Metal-organic transmembrane nanopores

Article abstract of DOI:10.1021/ja310425j

A stable tetraporphyrin metallacycle with Re(I) corners (1) is capable of forming nanopores in a liposomial membrane, provided that the porphyrin units are properly functionalized with peripheral carboxylic acid residues that, by establishing an hydrogen bond network, allow the formation of dimers that span the depth of the membrane.

Full text of DOI:10.1021/ja310425j

Communication
pubs.acs.org/JACS
Metal−Organic Transmembrane Nanopores
Mariangela Boccalon, Elisabetta Iengo,* and Paolo Tecilla*
Department of Chemical and Pharmaceutical Sciences, University of Trieste, via Giorgieri 1, I-34127, Trieste, Italy
S
* Supporting Information
work.11 However, coordination chemistry has been used only
occasionally for the design of synthetic ion channels and pores
and there are very few reports on the application of the so
called “Fujita-Stang” motif in this field.12 With the exception of
the copper based metal−organic polyedra reported by Kim,13 in
all the systems so far investigated by the groups of T. M. Fyles
and S. J. Webb, the metal ion used is Pd(II), which gives
kinetically labile complexes.14 Usually, this feature ensures that
the thermodynamically most stable adduct is formed, but in the
complex membrane environment, in which the concentration of
the different species is influenced by their partition between
bulk water and membrane, this may lead to complex mixtures
of species with different stoichiometry. For example, recently,
Webb and co-workers reported ionophoric activity in the
presence of a bis(meso-3-pyridyl)porphyrin and Pd(II) in
liposomes.14b However, it was impossible to identify the active
species among the different linear and cyclic adducts that may
form upon self-assembly of the porphyrin with the palladium
center.
On this ground, and taking inspiration from the work of
Kobuke, we decided to investigate the formation of trans-
membrane nanopores using stable metal-mediated macrocycles
of porphyrins. In our design, we chose a 10,20-meso-
dipyridylporphyrin, a linear ligand that upon binding to a cis-
coordinating metal fragment forms a 4 + 4 metallacycle about 2
nm in size. As corners we used low spin d6 fac-{ReIBr(CO)3}
metal fragments that are known to form kinetically inert and
thermodynamically stable bonds with pyridyl ligands.15 The
neutral metallacycle is therefore expected to be robust enough
as to preserve its structure intact also in the phospholipid
membrane, avoiding ligand exchange or scrambling. Moreover,
the bulky substituents on the two other meso positions of each
porphyrin hinder the rotation around the metal−pyridine bond
ensuring an average parallel disposition of the four tetrapyrrolic
rings, thus roughly defining an empty cube that spans ca. half of
the thickness of the phospholipid bilayer. Therefore, to
promote an efficient dimerization of the metallacycle and,
thus, the formation of a robust transmembrane nanopore, each
porphyrin unit was equipped with two peripheral 4-
carboxyphenyl groups at the 5,15-meso positions. In the
resulting tetraporphyrin metallacycle 1, four carboxylic groups
should be pointing upward and four downward, therefore,
assisting a hydrogen-bonding driven dimerization of the
metallacycle once it is inserted in the lipophilic membrane
(Figure 1). Here, we report on the synthesis, characterization,
and ionophoric properties of 1 in comparison with its methyl-
ester analogue 2 (Scheme 1).
ABSTRACT: A stable tetraporphyrin metallacycle with
Re(I) corners (1) is capable of forming nanopores in a
liposomial membrane, provided that the porphyrin units
are properly functionalized with peripheral carboxylic acid
residues that, by establishing an hydrogen bond network,
allow the formation of dimers that span the depth of the
membrane.
ransmembrane molecular nanopores have attracted
T
considerable attention for their potential application in
several fields, ranging from sensing to catalysis.1 Natural
nanopores are normally made by complex protein architectures,
with α-Hemolysin (αHL) being a well-known and representa-
tive example.2 This bacterial toxina monomeric, 293-residue,
polypeptideforms robust and wide (about 2 nm inner
diameter) heptameric pores in lipid bilayers that allow
permeation of large molecules and ions showing high electrical
conductance.3 Functionalization of the inner walls of this pore
led to systems capable of sensing diverse substrates, ranging
from transition metal ions to biologically relevant molecules
such as DNA.4 Taking inspiration from the natural examples,
the classical approach to synthetic ion channels is based on the
self-assembling of cyclic or linear oligomers.5 However, only
few systems capable of forming large and stable pores are
known.6 The most prominent example is the barrel-stave pore
formed by the self-assembling of p-octiphenyl based monomers
described by Matile and co-workers.7 The size and stability
make this artificial channel really competitive with the natural
ones, whereas the synthetic tunability of its inner chemical
space allowed to design several catalytically active or stimuli-
responsive systems.8
To this respect, the possibility that the walls of the pore
contain porphyrin moieties is particularly attractive as these
molecules, beside defining a relatively large, rigid and flat
aromatic surface, can be easily functionalized at the periphery as
well as metallated at the inner core. Recently, Kobuke and co-
workers described a tris-porphyrin covalent system that self-
assembles to form a cylindrical shaped macrocycle with an
inner diameter and a height of about 2.1 nm.9 The upper and
lower rims of the macrocycle are decorated with carboxylic acid
residues that promote a hydrogen-bonding driven dimerization
in the lipid bilayer resulting in a large transmembrane pore.
The metal-mediated directional-bonding approach is a
convenient alternative to the covalent one for the construction
of complex and functional molecular architectures.10 Indeed,
the combination of appropriate metal and organic fragments
allows for a precise control of the geometry, shape, kinetic and
thermodynamic stability of the resulting metal−organic frame-
Received: October 22, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja310425j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
%) from the 5-(4-pyridyl)dipyrromethane (3) and methyl 4-
formylbenzoate (4) under optimized conditions using an excess
of TFA, followed by oxidation with DDQ. The porphyrin was
then treated with Zn(II) acetate to obtain the [5,15-bis(4-
carboxymethylphenyl)-10,20-bis(pyridyl)porphinato]zinc(II)
(5a). The synthesis of 1 followed the procedure reported by
Hupp and coworkers for similar assemblies.16 Porphyrin 5a was
treated with ReBr(CO)5 in a 4:1 tetrahydrofuran/toluene
mixture under reflux for 48 h. After purification by
recrystallizzation and size exclusion chromatography, the
desired product 2 was recovered in almost quantitative yield.
The carboxylate derivative 1 was eventually obtained by alkaline
hydrolysis of the methyl esters in 48 % yield. Both metallacycles
1 and 2 were fully characterized by NMR, IR, and UV−vis
spectroscopy and the data are in good agreements with those
reported by Hupp and co-workers for similar systems (see
Supporting Information (SI)).16,17 The neutral tetraporphyrin
metallacycles are poorly soluble in many solvents, including
water. Improved solubility was obtained using zincated
porphyrins instead of the free-bases. Moreover, the presence
of the Zn(II) centers may assist ion transport by making the
inner walls of the channel more hydrophilic.
Figure 1. Schematic representation of the porphyrin metallacycle and
formation of a transmembrane nanopore upon hydrogen-bonding
driven dimerization.
a
Scheme 1. Synthesis of the Tetraporphyrin Metallacycle 1
The ionophoric activity of compounds 1 and 2 was
investigated with a standard base-pulse assay.18 The pH-
sensitive dye HPTS (8-hydroxypyrene-1,3,6-trisulfonic acid,
pKa = 7.2) was trapped inside 100 nm diameter liposomes
buffered at pH = 7. An aliquot of ionophore solution in DMSO
was added and, after 20 min of incubation time, a pH-gradient
of 0.6 units was applied by external addition of NaOH. The
collapse of the transmembrane pH-gradient implies basification
of the liposome inner water pool, which is signaled by an
increase of the HPTS fluorescence emission. Therefore, the
assay gives direct information on the transportation of H+
(efflux) or OH‑ (influx) and indirect information on the
correlated symport/antiport of counterions. Figure 2a shows
the typical kinetic profiles obtained with this experiment: at 200
s, the base-pulse is applied, and at 1400 s, an excess of
Figure 2. (a) Normalized fluorescence change in HPTS emission as a
function of time after addition of the base (50 μL of 0.5 M NaOH) to
95:5 EYPC/EYPG LUVs (100 nm diameter) loaded with HPTS (0.1
mM HPTS, 0.17 mM total lipid concentration, 25 mM HEPES, 100
mM NaCl, pH 7.0, total volume 3 mL), in the presence of
metallacycles 1, 2 and porphyrin 5b. The concentrations of
ionophores, given in percent with respect to the total concentration
of lipids, are indicated along the corresponding traces. (b) Depend-
ence of the observed rate constant for the transport process (kt, s−1) vs
concentration of 1 (HEPES, 4-(2-hydroxyethyl)-1-piperazine ethane-
sulfonic acid; EYPC, egg yolk phosphatidyl choline; EYPG, egg yolk
phosphatidyl glycerol).
a
Conditions: (1) trifluoroacetic acid (TFA, 43 equiv), dichloro-
methane (DCM), 0 °C, 20 min; (2) 2,3-dichloro-5,6-dicyano-1,4-
benzoquinone (DDQ, 2 equiv), rt, 1 h (39% from 3); (3) Zn(OAc)2
(3.2 equiv), CHCl3, rt, overnight (93%); (4) Re(CO)5Br (1 equiv),
THF/toluene 4:1, reflux, 48 h; (5) NaOH (8 equiv), THF/methanol
2:1, rt, overnight (48%).
The synthesis of the tetraporphyrin metallacycle 1 started
with the preparation of the 5,15-bis(4-carboxymethylphenyl)-
10,20-dipyridylporphyrin. This was obtained in good yield (39
B
dx.doi.org/10.1021/ja310425j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
TRITON X-100 is added, which lyses the liposomes leading to
the complete collapse of the pH gradient.
The tetraporphyrin metallacycle 1 shows high ionophoric
activity: the pH gradient is fully discharged after less than 2 min
in the presence of 1% ionophore and the activity is still
remarkable even at a concentration 10 times lower. Although
the different experimental conditions make comparisons
difficult, the kinetic profiles of 1 are similar to those reported
by Matile and co-workers for a barrel-stave channel,19 and its
activity is intermediate between those of amphotericin B and
gramicidin D, taken as examples of pore forming natural
compounds (Table S1). In contrast, the corresponding methyl
ester 2 (1%) and the single wall of metallacycle 1, that is, the
bis(4-carboxyphenyl)porphyrin 5b (4%), are almost ineffective
and the corresponding kinetic traces are close to the control
experiment recorded in the absence of ionophore. Thus, both
the preorganization of the metallacycle and the presence of the
properly oriented peripheral carboxylic acid groups are crucial
for the observed activity. The fitting of the kinetic traces
obtained at different concentrations of 1 gives the apparent
first-order rate constants (kt, s−1) for the transport process,
which show a roughly linear increase with the concentration of
metallacycle (Figure 2b). This linear dependence in the
activity/concentration profile may suggest that the active
species is either a monomer or a dimer with a dimerization
equilibrium largely shifted toward the dimer.20 In the first case,
the different activity of 1 compared to 2 might be related to a
better amphiphilic character of the acid derivative with respect
to the corresponding ester. However, control experiments on
the model porphyrins 5a and 5b show that the ester derivative
5a is more active than the parent acid 5b (Figure S7), thus
reasonably excluding a positive effect of the different
amphiphilicity on the ionophoric activity. Therefore, consider-
ing all the experimental evidences, and in analogy with the
model proposed by Kobuke,9 these results indicate that 1 is
indeed capable of forming the proposed transmembrane
nanopore via hydrogen bonding dimerization (Figure 1). In
this model, each metallacycle has four carboxylic groups
pointing inside the membrane involved in the hydrogen
bonding network that ensures dimerization of 1, while the
other four point outside toward the (inner or outer) bulk water
and are most likely deprotonated, thus efficiently directing the
correct orientation of the metallacycle in the membrane. As
expected for a nanopore with a large diameter and an empty
internal space, the system does not show selectivity toward first
Figure 3. Channel blockage experiments using a cystamine core
PAMAM dendrimer. Left: kinetic traces for the transport process
registered in the presence of the metallacycle 1 (0.75 %) and a G0−G2
dendrimer (1:1 or 0.5:1 molar ratio with respect to 1). In all the runs,
the dendrimers are added before the incubation time, with the
exception of the blue trace resulting from the addition of G2-
dendrimer at 300 s. For experimental conditions see Figure 2. Right:
structures of the generation 0-2 PAMAM dendrimers.
to the dimension of the PAMAM dendrimer and to the number
of ionizable amino groups: G0 (1:1 molar ratio, diameter ca. 1.5
nm) is almost ineffective, G1 (1:1 molar ratio, diameter ca. 2.2
nm) reduces only partially the activity, while G2 (1:1 molar
ratio, diameter ca. 2.9 nm) completely inactivates the channel.
Interestingly, the addition of G2 during the kinetic run blocks
immediately the transport process (blue trace in Figure 3)
indicating that the dendrimer has a very rapid effect.
These experiments clearly show that G2, and in part G1, act
as channel blockers. At pH close to 7, the amino groups of the
dendrimers are protonated, while the carboxylic groups of
porphyrin metallacycle 1 are deprotonated and exposed to the
bulk water. Therefore, most likely, a favorable electrostatic
interaction between the polycationic dendrimers and the
anionic portal of the channel leads to the observed blockage
effect. The dendrimer occupies the entrance of the channel
acting as a stopper and thus hampering the ion flux. Clearly, G0
(about 1.5 nm wide) is smaller and, having also a lower charge,
binds less efficiently to the channel. Therefore, it is unable to
occlude the channel, whereas the effect becomes more and
more important with increase of the size and the number of
ionizables groups of the dendrimer, until the complete channel
blockage.
In conclusion, we have described the first example of a stable
tetraporphyrin metallacycle (1), obtained via the metal-
mediated directional-bonding approach, capable of forming
nanopores in a liposomial membrane. The Re(I) corners ensure
that 1 is thermodynamically stable and kinetically inert. The
pore is likely to be dimeric and it is stabilized by an hydrogen
bond network between the appended carboxylic acid residues
on each porphyrin of the metallacycle. Efforts are being made
to establish the detailed mechanism behind the formation of
the transmembrane nanopore and the ion transport activity. We
are also modifying the structure of the porphyrin units in order
to further stabilize the dimeric assembly and to exploit the
functionalization of the inner space of the nanopore (e.g., with
metal ions other than Zn2+) for application in the field of
sensing. The intrinsic modular nature of the metallacycle will, in
principle, enable us to obtain a library of compounds by
changing the nature of the porphyrin unit and of the metal
corners. We strongly believe that our work has demonstrated
the suitability of the “Fujita-Stang” approach for the
construction of stable self-assembled nanostructures which are
group cations or inorganic anions (Cl−, Br−, I−, NO3 , ClO4 ,
see SI). Conversely, the large polyanionic calcein dye remains
trapped inside the liposome most likely as a consequence of
unfavorable electrostatic interactions with the negatively
charged portal of the channel (i.e., the deprotonated carboxylic
acids, Figure S10). Moreover, this last experiment suggests that
the ionophoric activity observed for 1 is specifically related to
the formation of a channel and is not due to lytic effects on the
membrane or to leakage of the HPTS dye.
To gain further insight into the nanopore formation, we
investigated the possibility to block the channel activity using a
cystamine core PAMAM dendrimer21 (Figure 3).22 Addition of
the G2-dendrimer, at 0.5 molar ratio with respect to the
tetraporphyrin metallacycle 1, blocks almost completely the
ionophoric activity, which is totally suppressed by increasing
the concentration of G2-dendrimer to a 1:1 molar ratio (the
kinetic trace is practically superimposable to the control
experiment). We found that the blockage is strongly correlated
−
−
C
dx.doi.org/10.1021/ja310425j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Pantos,
̧
G. D.; Sanders, J. K. M.; Orlandi, O.; Chiorboli, C.; Fracasso,
active in biological membranes and that it may stimulate new
S.; Scandola., F. Chem. Sci. 2011, 2, 676. (e) Wiester, M. J.; Ulmann, P.
A.; Mirkin, C. A. Angew. Chem., Int. Ed. 2010, 49, 2. (f) Yoshizawa, M.;
Klosterman, J. K.; Fujita, M. Angew. Chem., Int. Ed. 2009, 48, 3418.
(g) Scandola, F.; Chiorboli, C.; Prodi, A.; Iengo, E.; Alessio, E. Coord.
Chem. Rev. 2006, 250, 1471.
(11) (a) Chakrabarty, R.; Mukherjee, P. S.; Stang, P. J. Chem. Rev.
2011, 111, 6810. (b) Fujita, M.; Umemoto, K.; Yoshizawa, M.; Fujita,
N.; Kusukawa, T.; Biradha, K. Chem. Commun. 2001, 509.
(12) Sakai, N.; Matile, S. Angew. Chem., Int. Ed. 2008, 47, 9603.
(13) Jung, M.; Kim, H.; Baek, K.; Kim, K. Angew. Chem., Int. Ed.
2008, 47, 5755.
developments in the field of artificial ion transporters.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details of synthesis and characterization of
porphyrin components and tetraporphyrin metallacycles.
Procedure for the HPTS assay and results for cation and
anion selectivity of the nanopore. This material is available free
of charge via the Internet at http://pubs.acs.org.
(14) (a) Fyles, T. M.; Tong, C. C. New J. Chem. 2007, 31, 655.
(b) Devi, U.; Brown, J. R. D.; Almond, A.; Webb, S. J. Langmuir 2011,
27, 1448. (c) Wilson, C. P.; Boglio, C.; Ma, L.; Cockroft, S. L.; Webb,
S. J. Chem.Eur. J. 2011, 17, 3465. (d) Wilson, C. P.; Webb, S. J.
Chem. Commun. 2008, 4007.
(15) (a) Hupp, J. T. Struct. Bonding 2006, 121, 145. (b) Slone, R. V.;
́
Benkestein, K. D.; Belanger, S.; Hupp, J. T.; Guzei, I. A.; Rheingold, A.
L. Coord. Chem. Rev. 1998, 171, 221.
(16) Slone, R. V.; Hupp, J. T. Inorg. Chem. 1997, 36, 5422.
(17) Otto, W. H.; Keefe, M. H.; Splan, K. E.; Hupp, J. T.; Larive, C.
K. Inorg. Chem. 2002, 41, 6172.
(18) Sakai, N.; Matile, S. J. Phys. Org. Chem. 2006, 19, 452.
(19) Sakai, N.; Matile, S. J. Am. Chem. Soc. 2002, 124, 1184. See also
ref 13.
(20) Sakai, N.; Matile, S. Chem. Commun. 2003, 2514.
(21) Tomalia, D. A.; Huang, B.; Swanson, D. R.; Brothers, H.M.;
Klimashc, J.W. Tetrahedron 2003, 59, 3799.
AUTHOR INFORMATION
Corresponding Author
eiengo@units.it; ptecilla@units.it
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Dr. E. Iengo gratefully acknowledges the Financial support
from Reintegration Grant PERG02-GA-2007-224823, Prin2008
grant 20085ZXFEE, and Fondazione Beneficentia Stiftung. We
thank Prof. M. Prato and Dr. F. Giacalone of the University of
Trieste (Italy) for the kind gift of the cystamine core PAMAM
dendrimers, Mr. D. Milano for experimental help and Prof. E.
Alessio for helpful discussion.
(22) Dendrimers G0−G2 alone show a modest ionophoric activity
(Figure S11) in agreement with the reported inability of low-
generation PAMAM dendrimers to permeabilize lipid membranes:
Tiriveedhi, V.; Kitchens, K. M.; Nevels, K. J.; Ghandehari, H.; Butko,
P. Biochim. Biophys. Acta 2011, 1808, 209.
REFERENCES
■
(1) (a) Baudry, Y.; Bollot, G.; Gorteau, V.; Litvinchuk, S.; Mareda, J.;
Nishihara, M.; Pasini, D.; Perret, F.; Ronan, D.; Sakai, N.; Shah, M. R.;
Som, A.; Sorde, N.; Talukdar, P.; Tran, D.-H.; Matile, S. Adv. Funct.
́
Mater. 2006, 16, 169. (b) Howorka, S.; Siwy, Z. Chem. Soc. Rev. 2009,
38, 2360.
(2) Song, L.; Hobaugh, M. R.; Shustak, C.; Cheley, S.; Bayley, H.;
Gouaux, J. E. Science 1996, 274, 1859.
(3) Bayley, H.; Braha, O.; Gu, L. Q. Adv. Mater. 2000, 12, 139.
(4) (a) Braha, O.; Walker, B.; Cheley, S.; Kasianowicz, J. J.; Song, L.;
Gouaux, J. E.; Bayley, H. Chem. Biol. 1997, 4, 497. (b) Cheley, S.; Gu,
L.-Q.; Bayley, H. Chem. Biol. 2002, 9, 829. (c) Guan, X.; Gu, L.-Q.;
Cheley, S.; Braha, O.; Bayley, H. ChemBioChem 2005, 6, 1875.
(5) (a) Matile, S.; Jentzsch, A. V.; Montenegro, J.; Fin, A. Chem. Soc.
Rev. 2011, 40, 2453. (b) Davis, J. T.; Okunola, O.; Quesada, R. Chem.
Soc. Rev. 2010, 39, 3843. (c) Licen, S.; De Riccardis, F.; Izzo, I.; Tecilla,
P. Curr. Drug Discovery Technol. 2008, 5, 86. (d) Gokel, G. W.;
Daschbach, M. M. Coord. Chem. Rev. 2008, 252, 886. (e) Fyles, T. M.
Chem. Soc. Rev. 2007, 36, 335. (f) Davis, A. P.; Sheppard, D. N.; Smith,
B. D. Chem. Soc. Rev. 2007, 36, 348. (g) Sisson, A. L.; Shah, M. R.;
Bhosale, S.; Matile, S. Chem. Soc. Rev. 2006, 35, 1269. (h) Bhosale, S.;
Sisson, A. L.; Sakai, N.; Matile, S. Org. Biomol. Chem. 2006, 4, 3031.
(i) Matile, S.; Som, A.; Sorde, N. Tetrahedron 2004, 60, 6405.
(6) (a) Fernandez-Lopez, S.; Kim, H.-S.; Choi, E. C.; Delgado, M.;
Granja, J. R.; Khasanov, A.; Kraehenbuehl, K.; Long, G.; Weinberger,
D. A.; Wilcoxen, K. M.; Ghadiri, M. R. Nature 2001, 412, 452. (b) Ma,
L.; Melegari, M.; Colombini, M.; Davis, J. T. J. Am. Chem. Soc. 2008,
130, 2938. (c) Helsel, A.J.; Brown, A.L.; Yamato, K.; Feng, W.; Yuan,
L.; Clements, A.J.; Harding, S.V.; Szabo, G.; Shao, Z.; Gong, B. J. Am.
Chem. Soc. 2008, 130, 15784.
(7) Sakai, N.; Mareda, J.; Matile, S. Acc. Chem. Res. 2008, 41, 1354.
́
(8) (a) Sakai, N.; Sorde, N.; Matile, S. J. Am. Chem. Soc. 2003, 125,
7776. (b) Das, P.; Sakai, N.; Matile, S. Chirality 2002, 14, 18.
(c) Baumeister, B.; Sakai, N.; Matile, S. Org. Lett. 2001, 3, 4229.
(9) Satake, A.; Yamamura, M.; Oda, M.; Kobuke, Y. J. Am. Chem. Soc.
2008, 130, 6314.
(10) (a) Amouri, H.; Desmarets, C.; Moussa, J. Chem. Rev. 2012, 112,
2015. (b) Beves, J. E.; Blight, B. A.; Campbell, C. J.; Leigh, D. A.;
McBurney, R. T. Angew. Chem., Int. Ed. 2011, 50, 9260. (c) Breiner, B.;
Clegg, J. K.; Nitschke, J. R. Chem. Sci. 2011, 2, 51. (d) Iengo, E.;
D
dx.doi.org/10.1021/ja310425j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX