of linear peptides 19–21
Linear peptide
Coupling reagenta
Concentration (mol L-1)
Solvent
Product
Yield (%)
19
19
19
19
19
20
21
FDPP
FDPP
FDPP
FDPP
PyBOP
DMTMM
FDPP
0.005
0.01
0.05
0.05
0.05
0.05
0.05
DMF
DMF
DMF
CH3CN
DMF
DMF
DMF
22
22
22
22
22
23
24a
34
35
48
23
40
51
38b
a Abbreviations for coupling reagent names: FDPP: pentafluorophenyl diphenylphosphinate; PyBOP: benztriazol-1-yl-oxytrispyrrolidino-phosphonium
hexafluorophosphate; DMTMM: 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium tetrafluoroborate. b Linear precursor was cyclised without
purification.
Scheme 2 Macrolactamisation of linear precursors 19–21. General conditions: (i) Coupling reagent (3 equiv.), Hu¨nig’s base, DMF, 72 h. See Table 1 for
cyclisation yields.
The linear precursors to cyclic peptides 3,
5
and
6
at room temperature, with PyBOP giving slightly lower cyclisation
yields than either of these reagents under identical conditions.
Under these conditions the desired cyclic peptides 22, 23 and 24a
were constructed using Fmoc-strategy SPPS, utilising 2-(1H-
7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluoro-
phosphate methanaminium (HATU) as the coupling reagent. The
acid-labile 2-chlorotrityl chloride resin was chosen as the solid
support to enable cleavage of the linear peptides from the resin with
the side chain protecting groups intact. Due to the limited supply
of the oxazole-based amino acids 16–18, a manual SPPS procedure
was devised that enabled complete couplings with the use of just 1.5
equivalents of each amino acid building block. This was achieved
by using two couplings (1 equiv. of building block followed by
a further 0.5 equiv.) and extended reaction times of 24 h. The
side chain-protected peptides were then cleaved from the resin
using a mixture of hexafluoroisopropanol and dichloromethane
(1 : 4 v/v). LCMS analyses of the final cleaved materials indicated
formation of the desired linear precursors 19–21 (Scheme 2)
in approximately 70–80% yield. However, purification of linear
precursors 19 and 20 by preparative RP-HPLC was problematic
due to the insolubility of the crude materials in the required
solvent system (water–acetonitrile). This resulted in precipitation
of the crude linear peptides on the column and subsequent loss of
material. As such, the linear peptides 19 and 20 were isolated in
much lower yields (51% and 27% respectively) than expected from
the LCMS analyses of the crude peptides. To avoid further loss of
material, 21 was cyclised without further purification. A range of
conditions for the macrolactamisation of linear precursors 19–21
was explored (Table 1) and the optimum conditions were found
to be stirring a 0.05 M solution of linear peptide in DMF in the
presence of either FDPP or DMTMM coupling reagent for 3 days
were formed as the major products in 38–51% yields. 1H and 13
C
NMR spectroscopy of cyclic peptide 24a indicated C2-symmetry
as expected, whereas the non-symmetrical nature of cyclic peptides
22 and 23 resulted in more complex spectra (see supplementary
information†).
Cyclodimerisation approach
The solid-phase assembly of cyclic peptides 22, 23 and 24a
was successful, enabling the synthesis of scaffolds with different
arrangements of side chain functional groups. However, the
cyclisation yields were lower than expected, and because of the
small reaction scales permitted by SPPS, we were unable to
prepare the cyclic peptide scaffolds in large quantities. Therefore,
we investigated a more convergent approach to the C2-symmetric
receptors 1, 3 and 4, which involved cyclodimerisation of a
fully-deprotected linear peptide, which is in turn prepared by
coupling two oxazole-based building blocks in solution-phase.
This approach has previously been shown to afford azole-modified
cyclic octapeptides in significantly improved yields and higher
purity compared to the macrolactamisation approach.15,22,23
Synthesis of the required linear precursors to cyclodimerisation
commenced with acidic removal of the N-terminal Boc protecting
group of oxazole 12 to afford the hydrochloride salt 25, which
was then coupled to the free acid building blocks 16 or 17 in
solution using HBTU/HOBt as the coupling reagents to afford
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3471–3483 | 3473
©

View Article Online
Scheme 3 Solution-phase synthesis of C2-symmetric cyclic peptides 24a, 33a and 34a. Conditions: (i) For 26: 16, 25, HBTU (1.2 equiv), HOBt (1.2 equiv),
Hu¨nig’s base, DMF; For 27: 17, 25, HBTU (1.2 equiv), HOBt (1.2 equiv), Hu¨nig’s base, DMF; For 32: 18, 31, HBTU (1.2 equiv), HOBt (1.2 equiv),
Hu¨nig’s base, DMF; (ii) NaOH, MeOH, THF, H2O; (iii) FDPP, Hu¨nig’s base, DMF (see Table 2 for cyclisation yields).
the protected linear peptides 26 and 27 in good yields (Scheme 3).
Simultaneous removal of the N-terminal Fmoc and C-terminal
methyl ester protecting groups of 26 and 27 using 1 M NaOH
provided the linear precursors 28 and 29 in 94% and 89% yields,
respectively.
Using the same methodology, the alanine-derived linear precur-
sor 30 was synthesised to enable the construction of the previously
reported receptor 1 via the cyclodimerisation approach, thereby
providing a comparison to the original stepwise synthesis.24 The
previously prepared hydrochloride salt 3124 was coupled to the
free acid 18 in solution to provide linear peptide 32 in moderate
yield. Subsequent removal of the N- and C-terminal protecting
groups under the basic conditions described above gave the linear
precursor 30 in 90% yield.
Treatment of linear peptides 28–30 with FDPP coupling reagent
in the presence of Hu¨nig’s base under high dilution (0.01–0.05 M)
in DMF resulted in the formation of the desired cyclic dimers 24a,
33a and 34a (n = 1) as the major products (26–50%) (Scheme 3).
1H and 13C spectra of the cyclic dimers 24a, 33a and 34a exhibit
the expected number of signals for compounds with C2-symmetry.
As anticipated, the cyclic trimers 24b, 33b and 34b (n = 2) were
also observed as minor products in comparatively lower yields (5–
12%), and trace amounts (< 3%) of higher cyclic oligomers were
also obtained as evidenced by mass spectra of the crude reaction
mixture.
We explored the effect of different concentrations and reaction
times on the yields of cyclic peptides 24a, 33a and 34a. Table 2
summarises the reaction conditions and corresponding yields of
Table 2 Conditions for the cyclodimerisation of linear peptides 28–30
Cyclooligomerisation products
Linear peptide
Concentration (mol L-1)
Reaction time (h)
Dimer
Yield (%)
Trimer
Yield (%)
28
28
28
28
28
29
29
30
30
0.01
0.025
0.05
0.05
0.05
0.025
0.05
0.025
0.05
72
72
72
48
24
48
48
48
48
24a
24a
24a
24a
24a
33a
33a
34a
34a
26
33
44
42–48
39
36
40–50
28
24b
24b
24b
24b
24b
33b
33b
34b
34b
8
8
5
5
5
n/ia
n/ia
n/ia
12
36
a n/i: not isolated
3474 | Org. Biomol. Chem., 2011, 9, 3471–3483
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme 4 Side chain functionalisation of cyclic peptides 22, 23, 24a, 33a and 34a. Yields are reported over 3 steps. Conditions: (i) 45% HBr in AcOH;
(ii) 0.3 M NaOH; (iii) 2-pyridinecarboxaldehyde, NaHB(OAc)3, DMF.
the desired cyclic dimers and undesired cyclic trimers obtained
after purification by preparative reverse-phase HPLC.
Having prepared sufficient quantities of the cyclic peptide
scaffolds 22, 23, 24a, 33a and 34a, our attention turned to their
functionalisation with the dipicolylamino groups required for
anion binding studies. Cleavage of the Cbz protecting groups of
cyclic peptides 22, 23, 24a, 33a and 34a using a solution of 45%
hydrogen bromide in acetic acid gave the corresponding amines
as the dihydrobromide salts in quantitative yields (Scheme 4).
The hydrobromide salts were neutralised by a basic work-up
procedure using 0.3 M sodium hydroxide solution to give the
free amines, which were then subjected to reductive amination
with 2-pyridinecarboxaldehyde using our previously established
conditions,9 to provide the desired Dpa-functionalised cyclic
peptides 1 and 3–6 in good to excellent yields.
Preliminary anion binding studies for the bis-Zn(II) complexes
of compounds 1 and 3–6 [prepared by addition of 2 equiv. of
Zn(NO3)2 to each cyclic peptide]9 were undertaken in aqueous
solution (HEPES buffer, 5 mM, pH 7.4) at 25 ◦C, using our
previously reported indicator displacement assay26 with the fluo-
rescent coumarin derivative 35.9,12,27,28 Apparent stability constants
of 1·Zn2 and 3–6·Zn2 for 35 were of similar magnitude, ranging
from log Kin 6.2–6.9 ( 0.1), with the leucine derivatives 3·Zn2 and
5·Zn2 having the lowest and highest affinity for 35, respectively.
Titration of 1 : 1 solutions of the indicator: receptor complexes
(10 mM) with aliquots of the sodium salts of pyrophosphate,
ATP and ADP resulted in increases in fluorescence indicating
displacement of the indicator, while hydrogenphosphate was not
able to displace the indicator from any of the receptors to an
appreciable extent (<10% fluorescence recovery upon addition of
10 equiv hydrogenphosphate) indicating the selectivity of these
receptors for di- and tri-phosphate anions over monophosphate
anions. The receptors showed some differences in their selectivity
towards PPi, ATP and ADP. Notably, while 1·Zn2 displayed
selectivity for PPi (log Ka 7.2 0.1) over ATP and ADP (log Ka
6.5 0.1 and 6.3 0.1, respectively), increasing the steric bulk of
the non-binding side chains resulted in loss of this selectivity, with
the leucine derived receptor 3·Zn2 displaying similar affinity for
all three anions [log Ka 7.4–7.6 ( 0.2)]. Increased discrimination
between PPi and ATP or ADP was observed when the binding
sites were moved closer together. For example, the phenylalanine
Initially, a 0.01 M solution of linear peptide 28 was treated
with three equivalents of FDPP in the presence of Hu¨nig’s
base and the reaction mixture was stirred at room temperature
for 72 h, affording the cyclic dimer 24a and cyclic trimer 24b
in 26% and 8% yields, respectively. Significant improvement in
the yield of 24a was achieved using a higher concentration of
0.05 M. Additionally, it was found that a greater ratio of the
desired dimer to the undesired trimer (9 : 1) was obtained using
a concentration of 0.05 M, compared to when 0.025 M (4 : 1)
and 0.01 M (3 : 1) solutions of the linear peptide 28 were used. A
reduced reaction time of 48 h resulted in formation of 24a in
the highest yields obtained (42–48%). A shorter reaction time
of 24 h resulted in a slightly lower yield of the cyclic dimer
(39%), suggesting that the cyclodimerisation had not reached
completion.
The optimum conditions for cyclodimerisation were found to be
stirring a 0.05 M solution of linear peptide in DMF using FDPP
activation at room temperature for 48 h. Using these conditions,
cyclic dimers 33a and 34a were formed in 50% and 36% yields,
respectively. The cyclodimerisation of linear peptide 29 afforded
cyclic products in the ratio 8 : 1 (dimer/trimer), as determined
by LCMS analysis. As such, insufficient amounts of the cyclic
trimer 33b were isolated for its characterisation. In contrast, the
cyclisation of linear fragment 30 gave a lower ratio of the desired
cyclic dimer 34a to the cyclic trimer 34b (3 : 1). Therefore, the
isolation of the cyclic trimer was possible. In both cases, the use of
higher dilution (0.025 M) resulted in slightly diminished yields of
the desired cyclic dimers 33a and 34a.
Thus, an efficient route to the C2-symmetric cyclic peptides 24a,
33a and 34a has been developed. The cyclodimerisation reaction
was optimised to provide cyclic peptides 24a, 33a and 34a in 32%,
30% and 16% overall yields, starting from amino acid building
block 25, and in the case of scaffold 34a, building block 31.
In comparison, the overall yield of cyclic peptide 24a via the
macrolactamisation route amounts to only 17%, indicating that
cyclooligomerisation with a lower number of steps in the linear
sequence is the preferred route.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3471–3483 | 3475
©

View Article Online
derived receptor 4·Zn2 binds PPi, ATP and ADP with log Kas of
8.4, 8.6 and 7.5 ( 0.2), respectively, whereas 6·Zn2, which bears the
same side chains but with the two DPA functionalised sidechains
at proximal rather than distal positions as in 4·Zn2, binds PPi (log
Ka 7.9 0.1) with greater affinity than either ATP (log Ka 7.2
0.1) or ADP (log Ka 7.0 0.1).
Fourier Transform Ion Cyclotron Resonance mass spectrometer
(FT-ICR) with an analytical ESI source, operating at 4.7 T.
Analytical thin layer chromatography (TLC) was performed using
pre-coated silica gel plates (Merck Kieselgel 60 F254). Preparative
flash chromatography was carried out using Merck Kieselgel Silica
Gel 60 (particle size 0.040–0.065 mm) with the indicated ratio of
solvents (volume/volume) which were mixed as specified. Liquid
chromatography mass spectrometry (LCMS) was performed on
a Thermo Separation Products: Spectra System using a P400
Pump and a UV6000LP photodiode array detector. Separation
was achieved using a Phenomenex Jupiter column (5 mm, 150 ¥
2.1 mm ID) which was coupled to a Thermoquest Finnigan
LCQ Deca mass spectrometer (ESI). Flow rate was maintained
at 0.2 mL min-1 over a linear gradient from 0% to 100% solvent
B (solvent A: 100 : 0.1 v/v Milli-Q water/formic acid, solvent B:
100 : 0.1 v/v acetonitrile/formic acid) over 30 min. Preparative
RP-HPLC was performed on a Waters 600E multisolvent delivery
system with a Waters U6K injector, Waters 490E programmable
multiwavelength detector, Waters busSAT/IN module and Waters
Empower 2 software. Separation was achieved on a SunfireTM
PrepC18 OBDTM column (5 mm, 150 ¥ 19 mm ID). Compounds
were eluted at a flow rate maintained at 7.0 mL min-1 over the
stated linear gradient, using solvent A: 100 : 0.05 v/v Milli-Q
water/TFA, and solvent B: 100 : 0.05 v/v acetonitrile/TFA, unless
otherwise stated. Dichloromethane, methanol and acetonitrile
were distilled from calcium hydride. Tetrahydrofuran (THF) was
dried over sodium wire. HPLC grade N,N-dimethylformamide
was obtained from LabScan and used without further purification.
All other reagents were commercially available and were used as
supplied.
Conclusions
In summary, two viable approaches to the synthesis of Lissoclinum-
type cyclic peptide-based anion receptors 1 and 3–6 were imple-
mented: the stepwise macrolactamisation pathway was useful for
the preparation of the non-C2 symmetric cyclic peptides 5 and
6 bearing different side chain functionalities, whereas the more
convergent cyclodimerisation pathway was preferreed to provide
access to substantial quantities of C2-symmetric cyclic peptides 1,
3 and 4. Through the combined use of these approaches, we have
synthesised sufficient amounts of compounds 1 and 3–6 to allow
a comprehensive investigation of the anion binding capabilities
of their bis-zinc(II) complexes 1·Zn2 and 3·Zn2–6·Zn2 (Fig. 1) in
aqueous solution. Preliminary studies indicated that all of the
receptors bind di- and tri-phosphate anions with selectivity over
hydrogenphosphate and that the ‘non-binding’ side chains have a
significant influence on the ability of the receptors to discriminate
between di- and tri-phosphate derivatives. The results of more
extensive anion binding studies will be reported in due course.
General procedures for solid-phase peptide synthesis
1. Loading of amino acid onto 2-chlorotrityl chloride resin.
Under an atmosphere of nitrogen, 2-chlorotrityl chloride resin
(resin capacity 1.01–1.4 mmol g-1) was swollen in a sinter-fitted
syringe in anhydrous dichloromethane for 1 h. The resin was
drained and treated with a solution of Fmoc-protected amino
acid (1.5 equiv. relative to resin capacity) and Hu¨nig’s base (5
equiv. relative to resin capacity) in anhydrous dichloromethane–
DMF (9 : 1 v/v, 0.1 M) and the resulting suspension was gently
agitated at rt for 24 h under an atmosphere of nitrogen. At this time
the resin was drained and then treated with a mixture of methanol-
Hu¨nig’s base-dichloromethane (2 : 1 : 17 v/v/v, 2 ¥ 5 mL ¥ 15 min).
The resin was then washed sequentially with DMF (2 ¥ 10 mL),
dichloromethane (3 ¥ 10 mL) and diethyl ether (3 ¥ 10 mL). The
residual solvent was removed under reduced pressure.
Experimental
General procedures
Melting points were obtained using a Stanford Research Systems
Optimelt melting point apparatus and are uncorrected. Optical ro-
tations were obtained using a Perkin Elmer model 341 polarimeter
at 589 nm and 20 ◦C, using the indicated spectroscopic grade
solvent. 1H nuclear magnetic resonance spectra were recorded
using a Bruker Avance DPX 400 spectrometer at a frequency
of 400.13 MHz or a Bruker Avance DPX 300 spectrometer at
a frequency of 300.13 MHz, and are reported in parts per million
(ppm) relative to the residual isotopomer with one less deuterium
than the quoted perdeuterated solvent. The data is reported as
chemical shift (d), multiplicity (br = broad, s = singlet, d =
doublet, t = triplet, q = quartet, dd = doublet of doublets, ddd =
doublet of doublet of doublets, m = multiplet), coupling constant
(J Hz) and relative integral. 13C nuclear magnetic resonance
spectra were recorded on a Bruker Avance DPX 400 spectrometer
at a frequency of 100.61 MHz, or a Bruker Avance DPX 300
spectrometer at a frequency of 75.47 MHz and are reported in
parts per million (ppm) relative to the internal perdeuterated
solvent resonance, unless otherwise stated. Low resolution mass
spectra were recorded on a Thermo Finnigan LCQ Deca Ion
Trap mass spectrometer using electrospray ionisation (ESI). High
resolution mass spectra were recorded on a Bruker BioApex
2. N-terminal Fmoc deprotection. The resin-bound peptide
was agitated in a solution of 10% piperidine in DMF (2 ¥ 5 mL ¥
15 min) and then drained and washed sequentially with DMF
(2 ¥ 5 mL), dichloromethane (3 ¥ 5 mL) and DMF (3 ¥ 5 mL).
The resulting resin-bound amine was used immediately in the next
peptide coupling step.
3. Estimation of resin loading. The drained Fmoc depro-
tection solution was diluted with a solution of 10% piperidine
in DMF so that the maximum concentration of the fulvene-
piperidine adduct was in the range of 2.5–7.5 ¥ 10-5 M. A sample
of this solution (2 ¥ 3 mL) was transferred to two matched
3476 | Org. Biomol. Chem., 2011, 9, 3471–3483
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
1 cm quartz glass cuvettes and the UV-Vis absorbance at 301 nm
was measured, using the solution of 10% piperidine in DMF as
a reference. An average of the two absorbance values was used to
calculate the loading, using e = 7800 M-1 cm-1.
of starting material), then dried (MgSO4) and concentrated under
reduced pressure to give the crude oxazoline. The crude oxazoline
(1 equiv.) was dissolved in anhydrous dichloromethane (0.15 M)
and cooled to 0 ◦C under an atmosphere of nitrogen. The solution
was treated with BrCCl3 (2 equiv.) and DBU (2 equiv.) and the
resulting mixture was stirred at rt for 7 h. The mixture was
quenched by the addition of hydrochloric acid (0.3 M, pH 6)
and extracted with chloroform-isopropanol (3 : 1 v/v, 5 mL per
mmol of starting material ¥ 3). The combined organic fractions
were washed with half-strength brine solution (10 mL per mmol
of starting material), then dried (MgSO4) and concentrated under
reduced pressure.
4. Manual SPPS peptide coupling. Under an atmosphere of
nitrogen, a solution of Fmoc-protected oxazole (1.5 equiv. relative
to loading), HATU (2 equiv. relative to peptide) and Hu¨nig’s base
(3 equiv. relative to peptide) in anhydrous DMF (0.1 M relative
to dipeptide) was added to the resin and the resulting suspension
was agitated at rt for 24 h. The resin was then washed sequentially
with DMF (2 ¥ 5 mL), dichloromethane (3 ¥ 10 mL) and DMF
(3 ¥ 5 mL).
9. N-terminal Fmoc protection. To a stirred solution of the
amine (1 equiv.) and sodium carbonate (3 equiv.) in distilled water
(0.3 M) cooled to 0 ◦C, was added a solution of Fmoc succinimide
(1.5 equiv.) in 1,4-dioxane (0.3 M). The resulting mixture was
slowly warmed to rt and stirred for 24 h. At this time the mixture
was quenched with hydrochloric acid (0.3 M, pH 5) and the organic
solvent was removed under reduced pressure. The remaining
solution was extracted with chloroform-isopropanol (3 : 1 v/v,
5 mL per mmol of starting material ¥ 3) and the combined organic
fractions were washed with half-strength brine solution (10 mL per
mmol of starting material), then dried (MgSO4) and the solvent
was removed under reduced pressure.
5. Cleavage of peptides from the resin. A mixture of
hexafluoroisopropanol-dichloromethane (1 : 4 v/v, 3 ¥ 5 mL) was
added to the resin and the resulting suspension was agitated at
rt for 15 min. The solutions were drained and combined and the
solvent was removed under reduced pressure to afford the crude
linear peptide.
6. Macrolactamisation. To a solution of the linear tetraoxa-
zole (1 equiv.) in anhydrous DMF (0.05 M) was added coupling
reagent (3 equiv.) (see Table 1 for details) and Hu¨nig’s base (5
equiv.) and the resulting mixture was stirred at rt for 72 h under an
atmosphere of nitrogen. The mixture was acidified by the addition
of hydrochloric acid (0.3 M, pH 5) and then partitioned between
water (20 mL) and chloroform-isopropanol (3 : 1 v/v, 20 mL). The
aqueous phase was extracted with chloroform-isopropanol (3 : 1
v/v, 3 ¥ 20 mL) and the combined organic fractions were dried
(MgSO4) and concentrated under reduced pressure.
10. N-terminal Fmoc and C-terminal methyl ester deprotection.
To a solution of the bisoxazole (1 equiv.) in methanol–THF (1 : 1
v/v, 0.15 M) was added a solution of NaOH (10 equiv.) in water
(0.15 M) and the resulting mixture was stirred at rt for 5 h. The
mixture was then acidified by the addition of hydrochloric acid
(0.3 M, pH 6) and the organic solvent was removed under reduced
pressure. The residue thus obtained was partitioned between
chloroform-isopropanol (3 : 1 v/v, 5 mL per mmol of starting
material) and water (5 mL per mmol of starting material) and
the aqueous phase was extracted with chloroform-isopropanol
(3 : 1 v/v, 5 mL per mmol of starting material ¥ 3). The combined
organic fractions were washed with half-strength brine solution
(10 mL per mmol of starting material), then dried (MgSO4) and
concentrated under reduced pressure.
General procedures for solution-phase synthesis
7. Peptide coupling. To a stirred solution of the acid (1 equiv.)
and the amine hydrochloride salt (1 equiv.) in anhydrous DMF
(0.15 M) was added HBTU (1.2 equiv.), HOBt (1.2 equiv.), then
Hu¨nig’s base (3.2 equiv.) under an atmosphere of nitrogen. The
resulting mixture was stirred at rt for 22 h. At this time, the
mixture was concentrated under reduced pressure to half the initial
volume and then partitioned between sat. aq. NH4Cl solution
(5 mL per mmol of starting material) and chloroform-isopropanol
(5 mL per mmol of starting material). The organic phase was
extracted with water (5 mL per mmol of starting material ¥ 3)
and the combined aqueous fractions were back-extracted with
chloroform-isopropanol (3 : 1 v/v, 10 mL per mmol of starting
material ¥ 2). The combined organic fractions were washed with
half-strength brine solution (10 mL per mmol of starting material),
then dried (MgSO4) and the solvent was removed under reduced
pressure.
11. Cyclodimerisation. To a stirred solution of the bisoxazole
(1 equiv.) in anhydrous DMF (0.05 M) was added FDPP (3 equiv.)
and Hu¨nig’s base (5 equiv.) under an atmosphere of nitrogen. The
resulting mixture was stirred at rt for 48 h after which time analysis
by LCMS revealed complete conversion of starting material. The
mixture was concentrated under reduced pressure to a quarter
of the initial volume and then partitioned between chloroform-
isopropanol (3 : 1 v/v, 5 mL per mmol of starting material) and
hydrochloric acid (0.3 M, pH 3). The aqueous phase was extracted
with chloroform-isopropanol (3 : 1 v/v, 5 mL per mmol of starting
material ¥ 3) and the combined organic fractions were washed with
half-strength brine solution (10 mL per mmol of starting material),
then dried (MgSO4) and concentrated under reduced pressure.
8. Oxazole formation. Under an atmosphere of nitrogen,
Deoxo-fluor (1.2 equiv.) was added dropwise to a stirred solution
of the dipeptide (1 equiv.) in anhydrous dichloromethane (0.10 M)
◦
cooled to -20 C (EtOH/ice/CO2 chips). The resulting mixture
was stirred at -20 ◦C for 1 h, then slowly warmed to rt and stirred
for a further 6 h. At this time the mixture was cooled to -5 ◦C
and quenched with sat. aq. NaHCO3 solution (5 mL per mmol
of starting material) and then extracted with chloroform (5 mL
per mmol of starting material ¥ 3). The combined organic fractions
were washed with half-strength brine solution (10 mL per mmol
12. Functionalisation of cyclic peptide side chains.
Step 1. Under an atmosphere of nitrogen, a solution of
hydrogen bromide in acetic acid (33% v/v, 2.0 mL) was added
to the carbamate (1 equiv.) and the resulting mixture was stirred
at rt for 24 h. Anhydrous ether (25 mL) was added to give a
cream coloured precipitate which was condensed by centrifuge.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3471–3483 | 3477
©

View Article Online
Subsequent trituration of the precipitate with anhydrous ether
(10 ¥ 25 mL) and removal of the final clear ethereal layer under
reduced pressure, gave the dihydrobromide salt. This material was
partitioned between chloroform-isopropanol (3 : 1 v/v, 20 mL)
and NaOH (0.3 M, 20 mL) and the aqueous phase was extracted
with chloroform-isopropanol (3 : 1 v/v, 4 ¥ 20 mL). The combined
organic fractions were dried (MgSO4) and the solvent was removed
under reduced pressure to give the free diamine.
chromatography (silica gel: CHCl3–MeOH, 95 : 5 v/v) gave the
desired dipeptide 9 (11.1 g, 87%) as a colourless solid. m.p. 67–
68 ◦C; [a]D20 = -2.09 (c 1.1, MeOH); 1H NMR (300 MHz, CDCl3):
d 1.42 (s, 9H), 1.62 (m, 3H), 1.86 (m, 1H), 3.16 (m, 1H), 3.34 (m,
1H), 3.42 (br s, 1H), 3.74 (s, 3H), 3.93 (m, 2H), 4.27 (br m, 1H),
4.64 (m, 1H), 5.08 (AB quartet, 2H), 5.15 (t, J = 6.6 Hz, 1H), 5.39
(d, J = 8.1 Hz, 1H), 7.33 (m, 5H), O–H not observed; 13C NMR
(75.5 MHz, CDCl3): d 26.0, 28.4, 29.9, 40.1, 52.7, 53.7, 54.9, 62.8,
66.9, 80.4, 128.2, 128.6, 136.6, 156.1, 157.2, 171.0, 172.7, one signal
obscured or overlapping; HRMS (ESI) calcd. for C22H33N3O8Na
[M + Na]+ 490.2160, found 490.2157.
Step 2. To a stirred solution of the free diamine in anhydrous
DMF (2.0 mL) was added 2-pyridinecarboxaldehyde (20 equiv.)
and sodium triacetoxyborohydride (25 equiv.) under an atmo-
sphere of nitrogen. The resulting mixture was deoxygenated and
refilled with nitrogen three times, then slowly warmed to 35 ◦C
and stirred for 72 h. The mixture was concentrated to almost
dryness and then partitioned between chloroform-isopropanol
(3 : 1 v/v, 20 mL) and NaOH (0.3 M, pH 8). The aqueous phase was
extracted with chloroform-isopropanol (3 : 1 v/v, 4 ¥ 20 mL) and
the combined organic fractions were washed with half-strength
brine solution (40 mL), then dried (MgSO4) and the solvent was
removed under reduced pressure.
Cbz-Leu-Oxz(Ser)-OBn (10). The dipeptide
7 (9.73 g,
22.0 mmol) was converted to the crude oxazole 10 according to
general procedure 8. The crude material was purified by flash
chromatography (silica gel; hexane–EtOAc, 4 : 1 v/v) to give the
desired oxazole 10 (6.94 g, 75% over two steps) as a colourless solid.
m.p. 75–77 ◦C; [a]D20 = -4.35 (c 1.0, CHCl3); 1H NMR (400 MHz,
CD3OD): d 0.93 (d, J = 6.6 Hz, 3H), 0.96 (d, J = 6.6 Hz, 3H),
1.58–1.84 (m, 3H), 4.92 (dd, J = 6.6, 8.8 Hz, 1H), 5.08 (AB quartet,
2H), 5.34 (s, 2H), 7.19–7.48 (m, 10H), 8.48 (s, 1H), N–H signal not
observed; 13C NMR (100.6 MHz, CD3OD): d 22.0, 23.1, 25.8, 42.9,
67.8, 128.8, 129.0, 129.4(5), 129.4(7), 129.6, 134.0, 137.1, 138.1,
146.2, 158.3, 162.3, 167.4, three signals obscured or overlapping;
HRMS (ESI) calcd. for C24H26N2O5Na [M + Na]+ 445.1734, found
445.1733.
Synthesis
Cbz-Leu-Ser-OBn (7). The acid Cbz-Leu-OH (15.0 g,
56.5 mmol) and the hydrochloride salt of H-Ser-OBn (13.1 g,
56.5 mmol) were coupled in solution according to general proce-
dure 7. The crude material was purified by flash chromatography
(dry loaded onto silica gel; eluting first with EtOAc–hexane, 3 : 1
v/v, then neat EtOAc) to give the desired dipeptide 7 (16.0 g,
Cbz-Phe-Oxz(Ser)-OBn (11). The dipeptide
8 (4.26 g,
8.94 mmol) was converted to the crude oxazole 11 according to
general procedure 8. Subjection of the crude material to flash
chromatography (dissolved in CHCl3 and loaded to silica gel;
◦
64%) as a colourless solid. m.p. 122–124 C; [a]2D0 = -1.50 (c 1.0,
CHCl3); 1H NMR (400 MHz, CDCl3): d 0.91 (d, J = 6.4 Hz, 6H),
1.52 (m, 1H), 1.59–1.72 (m, 2H), 2.48 (br s, 1H), 3.93 (m, 2H),
4.24 (m, 1H), 4.68 (m, 1H), 5.02 (d, J = 12.2 Hz, 1H), 5.10 (d,
J = 12.2 Hz, 1H), 5.19 (s, 2H), 5.42 (d, J = 8.1 Hz, 1H), 7.07
(d, J = 7.8 Hz, 1H), 7.29–7.37 (m, 10H); 13C NMR (75.5 MHz,
CDCl3): d 23.9, 24.8, 26.6, 43.3, 55.7, 56.8, 64.7, 69.2, 69.5, 130.0,
130.1(8), 130.2(1), 130.5(0), 130.5(2), 130.6, 137.1, 138.0, 158.5,
172.1, 174.6; HRMS (ESI) calcd. for C24H30N2O6Na [M + Na]+
465.1996, found 465.1989.
hexane–EtOAc, 2 : 1 v/v) gave the desired oxazole 11 (2.95 g, 72%
◦
over two steps) as a colourless solid. m.p. 117–119 C; [a]D20
=
1
-2.75 (c 0.8, CHCl3); H NMR (400 MHz, CDCl3): d 3.25 (m,
2H), 5.09 (AB quartet, 2H), 5.28 (m, 1H), 5.36 (s, 2H), 5.55 (d, J =
8.8 Hz, 1H), 7.03 (m, 2H), 7.23 (m, 2H), 7.28–7.46 (m, 11H), 8.11
(s, 1H); 13C NMR (75.5 MHz, CDCl3): d 40.3, 50.8, 67.0, 67.3,
127.3, 128.2, 128.3, 128.6(5), 128.6(9), 128.8(0), 128.8(1), 129.4,
133.4, 135.4, 135.5, 136.2, 144.2, 155.6, 160.9, 164.5, one signal
obscured or overlapping; HRMS (ESI) calcd. for C27H28N2O6Na
[M + Na]+ 479.1583, found 479.1578.
Cbz-Phe-Ser-OBn (8). The acid Cbz-Phe-OH (10.0 g,
33.4 mmol) and the hydrochloride salt of H-Ser-OBn (7.74
g, 33.4 mmol) were coupled in solution according to general
procedure 7. The crude material was triturated with ethyl acetate
(3 ¥ 200 mL) to give the desir◦ed dipeptide 8 (12.0 g,◦75%) as a
Boc-Orn(Cbz)-Oxz(Ser)-OMe (12). The dipeptide 9 (12.5 g,
26.7 mmol) was converted to the crude oxazole 12 according to
general procedure 8. The crude material was purified by flash
chromatography (silica gel; eluting first with EtOAc–hexane, 1 : 1
v/v, then EtOAc–hexane, 2 : 3 v/v) to give the desired oxazole 12
(8.63 g, 69% over two steps) as a colourless solid. m.p. 83–84 ◦C;
colourless solid. m.p. 115–117 C [lit.25 m.p. 115–118 C]; [a]2D0
=
+3.20 (c 1.0, CHCl3); 1H NMR (300 MHz, CD3OD): d 2.81 (dd,
J = 9.5, 13.9 Hz, 1H), 3.12 (dd, J = 4.8, 13.9 Hz, 1H), 3.82 (dd,
J = 4.4, 11.3 Hz, 1H), 3.93 (dd, J = 4.4, 11.3 Hz, 1H), 4.47 (dd,
J = 4.8, 9.5 Hz, 1H), 4.57 (m, 1H), 4.99 (AB quartet, 2H), 5.18 (s,
1
[a]2D0 = -9.40 (c 1.0, MeOH); H NMR (300 MHz, CD3OD): d
1.43 (s, 9H), 1.47–1.67 (m, 2H), 1.75–2.02 (m, 2H), 3.15 (m, 2H),
3.87 (s, 3H), 4.79 (m, 1H), 5.05 (s, 2H), 7.02 (t, J = 6.6 Hz, 1H),
7.27–7.34 (m, 4H), 8.46 (s, 1H), N–H signal not observed; 13C
NMR (75.5 MHz, CD3OD): d 27.2, 28.7, 31.4, 41.2, 41.3, 52.5,
67.4, 80.8, 128.8, 128.9, 129.4, 133.9, 138.4, 146.0, 157.7, 158.9,
163.0, 167.2; HRMS (ESI) calcd. for C22H29N3O7Na [M + Na]+
470.4713, found 470.1897.
2H), 7.13–7.42 (m, 15H), N–H and O–H signals not observed; 13
C
NMR (75.5 MHz, CD3OD): d 39.1, 56.3, 57.6, 62.8, 67.6, 68.1,
127.7, 128.6, 128.9, 129.2, 129.3, 129.4(0), 129.4(2), 129.6, 130.3,
137.1, 138.1, 138.5, 158.3, 171.4, 174.3; HRMS (ESI) calcd. for
C27H28N2O6Na [M + Na]+ 499.1844, found 499.1840.
Boc-Orn(Cbz)-Ser-OMe (9). The acid Boc-Orn(Cbz)-OH
(10.0 g, 27.3 mmol) and the hydrochloride salt of H-Ser-OMe
(4.25 g, 27.3 mmol) were coupled in solution according to
general procedure 7. Subjection of the crude material to flash
H2N-Leu-Oxz(Ser)-OH (13). To a solution of the oxazole 10
(6.12 g, 14.5 mmol) in anhydrous methanol (145 mL) was added
Pd/C catalyst (10 wt%) and the mixture was stirred under an
atmosphere of hydrogen for 20 h, after which time a colourless
3478 | Org. Biomol. Chem., 2011, 9, 3471–3483
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
precipitate had formed. The reaction mixture was filtered through
a pad of celite and washed with hot methanol (3 ¥ 100 mL)
followed by hot water-aq. ammonia (95 : 5 v/v, 3 ¥ 200 mL). The
filtrates were collected and the solvent was removed under reduced
pressure to give the fully-deprotected oxazole 13 (2.85 g, quant.)
as a colourless solid. m.p. 184–185 ◦C; 1H NMR (300 MHz,
CF3CO2D): d 0.91 (d, J = 6.5 Hz, 3H), 0.95 (d, J = 6.5 Hz, 3H),
1.53 (m, 1H), 2.01–2.15 (m, 2H), 5.03 (dd, J = 6.5, 9.9 Hz, 1H),
8.46, (s, 1H), N–H and O–H signals not observed; 13C NMR
(75.5 MHz, CF3CO2D): d 22.2, 23.2, 27.1, 42.9, 51.4, 135.0, 149.7,
166.9, one signal obscured or overlapping; HRMS (ESI) calcd. for
C9H13N2O3 [M - H]- 197.0932, found 197.0928.
chromatography (silica gel; eluting first with hexane–EtOAc, 5 : 1
v/v, then hexane–EtOAc, 2 : 1 v/v) to give the desired Fmoc-
protected oxazole 16 (7.04 g, 71%) as a colourless solid. m.p.
◦
134–136 C; [a]2D0 = -11.2 (c 1.0, MeOH); H NMR (400 MHz,
CD3OD): d 0.93 (d, J = 6.6 Hz, 3H), 0.96 (d, J = 6.6 Hz, 3H), 1.50–
1.83 (m, 3H), 4.19 (t, J = 6.9 Hz, 1H), 4.36 (dd, J = 6.9, 10.5 Hz,
1H), 4.41 (dd, J = 6.9, 10.5 Hz 1H), 4.90 (partially obscured m,
1H), 7.28 (t, J = 7.3 Hz, 2H), 7.37 (t, J = 7.6 Hz, 2H), 7.64 (dd, J =
3.4, 7.6 Hz, 2H), 7.77 (d, J = 7.3 Hz, 2H), 8.41 (s, 1H), N–H and
O–H signals not observed; 13C NMR (100.6 MHz, CD3OD): d
22.0, 23.1, 25.7, 42.8, 48.9, 67.7, 120.8, 126.1, 128.0, 128.7, 134.5,
142.5, 145.0, 145.7, 158.2, 163.8, 167.1, one signal obscured or
overlapping; HRMS (ESI) calcd. for C24H24N2O5Na [M + Na]+
443.1583, found 443.1584.
1
H2N-Phe-Oxz(Ser)-OH (14). To a solution of the oxazole 11
(5.36 g, 11.7 mmol) in anhydrous methanol–THF (2 : 1 v/v, 90 mL)
was added Pd/C catalyst (10 wt%) and the mixture was stirred
under an atmosphere of hydrogen for 20 h, after which time
a colourless precipitate had formed. The reaction mixture was
filtered through a pad of celite and washed with hot methanol
(3 ¥ 100 mL) followed by hot water-aq. ammonia (98 : 2 v/v, 3 ¥
100 mL). The filtrates were collected and the solvent was removed
under reduced pressure to give the fully-deprotected oxazole 14
(2.72 g, quant.) as a colourless solid. m.p. 245–246 ◦C; 1H NMR
(400 MHz, CD3OD): d 3.34 (dd, J = 6.9, 13.9 Hz, 1H), 3.39 (dd,
J = 7.8, 13.9 Hz, 1H), 4.87 (dd, J = 6.9, 7.8 Hz, 1H), 7.15 (m,
2H), 7.24–7.33 (m, 3H), 8.49 (s, 1H), N–H and O–H signals not
observed; 13C NMR (75.5 MHz, CF3CO2D): d 39.5, 53.5, 130.9,
131.0, 131.6, 133.1, 134.5, 149.4, 162.6, 166.5. NMR data is in
good agreement with that previously reported.21 HRMS (ESI)
calcd. for C12H12N2O3Na [M + Na]+ 255.0740, found 255.0743.
Fmoc-Phe-Oxz(Ser)-OH (17). Compound 14 (2.72 g,
11.7 mmol) was Fmoc-protected according to general procedure
9. Subjection of the crude material to flash chromatography (silica
gel; CHCl3–MeOH, 98 : 2 v/v) gave the desired dipeptide (Rf 0.2)
which had co-eluted with an impurity (Rf 0.3). The appropriate
fractions were concentrated under reduced pressure and triturated
with diethyl ether (3 ¥ 30 mL) to afford the desired Fmoc-protec◦ted
oxazole 17 (3.40 g, 64%) as a colourless solid. m.p. 174–176 C;
[a]2D0 = -11.8 (c 0.8, CHCl3); 1H NMR (400 MHz, CDCl3): d 3.15
(dd, J = 7.2, 13.9 Hz, 1H), 3.21 (dd, J = 6.9, 13.9 Hz, 1H), 4.12 (t,
J = 6.9 Hz, 1H), 4.26 (dd, J = 6.9, 10.8 Hz, 1H), 4.36 (dd, J = 6.9,
10.8 Hz, 1H), 5.16 (dd, J = 6.9 Hz, 1H), 7.00 (d, J = 6.9 Hz, 2H),
7.14–7.28 (m, 5H), 7.34 (t, J = 7.3 Hz, 2H), 7.51 (t, J = 7.6 Hz,
2H), 7.70 (d, J = 7.3 Hz, 2H), 8.10 (s, 1H), N–H and O–H signals
not observed; 13C NMR (100.6 MHz, CDCl3): d 39.9, 47.1, 50.6,
67.0, 120.0, 125.1(0), 125.1(3), 127.1, 127.7, 128.6, 129.2, 133.6,
135.6, 141.3, 143.8, 144.1, 156.0, 162.8, 164.5; HRMS (ESI) calcd.
for C27H22N2O5Na [M + Na]+ 477.1421, found 477.1420.
HCl.H2N-Orn(Cbz)-Oxz(Ser)-OH (15). To a solution of the
oxazole 12 (1.00 g, 2.24 mmol) in methanol (16 mL) was added a
solution of NaOH (0.894 g, 22.4 mmol) in water (6 mL) and the
resulting mixture was stirred at rt for 18 h under an atmosphere
of nitrogen. The mixture was acidified (pH 6) by the addition of
hydrochloric acid (30 mL) and the organic solvent was removed
under reduced pressure. The residue thus obtained was partitioned
between water (60 mL) and chloroform-isopropanol (3 : 1 v/v,
100 mL). The aqueous phase was extracted with chloroform-
isopropanol (3 : 1 v/v, 3 ¥ 100 mL) and the combined organic
fractions were dried (MgSO4) and concentrated under reduced
pressure to afford the carboxylic acid (0.973 g, quant.) as a
colourless foam. The carboxylic acid (0.973 g, 2.24 mmol) was
dissolved in anhydrous dichloromethane (7.5 mL) and treated
with a solution of hydrochloric acid in dioxane (4 M, 8.4 mL,
33.6 mmol) under an atmosphere of nitrogen. The solution was
stirred at rt for 4 h, after which time analysis by TLC revealed
complete conversion of starting material. The reaction mixture
was concentrated under reduced pressure to give the desired
Fmoc-Orn(Cbz)-Oxz(Ser)-OH (18). Compound 15 (0.793 g,
2.24 mmol) was Fmoc-protected according to general procedure
9. The crude material thus obtained was triturated in anhydrous
diethyl ether (3 ¥ 40 mL) to afford the desired Fmoc-protected
oxazole 18 (1.05 g, 84%) as colourless solid. m.p. 131–133 ◦C;
1
[a]2D0 = -5.30 (c 1.1, MeOH); H NMR (400 MHz, DMSO-d6):
d 1.25–1.54 (m, 2H), 1.65–1.91 (m, 2H), 3.00 (m, 2H), 4.15–4.44
(m, 3H), 4.72 (m, 1H) 5.00 (m, 2H), 7.24–7.45 (m, 10H), 7.69
(m, 2H), 7.88 (m, 2H), 8.05 (m, 2H), O–H signal not observed;
13C NMR (100.6 MHz, DMSO-d6): d 25.9, 30.5, 46.7, 48.7, 65.2,
65.6, 120.1, 125.2, 125.3, 127.1, 127.6, 127.7, 128.4, 137.2, 140.1,
140.7, 143.7, 143.8, 155.9, 156.2, 163.1, 164.4, one signal obscured
or overlapping; HRMS (ESI) calcd. for C31H29N3O7Na [M + Na]+
578.1898, found 578.1914.
CF3CO2H.H2N-[Phe-Oxz(Ser)]2-[Orn(Cbz)-Oxz(Ser)]2-OH
(19). The oxazole 18 was loaded onto 2-chlorotrityl chloride
resin (0.275 g, resin capacity 1.01 mmol g-1) according to general
procedure 1 (loading 0.98 mmol g-1). The appropriate Fmoc-
protected oxazole building blocks 18 and 17 were coupled to
the resin-bound amine according to the manual SPPS procedure
4 to give the resin-bound tetraoxazole. The N-terminal Fmoc
protecting group was then removed (general procedure 2) and
the resulting amine was cleaved from the resin (general procedure
5) to give a colourless solid, which was purified by preparative
RP-HPLC [gradient of 20% to 70% acetonitrile (0.05% TFA) in
1
hydrochloride salt 15 (0.793 g, quant.) as a colourless solid. H
NMR (400 MHz, CD3OD): d 1.60 (m, 2H), 2.07 (m, 2H), 3.16 (t,
J = 6.6 Hz, 2H), 4.69 (t, J = 6.6 Hz, 1H), 5.07 (s, 2H), 7.33 (m,
5H), 8.60 (s, 1H), N–H and O–H signals not observed; 13C NMR
(75.5 MHz, CD3OD): d 18.6, 26.5, 30.2, 40.8, 58.3, 128.8, 129.0,
129.5, 135.2, 138.3, 147.3, 159.0, 161.8, 163.6.
Fmoc-Leu-Oxz(Ser)-OH (16). Compound 13 (4.66 g,
23.5 mmol) was Fmoc-protected according to general procedure
9. The crude material thus obtained was purified by flash
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3471–3483 | 3479
©

View Article Online
water (0.05% TFA) over 60 min] to afford the TFA salt of the
Cyclo[Phe-Oxz(Ser)]2-[Orn(Cbz)-Oxz(Ser)]2 (22). The linear
tetraoxazole 19 (65 mg, 0.060 mmol) was cyclised according
to general procedure 6. Subjection of the crude material to
preparative RP-HPLC [gradient of 30% to 80% acetonitrile (0.05%
TFA) in water (0.05% TFA) over 60 min] gave the desired cyclic
tetraoxaz◦ole 22 (tR = 45.7 min, 31 mg, 48%) as a brown solid, m.p.
desired linear tetraoxazole 19 (tR = 38.8 min 0.150 g, 51%) as a
◦
colourless solid. m.p. 119–122 C; [a]2D0 = +16.4 (c 0.8, MeOH);
1H NMR (400 MHz, CD3OD): d 1.58 (m, 4H), 1.95–2.19 (m,
4H), 3.16 (m, 4H), 3.27 (partially obscured m, 2H), 3.43 (m, 2H),
4.70 (dd, J = 6.2, 8.2 Hz, 1H), 5.03 (s, 2H), 5.04 (s, 2H), 5.34
(dd, J = 5.9, 8.8 Hz, 1H), 5.54 (m, 2H), 7.15–7.33 (m, 20H), 8.26
(s 1H), 8.30 (s, 1H), 8.42 (s, 1H), 8.52 (s, 1H), N–H and O–H
signals not observed; 13C NMR (100.6 MHz, CD3OD): d 26.6,
27.3, 30.1, 30.8, 39.6, 40.7, 41.1, 49.6, 49.9, 50.0, 67.4, 67.5, 122.4,
128.0, 128.7, 128.8, 128.9(7), 129.0(3), 129.4(6), 129.4(7), 129.6,
129.7, 130.3, 134.7, 136.7, 137.2, 137.7(0), 137.7(1), 138.3, 138.4,
143.5, 143.8, 144.8, 146.1, 158.9(6), 159.0(0), 161.3, 162.0, 162.2,
162.3, 162.5, 163.9, 164.7, 165.1, 165.5, three signals obscured
or overlapping; HRMS (ESI) calcd. for C56H57N10O13 [M + H]+
1077.4101, found 1077.4093.
1
107–109 C; [a]2D0 = +22.3 (c 0.6, MeOH); H NMR (400 MHz,
CD3OD): d 1.52–1.71 (m, 4H), 1.95–2.12 (m, 4H), 3.25–3.45
(partially obscured m, 8H), 5.06 (s, 4H), 5.43 (m, 2H), 5.62 (t,
J = 7.3 Hz, 1H), 5.67 (dd, J = 6.4, 8.1 Hz, 1H), 7.16–7.34 (m,
20H), 8.24 (s 1H), 8.32 (s, 1H), 8.36 (s, 2H), N–H signals not
observed; 13C NMR (100.6 MHz, CD3OD): d 27.3, 27.4, 31.4, 31.5,
40.1, 40.2, 41.2, 47.5, 47.6, 67.4, 128.2, 128.8, 129.0, 129.5, 129.6,
130.4(0), 130.4(4), 136.7, 136.8, 137.4, 137.5, 138.4, 143.8, 143.9,
144.0, 159.0, 161.7, 161.8, 162.0, 162.1, 164.9, 165.0, 165.5, 165.6,
fourteen signals obscured or overlapping; HRMS (ESI) calcd. for
C56H54N10O12Na [M + Na]+ 1081.3815, found 1081.3817.
CF3CO2H.H2N-[Leu-Oxz(Ser)]2-[Orn(Cbz)-Oxz(Ser)]2-OH
(20). The oxazole 18 was loaded onto 2-chlorotrityl chloride
resin (0.300 g, resin capacity 1.01 mmol g-1) according to general
procedure 1 (loading 1.00 mmol g-1). The appropriate Fmoc-
protected oxazole building blocks 18 and 16 were coupled to
the resin-bound amine according to the manual SPPS procedure
4 to give the resin-bound tetraoxazole. The N-terminal Fmoc
protecting group was deprotected (general procedure 2) and the
resulting amine was cleaved from the resin (general procedure
5) to give a colourless solid, which was purified by preparative
RP-HPLC [gradient of 20% to 80% acetonitrile (0.05% TFA)
in water (0.05% TFA) over 60 min] to afford the TFA salt of
the desired linear tetraoxazole 20 (tR = 29.2 min, 81 mg, 27%)
as a colourless solid. m.p. 124–125 ◦C; 1H NMR (400 MHz,
CD3OD): d 0.92–1.03 (m, 12H), 1.51–1.70 (m, 6H), 1.81–2.17
(m, 8H), 3.11–3.22 (m, 4H), 4.71 (dd, J = 6.5, 8.8 Hz, 1H),
5.04 (s, 4H), 5.28–5.35 (m, 2H), 5.42 (dd, J = 5.9, 9.5 Hz, 1H),
7.23–7.34 (m, 10H), 8.35 (s 1H), 8.36(9) (s, 1H), 8.37(4) (s, 1H),
8.53 (s, 1H), N–H and O–H signals not observed; 13C NMR
(100.6 MHz, CD3OD): d 22.0, 22.1, 22.7, 23.2, 25.7, 25.9, 27.3,
30.9, 41.1, 41.9, 42.4, 46.8, 67.4, 128.8, 128.9, 129.5, 135.6, 136.8,
137.3, 138.4, 143.6, 144.7, 145.6, 158.9, 161.8, 162.0, 162.5, 162.6,
164.7, 165.3, 165.6, 165.7, twelve signals obscured or overlapping;
HRMS (ESI) calcd. for C50H61N10O13 [M + H]+ 1010.0785, found
1010.0751.
Cyclo[Leu-Oxz(Ser)]2-[Orn(Cbz)-Oxz(Ser)]2 (23). The linear
peptide 20 (81 mg, 0.080 mmol) was cyclised according to general
procedure 6. Subjection of the crude material to preparative RP-
HPLC [gradient of 30% to 80% acetonitrile (0.05% TFA) in water
(0.05% TFA) over 60 min] gave the desired cyclic tetraoxazole 23
(tR = 43.5 min, 40 mg, 51%) as a colourless oil. [a]2D0 = -32.8 (c 1.0,
MeOH); 1H NMR (400 MHz, CD3OD); d 0.99 (d, J = 6.4 Hz, 6H),
1.00 (d, J = 6.4 Hz, 6H), 1.50–1.74 (m, 7H), 1.84–2.17 (m, 7H),
3.19 (m, 4H), 4.99–5.07 (m, 5H), 5.25–5.45 (m, 2H), 5.49 (dd, J =
5.8, 9.5 Hz, 1H), 7.21–7.34 (m, 10H), 8.35 (s, 2H), 8.36 (s, 2H), N–
H signals not observed; 13C NMR (100.6 MHz, CD3OD): d 22.1,
23.1, 26.0, 27.3, 31.4, 31.5, 41.2, 43.1, 43.2, 46.0, 47.5, 67.4, 128.7,
128.9, 129.4, 136.8, 138.4, 143.8, 143.9(0), 143.9(2), 144.0, 158.9,
162.0, 162.1, 165.5(6), 165.6(2), 166.0(5), 166.0(8), sixteen signals
obscured or overlapping; HRMS (ESI) calcd. for C50H58N10O12Na
[M + Na]+ 1013.4128, found 1013.4117.
Cyclo[Leu-Oxz(Ser)-Orn(Cbz)-Oxz(Ser)]2 (24a).
Method A. The linear tetraoxazole 21 (34 mg, 0.033 mmol)
was cyclised according to general procedure 6. The crude material
was purified by preparative RP-HPLC [gradient of 10% to 80%
acetonitrile (0.05% TFA) in water (0.05% TFA) over 60 min] to
give the desired cyclic tetraoxazole 24a (tR = 45.3 min, 12 mg, 38%)
as a colourless solid. Characterisation data was identical to that
obtained below (fraction A).
Method B. Bisoxazole 28 (0.264 g, 0.514 mmol) was cy-
clodimerised according to general procedure 11. The crude
material was subjected to preparative RP-HPLC [gradient of 45%
to 70% acetonitrile (0.1% TFA) in water (0.1% TFA) over 50 min]
to give two fractions A and B.
H2N-[Leu-Oxz(Ser)-Orn(Cbz)-Oxz(Ser)]2-OH (21). The oxa-
zole building block 18 was loaded onto 2-chlorotrityl chloride
resin (80 mg, resin capacity 1.4 mmol g-1) according to general
procedure 1 (loading 0.73 mmol g-1). The appropriate Fmoc-
protected oxazoles 16 and 18 were coupled to the resin-bound
amine according to the manual SPPS procedure 4 to give the
resin-bound tetraoxazole. The N-terminal Fmoc protecting group
was then removed (general procedure 2) and the resulting amine
was cleaved from the resin according to general procedure 5 to
afford the desired linear tetraoxazole 21 (74 mg, quant.) as a
colourless solid, which was used in the next step without further
purification. 1H NMR (300 MHz, CD3OD): d 0.96 (m, 12H), 1.46–
2.16 (complex m, 12H), 3.15 (m, 4H), 5.04 (s, 4H), 5.37 (m, 4H),
7.30 (s, 10 H), 8.14 (s, 1H), 8.33 (s, 1H), 8.37 (s, 1H) 8.44 (s, 1H),
N–H and O–H signals not observed; LRMS (ESI) m/z = 1009
[M + H]+.
Concentration of fraction A (tR = 37.4 min) gave the desired
cyclic dimer 24a (0.122 g, 48%) as a colourless solid. m.p. 124–
◦
126 C; [a]2D0 = -26.9 (c 1.0, MeOH); 1H NMR (400 MHz,
CD3OD): d 0.99 (d, J = 6.6 Hz, 6H), 1.01 (d, J = 6.6 Hz, 6H), 1.53–
1.75 (m, 6H), 1.89 (m, 2H), 2.02–2.14 (m, 6H), 3.20 (m, 4H), 5.06
(s, 4H), 5.46 (m, 2H), 5.49 (dd, J = 6.0, 9.6 Hz, 2H), 7.31 (m, 10H),
8.36 (s, 2H), 8.37 (s, 2H), N–H signals not observed; 13C NMR
(100.6 MHz, CD3OD): d 22.1, 23.1, 25.9, 27.3, 31.4, 41.1, 43.1,
46.0, 47.4, 67.4, 128.7, 128.9, 129.4, 136.8, 138.4, 143.8, 144.0,
159.0, 162.1, 165.5, 166.2, two signals obscured or overlapping;
HRMS (ESI) calcd. for C50H58N10O12Na [M + Na]+ 1013.4128,
found 1013.4131.
3480 | Org. Biomol. Chem., 2011, 9, 3471–3483
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
[a]2D0 = -4.75 (c 0.8, MeOH); 1H NMR (300 MHz, CD3OD): d 0.96
(d, J = 6.6 Hz, 3H), 1.00 (d, J = 6.6 Hz, 3H), 1.52–1.73 (m, 3H),
1.85 (m, 1H), 1.96–2.20 (m, 3H), 3.18 (t, J = 6.6 Hz, 2H), 4.72
(dd, J = 6.2, 6.6 Hz, 1H), 5.05 (s, 2H), 5.34 (m, 1H), 7.32 (m, 5H),
8.36 (s, 1H), 8.53 (s, 1H), N–H and O–H signals not observed;
13C NMR (75.5 MHz, CD3OD): d 22.2, 22.7, 25.7, 27.2, 31.0,
41.1, 42.0, 67.4, 128.7, 128.9, 129.5, 136.0, 137.3, 138.4, 144.6,
145.3, 158.9, 161.8, 162.0, 165.0, 165.5, two signals obscured or
overlapping; HRMS (ESI) calcd. for C25H31N5O7Na [M + Na]+
536.2116, found 536.2119.
Concentration of fraction B (tR = 42.0 min) gave the cyclic trimer
24b (13 mg, 5%) as a colourless oil. 1H NMR (400 MHz, CD3OD):
d 0.97 (m, 18H), 1.49–1.73 (m, 10H), 1.86–2.21 (m, 11H), 3.17 (m,
6H), 5.03 (s, 6H), 5.25–5.43 (m, 6H), 7.23–7.36 (m, 15H), 8.3 (m,
6H), N–H signals not observed; 13C NMR (100 MHz, CD3OD):
d 22.9, 23.4, 25.9, 27.3, 30.8, 41.2, 42.5, 46.7, 47.4, 67.4, 128.8,
129.0, 129.5, 136.8, 136.9, 138.4, 143.6(9), 143.7(4), 158.9, 162.6,
162.7, 165.5, 166.0; HRMS (ESI) calcd. for C75H87N15O18Na [M +
Na]+ 1508.6246, found 1508.6250.
Fmoc-Leu-Oxz(Ser)-Orn(Cbz)-Oxz(Ser)-OMe (26). Oxazole
12 (0.973 g, 2.24 mmol) was dissolved in anhydrous
dichloromethane (7.5 mL) and treated with a solution of hy-
drochloric acid in dioxane (4 M, 8.4 mL, 33.6 mmol) under an
atmosphere of nitrogen. The solution was stirred at rt for 4 h, after
which time the mixture was concentrated under reduced pressure
to give the hydrochloride salt 25 (0.973 g, quant.). The acid 16
(0.291 g, 0.693 mmol) and the hydrochloride salt 25 (0.256 g,
0.693 mmol) were then coupled in solution according to general
procedure 6. The resulting crude material was purified by flash
chromatography (silica gel; EtOAc–hexane, 2 : 1 v/v) to give the
H2N-Phe-Oxz(Ser)-Orn(Cbz)-Oxz(Ser)-OH (29). The N-
terminal Fmoc and C-terminal methyl ester protecting groups
of bisoxazole 27 (1.66 g, 2.12 mmol) were removed according to
general procedure 10. The crude material was purified by flash
chromatography (silica gel; eluting first with neat CHCl3, then
CHCl3–MeOH, 80 : 20 v/v) to afford the bisoxazole with a free
C- and N-termini 29 (1.04 g, 89%) as a colourless solid. m.p.
◦
134–136 C; [a]2D0 = -0.25 (c 0.8, MeOH); H NMR (300 MHz,
CD3OD): d 1.58 (m, 2H), 1.94–2.19 (m, 2H), 3.17 (t, J = 5.9 Hz,
2H), 3.39 (m, 2H), 4.91 (partially obscured m, 2H), 5.05 (m, 2H),
7.17(m, 2H), 7.25–7.34 (m, 8H), 8.35(s, 1H), 8.45 (s, 1H), N–H and
O–H signals not observed; 13C NMR (75.5 MHz, CD3OD): d 27.2,
31.0, 39.2, 41.1, 51.2, 67.4, 128.7, 128.8(8), 128.9(4), 129.5, 130.1,
130.4, 135.2, 136.2, 137.2, 138.4, 144.5, 145.2, 158.9, 161.4, 162.0,
165.2, 165.4, one signal obscured or overlapping; HRMS (ESI)
calcd. for C28H29N5O7Na [M + Na]+ 570.1959, found 570.1963.
1
desired bisoxazole 26 (0.402 g, 77%) as a colourless foam. [a]D20
=
1
-12.4 (c 0.8, MeOH); H NMR (400 MHz, CD3OD): d 0.94 (d,
J = 6.4 Hz, 3H), 0.97 (d, J = 6.6 Hz, 3H), 1.61 (m, 4H), 1.77 (s,
2H), 1.96 (m, 1H), 2.10 (m, 1H), 3.17 (t, J = 6.6 Hz, 2H), 3.85 (s,
3H), 4.20 (t, J = 6.9 Hz, 1H), 4.37–4.45 (m, 2H), 4.90 (m, 1H), 5.04
(s, 2H), 5.29 (dd, J = 6.5, 9.1 Hz, 1H), 7.23–7.34 (m, 7H), 7.37 (t,
J = 7.3 Hz, 2H), 7.62 (d, J = 7.6 Hz, 1H), 7.66 (d, J = 7.6 Hz, 1H),
7.78 (d, J = 7.8 Hz, 2H), 8.33 (s, 1H), 8.41 (s, 1H), N–H signals not
observed; 13C NMR (75.5 MHz, CD3OD): d 22.0, 23.2, 25.7, 27.1,
30.9, 38.9, 41.1, 42.7, 48.2, 52.5, 67.3, 67.7, 120.9, 126.1, 128.1,
128.7, 128.9, 129.4, 134.0, 136.6, 138.4, 142.6, 143.4, 145.0, 145.2,
146.1, 158.3, 158.8, 162.6, 162.8, 165.9, 166.5; HRMS (ESI) calcd.
for C41H43N5O9Na [M + Na]+ 772.2953, found 772.2958.
H2N-Orn(Cbz)-Oxz(Ser)-Ala-Oxz(Ser)-OH (30). The N-
terminal Fmoc and C-terminal methyl ester protecting groups
of bisoxazole 32 (0.216 g, 0.305 mmol) were removed according
to general procedure 10. Subjection of the crude material to flash
chromatography (silica gel; eluting first with neat CHCl3, then
CHCl3–MeOH, 8 : 2 v/v) gave the fully-deprotected◦bisoxazole 30
1
(0.133 g, 90%) as a colourless solid. m.p. 152–155 C; H NMR
Fmoc-Phe-Oxz(Ser)-Orn(Cbz)-Oxz(Ser)-OMe (27). The acid
17 (1.52 g, 3.35 mmol) and the amine hydrochloride salt 25
(1.29 g, 3.35 mmol) were coupled in solution according to
general procedure 7. Subjection of the crude material to flash
chromatography (silica gel; EtOAc–hexane, 3 : 2 v/v) afforded the
desired bisoxazole 27 (1.78 g, 68%) as a colourless solid. m.p. 83–
85 ◦C; [a]D20 = -9.86 (c 0.7, MeOH); 1H NMR (300 MHz, CDCl3):
d 1.53–1.90 (m, 4H), 1.91–2.16 (m, 2H), 3.22 (m, 3H), 3.86 (s, 3H),
4.20 (m, 1H), 4.33–4.49 (m, 2H), 4.92 (br m, 1H), 5.07 (s, 2H), 5.23
(dd, J = 7.0, 14.3 Hz, 1H), 5.40 (dd, J = 8.4, 15.4 Hz, 1H), 5.53
(br d, J = 7.7 Hz, 1H), 7.01 (m, 2H), 7.21–7.42 (m, 12H), 7.52 (m,
2H), 7.76 (d, J = 7.3 Hz, 2H), 8.09 (s, 1H) 8.10 (s, 1H); 13C NMR
(75.5 MHz, CDCl3): d 26.3, 31.2, 39.8, 40.5, 46.9, 47.3, 51.0, 52.4,
66.9, 67.2, 120.2, 125.2, 127.2, 127.4, 127.9, 128.2(3), 128.2(5),
128.7, 128.8, 129.4, 133.5, 135.5, 136.6, 141.5, 141.9, 143.8, 143.9,
144.2, 155.7, 156.6, 160.2, 161.5, 163.4, 164.4; HRMS (ESI) calcd.
for C44H41N5O9Na [M + Na]+ 806.2797, found 806.2784.
(400 MHz, CD3OD): d 1.56 (m, 2H), 1.63 (d, J = 6.6 Hz, 3H), 2.03
(m, 2H), 3.15 (t, J = 6.6 Hz, 2H), 4.57 (t, J = 6.6 Hz, 1H), 5.04 (s,
2H), 5.36 (m, 1H), 7.25–7.33 (m, 5H), 8.14 (s, 1H), 8.44 (s, 1H), N–
H and O–H signals not observed; 13CNMR (100.6 MHz, CD3OD):
d 19.0, 26.7, 31.0, 40.9, 44.4, 67.4, 128.8, 129.0, 129.5, 137.1, 138.4,
143.4, 144.3, 159.0, 159.5, 161.9, 163.1, 165.4, 167.9, one signal
obscured or overlapping; HRMS (ESI) calcd. for C22H25N5O7Na
[M + Na]+ 494.1652, found 494.1644.
Fmoc-Orn(Cbz)-Oxz(Ser)-Ala-Oxz(Ser)-OMe (32). The acid
18 (0.784 g, 1.41 mmol) and the previously prepared amine
hydrochloride salt 31 (0.269 g, 1.41 mmol) were coupled in solution
according to general procedure 7. Subjection of the crude material
to flash chromatography (silica gel; hexane–EtOAc, 3 : 1 v/v) gave
the desired bisoxazole 32 (0.589 g, 59%) as a colourless foam.
1
[a]2D0 = -11.0 (c 0.7, MeOH); H NMR (400 MHz, CD3OD): d
1.56 (br m, 2H), 1.63 (d, J = 7.1 Hz, 3H), 1.79–2.03 (m, 2H), 3.15
(td, J = 2.7, 6.9 Hz, 2H), 3.85 (s, 3H), 4.21 (t, J = 6.6 Hz, 1H),
4.41 (d, J = 6.6 Hz, 2H), 4.84 (m, 1H), 5.05 (s, 2H), 5.35 (q, J =
7.1 Hz, 1H), 7.24–7.34 (m, 7H), 7.37 (t, J = 7.3 Hz, 2H), 7.64 (d,
J = 7.6 Hz, 1H), 7.66 (d, J = 7.6 Hz, 1H), 7.82 (d, J = 7.6 Hz, 2H),
8.33 (s, 1H), 8.42 (s, 1H), N–H signals not observed; 13C NMR
(100.6 MHz, CD3OD): d 18.6, 27.2, 31.1, 41.1, 44.3, 50.5, 52.5,
67.4, 67.8, 120.9, 126.2, 128.2, 128.8, 129.0, 129.5, 134.0, 136.7,
138.4, 142.6, 143.5, 145.1, 145.2, 146.2, 158.4, 159.0, 162.4, 162.9,
H2N-Leu-Oxz(Ser)-Orn(Cbz)-Oxz(Ser)-OH (28). The N-
terminal Fmoc and C-terminal methyl ester protecting groups
of bisoxazole 26 (1.14 g, 1.52 mmol) were removed according
to general procedure 10. The crude material was purified by
flash chromatography (silica gel; eluting first with neat CHCl3,
then CHCl3–MeOH, 80 : 20 v/v) to give the fully-deprotected
bisoxazole 28 (0.73 g, 94%) as a colourless solid. m.p. 135–136 ◦C;
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3471–3483 | 3481
©

View Article Online
166.0, 166.7, one signal obscured or overlapping; HRMS (ESI)
calcd. for C38H37N5O9Na [M + Na]+ 730.2489, found 730.2479.
25.9, 31.2, 42.8, 46.0, 46.9, 55.6, 58.5, 125.8(0), 125.8(4), 136.7,
136.8, 140.7, 144.0, 144.2, 149.3, 152.6, 162.0, 162.2, 165.0, 166.1;
HRMS (ESI) calcd. for C58H67N14O8 [M + H]+ 1087.5261, found
1087.5276.
Cyclo[Phe-Oxz(Ser)-Orn(Cbz)-Oxz(Ser)]2 (33a). The bisoxa-
zole 29 (0.250 g, 0.456 mmol) was cyclodimerised according
to general procedure 11. Subjection of the crude material to
preparative RP-HPLC [gradient of 45% to 70% acetonitrile (0.1%
TFA) in water (0.1% TFA) over 50 min] gave the desired cyclic
dimer 33a (tR◦ = 37.0 min, 0.112 g, 46%) as a colourless solid.
m.p. 125–127 C; 1H NMR (400 MHz, CD3OD); d 1.55–1.70 (m,
4H), 1.97–2.18 (m, 4H), 3.22 (t, J = 6.9 Hz, 4H), 3.29 (partially
obscured m, 2H), 3.38 (dd, J = 7.0, 13.9 Hz, 2H), 5.06 (s, 4H), 5.42
(dd, J = 6.4, 8.6 Hz, 2H), 5.64 (dd, J = 7.0, 7.8 Hz, 2H), 7.15–7.32
(m, 20H), 8.25 (s, 2H), 8.30 (s, 2H), N–H signals not observed;
13C NMR (100.6 MHz, CD3OD): d 27.4, 31.3, 40.2, 41.2, 47.4,
67.4, 128.2, 128.8, 129.0, 129.5, 129.6, 130.4, 136.7, 136.8, 137.4,
138.4, 143.8, 143.9, 159.0, 161.8, 162.0, 165.1, 165.4, one signal
obscured or overlapping; HRMS (ESI) calcd. for C56H54N10O12Na
[M + Na]+ 1081.3815, found 1081.3819.
Cyclo[Phe-Oxz(Ser)-Orn(Dpa)-Oxz(Ser)]2 (4). The Cbz-
protected side chains of cyclic peptide 33a (0.112 g, 0.106 mmol)
were functionalised with Dpa groups according to general
procedure 12. Subjection of the crude material to preparative RP-
HPLC [gradient of 5% to 40% acetonitrile (0.1% TFA) in water
(0.1% TFA) over 50 min] gave the desired Dpa-functionalised
cyclic peptide 4 (tR = 39.5 min, 93 mg, 89%) as a colourless solid.
m.p. 76–78 ◦C; [a]2D0 = -19.9 (c 0.7, MeOH); 1H NMR (400 MHz,
CD3OD); d 1.83–2.21 (m, 8H), 3.31 (m, 4H), 3.36 (dd, J = 8.5,
13.9 Hz, 2H), 3.46 (dd, J = 6.9, 13.9 Hz, 2H), 4.56 (s, 8H), 5.36
(dd, J = 6.1, 9.0 Hz, 2H), 5.67 (dd, J = 6.9, 8.5 Hz, 2H), 7.16–7.27
(m, 10H), 7.49 (dd, J = 5.4, 7.6 Hz, 4H), 7.54 (d, J = 7.6 Hz,
4H), 7.94 (ddd, J = 1.6, 7.6, 7.6 Hz, 4H), 8.31 (s, 2H), 8.35 (s,
2H), 8.65 (d, J = 5.4 Hz, 4H), N–H signals not observed; 13C
NMR (100.6 MHz, CD3OD): d 22.2, 31.1, 39.9, 46.9, 58.6, 125.7,
128.1, 129.6, 130.3, 136.7, 136.8, 137.6, 140.4, 144.1, 149.5, 152.5,
161.8, 162.1, 164.9, 165.2, four signals obscured or overlapping;
HRMS (ESI) calcd. for C64H63N14O8 [M + H]+ 1155.4948, found
1155.4931.
Cyclo[Ala-Oxz(Ser)-Orn(Cbz)-Oxz(Ser)]2 (34a). The bisoxa-
zole 30 (70 mg, 0.144 mmol) was cyclodimerised according
to general procedure 11. Subjection of the crude material to
preparative RP-HPLC [gradient of 20% to 70% acetonitrile (0.05%
TFA) in water (0.05% TFA) over 60 min] gave two fractions A and
B.
Cyclo[Leu-Oxz(Ser)]2-[Orn(Dpa)-Oxz(Ser)]2 (5). The Cbz-
protected side chains of cyclic peptide 23 (40 mg, 0.040 mmol) were
functionalised with Dpa groups according to general procedure
12. Subjection of the crude material to preparative RP-HPLC
[gradient of 5% to 50% acetonitrile (0.05% TFA) in water (0.05%
TFA) over 50 min] gave the desired Dpa-functionalised cyclic
Concentration of fraction A (tR = 44.8 min) gave the desired
cycli◦c dimer 34a (23 mg, 36%) as a colourless powder. m.p. 115–
◦
118 C [lit.24 m.p. 116–118 C]; [a]D25 = -101 (c 0.5, CHCl3) [lit.24
[a]2D5 = -108 (c 0.51, CHCl3)].
Concentration of fraction B (tR = 51.0 min) gave the cyclic trimer
1
34b (11 mg, 12%) as a colourless powder. H NMR (400 MHz,
peptide 5 (tR = 35.0 min, 38 mg, 87%) as a colourless oil. [a]D20
=
CD3OD): d 1.51–1.67 (m, 15H), 1.99 (m, 4H), 2.16 (m, 2H), 3.17
(m, 6H), 5.03 (s, 6H), 5.29–5.40 (m, 6H), 7.23–7.34 (m, 15H),
8.36 (m, 6H), N–H signals not observed; HRMS (ESI) calcd. for
C66H69N15O18Na [M + Na]+ 1382.4837, found 1382.4839.
1
-34.0 (c 0.8, MeOH); H NMR (400 MHz, CD3OD); d 0.98–
1.03 (m, 12H), 1.64–1.74 (m, 2H), 1.83–2.22 (m, 12H), 3.23–3.33
(partially obscured m, 4H), 4.53 (s, 8H), 5.33–5.39 (m, 2H), 5.46–
5.53 (m, 2H), 7.54 (m, 4H), 7.60 (d, J = 7.8 Hz, 4H), 8.01 (m,
4H), 8.36 (m, 2H), 8.38 (m, 2H), 8.68 (d, J = 5.1 Hz, 4H), N–H
signals not observed; 13C NMR (100.6 MHz, CD3OD): d 22.0(6),
22.1(1), 22.3, 23.1, 25.9, 31.0, 31.1, 43.0, 45.9, 46.1, 46.9, 55.5,
55.6, 58.4, 125.9, 126.0, 136.7, 136.8, 136.9, 141.1, 143.8, 144.1,
144.3, 149.0, 152.7, 161.3, 161.7, 162.0, 162.2, 162.3, 165.0, 165.1,
166.0, 166.1, ten signals obscured or overlapping; HRMS (ESI)
calcd. for C58H67N14O8 [M + H]+ 1087.5261, found 1087.5272
Cyclo[Ala-Oxz(Ser)-Orn(Dpa)-Oxz(Ser)]2 (1). The Cbz-
protected side chains of cyclic peptide 34a (18 mg, 0.020 mmol)
were functionalised with Dpa groups according to general
procedure 12. Subjection of the crude material to preparative RP-
HPLC [gradient of 5% to 40% acetonitrile (0.1% TFA) in water
(0.1% TFA) over 50 min] gave the desired Dpa-functionalised
cyclic peptide 1 (tR = 26.9 min, 11 mg, 55%) as a colourless oil.
[a]2D0 = -74.1 (c 0.60, MeOH) [lit.9 [a]2D0 = -86.8 (c 0.65, MeOH)].
Cyclo[Phe-Oxz(Ser)]2-[Orn(Dpa)-Oxz(Ser)]2 (6). The Cbz-
protected side chains of cyclic peptide 22 (38 mg, 0.035 mmol) were
functionalised with Dpa groups according to general procedure
12. Subjection of the crude material to preparative RP-HPLC
[gradient of 5% to 50% acetonitrile (0.05% TFA) in water (0.05%
TFA) over 50 min] gave the desired Dpa-functionalised cyclic
peptide 6 (tR = 33.8 min, 31 mg, 77%) as a colourless solid. m.p. 62–
64 ◦C; 1H NMR (400 MHz, CD3OD); d 1.83–2.23 (m, 8H), 3.32–
3.48 (partially obscured m, 8H), 4.56 (s, 4H), 4.58 (s, 4H), 5.36 (dd,
J = 6.1, 8.8 Hz, 1H), 5.41 (dd, J = 5.9, 9.1 Hz, 1H), 5.61 (m, 1H),
5.69 (m, 1H), 7.18–7.29 (m, 10H), 7.46 (m, 4H), 7.52 (d, J = 7.6 Hz,
4H), 7.87–7.94 (m, 4H), 8.23 (s 1H), 8.33 (s, 1H), 8.39 (m, 2H),
8.64 (m, 4H), N–H signals not observed; 13C NMR (100.6 MHz,
CD3OD): d 22.1(9), 22.2(1), 30.9(6), 31.0(1), 40.0, 40.1, 46.8, 47.0,
55.6, 58.7, 125.6(1), 125.6(2), 128.2, 129.6, 130.3, 130.4, 136.6(8),
136.7(3), 136.8, 137.5, 137.6, 140.0(8), 140.1(0), 143.8, 144.1,
Cyclo[Leu-Oxz(Ser)-Orn(Dpa)-Oxz(Ser)]2 (3). The Cbz-
protected side chains of cyclic peptide 24a (51 mg, 0.050 mmol)
were functionalised with Dpa groups according to general
procedure 12. The crude material was purified by preparative RP-
HPLC [gradient of 2% to 40% acetonitrile (0.1% TFA) in water
(0.1% TFA) over 50 min] to give the desired Dpa-functionalised
cyclic peptide 3 (tR = 40.3 min, 41 mg, 75%) as a colourless solid.
m.p. 81–83 ◦C; [a]2D0 = -17.7 (c 0.7, MeOH); 1H NMR (400 MHz,
CD3OD); d 1.00 (d, J = 6.7 Hz, 6H), 1.02 (d, J = 6.9 Hz, 6H),
1.70 (m, 2H), 1.83–2.20 (m, 12H), 3.30 (partially obscured m,
4H), 4.54 (s, 8H), 5.38 (dd, J = 5.6, 9.1 Hz, 2H), 5.49 (dd, J = 5.9,
9.7 Hz, 2H), 7.52 (ddd, J = 1.2, 5.1, 7.8 Hz, 4H), 7.57 (ddd, J =
0.9, 1.2, 7.8 Hz, 4H), 7.98 (ddd, J = 1.7, 7.8, 7.8 Hz, 4H), 8.38 (s,
2H), 8.39 (s, 2H), 8.66 (ddd, J = 0.9, 1.7, 5.1 Hz, 4H), N–H signals
not observed; 13C NMR (100.6 MHz, CD3OD): d 22.1, 22.2, 23.1,
3482 | Org. Biomol. Chem., 2011, 9, 3471–3483
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
144.2, 144.3, 149.7(9), 149.8(3), 152.4, 152.5, 161.6(7), 161.7(4),
162.1(7), 162.2(0), 164.9, 164.9(6), 165.0(0), ten signals obscured
or overlapping; HRMS (ESI) calcd. for C64H63N14O8 [M + H]+
1155.4948, found 1155.4939.
11 Y. Singh, M. J. Stoermer, A. J. Lucke, M. P. Glenn and D. P. Fairlie,
Org. Lett., 2002, 4, 3367.
12 J. Veliscek Carolan, S. J. Butler and K. A. Jolliffe, J. Org. Chem., 2009,
74, 2992.
13 D. E. Wilcox, Chem. Rev., 1996, 96, 2435.
14 D. H. Lee, S. Y. Kim and J.-I. Hong, Angew. Chem., Int. Ed., 2004, 43,
4777.
15 G. Haberhauer and F. Rominger, Eur. J. Org. Chem., 2003, 3209.
16 (a) A. Bertram, A. J. Blake, F. Gonzale´z-Lo´pez de Turiso, J. S. Hannam,
K. A. Jolliffe, G. Pattenden and M. Skae, Tetrahedron, 2003, 59, 6979;
(b) A. Bertram, J. S. Hannam, K. A. Jolliffe, F. Gonzale´z-Lo´pez de
Turiso and G. Pattenden, Synlett, 1999, 1723.
17 R. J. G. Black, V. J. Dungan, R. Y. T. Li, P. G. Young and K. A. Jolliffe,
Synlett, 2010, 4, 551.
Acknowledgements
The authors thank the Australian Research Council for financial
support (DP0877726) and Dr Kelvin Picker (The University of
Sydney) for technical support with HPLC purification.
18 S.-L. You and J. W. Kelly, Tetrahedron Lett., 2005, 46, 2567.
19 S.-L. You, S. Deechongkit and J. W. Kelly, Org. Lett., 2004, 6, 2627.
20 A. J. Phillips, Y. Uto, P. Wipf, M. J. Reno and D. R. Williams, Org.
Lett., 2000, 2, 1165.
21 E. Biron, J. Chatterjee and H. Kessler, Org. Lett., 2006, 8, 2417.
22 G. Haberhauer, A. Pinter, T. Oeser and F. Rominger, Eur. J. Org. Chem.,
2007, 1779.
Notes and references
1 A. Bertram and G. Pattenden, Nat. Prod. Rep., 2007, 24, 18.
2 G. Pattenden and T. Thompson, Tetrahedron Lett., 2002, 43, 2459.
3 L. Somogyi, G. Haberhauer and J. Rebek, Tetrahedron, 2001, 57,
1699.
4 A. J. Lucke, J. D. A. Tyndall, Y. Singh and D. P. Fairlie, J. Mol. Graphics
23 G. Haberhauer and F. Rominger, Tetrahedron Lett., 2002, 43, 6335.
24 S. J. Butler, K. A. Jolliffe, W. Y. G. Lee, M. J. McDonough and A. J.
Reynolds, Tetrahedron, 2011, 67, 1019.
25 M. I. Pita Boente, G. W. Kirby, G. L. Patrick and D. J. Robins, J. Chem.
Soc., Perkin Trans. 1, 1991, 1283.
26 S. L. Wiskur, H. Ait-Haddou, J. J. Lavigne and E. V. Anslyn, Acc. Chem.
Res., 2001, 34, 963.
27 R. G. Hanshaw, S. M. Hilkert, H. Jiang and B. D. Smith, Tetrahedron
Lett., 2004, 45, 8721.
28 A less competitive solvent was used in the present studies (previous
studies employed 5 mM HEPES buffer and 145 mM NaCl) because the
phenylalanine derivatives 4·Zn2 and 6·Zn2 were not sufficiently soluble
in the NaCl solution to allow binding constants to be determined .
Modell., 2003, 21, 341.
5 Y. Singh, N. Sokolenko, M. J. Kelso, L. R. Gahan, G. Abbenante and
D. P. Fairlie, J. Am. Chem. Soc., 2001, 123, 333.
6 D. Mink, S. Mecozzi and J. Rebek, Jr., Tetrahedron Lett., 1998, 39,
5709.
7 M. Schnopp, S. Ernst and G. Haberhauer, Eur. J. Org. Chem., 2009,
213.
8 K. A. Jolliffe, Supramol. Chem., 2005, 17, 81.
9 M. J. McDonough, A. J. Reynolds, W. Y. G. Lee and K. A. Jolliffe,
Chem. Commun., 2006, 2971.
10 G. Haberhauer, T. Oeser and F. Rominger, Chem.–Eur. J., 2005, 11,
6718.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3471–3483 | 3483
©