Side-chain hydrophobicity controls the activity of proton channel forming rigid rod-shaped polyols

Article abstract of DOI:10.1039/a708841h

Increased activity, facile incorporation into lipid bilayers and intact active structure and transport selectivity are the consequences of modifications of the side-chain hydro-phobicity of a proton channel-forming octa(p-phenylene).

Full text of DOI:10.1039/a708841h

View Article Online / Journal Homepage / Table of Contents for this issue
Side-chain hydrophobicity controls the activity of proton channel forming rigid
rod-shaped polyols
Chiyou Ni and Stefan Matile*†
Department of Chemistry, Georgetown University, Washington, D.C. 20057-1227, USA
Increased activity, facile incorporation into lipid bilayers
and intact active structure and transport selectivity are the
consequences of modifications of the side-chain hydro-
phobicity of a proton channel-forming octa(p-phenylene).
selectivity.2 It further may enlarge the ionophoric tube between
the rigid-rod scaffold and ‘proton wire’. The additional propyl
group at the terminus of each side-chain in 3 possibly covers the
hydrophilic ‘proton wire’ with an external lipophilic layer.
Syntheses of the octamers 2 and 3 are shown in Scheme 1. For
2, bromide 4 was subjected to Sharpless asymmetric dihydrox-
ylation,7‡ and the resulting diol 5 was converted to acetonide 6.
For 3, the diastereomeric mixture of triol 7 was selectively
tosylated, and protection of the resulting diol 8 afforded 9.
Williamson ether synthesis using 6/9 and the crude octaphenol
prepared by treatment of 10 with BBr3,1 followed by deprotec-
tion of 11/12, yielded the polyols 2/3, respectively. The
corresponding hexamers 13 and 14 were prepared identically
from hexaanisole 15.§
The interaction of the fluorophores 1–3 with lipid bilayers
was examiend using uniformly sized (68 ± 3 nm) EYPC-SUVs
labeled with either 2 mol% 5- or 12-DOXYL-PC.8,2¶ Quench-
ing of ca. 60% of the red-shifted emission of 2 and 3 by
5-DOXYL-PC labeled EYPC-SUVs proved facile incorpora-
tion compared to 1 (Fig. 2). Nearly identical quenching of 1–3
by 5-DOXYL-PC and 12-DOXYL-PC (not shown) confirmed
that variation of side-chain hydrophobicity does not disturb the
preferred transmembrane orientation of 1–3.
Rigid-rod molecule 11,2 represents a promising new class of
non-peptide ion channel models3–5 with unique properties, but
unfortunately it does not incorporate well into lipid bilayers
(Fig. 1). Although the incorporation of artificial ion channels
into lipid bilayers is a general problem, it is often bypassed by
using high temperatures or constructing the entire supramo-
lecular assay system in the presence of the test sample for every
new experiment. Rational design strategies to tune partition
without changing desired properties of a parent model have not
been established so far, in spite of the fact that poor
incorporation may prevent potential biological studies and
pharmaceutical applications. Here we present our efforts to
solve this problem by increasing the hydrophobicity of lateral
side-chains attached to oligo(p-phenylene)s without destructive
effects on active structure or transport selectivity.
In addition to the general aim to increase the lipophilicity of
octamer 1, the following considerations were taken into account
for the design of the new side-chains in 2 and 3 (Fig. 1). We
anticipated that length and flexibility of the propylene spacer in
polyol 2 could facilitate the arrangement of the diols to form the
transmembrane hydrogen-bonded chain6 essential for proton
In sharp contrast to the membrane-spanning octamers 1–3 (34
Å), the side-chain structure governs the organization of the
corresponding hexamers (26 Å) in EYPC-bilayers. Namely, 16
was found at the membrane/water interface (quenching by
12-DOXYL-PC < 5-DOXYL-PC),2 13 in transmembrane
orientation [12-DOXYL-PC (67%) ≈ 5-DOXYL-PC (69%)],
and 14 between the two leaflets of the lipid bilayer
[12-DOXYL-PC (49%) > 5-DOXYL-PC (26%)].
OH
OH OH
OO
OH OH
OO
OH OH
OH
O
OHOH
OHOH
OHOH
OHOH
O
O O
HO
OH
Br
Br
1
7 R = H
8 R = Ts
(e)
(f)
(a)
(b)
4
5
RO
OH
OH OH
OH OH
OH OH
OH
O
OH
O
C
OH
O
OH
OH
OHOH
OHOH
OHOH
A
B
O
O
Br
O
OO
OO
O O
TsO
6
9
D
2
OR
ROOR
ROOR
ROOR
RO
OH
OH OH
OO
OH OH
OH OH
O O
OH
O
10 R = Me
11 R = B
10 R = Me
12 R = D
3 R = C
(g)
(d)
(c)
(d)
OH
OH
OHOH
OHOH
OHOH
2
R = A
O
OO
OR
ROOR
ROOR
RO
3
15 R = Me
17 R = B
13 R = A
15 R = Me
18 R = D
14 R = C
(h)
(d)
(i)
34 Å
(d)
16 R = CH2CHOHCH2OH
Scheme
1 Reagents and conditions: (a) AD-mix-a, 71%; (b)
EYPC bilayer (ca. 36 Å)
2,2-dimethoxypropane, H2SO4, 80%; (c) (1) BBr3; (2) 6, Cs2CO3, 66%; (d)
TFA (quant.); (e) TsCl, pyridine, 65%; (f) 2,2-dimethoxypropane, H2SO4,
83%; (g) (1) BBr3; (2) 9, Cs2CO3, 27%; (h) see (c), 15%; (i) see (g),
27%
Fig. 1 In-scale planar structures of membrane-bound octa(p-phenylene)s
1–3. Structural modifications of the side-chains are emphasized with bold
lines.
Chem. Commun., 1998
755

View Article Online
393
(1–3), accelerated internal pH changes in the presence of 12 nm
valinomycin were observed as reported before for polyol 1 (not
shown).2 Thus, the difference in side-chain lipophilicity of 2
and 3 does not significantly alter the proton selectivity
previously observed for 1.∑
These results demonstrate that the incorporation of proton
channel forming rigid-rod octa(p-phenylene)s into lipid bilayers
can be precisely tuned without significant disturbance of active
structure and transport selectivity. The erratic interactions of
identically modified hexa(p-phenylene)s with lipid bilayers
further corroborate the importance of the length of the rigid-rod
scaffold for controlled, transmembrane binding of substituted
oligo(p-phenylene)s. Ion channel formation of octamers 1–3,
but not hexamer 16 in planar lipid bilayers supports these
conclusions and will be reported in due course.
We thank NIH (GM56147-01), the donors of the Petroleum
Research Fund, administered by the American Chemical
Society, Suntory Institute for Bioorganic Research (SUNBOR
Grant), and Georgetown University for support of this work.
Both authors thank Dr Naomi Sakai for invaluable discussions
and experimental advise.
387
(a)
400
100
80
60
40
20
0
(b)
(c)
(d)
(f)
(e)
350
400
450
500
l / nm
Fig. 2 Relative emission intensities [I/I (lemax) 3 100%, lex = 328 nm) of
5 mm solutions of 1 (a), 2 (b), and 3 (c) with 0.5 mm of unlabeled EYPC-
SUVs, and of 1 (d), 2 (e) and 3 (f) with 5-DOXYL-PC labeled EYPC-SUVs
(100 mm KCl, 100 mm HEPES, pH 7.1)
100
Notes and References
(a)
30
† E-mail: matiles@gusun.georgetown.edu
I
‡ The stereochemistry of the side-chain is with all likelihood irrelevant for
the transport activity of rigid-rod polyols.2 However, judging from the
outcome of osmium-catalyzed asymmetric dihydroxylation of terminal
olefins, formation of the S-enantiomer 5 in 78–97% ee can be expected
using AD-mix-a.7
I
20
(b)
I
(c)
t
I
10
0
§ All final products were purified by reverse-phase HPLC and gave
satisfactory spectroscopic data.
triton X-100
500 600
¶ Abbreviations: AmB: amphotericin B; 5-DOXYL-PC: 1-palmitoyl-
0
100
200
300
t / s
400
2-stearoyl(5-DOXYL)-sn-glycero-3-phosphocholine;
12-DOXYL-PC:
1-palmitoyl-2-stearoyl(12-DOXYL)-sn-glycero-3-phosphocholine; EYPC:
egg yolk phosphatidylcholine; HPTS: 8-hydroxypyrene-1,3,6-trisulfonic
acid; SUV: small unilamellar vesicle.
∑ In all experiments conducted with HPTS, the intensity change was
confirmed to be the consequence of internal pH change by simultaneous
measurement of the time course of emission intensity at 510 nm due to
excitation at 460 nm as well as 405 nm. The contribution of the negative
control was eliminated for the calculation of the initial first-order rate
constants. The effect of valinomycin further corroborates the absence of
HPTS-leakage.2
Fig. 3 Change in fluorescent intensity {[(It–I0)/(IH–I0)] 3 100%, lex = 460
nm, lem = 510 nm} of EYPC-SUV-entrapped HPTS (100 mm KCl, 100 mm
HEPES, pHin = 7.0, pHout = 7.6) as a function of time after the addition of
10 nmoles of octa(p-phenylene)s 1 (c), 2 (b) and 3 (a) in 20 ml MeOH
followed by 40 ml of 1.2% triton X-100¶
The ion transport activity of polyol 1 was originally assessed
in comparison with the structurally related antifungal polyol
amphotericin B (AmB).1 Both mediated intravesicular pH
changes with comparable exchange rates (1 > AmB) when
added to EYPC-SUVs having entrapped pH sensitive fluoro-
1 N. Sakai, K. C. Brennan, L. A. Weiss and S. Matile, J. Am. Chem. Soc.,
1997, 119, 8726.
2 L. A. Weiss, N. Sakai, B. Ghebremariam, C. Ni and S. Matile, J. Am.
Chem. Soc., 1997, 119, 12142.
3 G. W. Gokel and O. Murillo, Acc. Chem. Res., 1996, 29, 425 and citations
in ref. 2.
4 J.-C. Meillon and N. Voyer, Angew. Chem., Int. Ed. Engl., 1997, 36,
967.
phore HPTS and a transmembrane pH gradient.1 The K+/Na+
>
H+ selectivity of AmB was shown by accelerated internal pH
changes in the presence of a selective H+ carrier,9,1 while
similar enhancements induced by the presence of the K+ carrier
valinomycin demonstrated proton selectivity for 1.2
For direct comparison with incorporation efficiencies of
polyols 1–3 (Fig. 2), we conducted the activity measurements
summarized above under the conditions used for fluorescence
quenching, i.e. reduced polyol and increased lipid concentra-
tions compared to previous reports.1,2 Under these conditions,
the transport activity of 1 without additional valinomycin is
5 H. Wagner, K. Harms, U. Koert, S. Meder and G. Boheim, Angew.
Chem., Int. Ed. Engl., 1996, 35, 2643.
6 J. F. Nagle and S. Tristram-Nagle, J. Membr. Biol., 1983, 74, 1.
7 K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A. Crispino, J. Hartung,
K.-S. Jeong, H.-L. Kwong, K. Morikawa, Z.-M. Wang, D. Xu and
X. L. Zhang, J. Org. Chem., 1992, 57, 2768.
8 A. S. Ladokhin, Method Enzymol., 1997, 278, 462.
9 J. Bolard, P. Legrand, F. Heitz and B. Cybulska, Biochemistry, 1991, 30,
5707.
nearly identical with the negative control (k = 9.0 3 1026 s21
,
Fig. 3).∑ Improved incorporation (Fig. 2) resulted in sig-
nificantly increased ion flux rates for octamer 2 (k = 2.2 3
1024 s21) and 3 (k = 2.4 3 1024 s21, Fig. 3). For all octamers
Received in Corvallis, OR, USA, 9th December 1997; 7/08841H
756
Chem. Commun., 1998