Monoacylglycerols as transmembrane Cl- anion transporters

Article abstract of DOI:10.1039/c2cc18148g

We report that the amphiphilic natural product, monoacylglycerol 1, functions as a transmembrane Cl^-/NO₃^- anion transporter. The 1,2-diol group is crucial for the transport function since diacylglycerol and triacylglycerol analogs are not anion transporters. Furthermore, adding another hydrogen bond donor to the glycerol head-group and perfluorination of the acyl tail gave synthetic analogs with improved Cl ^- membrane transport properties.

Full text of DOI:10.1039/c2cc18148g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 4432–4434
www.rsc.org/chemcomm
COMMUNICATION
Monoacylglycerols as transmembrane Clꢀ anion transportersw
Soheila Bahmanjah, Ning Zhang and Jeffery T. Davis*
Received 29th December 2011, Accepted 2nd March 2012
DOI: 10.1039/c2cc18148g
We report that the amphiphilic natural product, monoacyl-
ꢀ
lipid tail might lead to improved anion transporters.
glycerol 1, functions as a transmembrane Clꢀ/NO3 anion
Our initial forays now show that (a) monoacylglycerols
ꢀ
transporter. The 1,2-diol group is crucial for the transport
function since diacylglycerol and triacylglycerol analogs are
not anion transporters. Furthermore, adding another hydrogen
bond donor to the glycerol head-group and perfluorination of the
acyl tail gave synthetic analogs with improved Clꢀ membrane
transport properties.
such as 1a and 1b transport Clꢀ and NO3 anions across
liposomes and (b) an additional hydrogen bond donor in the
head-group and fluorination of the tail further improves this
ion exchange activity relative to the parent monoacylglycerols
1a,b.
To determine whether the 1,2-diol is functionally important
we first evaluated whether acylglycerols 1b, 2 and 3 could
transport Clꢀ and NO3ꢀ anions across bilayers. Because Clꢀ is
one of the few biologically relevant anions (along with
HCO3ꢀ) we decided to restrict our initial studies in this
communication to evaluation of Clꢀ anion transport. We
prepared liposomes of ca. 100 nm diameter from egg-yolk
phosphatidylcholine (EYPC) that contained 1 mol% of com-
pounds 1–3 pre-incorporated into the vesicles. The liposomes
were filled with NaNO3 (225 mM, pH 7.4 with 20 mM
HEPES) and 1 mM of lucigenin (LG), a chloride-selective
fluorescent dye.9 A transmembrane NO3ꢀ/Clꢀ gradient was
induced by adding NaCl to the extravesicular solution.
Transporter-mediated influx of Clꢀ into the liposomes was
signaled by a decrease in LG fluorescence. Dynamic light
scattering measurements confirmed that the liposomes
remained intact and did not change in size upon addition of
the extravesicular NaCl (ESI Table 1w). As shown in Fig. 1,
only monoacylglycerol 1b promoted Clꢀ influx, whereas
diacylglycerol 2 and triacylglycerol 3 did not have such Clꢀ
transport ability. Apparently, the 1,2-diol group is needed for
function. This is the first report, to our knowledge, that this
class of natural products can promote anion transport across
phospholipid bilayers.10
Over the past decade, there has been a growing interest
in small molecules that facilitate transmembrane anion trans-
port, particularly for Clꢀ anion, and many synthetic com-
pounds have now been shown to carry out that function.1 This
interest in synthetic chloride transporters is motivated, in large
part, by the realization that misregulation of chloride trans-
port across cell membranes can lead to diseases.2 In contrast
to the array of synthetic anion transporters,1 only a few
natural products, such as the prodigiosins3,4 and the sphingo-
lipid ceramide,5,6 have been so far characterized as trans-
membrane anion transporters. With an eye toward identifying
other natural products that might catalyze anion transport
we considered monoacylglycerols as candidates. We reasoned
that the 1,2-diol in the head-group of monoacylglycerol
1 should provide a good anion-binding site and that the
C3-acyl chain should provide the lipophilicity needed
to function in a phospholipid membrane.7,8 We also envi-
sioned that subsequent modifications of the head-group and
We next sought to improve the anion binding and transport
properties of the parent acylglycerol 1a by preparing the
3-amido-1,2-diols 4a and 5. We reasoned that substituting
an amide NH at the glycerol C3 position, next to the 1,2-diol,
would make amide 4a a better anion binder, and possibly a
better transporter, than ester 1a.11 We made perfluoroamide 5
to examine the effect of attaching a perfluorinated acyl chain
to the anion-binding head-group, given the dramatic influence
that fluorination often has on the membrane activity of other
amphiphiles.12,13
Negative-mode ESI-MS analysis of a solution of the
3-amido-1,2-propanediol 4a and tetrabutylammonium chloride
(TBACl) showed a strong signal for the 4aꢁClꢀ adduct (m/z =
252.137), whereas the ester 1a did not show such an adduct
with Clꢀ anion.14 NMR spectroscopy provided molecular-level
Department of Chemistry and Biochemistry, University of Maryland,
College Park, MD 20742, USA. E-mail: jdavis@umd.edu;
Fax: 001-301-314-9121; Tel: 001-301-405-1845
w Electronic supplementary information (ESI) available: Experimental
details, synthetic procedures, representative spectra and NMR binding
data. See DOI: 10.1039/c2cc18148g
c
4432 Chem. Commun., 2012, 48, 4432–4434
This journal is The Royal Society of Chemistry 2012

View Article Online
in non-polar solution and the anion transport across a hydro-
phobic membrane.
Having found that both amide analogs 4a and 5 bind Clꢀ
better than the parent ester 1a, we next compared the trans-
membrane anion transport for these three analogs. Again, we
used liposomes made from EYPC (d = 100 nm) that now
contained 0.05 mol% of the transporters 1a, 4a or 5 pre-
incorporated into the phospholipid vesicles.17 Typical traces
for LG quenching, obtained after adding NaCl to these
nitrate-filled liposomes, are shown in Fig. 2. Once again,
dynamic light scattering experiments confirmed that the
addition of extravesicular NaCl solution did not alter the
stability or structure of vesicles containing compounds 1a,
4a or 5 (see ESI Table 1 and ESI Fig. S28w). This transport
data in Fig. 2 indicates that replacing the ester in the head-
group of 1a with a secondary amide (1a vs. 4a) enhances anion
transport.18 Further, Fig. 2 also shows that perfluorination of
the amide’s acyl tail further increased the transport activity of
5 relative to its unfluorinated analog 4a under these condi-
tions. To gain further insight into the transport mechanism we
also monitored whether these compounds could facilitate Clꢀ
influx when the liposomes were filled with the doubly charged
SO42ꢀ anions rather than with NO3ꢀ. This is a fairly common
test used to probe whether a transporter functions by catalyzing
anion exchange.13,16 After addition of NaCl to sulfate-filled
EYPC liposomes, containing either pre-incorporated 1b, 4b or
5 at concentrations up to 5 mol% (100 mM Na2SO4, 1 mM
LG, 20 mM HEPES, pH 7.4), we saw no Clꢀ influx, above
background, in any of the experiments (ESI Fig. S23w).
Since sulfate has a much higher dehydration energy (DG =
ꢀ1080 kJ molꢀ1) than both Clꢀ (DG = ꢀ340 kJ molꢀ1) and
Fig. 1 Chloride influx promoted by 1 mol% of monoacylglycerol 1b
(blue), diacylglycerol 2 (red), triacylglycerol 3 (green) incorporated
into EYPC liposomes, filled with 225 mM NaNO3 and buffered to
pH 7.4 with 20 mM HEPES buffer. At t = 20 s NaCl solution was
added (225 mM).
Table 1 Clꢀ binding constants, EC50 values, Hill coefficients (n) for
chloride transport and log P values for monoacylglycerol analogs 1a,
4a and 5
Ester 1a
Amide 4aa
Perfluoro 5
Ka (Mꢀ1) CDCl3
Ka (Mꢀ1) CD3CN
EC50 270 s (mM)
Hill coefficient
c log P
92 ꢂ 1.7
—
0.195 ꢂ 0.01
1.45 ꢂ 0.17
1.64
1474 ꢂ 32
137 ꢂ 1.8
0.078 ꢂ .002
1.12 ꢂ 0.045
0.97
—
193 ꢂ 7.6
0.011 ꢂ 0.002
0.83 ꢂ 0.20
2.99
a
Hill analysis of the C18 amide 4b showed that the length of the acyl
chain in 4a (C8) vs. 4b (C18) had little influence on transport efficiency,
as 4b had an EC50 value of 0.076 ꢂ 0.004 (see ESI Fig. S26).
NO3 (DG = ꢀ300 kJ molꢀ1),19 it is far more difficult to
ꢀ
details about how compounds 1a, 4a and 5 bind Clꢀ in non-
polar solvents such as CDCl3 and CD3CN. Job plots showed
that each analog formed a 1 : 1 complex with TBACl in CDCl3
or CD3CN solution. Changes in chemical shifts of the NH
and/or OH protons upon titration of TBACl into CDCl3
or CD3CN solutions of ester 1a, amide 4a and perfluoro-
amide 5 confirmed that the acidic hydrogen’s in the glycerol
head-group bind Clꢀ. In CDCl3, the two OH resonances for
monoacylglycerol 1a appeared together as a broad signal.
Importantly, the NMR signals for these two OH protons
separated upon addition of TBACl (Dd = 2.53 and 2.21).
Similarly, addition of TBACl to a solution of the amide 4a
in CDCl3 resulted in significant downfield shifts for the NH
(Dd = 2.91) and both OH signals (Dd = 2.39 and 2.82,
respectively), consistent with hydrogen bonding between
chloride and these three acidic hydrogen atoms. The change
in chemical shift of NH and OH signals was fit to a 1 : 1
binding model to give values of Ka = 1474 Mꢀ1 for amide 4a
and Ka = 92 Mꢀ1 for ester 1a.15 These comparative NMR
data demonstrate that amide 4a binds chloride significantly
better than does ester 1a in non-polar solution. Data obtained
in CD3CN for 4a (137 Mꢀ1) and 5 (193 Mꢀ1) indicated
that perfluorination of the acyl chain had a relatively modest
effect on the receptor’s Clꢀ binding strength, but showed
that perfluorination of the chain did increase the affinity
for Clꢀ in solution. Oftentimes,13,16 there is a positive
correlation between a compound’s affinity for binding anions
dehydrate and transport the doubly-charged sulfate than a
singly charged anion across a hydrophobic bilayer. That Clꢀꢀ
transport was observed only with a transmembrane Clꢀ/NO3
gradient, but did not occur with intravesicular SO42ꢀ, indicates
that compounds 1b, 4b and 5 all function by promoting
transmembrane anion exchange.20
We next quantified the transport abilities of 1a, 4a and 5
by performing concentration-dependent Hill analyses for
Fig. 2 Chloride influx promoted by 0.05 mol% of ester 1a (red),
amide 4a (green) and fluorinated amide 5 (purple) incorporated into
EYPC vesicles (d = 100 nm), loaded with 225 mM NaNO3 and
buffered to pH 7.4 with 20 mM HEPES buffer. At t = 20 s NaCl
solution was added (225 mM). At the end of the experiment, Triton
detergent was added to lyse the vesicles.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 4432–4434 4433

View Article Online
transmembrane Clꢀ/NO3 exchange.21 This enabled us to
3 Prodigiosins: T. Sato, H. Konno, Y. Tanaka, T. Kataoka,
K. Nagai, H. H. Wasserman and S. Ohkuma, J. Biol. Chem.,
1998, 273, 21455.
ꢀ
calculate EC50 values, defined as the concentration of trans-
porter needed to achieve 50% of maximal chloride influx at
t = 270 s (Table 1).22 The relatively low EC50 values for the
perfluoro amide 5 confirms that combined addition of a
NH hydrogen bond donor and fluorination of the 30-acyl
chain has a significant impact on transport efficiency, since 5
(0.011 mM) has an EC50 value that is about 20-fold lower
than the parent monoacylglycerol 1a (0.195 mM). The Hill
coefficients that are close to 1 for both 4a and 5 suggest that
the active species in anion transport for 4a and 5 are
monomers.21 Calculation of log P values (clog P) show that
perfluorination of the acyl tail increased the hydrophobicity of
compound 5 by 100-fold, relative to its analog amide 4a
(Table 1).23 This enhanced hydrophobicity, and the known
incompatibility of mixing for fluorinated and alkyl chains,
may well contribute to the faster anion transport rates shown
by perfluoro-amide 5.24 Certainly, one wonders whether
the increased permeability of Clꢀ in the presence of
perfluoro-amide 5 is due to increased flip-flop rates for the
perfluorinated analogs relatively to the unfluorinated analog
4a. In future studies on these compounds we will focus on such
mechanistic issues and we will also try to determine how
compound 5 might be distributed in the bilayer membrane
of the liposomes.
4 For transmembrane Clꢀ anion transport properties of synthetic
prodigiosin: (a) R. I. Saez Dıaz, J. Regourd, P. V. Santacroce,
´ ´
J. T. Davis, D. L. Jakeman and A. Thompson, Chem. Commun.,
2007, 2701; (b) P. I. Hernandez, D. Moreno, A. A. Javier,
T. Torroba, R. Perez-Tomas and R. Quesada, Chem. Commun.,
2012, 48, 1556.
5 W. A. Harrell, M. L. Bergmeyer, P. Y. Zavalij and J. T. Davis,
Chem. Commun., 2010, 46, 3950.
6 R. Pajewski, N. Djedovic, E. Harder, R. Ferdani, P. H. Schlesinger
and G. W. Gokel, Bioorg. Med. Chem., 2004, 13, 29.
7 S. K. Berezin and J. T. Davis, J. Am. Chem. Soc., 2009, 131,
2458.
8 (a) A. Ganem-Quintanaar, D. Quintanaar-Guerrero and P. Buri,
Drug Dev. Ind. Pharm., 2000, 26, 809; (b) C. V. Kulkarni,
W. Wachter, G. Iglesias-Salto, S. Engelskirchen and S. Ahualli,
Phys. Chem. Chem. Phys., 2011, 13, 3004.
9 B. A. McNally, A. V. Koulov, B. D. Smith, J. B. Joos and
A. P. Davis, Chem. Commun., 2005, 1087.
10 Szostak, Mansy and colleagues have proposed that monoacyl-
glycerides can facilitate the transmembrane transport of nucleotides:
S. S. Mansy, J. P. Schrum, M. Krishnamurthy, S. Tobe, D. A. Treco
´
and J. W. Szostak, Nature, 2008, 454, 122.
11 Hydrogen bonding of glycerol esters and amides influence their
intermolecular interactions: (a) U. Gehlert, G. Weidemann,
D. Vollhardt, G. Brezesinski, R. Wagner and H. Mohwald,
¨
Langmuir, 1998, 14, 2112; (b) K. Thirumoorthy, N. Nandi and
D. Vollhardt, J. Phys. Chem. B, 2005, 109, 10820.
12 (a) M. P. Krafft, Adv. Drug Delivery Rev., 2001, 47, 209;
(b) L. Dafik, V. Kalsani, A. K. L. Leung and K. Kumar, J. Am.
Chem. Soc., 2009, 131, 12091; (c) G. Godeau, H. Arnion, C. Brun,
C. Staedel and P. Barthelemy, Med. Chem. Commun., 2010,
1, 76.
13 P. A. Gale, C. C. tong, O. Adeosun, D. E. Gross, E. Karnas,
E. M. Sedenburg, R. Quesada and J. L. Sessler, J. Am. Chem. Soc.,
2010, 132, 3240.
14 ESI-MS indicated that glycerol amide 4a outcompetes ester 1a for
binding to Clꢀ anion.
15 Binding constants were determined using WINEQNMR2 software
assuming 1 : 1 binding: M. J. Hynes, J. Chem. Soc., Dalton Trans.,
1993, 311.
16 B. A. McNally, A. V. Koulov, T. N. Lambert, B. D. Smith,
J. B. Joos, A. L. Sisson, J. P. Clare, V. Sgarlata, L. W. Judd,
G. Magro and A. P. Davis, Chem.–Eur. J., 2008, 14, 9599.
17 Chloride transport activity was sometimes, but not usually,
observed when aliquots of MeOH solutions of 1a or 4a were added
to aqeous solutions of EYPC liposomes.
18 Transport experiments with the enantiomeric N-tetradecanoyl-(R)-
3-amino-1,2-propanediol (R)-6 and N-tetradecanoyl-(S)-3-amino-
1,2-propanediol (S)-6 showed no significant differences (see ESI
Fig. S22w).
19 Y. Marcus, J. Chem. Soc., Faraday Trans., 1991, 87, 2995.
20 That there was no increase in the lucigenin fluorescence in the
presence of intravesicular sulfate also indicates that compounds 1a,
4a and 5 do not rupture the vesicles.
21 S. Bhosale and S. Matile, Chirality, 2006, 18, 849.
22 N. J. Andrews, C. J. E. Haynes, M. E. Light, S. J. Moore,
C. C. Tong, J. T. Davis, W. A. Harrell and P. A. Gale, Chem.
Sci., 2011, 2(2), 256–260.
In this study, we have found that natural product mono-
acylglycerols facilitate the transmembrane transport of
chloride anion across phospholipid bilayers. One critical
structural feature that enables anion binding and transport
is the 1,2-diol in the glycerol head-group. Furthermore,
synthetic permutations to both the head-group and the acyl
tail gave new glycerol-based amphiphiles that showed
enhanced rates of anion transport. Thus, adding another
hydrogen bond donor to the monoacylglycerol’s head-
group and increasing the lipophilicity of the acyl tail by
perfluorination gave a synthetic analog 5 that is almost
20-fold better at transporting chloride than the parent
monoacylglycerol 1a. Our results raise the possibility
that naturally occurring monoacylglycerides may help
regulate transmembrane concentrations of physiologically
essential anions such as Clꢀ. Furthermore, we have shown
that rational synthetic modifications to these natural products
can provide leads toward new and improved anion
transporters.
We thank the NSF (CHE-0822244) and DOE (DE-FG02-
98ER14888) for financial support.
23 The clog P values were calculated with Spartan ’08 for Macintosh;
Wavefunction Inc.: Irvine, CA using the Ghose-Crippen model:
A. K. Ghose, A. Pritchett and G. M. Crippen, J. Comput. Chem.,
1988, 9, 80–90.
24 N. Busschaert, M. Wenzel, M. E. Light, P. Iglesias-Hernandez,
R. Perez-Tomas and P. A. Gale, J. Am. Chem. Soc., 2011,
133, 14136.
Notes and references
1 Reviews: (a) A. P. Davis, D. N. Sheppard and B. D. Smith, Chem.
Soc. Rev., 2007, 36, 348; (b) J. T. Davis, O. Okunola and
R. Quesada, Chem. Soc. Rev., 2010, 39, 3843.
2 F. M. Ashcroft, Ion Channels and Disease, Academic Press,
San Diego and London, 2000.
c
4434 Chem. Commun., 2012, 48, 4432–4434
This journal is The Royal Society of Chemistry 2012