Cationic cyclocholamides; toroidal facial amphiphiles with potential for anion transport

Article abstract of DOI:10.1039/b805777j

Cholic acid has been transformed into cyclotrimeric and cyclotetrameric toroidal amphiphiles with inward-directed ammonium substituents; the cyclotrimer 3b effects the transport of chloride anions across vesicle bilayer membranes. The Royal Society of Che

Full text of DOI:10.1039/b805777j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Cationic cyclocholamides; toroidal facial amphiphiles with potential
for anion transportw
Samuel D. Whitmarsh, Adrian P. Redmond, Valentina Sgarlata and Anthony P. Davis*
Received (in Cambridge, UK) 7th April 2008, Accepted 8th May 2008
First published as an Advance Article on the web 17th June 2008
DOI: 10.1039/b805777j
Cholic acid has been transformed into cyclotrimeric and cyclo-
tetrameric toroidal amphiphiles with inward-directed ammonium
substituents; the cyclotrimer 3b effects the transport of chloride
anions across vesicle bilayer membranes.
small cavity) while allowing cyclotrimerisation and the forma-
tion of higher oligomers. The benzyl chromophores would also
aid separation by HPLC.
Amphiphiles are molecules with both hydrophilic and hydro-
phobic moieties. The more common amphiphiles possess small
polar head-groups and long hydrocarbon tails, but other
geometries are possible. For example, there is widespread
interest in ‘‘facial amphiphiles’’ (FAs),1,2 rigid molecules in
which one face is lipophilic and the other polar. These systems
have potential applications in controlled self-assembly,1b,2a–d
anti-microbial chemotherapy,1d,e,2e–g gene delivery,1f,g vesicle
fusion1g and the stabilisation of membrane proteins.1h We
have previously shown that cholic acid 1 can be converted to
triamino esters which, after tris-protonation, exist as the
cationic, high-contrast facial amphiphiles 2.1g,3 We have now
used 2 to extend the FA concept, by preparing macrocyclic
oligomers (cyclocholamides)4 3. These ‘‘toroidal facial amphi-
philes’’ combine lipophilic exteriors with highly hydrophilic,
polycationic interiors, and show potential for anion transport
across phospholipid membranes.
Steroidal azide 5 was available as starting material via our
previously reported large-scale procedure.6 As shown in
Scheme 1,w the N-protection was first adjusted by installing
Cbz at positions 7 and 12, and Boc at position 3. Ester
hydrolysis and treatment with pentafluorophenol–DIC gave
activated ester 7. The Boc group was removed with TFA, and
the resulting salt was added slowly to DMAP in THF (final
concentration 0.8 mM). Under these high dilution conditions
cyclisation took place to give a mixture of 8b–d,7 separable by
HPLC. Typical yields were 34%, 16% and 2%, respectively.
Removal of the Cbz groups by hydrogenolysis proved difficult,
probably due to steric congestion, but cleavage with HBr–
AcOH proceeded smoothly. Toroidal facial amphiphiles 3bꢀ
6Brꢁ and 3cꢀ8Brꢁ were produced in almost quantitative yields.
The geometries and conformational properties of 3b and 3c
were assessed using molecular modelling (see ESIw). Both
macrocycles possess flexibility due to the steroidal C20–C24
side chain (for numbering, see 4). However, the calculations
Our synthetic approach to 3 involved cyclo-oligomerisation
of intermediate 4 at high dilution, followed by 7,12-N-depro-
tection. Molecular modelling and precedent5 suggested that
control over ring size should be possible through choice of the
7,12-protecting groups. By using the moderately bulky Cbz,
we hoped to avoid dimerisation to 3a (of less interest due to its
School of Chemistry, University of Bristol, Cantock’s Close, Bristol,
UK BS8 1TS. E-mail: Anthony.Davis@bristol.ac.uk
w Electronic supplementary information (ESI) available: Further de-
tails concerning the syntheses of macrocycles 3, molecular modelling
and electron microscopy of 3b/c, and studies of ion transport by 3b.
See DOI: 10.1039/b805777j
suggested that cyclotrimer 3b is strongly biased towards con-
1
formations with inward directed NH3 . This may be
ꢂc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 3669–3671 | 3669

View Article Online
distances.w The global minimum was a C3-symmetric structure
with a cavity B9 A in diameter (see Fig. 1a). Cyclotetramer 3c
was found to possess greater freedom of movement, but open
1
structures with inward-directed NH3 predominated at low
energies (see Fig. 1b). The global minimum structure sur-
rounded a cavity with an average internal diameter of B14 A.
Amphiphile geometry has a major influence on self-assembly
in aqueous media.2a The aggregation properties of 3bꢀ6Brꢁ and
3cꢀ8Brꢁ were studied in water–methanol mixtures. Both macro-
cycles were freely soluble in methanol and gave well-resolved 1H
NMR spectra in CD3OD, suggesting minimal aggregation.
Addition of water caused no visible effect up to H2O–MeOH
(2 : 1), but when the resulting mixtures were left to stand
(presumably with loss of some MeOH) the solutions turned
cloudy. The aggregates were analysed by transmission electron
microscopy (TEM). Cyclotrimer 3bꢀ6Brꢁ produced elongated
cone-shaped structures, often nested or clustered (see Fig. 2).
The origin of these shapes is obscure, but they demonstrate that
the toroidal FA architecture can lead to distinctive morpholo-
gies. Cyclotetramer 3cꢀ8Brꢁ yielded simple spheres,w possibly
reflecting its greater flexibility and, correspondingly, a lesser
inclination towards organised self-assembly.
Scheme 1 Synthetic route to macrocycles 3. Reagents and conditions:
(i) TFA, DCM; (ii) BnOCOCl, NaHCO3 aq., THF; (iii) PMe3, THF,
then H2O; (iv) (Boc)2O, NaHCO3 aq., THF; (v) NaOH, MeOH, H2O;
(vi) C6F5OH, N,N0-diisopropylcarbodiimide (DIC), THF; (vii) TFA,
DCM; (viii) DMAP, THF, high dilution; (ix) HBr, AcOH.
Toroidal FAs with hydrophobic exteriors have potential for
membrane transport. Locating within the bilayer, they can
self-assemble to form channels or act as carriers for polar
species. In the case of 3, the cationic interiors should give
selectivity for anionic substrates.8 To test this possibility we
studied the ability of 3b to promote chloride efflux from
unilamellar vesicles, using the chloride electrode method pre-
viously employed for neutral steroid-based transporters.9 In a
typical experiment, vesicles were formed from egg yolk phos-
phatidylcholine (EYPC) and aqueous NaCl (500 mM) by
extrusion through a filter membrane (pore size 0.2 mm).
External NaCl was exchanged for NaNO3 (500 mM) by
dialysis. Macrocycle 3b was added as the hexakis-trifluoro-
acetate salt (final concentration 11 mM), and the solution
external to the vesicles was monitored using a chloride-selec-
tive electrode. The experiment was terminated by addition of
detergent (octaethylene glycol monodecyl ether) which
1
attributed to repulsion between NH3 groups; conformations
1
with outward-directed NH3 possess smaller inter-cation
Fig. 2 Aggregates formed from 3bꢀ6Brꢁ in methanol–water, as
observed by TEM. The same assembly is shown at two levels of
magnification. These elongated, cone-shaped structures were found
throughout the sample, sometimes singly and sometimes clustered as
shown.
Fig. 1 Global minima for (a) cyclotrimer 3b, and (b) cyclotetramer 3c
as predicted by Monte Carlo molecular mechanics [Macromodel 9.1,
MMFFs force field, GB/SA (water) continuum solvation].
ꢂc
This journal is The Royal Society of Chemistry 2008
3670 | Chem. Commun., 2008, 3669–3671

View Article Online
liberated all remaining chloride. The results showed that the
addition of 3b initiated chloride efflux, and that ca. 80% of the
chloride was released within 5 minutes.w
to membrane-based applications, and we have shown that
cyclotrimer 3b can transport chloride ions across phospholipid
bilayers. There is potential for the recognition and transport of
other anionic species, and this will be the subject of future
research.
Further experiments with 3b revealed that the rate of trans-
port increased roughly linearly with concentration, and that the
use of less fluid phospholipid–cholesterol (7 : 3) vesicle mem-
branes lowered transport rates.w Both results tend to suggest
that transport is due to a carrier mechanism, as opposed to a
static, multicomponent channel. Addition of phosphate buffer
(pH ¼ 6.5–8) produced a further lowering of transport rates,
possibly due to competition for the binding site. Monomeric
control compounds 9 were tested to confirm the importance of
the toroidal architecture. Neither promoted chloride transport
to a measurable extent, suggesting that the macrocyclic structure
is necessary to shield the chloride from the membrane hydro-
carbon. 3b was also tested for bromide transport, by replacing
the KCl in the vesicles by KBr and employing a bromide-
selective electrode. Interestingly, rates were lower by a factor
of B2.4 at steady state,w perhaps due to slow release of the
lipophilic bromide anion. Finally, anion vs. cation selectivity was
tested by encapsulating KCl in the vesicles, suspending in
NaNO3, adding 3b, then following both Clꢁ and K1 efflux
using appropriate ion-selective electrodes. While Clꢁ emerged
from the vesicles as expected, the efflux of K1 was negligible
(Fig. 3). This experiment confirms the expected anion-selectivity,
and also shows that the macrocycle does not disrupt the vesicles.
Financial support from the BBSRC (BBS/B/11044), the
EPSRC (EP/C528859/1) and the University of Bristol is grate-
fully acknowledged. Dr Sean A. Davis and Jon A. Jones are
thanked for help with TEM imaging.
Notes and references
1 Facial amphiphiles based on steroids: (a) Y. Cheng, D. M. Ho, C.
R. Gottlieb, D. Kahne and M. A. Bruck, J. Am. Chem. Soc., 1992,
114, 7319; (b) U. Maitra, S. Mukhopadhyay, A. Sarkar, P. Rao
and S. S. Indi, Angew. Chem., Int. Ed., 2001, 40, 2281; (c) Z. Q.
Zhong, J. Yan and Y. Zhao, Langmuir, 2005, 21, 6235; (d) B. W.
Ding, N. Yin, Y. Liu, J. Cardenas-Garcia, R. Evanson, T. Cirsak,
M. J. Fan, G. Turin and P. B. Savage, J. Am. Chem. Soc., 2004,
126, 13642; (e) H. M. Willemen, L. de Smet, A. Koudijs, M. C. A.
Stuart, I. Heikamp-de Jong, A. T. M. Marcelis and E. J. R.
Sudholter, Angew. Chem., Int. Ed., 2002, 41, 4275; (f) S. Walker,
M. J. Sofia, R. Kakarla, N. A. Kogan, L. Wierichs, C. B. Longley,
K. Bruker, H. R. Axelrod, S. Midha, S. Babu and D. Kahne, Proc.
Natl. Acad. Sci. U. S. A., 1996, 93, 1585; (g) Y. R. Vandenburg, B.
D. Smith, M. N. Perez-Payan and A. P. Davis, J. Am. Chem. Soc.,
´ ´
2000, 122, 3252; (h) Q. H. Zhang, X. Q. Ma, A. Ward, W. X. Hong,
V. P. Jaakola, R. C. Stevens, M. G. Finn and G. Chang, Angew.
Chem., Int. Ed., 2007, 46, 7023.
2 Facial amphiphiles based on other scaffolds: (a) T. M. Stein and S.
H. Gellman, J. Am. Chem. Soc., 1992, 114, 3943; (b) J. Elemans, R.
R. J. Slangen, A. E. Rowan and R. J. M. Nolte, J. Org. Chem.,
2003, 68, 9040; (c) F. M. Menger and J. L. Sorrells, J. Am. Chem.
Soc., 2006, 128, 4960; (d) D. J. Hong, E. Lee and M. Lee, Chem.
Commun., 2007, 1801; (e) E. A. Porter, B. Weisblum and S. H.
Gellman, J. Am. Chem. Soc., 2002, 124, 7324; (f) D. H. Liu, S.
Choi, B. Chen, R. J. Doerksen, D. J. Clements, J. D. Winkler, M.
L. Klein and W. F. DeGrado, Angew. Chem., Int. Ed., 2004, 43,
1158; (g) Y. R. Vandenburg, B. D. Smith, E. Biron and N. Voyer,
Chem. Commun., 2002, 1694.
3 (a) S. Broderick, A. P. Davis and R. P. Williams, Tetrahedron
Lett., 1998, 39, 6083; (b) A. P. Davis and M. N. Pe
Synlett, 1999, 991; (c) V. del Amo, K. Bhattarai, M. Nissinen, K.
Rissanen, M. N. Perez-Payan and A. P. Davis, Synlett, 2005, 1319.
4 We have used this term for bile acid cyclooligomers linked by
annular amides. See, for example, A. P. Davis and J. J. Walsh,
Chem. Commun., 1996, 449.
5 R. P. Bonar-Law and J. K. M. Sanders, Tetrahedron Lett., 1992,
33, 2071; P. A. Brady, R. P. Bonar-Law, S. J. Rowan, C. J.
Suckling and J. K. M. Sanders, Chem. Commun., 1996, 319.
6 V. del Amo, L. Siracusa, T. Markidis, B. Baragana, K. M.
rez-Payan,
´ ´
In conclusion we have shown that cholic acid 1 can be used
to prepare a new class of amphiphilic molecules, by enhancing
facial amphiphilicity (converting –OH to –NH31) then cyclo-
oligomerising. The resulting systems 3 are toroidal facial
amphiphiles with hydrophobic outer surfaces and strongly
hydrophilic interiors. This combination of properties points
´
´
Bhattarai, M. Galobardes, G. Naredo, M. N. Perez-Payan and
´ ´
A. P. Davis, Org. Biomol. Chem., 2004, 2, 3320.
7 Some experiments also yielded a small amount (ca. 2%) of 8a,
identified by ESMS.
8 For a review of synthetic anion transporters, see: A. P. Davis, D.
N. Sheppard and B. D. Smith, Chem. Soc. Rev., 2007, 36, 348.
Recent examples: V. Gorteau, G. Bollot, J. Mareda, A. Perez-
Velasco and S. Matile, J. Am. Chem. Soc., 2006, 128, 14788; P. V.
Santacroce, J. T. Davis, M. E. Light, P. A. Gale, J. C. Iglesias-
Sanchez, P. Prados and R. Quesada, J. Am. Chem. Soc., 2007, 129,
1886; X. Li, B. Shen, X. Q. Yao and D. Yang, J. Am. Chem. Soc.,
2007, 129, 7264; L. You, R. Ferdani, R. Q. Li, J. P. Kramer, R. E.
K. Winter and G. W. Gokel, Chem.–Eur. J., 2008, 14, 382.
9 A. V. Koulov, T. N. Lambert, R. Shukla, M. Jain, J. M. Boon, B.
D. Smith, H. Y. Li, D. N. Sheppard, J. B. Joos, J. P. Clare and A.
P. Davis, Angew. Chem., Int. Ed., 2003, 42, 4931. See also P. H.
Schlesinger, R. Ferdani, J. Liu, J. Pajewska, R. Pajewski, M. Saito,
H. Shabany and G. W. Gokel, J. Am. Chem. Soc., 2002, 124, 1848.
10 Chloride efflux was slower than in some other experiments, due to
the presence of buffer and the incorporation of cholesterol in the
membranes (see discussion in text, and Fig. S7 and S8 in the ESIw).
ꢁ
Fig. 3 Anion-selective transport by 3bꢀ(CF3CO2
)
6
through vesicle
3. Inside: KCl
membranes.w10 Vesicles: EYPC–cholesterol,
7
:
(500 mM)–phosphate buffer (pH ¼ 7, 10 mM). Outside: NaNO3 (500
mM)–phosphate buffer. Changes to Clꢁ and K1 external concentrations
were monitored simultaneously using ion-selective electrodes.
ꢂc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 3669–3671 | 3671