Synthesis, evaluation and incorporation into liposomes of 4-functionalised-2,5-diphenyloxazole derivatives for application in scintillation proximity assays

Article abstract of DOI:10.1016/j.tetlet.2004.01.034

The preparation of a series of 4-functionalised-2,5-diphenyloxazoles is described. The scintillating efficiency of each of these 'scintilipid' molecules has been evaluated in the presence of ionising radiation. Each 'scintilipid' has been assessed for the ability to assemble, with other lipids, into liposomes, under a variety of preparative conditions. Each liposomal preparation has been monitored for the ability to scintillate in the presence of ionising radiation. The optimal 'scintilipid', both in terms of effective liposomal formation and scintillation efficiency, has been determined.

Full text of DOI:10.1016/j.tetlet.2004.01.034

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 2163–2166
Synthesis, evaluation and incorporation into liposomes of
4-functionalised-2,5-diphenyloxazole derivatives for application in
scintillation proximity assays
Mark C. McCairn,a Steven J. Culliford,c Roland Z. Kozlowskib and
Andrew J. Sutherlanda,*
aChemical Engineering and Applied Chemistry, Aston University, Aston Triangle, Birmingham B4 7ET, UK
bDepartment of Pharmacology, School of Medical Sciences, University Walk, Clifton, Bristol BS8 1TD, UK
cProcognia Ltd, Unit 4, The Switchback, Gardener Road, Maidenhead, Berks SL6 7RJ, UK
Received 9 December 2003; accepted 9 January 2004
Abstract—The preparation of a series of 4-functionalised-2,5-diphenyloxazoles is described. The scintillating efficiency of each of
these ꢀscintilipidꢁ molecules has been evaluated in the presence of ionising radiation. Each ꢀscintilipidꢁ has been assessed for the ability
to assemble, with other lipids, into liposomes, under a variety of preparative conditions. Each liposomal preparation has been
monitored for the ability to scintillate in the presence of ionising radiation. The optimal ꢀscintilipidꢁ, both in terms of effective
liposomal formation and scintillation efficiency, has been determined.
Ó 2004 Elsevier Ltd. All rights reserved.
The scintillation proximity assay (SPA), developed and
marketed by Amersham Biosciences, is a real-time,
quantitative and dynamic assay for the evaluation of
receptor/ligand binding interactions.1 In a typical SPA
procedure, the receptor is labelled by immobilisation
upon the surfaces of SPA beads, microspheres that
encapsulate scintillant molecules, whilst the ligand is
radiolabelled with a short pathlength emitter such as
tritium. When the ligand is bound by the receptor, the
radiolabel is brought into close proximity with the
scintillant molecules within the SPA bead resulting in
the emission of light. The emitted light may be detected
readily and quantified in a scintillation counter. If no
binding occurs, the majority of the radioactivity remains
too remote from the encapsulated scintillant to elicit a
significant signal. The SPA has also been adapted to the
study of whole cells in microtitre plate format.2 Here,
scintillant molecules, such as 2,5-diphenyloxazole, are
incorporated into the plastic comprising the base of the
microtitre plate. Cells are then grown on the bottom
surfaces of the plate and incubated with a radiolabelled
molecule. If the cells in a particular well take-up the
radiolabelled molecule the base of the well will emit light
in an analogous fashion to the bead-based assay.2
As part of our on-going program investigating scintil-
lation phenomena, we were intrigued by the possibility
of creating cells, which functioned directly as ꢀliving SPA
beadsꢁ. We reasoned that liposomal delivery of appro-
priately derivatised scintillant molecules should enable
their incorporation into the membranes of living cells.
Herein, we report the synthesis of a series of 4-func-
tionalised-2,5-diphenyloxazoles and the evaluation of
each derivative in terms of its scintillation efficiency.3
We also report the suitability of each derivative for use
in forming liposomes under standard conditions4 and, in
each case, have determined the scintillating efficiency of
the resultant liposomes.
Previous work3;5 has demonstrated that 2,5-diphenyl-
oxazole 1 may be converted into 2,5-diphenyl-4-hy-
droxymethyloxazole
3
and 4-bromomethyl-2,5-di-
phenyloxazole 4 relatively easily. In addition, ethyl-2,5-
diphenyl-4-oxazolecarboxylate 2 may be synthesised
readily.3 Scheme 1 outlines the divergent synthetic
strategy that has been adopted, using these readily
available precursors, to access a wide number of differ-
ent 4-functionalised-2,5-diphenyloxazoles (ꢀscintilipidsꢁ)
as efficiently as possible.
Keywords: SPA; Scintillation; Oxazoles.
* Corresponding author. Tel.: +44-121-359-3611; fax: +44-121-359-
4094; e-mail: a.j.sutherland@aston.ac.uk
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.01.034

2164
M. C. McCairn et al. / Tetrahedron Letters 45 (2004) 2163–2166
R
N
Ph
Ph
O
O
O
H
O
OH
O
O
O
O
O
15 R =
16 R =
1 R =
2 R =
3 R =
4 R =
5 R =
6 R =
7 R =
8 R =
9 R =
O
(CH2)17CH3
O
O
OH
OH
Br
10 R =
11 R =
12 R =
13 R =
14 R =
O
O
17 R =
18 R =
O
O
NEt3 Br
O
O
NH2
N
H
O
O
OH
OH
O
O
(CH2)7CH3
(CH2)9CH3
19 R =
O
O
O
Cl
NH3
N
H
d
m
2
5
14
6
b
r
7
9
e
15
16
17
19
f
i
n
o
a
g
1
3
8
p
h
c
10
11
12
13
j
l
k
q
4
18
Scheme 1. Reagents and conditions: (a) (1) sec-BuLi, TMP, THF, )78 to 0 °C, DMF/10 h, rt; (2) NaBH4, MeOH, 33%. (b) LiBH4, LiEt3BH, THF,
98%. (c) PBr3, DCM, 98%. (d) NH2NH2ÆxH2O, EtOH, 43%. (e) NaH, C8H17Br, DMF, 76%. (f) NaH, C10H21Br, DMF, 72%. (g) (1) NaH,
C6H12BrOSiButMe2, DMF; (2) TBAF, THF, 40%. (h) NaH, C18H37Br, DMF, 76%. (i) (1) NaH, C8H17BrOSiButMe2, DMF; (2) TBAF, THF, 28%.
(j) Et3N, CH3COCH3, 66%. (k) HOPh(CH2)3OH, K2CO3, 18-crown-6, CH3CN, 55%. (l) NaH, trans-CH3(CH2)4CH@CH(CH2)3OH, DMF, 12%.
(m) HCl , CHCl3, 100%. (n) PCC/silica gel (1:1), DCM, 84%. (o) CH3(CH2)2COCl, Et3N, DCM, 45%. (p) PhCOCl, Et3N, DCM, 52%. (q) PhCOCl,
ðgÞ
Et3N, DCM, 65%. (r) Ag2O, NaOH , MeOH, 77%.
ðaqÞ
The fusion of a cell membrane with a liposome is
facilitated when the liposome is constructed from cat-
ionic lipids.6 Accordingly, the first target was the cat-
ionic tetraalkylammonium salt 11 that was synthesised,
in one step, by treating a solution of bromide 4 in ace-
tone with triethylamine. A second cationic scintillant-
containing compound was synthesised by treating a
solution of ester 2 in ethanol with excess hydrazine
hydrate to generate hydrazide 5. Subsequently, excess
hydrogen chloride gas was bubbled through a solution
of this hydrazide in chloroform to precipitate the
hydrochloride salt 14.
presence of ionising radiation (Table 1). A possible
explanation for the low scintillation efficiencies may be
the close proximity of the oxazole ring to the electron
density of the carbonyl group or a formal charge, which
may cause quenching of scintillation through a charge
transfer process. The prediction of scintillation efficiency
in these systems is nontrivial and is highly dependent
upon the type of chemical functionality attached at the
4-position of the oxazole. For example, previous work3a
had shown that ester 2 scintillates with very high effi-
ciency but aldehydic and carboxylic acid functionality
at the 4-position of the 2,5-diphenyloxazole skeleton
results in very poor scintillation.
Unfortunately, hydrazide 5 and both scintillant-con-
taining cationic compounds 11 and14 and their corre-
sponding liposomes (ꢀscintiliposomesꢁ) all exhibited
relatively low scintillation counting efficiencies in the
In previous work3a an ether linkage had been used to
attach substituents at the 4-position of 2,5-diphenyl-
oxazole, whilst retaining very high scintillation counting

M. C. McCairn et al. / Tetrahedron Letters 45 (2004) 2163–2166
2165
Table 1. Scintillation efficiencies of scintilipids 5–14, 16–19 and scintillation counting data for scintiliposomes incorporating these lipids
Scintilipid
Scintillation efficiencya;b
Scintillation counts obtained for each liposomal composition/cpm
PCa;c
DOTAPa;c
DOTAP/DOPE (1:1)a;c
5
0.6
49
3
4
3
8
10
11
6
7
47
73
96
60
161
12
214
308
12
18
9
12
76
11
62
8
0.9
43
9
285
129
10
199
172
8
10
11
12
13
14
16
17
18
19
21
0.1
14
469
128
8
506
518
9
19
0.5
17
30
20
44
66
13
11
8
12
234
58
146
20
93
a Values are background corrected.
b To a solution of scintilipid compound in toluene (200 lL, 20 mM) was added [3H]hexadecane (200 lL, 0.5 lCi/mL), followed by monitoring in a
scintillation counter. Values correspond to the cpm as a percentage of the cpm detected with the addition of commercial scintillation fluid
Microscint 20. All values have been adjusted to accommodate the fact that a system based solely on 2,5-diphenyloxazole 1 gives 70% scintillation
counting efficiency when compared with commercial scintillation fluid.3a
c
[
14C]taurine (0.1 lCi) was added to the scintiliposome preparation (500 lL, 10 mg/mL, 10:1, lipid/scintilipid) and the assay solution monitored in a
scintillation counter.
efficiency in the presence of ionising radiation. The ether
linkage was also deemed attractive due to its chemical
robustness and ready accessibility via either scintillant
alcohol 3 or scintillant bromide 4. Accordingly, all
compounds synthesised subsequently, that is com-
pounds 6–10, 12, 13 and 15–19, were constructed with
an ether linkage between the 4-position of the 2,5-
diphenyloxazole unit and the side chain.
In a further attempt to try and prevent the deposition of
crystalline scintillant within the liposomes, the relatively
rigid aromatic-containing side chains, present in scin-
tilipids 12 and 18, were replaced with side chains that
were entirely aliphatic to give scintilipids 6, 7 and 9.
Each scintilipid 6, 7 and 9 was constructed in a Wil-
liamson-type ether synthesis reaction between alcohol 3
and the appropriate alkyl halide. The resultant scinti-
lipids all exhibited high scintillation counting efficiencies
(Table 1). Unfortunately, incorporation of scintilipids 6,
7 and 9 into liposomes resulted, in each case, in a sig-
nificant reduction in scintillation counting efficiency and
the deposition of crystalline scintilipid within the lipo-
somes.
Alcohol 12 was synthesised successfully by a potassium
carbonate/18-crown-6 mediated Williamson-type ether
synthesis between 3-(4-hydroxyphenyl)-propan-1-ol and
bromide 4. Alcohol 12 exhibited a modest but a suffi-
ciently high scintillation efficiency (Table 1) for further
evaluation. Accordingly, alcohol 12 was incorporated
into standard liposomal preparations to provide the
corresponding ꢀscintiliposomesꢁ. These scintiliposomes,
12/dioleoyltrimethylammonium propane (DOTAP)
(469 cpm), 12/phosphatidylcholine (PC) (214 cpm) and
12/DOTAP/dioleoylphosphatidylethanolamine (DOPE)
(506 cpm) exhibited a significant number of scintillation
counts compared with all other scintiliposome prepa-
rations evaluated (Table 1). Unfortunately, microscopic
examination of liposomal preparations incorporating
alcohol 12 revealed crystalline regions of this scintilipid
within all three scintiliposomes, which rendered them
unusable in subsequent cell-based assays.
To try and circumvent the problem of crystallisation it
was next decided to evaluate a scintilipid with a degree
of unsaturation in the side chain. A Williamson-type
ether synthesis between trans-4-decenol (obtained in
quantitative yield by reducing a sample of commercially
available trans-4-decenal with sodium borohydride) and
scintillant bromide 4 furnished scintillant alkene 13.
Whilst the yield for this reaction was disappointingly
low (12%), sufficient material was produced for evalua-
tion purposes. Alkene 13 exhibited only a moderate
scintillation counting efficiency (19%) in the presence of
ionising radiation. However, incorporation of the scin-
tillant alkene 13 into liposome preparation provided
scintiliposomes 13/DOTAP (128 cpm), 13/PC (308 cpm)
and 13/DOTAP/DOPE (518 cpm) that exhibited a sig-
nificant number of scintillation counts compared with
other liposome preparations evaluated (Table 1). In fact,
13/PC and 13/DOTAP/DOPE exhibited greater num-
bers of counts than any other scintiliposomes, of anal-
ogous composition, that were evaluated. Unfortunately,
microscopic analysis revealed that each scintiliposome
contained crystalline regions of scintilipid 13.
In an attempt to circumvent the problem of crystallisa-
tion within the scintiliposomes, alcohol 12 was modified
further. Esterification with benzoyl chloride under
standard conditions gave ester 18 in good yield. Ester 18
and scintiliposomes incorporating this scintilipid exhib-
ited relatively high scintillation counting efficiencies but,
unfortunately, microscopic examination of the scinti-
liposomes again revealed crystalline regions of the
scintilipid.

2166
M. C. McCairn et al. / Tetrahedron Letters 45 (2004) 2163–2166
In a further attempt, to prevent crystallisation of the
scintilipids within the liposomes, a set of scintilipids that
contained functionalised but fully saturated side chains
were prepared. Alcohol 8 was synthesised by a Wil-
liamson-type ether synthesis between a silylated deriva-
tive of 6-bromo-1-hexanol and scintillant alcohol 3.
Initially, the hydroxyl functionality of 6-bromo-1-hex-
anol was protected by silylation with tert-butyldimeth-
ylsilyl chloride. The resultant silyl-ether was added to a
solution of scintillant alkoxide in DMF, generated by
treatment of 3 with sodium hydride, to give the coupled
scintillant silyl-ether adduct, which was then desilylated
by treatment with tetrabutylammonium fluoride
(TBAF) to provide alcohol 8. Surprisingly, alcohol 8
gave only a low scintillation counting efficiency (0.9%)
but still furnished liposomal preparations, which gave
significant scintillation counts, 8/PC (96 cpm), 8/DO-
TAP (76 cpm), 8/DOTAP/DOPE (62 cpm). This
encouraging result prompted the derivatisation of alco-
hol 8 into a number of other scintillant-containing
compounds 15, 16 and 17.
an aliphatic chain length two methylene units longer in
length than alcohol 8. Gratifyingly, alcohol 10 exhibited
a scintillation counting efficiency (21%) that is 25 times
greater than alcohol 8 (0.9%). More importantly how-
ever, alcohol 10 produced approximately twice the
number of scintillation counts, compared with 8, when
incorporated into liposomal preparations. Scintilipo-
somes incorporating alcohol 10 have been used to
introduce scintilipid 10 into the membrane of living
HeLa cells. These cells have been used successfully in a
cell-based SPA to detect and quantify the uptake of
[14C]methionine.4
In conclusion, we have derivatised the 4-position of 2,5-
diphenyloxazole systematically to generate a series of
scintilipids. It has been demonstrated that through an
iterative process of synthesis, assay and evaluation it has
been possible to develop a ꢀlipid-likeꢁ scintillant-con-
taining molecule that scintillates efficiently in the pres-
ence of ionising radiation either in solution or within
liposomal membranes. Currently we are investigating
further cell-based SPA style assays employing scintilipid
10 and will report our findings in due course.
Long, linear aliphatic carboxylic acids dissolved in a
solvent above their critical micelle concentration fre-
quently assemble into micelles.7 Consequently it was
anticipated that 2,5-diphenyloxazole tagged with an
aliphatic carboxylic acid moiety would be incorporated
readily into liposomes. Attempts to oxidise alcohol 8
directly into the corresponding carboxylic acid 19 using
either Jones reagent (CrO3/AcOH) or KMnO4 generated
a variety of oxidised products. Consequently, a less
vigorous two-step oxidation strategy was employed to
convert alcohol 8 into carboxylic acid 19 successfully via
intermediate aldehyde 15. Contrary to a previous find-
ing that indicated that carboxylic acids quench scintil-
lation significantly,3a carboxylic acid 19 exhibited the
greatest scintillation efficiency (93%) of all the scinti-
lipids evaluated! Unfortunately, incorporation of the
carboxylic acid 19 into various liposomal preparations
produced scintiliposomes that only exhibited compara-
tively low scintillation counts (Table 1).
Acknowledgements
We thank Dr. M. C. Perry and Mrs. K. C. Farrow (both
of Aston University) for NMR spectra and mass spec-
tra, respectively. We are grateful to Aston University for
the award of a Research Studentship to MCM.
References and notes
1. (a) Amersham Biosciences Catalogue 2002; (b) Hertzberg,
R. P.; Pope, A. J. Curr. Opin. Chem. Biol. 2000, 4, 445;
(c) Bertaglio-Matte, J. H. U.S. Patent 4,568,649, 1986;
(d) Hart, H. E.; Greenwald, E. G. Mol. Immunol. 1979, 16,
265.
2. (a) Jessop, R.; Smith, T. Phys. World 1997, 10, 26; (b)
Alcohol 8 was further derivatised by esterification with
butyryl chloride and benzoyl chloride to provide the
corresponding esters 16 (45%) and 17 (52%), respec-
tively. Esters 16 and 17 have scintillation counting effi-
ciencies (17% and 13%, respectively) that are
approximately 20 and 15 times greater than alcohol 8
(0.9%). Unfortunately however, incorporation of either
of these esters into liposomal preparations gave lipo-
somes that exhibited scintillation counts below those
obtained for liposomes incorporating alcohol 8.
ꢀ
€
€
Bonge, H.; Hallen, S.; Fryklund, J.; Sjostrom, J. E. Anal.
Biochem. 2000, 282, 94; (c) Williams, J. B.; Mallorga, P. J.;
Lemaire, W.; Williams, D. L.; Na, S.; Patel, S.; Conn, J. P.;
Pettibone, D. J.; Austin, C.; Sur, C. Anal. Biochem. 2003,
321, 31.
3. (a) Clapham, B.; Richards, A. J.; Wood, M. L.; Sutherland,
A. J. Tetrahedron Lett. 1997, 38, 9061; (b) McCairn, M. C.;
Hine, A. V.; Sutherland, A. J. J. Mater. Chem. 2003, 13,
225.
4. Culliford, S. J.; McCauley, P.; Sutherland, A. J.; McCairn,
M. C.; Sutherland, J.; Blackburn, J.; Kozlowski, R. Z.
Biochem. Biophys. Res. Commun. 2002, 296, 857.
5. (a) Whitney, S. E.; Rickborn, B. J. Org. Chem. 1991, 56,
3058; (b) Clapham, B.; Sutherland, A. J. Tetrahedron Lett.
2000, 2253; (c) Clapham, B.; Sutherland, A. J. J. Org.
Chem. 2001, 66, 9033.
6. Wrobel, I.; Collins, D. Biochim. Biophys. Acta 1995, 1235,
296.
7. Shaw, D. J. Introduction to Colloid and Surface Chemistry,
4th ed.; Butterworth-Heinemann Ltd: Oxford, 1992; pp 84–
93.
Since derivatives of alcohol 8 gave no significant
improvements over using the precursor alcohol, an
attempt was made to investigate an analogue of alcohol
8, which differed only in the length of the alkyl chain
between the hydroxyl group and the oxazole ring.
Accordingly, a sodium hydride mediated Williamson-
type ether synthesis reaction between scintillant alcohol
3 and silylated-8-bromo-1-octanol, followed by TBAF
desilylation of the product, gave alcohol 10, which has