Synthetic pores with sticky π-clamps

Article abstract of DOI:10.1039/b702255g

In this report, we describe design, synthesis, evaluation and molecular dynamics simulations of synthetic multifunctional pores with π-acidic naphthalenediimide clamps. Experimental evidence is provided for the formation of unstable but inert, heterogeneo

Full text of DOI:10.1039/b702255g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthetic pores with sticky p-clamps†
Hiroyuki Tanaka, Guillaume Bollot, Jiri Mareda, Svetlana Litvinchuk, Duy-Hien Tran, Naomi Sakai and
Stefan Matile*
Received 12th February 2007, Accepted 19th March 2007
First published as an Advance Article on the web 3rd April 2007
DOI: 10.1039/b702255g
In this report, we describe design, synthesis, evaluation and molecular dynamics simulations of
synthetic multifunctional pores with p-acidic naphthalenediimide clamps. Experimental evidence is
provided for the formation of unstable but inert, heterogeneous and acid-insensitive dynamic tetrameric
pores that are sensitive to base and ionic strength. Blockage experiments reveal that the introduction of
aromatic electron donor–acceptor interactions provides access to the selective recognition of p-basic
intercalators within the pore. This breakthrough is important for the application of synthetic pores as
multianalyte sensors.
assembly of p-clamps into stacked rosettes reveals the close
Introduction
relation between p-clamping and broad field of intercalation.14
Bifunctional acyclic receptors, a.k.a. molecular clamps,1 tweezers,2
clips,3,4 clefts5 or jaws6 have stimulated the creativity of chemists
Ion-pair assisted p-clamping of nucleotides on one side of a
synthetic b-sheet has been achieved with a diagonal tryptophan
since decades.1–12 Molecular clamps continue to attract scientific
attention because they provide access to the acyclic version of
the inclusion complexes formed by macrocyclic receptors such
as cyclophanes, carcerands, cryptophanes, calixarenes or larger
architectures like biological and synthetic a-helix bundles or b-
barrels. Bidentate coordination within molecular tweezers12 as
well as bifunctional intercalation into molecular p-clamps1–11
has been particularly productive. p-Clamping with Whitlock’s
pioneering caffeine tweezers was found early on to increase with
tweezer rigidity, improving molecular recognition of benzoates
and naphthoates by about two orders of magnitude.2 The reversed
p-clamping of caffeine with the black tea polyphenol theaflavin
is a fine example for the quite frequently observed p-clamping in
biology and medicine.7
Other highlights in the history of p-clamping include Zim-
merman’s application of rigid acridine and phenanthrene tweez-
ers to affinity chromatography of nitrated polycyclic aromatic
hydrocarbons8 or Nolte’s glycoluryl clips.3 The more recent
Kla¨rner series of tetra-, tri-, and dimethylene-bridged naphthalene
and anthracene tweezers and clips revealed, inter alia, a rich
collection of intercalators, usefulness of charge transfer for
effective p-clamping, and stunning template effects that perfectly
imitate the pinching action of tweezers on the molecular level.4
Colquhoun has introduced pyrene tweezers to p-clamp macro-
cyclic naphthalenediimide (NDI) dimers.9 Contributions from
aromatic donor–acceptor interactions to adhesive p-clamping
have been found recently for pentafluorobenzene dimers.10 p-
Clamping with acylic (and cyclic)13 porphyrin dimers has been
studied by several groups with regard to the recognition of
fullerenes,7,13 nucleotides5 or alkyl viologens.11 The programmed
(W) tweezer with flanking lysines (K).15,16 It was this breakthrough
by Butterfield and Waters that identified p-clamping as ideal to
expand molecular recognition within synthetic pores1,17 beyond
ion pairing.
Synthetic multifunctional pores such as 1 are rigid-rod b-barrels
(Fig. 1).17–21 These barrel-stave supramolecules are composed of
rigid-rod staves and b-sheet hoops. During self-assembly from
monomeric p-oligophenyls (see Scheme 1), the non-planar staves
help to roll the planar b-sheets into the cylindrical oligomers. In the
final barrel, the b-sheet hoops help to position functional groups
at outer and inner pore surfaces to maximize interactions with
the surrounding bilayer and molecules passing through the pore,
respectively. So far, internal molecular recognition has focused
exclusively on topologically precise ion pairing. This simple and
sloppy recognition motif within synthetic pores was already
sufficient to develop applications in catalysis and sensing.17–21
Highlights include synthetic pores as general optical transducers
of chemical reactions18 or multicomponent sensing in complex
matrixes19 exemplified with sugar sensing in Coke.20 Refined pore
designs have been realized to probe the importance of guest
inclusion and template effects as well as the depth of inclusion.21
The introduction of recognition motifs beyond ion pairing
is of very high interest considering the possible practical ap-
plications of synthetic ion channels and pores (for selected
reviews and selected highlights on the topic of synthetic ion
channels and pores, please see ref. 22–32). p-Clamping within
synthetic multifunctional pores was particularly attractive because
characteristics such as the salt effects are orthogonal to ion pairing.
NDIs, unique, abiotic, compact, organizable,33–43 colorizable42,43
and functionalizable36–38,42,43 n-semiconductors39,40 were selected
for p-clamping within synthetic multifunctional pore 2. With a
quadrupole moment of +19 B (Buckinghams),38 the p-acidity of
NDIs exceeds that of hexafluorobenzene as well as the p-basicity
of higher aromatics such as pyrene (−14 B).38 This high p-acidity
is ideal for both self-assembly as well as the formation of face-to-
face aromatic electron donor–acceptor (AEDA) complexes.33–36
Department of Organic Chemistry, University of Geneva, Geneva, Switzer-
land. E-mail: stefan.matile@chiorg.unige.ch; Fax: +41 22 379 5123; Tel: +41
22 379 6523
† Electronic supplementary information (ESI) available: Experimental
details. See DOI: 10.1039/b702255g
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1369–1380 | 1369
©

View Article Online
Fig. 1 Rigid-rod b-barrel pores 1 and 2 (top) with indication of possible p-clamp motions within the p-acidic pore 2 (arrows, middle) to catch the
⊃
p-basic blocker 3 (red) in inclusion complex 2 3, featuring the classical face-to-face DAN-NDI charge-transfer complex with operational aromatic
electron donor–acceptor interactions (red half-arrows, bottom). b-Sheets are shown as solid (backbone) and dotted lines (hydrogen bonds, axial views)
or as arrows (N → C, side view); external amino acid residues are dark on white, internal ones white on dark (single-letter abbreviations), see Fig. 2 and
⊃
3 for molecular models of pore 2 and pore-blocker complexes 2 3, respectively.
Pioneered by Iverson with elegant foldamers,33,34 AEDA com-
plexes with the complementary dialkoxynaphthalenes (DANs)
received particular attention for the creation of double helices,34
barrel-stave supramolecules,37,38 rotaxanes35 or catenanes35 that
can act as shuttles,35 stimuli-responsive gels,36 ligand-gated ion
channels37 or anion-p slides.38 NDI intercalation into p-stacks
beyond DAN reaching from DNA duplexes41 to pyrene tweezers9
has been reported as well. The previous use of DAN-NDI
interactions to open up helical rigid-rod p-stacks into small ion
channels37 can be considered as complementary to the present
use to close large pores formed by rigid-rod b-barrels. The
application of the underlying p-stack architecture for artificial
photosynthesis42,43 is conceptually complementary to the sensing
applications of the b-barrel architecture used in this study.18–20
NDI clamps within synthetic pores were also of interest because
they introduce rigid-rod b-barrels with artificial amino acids and
are thus beyond reach or difficult to access with biological or
bioengineered pores, respectively. Functional internal and external
1370 | Org. Biomol. Chem., 2007, 5, 1369–1380
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Scheme 1 Synthesis of multifunctional pore 2. (a) C6H5I(OAc)2, 92%.67 (b) 1. NaOH, H2O; 2. 12, H3PO4, pH 5–7, 110 ◦C. (c) ZCl, Na2CO3, 48%.
◦
(d) TBDPSCl, imidazole, CH2Cl2, 0 C, 1 h. (e) Pd(OH)2/C, H2, EtOH, quant.68–70 (f) DMF, 110 ◦C; 70% from 12. (g) 4 steps, see ref. 71 and 72.
(h) HBTU, Et3N, DMF, 84%. (i) Pd(OH)2/C, H2, DMF–MeOH (5 : 1), 80%. (j) HBTU, Et3N, DMF, 70%. (k) 5% Piperidine–DMF, 64%. l) 9 steps, see
ref. 63. (m) HATU, Et3N, DMA, 99%. (n) 1. HF–Py, DMA, 2. TFA, 62%. (o) Self-assembly in lipid bilayer membranes.
rings of aromatics occur, however, quite frequently in biological
and bioengineered pores. The aromatic rings at the pore exterior
and at the membrane–water interface are unrelated to the topic
of this study.44–47 Internal aromatic rings, however, exist as well
and play roles in protein translocation through and effective
blockage (nanomolar IC50s) of the Anthrax pore,48 the blockage
of potassium channels,49 the selectivity of ammonia channels,50
pH gating and blockage of the M2 channel from influenza-A
virus with the antiviral amantadine (aminoadamantane),51 and
so on, often by cation–p interactions rather than p-clamping.
Poor TNT recognition by aromatic rings engineered into the a-
hemolysin pore (molar IC50’s) illustrates that the introduction of
aromatic rings into pores does not automatically result in effective
p-clamping.52
are sensitive to base, ionic strength and the selective recognition
of nucleotides, p-basic naphthalenes and other intercalators by
ion-pair assisted adhesive p-clamping (e.g., 3, Fig. 1). Selected
highlights have been reported in the preliminary communication
on the topic.1
Results and discussion
Design
Synthetic multifunctional pore 1 is one of the best explored rigid-
rod b-barrel pores.20,53,54 Leucines are placed at the outer surface as
an ideal compromise between b-propensity and bilayer affinity.17,44
Internal histidine-lysine (HK) dyads were introduced as binding
sites for anionic molecules, thus allowing us to discriminate ATP
and ADP with the “naked eye.” This important function was of
use to sense sugar in soft drinks with pore 1.20 Single-molecule
images of pore-blocker complexes are also available.54
In the following, we describe design, synthesis, evaluation
and molecular dynamics simulations of synthetic multifunctional
pores 2 with p-acidic NDI p-clamps. Experimental evidence
is reported for the formation of unstable but relatively inert,
heterogeneous and acid insensitive dynamic tetrameric pores that
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1369–1380 | 1371
©

View Article Online
Synthetic multifunctional pore 2 was envisioned to expand
molecular recognition beyond the dominant ion pairing in 1
and related rigid-rod b-barrel pores. For adhesive p-clamping,
that is for molecular recognition by aromatic donor–acceptor
interactions within pore 2, pore 1 was changed as little as possible
to assure maximal comparability. Histidine was replaced by the
artificial NDI amino acid pA (this arbitrary abbreviation was made
to indicate a “p-acidic” amino-acid side chain, or a “p-acceptor”).
The sequence of the internal residues was reversed to introduce the
artificial NDI amino acid as late as possible in the pentapeptide
synthesis. The ethanolamine as an NDI terminus was introduced
to be able to solubilize intractable synthetic intermediates as bulky
TBDPS derivatives (see below).
The assumption that the new pore 2 would be a tetramer was
made based on extensive data available on this motif including pore
1.17–21 Namely, Hill plots for the self-assembly of the (unstable)53,55
rigid-rod b-barrels with pentapeptide b-sheets including 1 have
demonstrated self-assembly into tetramers. Hill plots of rigid-
rod b-barrel 2 suggest that increasing internal crowding with the
bulky NDIs is, in contrast to other examples,56 insufficient to
cause supramolecular barrel expansion into higher oligomers (see
below).
Molecular dynamics simulations were compatible with
tetrameric active structures (Fig. 2) as well as selective pore block-
age by aromatic electron donor–acceptor interactions (Fig. 3). The
structures shown were obtained following previously established
protocols.57 In brief, MacroModel version 7.058,59 coupled with
a Maestro 4.1 graphical interface60 was used first to assemble
and preoptimize pore 2. Close contacts were eliminated with the
MMFF94s force field61 and the Polak–Ribiere conjugate gradient
(PRCG) algorithm. The empty barrel 2 was submitted to 2 ns MD
simulations using AMBER 8.62 Additional molecular dynamics
of 2 ns were performed for barrels where the p-clamping with
an aromatic electron donor was modeled. Such simulation times
were adequate for equilibration of these systems in the gas
phase.
In agreement with pH profiles (see below), lysine residues were
fully protonated to assure the internal charge repulsion needed to
stabilize internal space and prevent its collapse.53,57 In the modeled
pore, the bulky NDIs were found to preferably reside in loosely
packed p-stacks located towards the pore walls and over the p-
octiphenyl turns. Consistent with high activity (see below), this
architecture yielded a wide-open pore. The minimal inner pore
diameter was d = 1.0 nm. This diameter was in agreement with
the observed efflux of CF (5(6)-carboxyfluorescein, minimal outer
diameter = 1.0 nm) and the observed high conductance of single
pores (see below). The P-helicity of 21◦ was still small enough to be
compatible with the classification as barrel-stave supramolecule.57
The NDI stacking in the p-octiphenyl turns of this widely open
pore architecture was, however, possible only at the cost of
some local distortion of the b-sheets. This local destabilization
may account for the reduced inertness and apparent two-state
behavior of pore 2 in single-pore conductance experiments (see
below). Among several local minima, this particular architecture
corresponds to the most stable conformer.
Fig. 2 Molecular dynamics simulations of synthetic multifunctional pore
2 in axial (top) and side view (bottom) with rigid-rod staves in dark gray,
b-sheet hoops (ribbons) in yellow, lysine side-chains in blue (100% amine
protonation) and NDI clamps in blue.
toward the middle of the pore, where the p-stacks were again
formed. In this second possible NDI architecture, the pore is
closed. The observed high pore activity (see below), together
with clearly poorer stability in molecular dynamics simulations,
suggested that this closed NDI architecture is less favorable than
the open architecture with NDI stacks next to the rigid-rod turns
(Fig. 2).
To simulate adhesive p-clamping within p-acidic pore 2, the
tetra-anionic version 3 of the p-basic dialkoxynaphthyl partner
DAN in the classical DAN-NDI AEDA complexes was selected
because of its excellent blockage efficiency and potential use for
umami sensing (see below for experimental data and discussion).
To p-clamp the DAN blocker, two of the NDIs initially folded
back over p-octiphenyl turns are rotated around the Ca–Cb bond
The minimal length of methylene spacer between b-sheet and
NDI was found to determine the poor mobility of internal p-
clamps. Nevertheless, the rotation of about 100◦ around the Ca–Cb
single bond caused dramatic NDI movements to end up pointing
1372 | Org. Biomol. Chem., 2007, 5, 1369–1380
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
⊃
⊃
⊃
Fig. 3 Molecular dynamics simulations of pore-blocker complex 2 3 in axial (top) and side view (bottom) with one (2 31, left) and two (2 32, right)
bound blockers 3 in red (100% carboxylic acid deprotonation). The rigid-rod staves of pore 2 are in dark grey, b-sheet hoops (ribbons) in orange (lysine,
100% amine protonation), yellow (leucine) and blue (NDI clamps).
of the sheet-clamp spacer. In the resulting, energy-minimized
AEDA complex, the DAN blocker is caught between two sticky
NDIs like a mosquito between two hands clapped together. The
structures found after dynamics simulations were compatible with
the existence of the experimentally confirmed adhesive p-clamping
between DAN donor and NDI acceptors within 2 (Fig. 3). The
carboxylate ion pairs with ammonium cations from surrounding
lysines are embedded in a network of supportive hydrogen
bonds.
effects have been observed previously with rigid-rod b-barrels
pores, both in computational models as well as in AFM images.54,57
⊃
In the inclusion complex 2
32 with two DAN blockers, the
minimal inner pore diameter further decreases to d = 0.4 nm.
Both complexes are consistent with blockage of CF efflux, the
observed Hill coefficients for DAN blockers are ambiguous,
located consistently between n = 1 and n = 2 (see below).
The molecular dynamics simulations revealed the capacity of
rigid-rod b-barrel 2 to accommodate in their internal space up
⊃
Synthesis
to two DAN blockers. In the inclusion complex 2
31 with
Rigid rod 4 for self-assembly into pore 2 was synthesized from
scratch in 23 steps overall (Scheme 1).1 The p-octiphenyl scaffold
5 was prepared in nine steps from the commercially available
biphenyl 6 and bromoacetate 7 following our optimized procedure
without change.63
one DAN blocker, the pore experiences a small distortion, a
guest → host template effect that preserves the P-helicity of
20◦ and contributes to the reduction of the minimal inner pore
diameter to d = 0.7 nm. More pronounced guest → host template
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1369–1380 | 1373
©

View Article Online
Several approaches to NDI amino acids have been reported
previously.64–66 We explored four different routes to synthesize NDI
amino acid 8. The best approach, in our hands, uses aspartate 9,
ethanolamine 10 and naphthalene dianhydride 11 as commercially
available starting materials. The previously reported Hofmann
rearrangement of aspartate 9 with iodosobenzene diacetate67
provided facile and rapid access to Z-protected diaminopropionic
acid (Dap) 12. Reaction of Dap 12 with partially hydrolyzed
dianhydride 11 under slightly acidic conditions afforded imide
13 in quantitative yield. The ethanolamine 14 with a removable
TBDPS solubilizer was selected to introduce the second imide in
NDI amino acid 8. It was readily accessible from ethanolamine 10
in three simple steps.68–70 Chemoselective Z-protection gave alcohol
15, which was silylated to give 16. Z-Deprotection afforded the
desired amine 14, reaction with diacid 13 produced the NDI amino
acid 8 in 70% yield.
The solution-phase synthesis of tripeptide 17 was neither
new71,72 nor difficult. Attachment of the HBTU-activated NDI
amino acid 8 following routine peptide synthesis procedures
gave tetrapeptide 18. Hydrogenolysis to deprotect the N-terminus
afforded peptide 19. Coupling with leucine 20 and deprotection of
the N-terminus of NDI pentapeptide 21 with piperidine gave the
key intermediate 22.
Pentapeptide 22 was purified by semi-preparative HPLC before
attachment to the freshly prepared p-octiphenyl 5. Using 11.2
equivalents of pentapeptide 22, 12 equivalents of HATU as
coupling reagent and N,N-dimethylacetamide (DMA) as solvent,
full conversion of octaacid 5 to peptide-rod conjugate 23 was
accomplished within 1 h at room temperature in nearly quantita-
tive yield. Removal of the TBDPS-solubilizers with HF–pyridine
overnight at 0 ◦C (TBAF caused decomposition) followed by
lysine deprotection with TFA gave the final rod 4. The target
molecule was purified by Sephadex LH-20 and reverse-phase
HPLC before characterization and use. ESI MS, NMR spectra
and RP-HPLC were all consistent with expected structure and
sample homogeneity.
Pore activity in fluorogenic vesicles
Arguably the simplest and most reliable approach to zoom in
on activity and selectivity of synthetic ion channels and pores
begins with the determination of pH profile and Hill plot in
fluorogenic vesicles.17–32 Egg yolk phosphatidylcholine large unil-
amellar vesicles loaded with fluorophore 8-aminonaphthalene-
1,3,6-trisulfonate and quencher p-xylene-bis-pyridinium bromide
⊃
(EYPC-LUVs
ANTS–DPX) were used to measure the pH
profile of pore 2. In the ANTS–DPX assay, pore activity is
observed as an increase in ANTS emission due to efflux of the
cationic quencher DPX, the anionic fluorophore ANTS, or both.
This assay is ideal to record pH profiles because of the poor
sensitivity to pH changes and the ion selectivity of the pore.53,74
This is important because it assures that the reported changes
really originate from changes in the activity of the pore with pH
and not from changes in the ion selectivity of the pore with pH or
changes in the properties of the assay itself with pH.
The pH profile revealed that pore 2 closes at high pH as expected
for pore collapse without internal charge repulsion between at least
partially protonated lysine amines (Fig. 5, effective pH50 = 8.8).74–76
However, different to pore 1, pore 2 was insensitive to acid. This
finding confirmed the suspicion that in pore 1, partially or fully
protonated histidines (intrinsic pKa ∼ 6) are responsible for the
charge repulsion leading to full opening of pore around pH 5 or
inactivation by overcharging at pH < 5, respectively.53,74–76 This
additional possibility to charge and overcharge pores disappears
with the replacement of these histidines in pore 1 with the NDI
clamps in pore 2. Acid insensitivity, therefore, suggested that pore
2 may be permanently undercharged and never reach full activity.
This interpretation was in agreement with decreasing activity with
increasing ionic strength (Fig. 6).
The key to the successful synthesis of NDI pore 2 was the
introduction of TBDPS solubilizers. The NDI pentapeptide 22
with TBDPS was soluble in chloroform–methanol mixtures, DMF
(>50 mg ml−1) and N,N-dimethylacetamide (>100 mg ml−1). The
corresponding NDI pentapetide 24 without TBDPS solubilizer
was insoluble in chloroform–methanol mixtures and had low
solubility in DMF (10–20 mg ml−1) and N,N-dimethylacetamide
(25–50 mg ml−1, Fig. 4). The enormous usefulness of bulky spheres
to solubilize difficult synthesis intermediates is well documented
in the literature. The tert-butyl group in 7, for example, has been
the key for success in the initial synthesis of p-octiphenyl 5.73
Fig. 5 pH profile of pore 2. Fractional ANTS emission intensity Y (kex
353 nm, kem 510 nm) 4 min after the addition of monomer 4 (200 nM
⊃
final) to EYPC-LUVs ANTS–DPX [∼125 lM EYPC; inside: 12.5 mM
ANTS, 45 mM DPX, 5 mM TES, 20 mM NaCl, pH 7; outside: 100 mM
NaCl, 10 mM MES (᭹), HEPES (᭺) or AMPSO (ꢀ)] as a function of pH
(25 ◦C, calibrated by final lysis).
The concentration dependence of pore activity53,55,77 was de-
⊃
termined in EYPC-LUVs
CF. Similar to the ANTS–DPX
assay, pore activity in the CF assay is reported as an increase
in CF emission, because CF dilution by CF efflux from the
vesicles reduces CF self-quenching. Compared to the ANTS–
DPX assay, the CF assay is cheaper and more sensitive but has
limited applicability to anion-transporting large pores and works
at pH ≥ 6.5 only. The Hill plot of pore 2 revealed non-linear
dependence of pore activity on the concentration of monomer 4
Fig. 4 Structure of NDI pentapeptide 24, similar to 22 but without
TBDPS solubilizers (synthesis not shown).
1374 | Org. Biomol. Chem., 2007, 5, 1369–1380
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Fig. 7 Representative planar EYPC bilayer conductance in the presence
of rod 4 (0.08 mol%) at +50 mV in 2 M KCl. Conductance levels include
a) high conductance with a long lifetime, b) several lower conductance
levels and c) frequent high conductance with a very short lifetime.
Fig. 6 Hill plot of pore 2 with dependence on ionic strength [107 mM (᭹),
200 mM (᭺) and 300 mM NaCl (ꢁ)]. Fractional CF emission intensity Y
(kex 492 nm, kem 517 nm) 4 min after the addition of monomer 4 (0–2 lM
⊃
final) to EYPC-LUVs CF [∼125 lM EYPC, 10 mM HEPES; inside:
One dominant current level was relatively long-lived and had the
high conductance expected for the large inner diameter of rigid-
rod b-barrel 2 with internal NDI oriented along the p-octiphenyl
scaffolds (Fig. 7a, 1 and 2). Many interpretations for the observable
levels of lower conductance were conceivable, including reduction
of the apparent internal diameter by NDIs oriented toward the
center of the pore (Fig. 7b and 1). A high-conducting rapid
flickering occurred quite frequently (Fig. 7c). The similarity in high
conductance between these very labile single-pore currents and the
very inert single-pore currents mentioned above was reminiscent
of the Engelman two-state model,79 where the formation of the
inert rigid-rod b-barrel 2 would be preceded by a labile bundle
of transmembrane rods of similar structure but without the
stabilizing b-sheets. Similar observations in support of a two-
state model have been made previously.80 Molecular dynamics
simulations confirmed that internal NDI clamps may destabilize
and distort the b-sheets of the rigid-rod b-barrel (Fig. 2).
While the appearance of single-pore currents was well repro-
ducible, their characteristics could change from experiment to
experiment without a clear pattern. The observed comparably high
heterogeneity may indicate that the introduction of internal NDI
clamps increases the number of different active suprastructures,
either by the b-sheet destabilization indicated by Engelman-type
two-state behavior and molecular dynamics simulations, the ac-
cumulation of otherwise imperfect barrels, barrel contraction into
trimers or flattened tetramers,54 or large motions of internal NDIs
within otherwise perfect rigid-rod b-barrels (Fig. 1). In any case,
further investigation and (over)interpretation was not meaningful
at this stage given the complexity of the single-pore characteristics.
Perhaps because of their dynamic, adaptable suprastructure, rigid-
rod b-barrel pores with irregular single-pore characteristics were
identified previously as excellent optical transducers of chemical
reactions, enzyme detectors and biosensors.18,81 Results described
in the following sections are similarly promising.
50 mM CF, pH 7.4, X-100 mM NaCl; outside: pH 6.5, X mM NaCl, X =
107 (᭹), 200 (᭺) or 300 (ꢀ)] as a function of the concentration of monomer
4 (25 ◦C, calibrated by final lysis).
(Fig. 6᭹). This demonstrated that more than one rod 4 is needed to
form an active pore, i.e., the presence of supramolecular function.
The observed Hill coefficient n = 3.5 0.1 was compatible with
tetramer 2, a contracted trimer, a mixture of both, or, naturally,
the average of any more complex mixture. The non-linear Hill
plot further demonstrated that the self-assembly of pore 2 is
endergonic.53,55 Thermodynamic instability and tri- to tetrameric
active suprastructure of pore 2 were as with pore 1. With an
effective concentration EC50 = 47.5 0.5 nM (i.e., 190 2 nM
monomer 4), tetramer 2 was about one order of magnitude more
active than tetramer 1 and overall one of the most active rigid-rod
b-barrel pores reported so far. For reasons discussed elsewhere,53
high activity coupled with thermodynamic instability are ideal
characteristics for sensing applications.
The activity of pore 2 decreased with increasing ionic strength.
The effective concentration EC50 = 47.5 0.5 nM for tetramer 2
in 107 mM NaCl increased to EC50 ≈ 170 nM in 200 mM NaCl
(Fig. 6᭺) and EC50 ≈ 500 nM in 300 mM NaCl (Fig. 6ꢀ). This
salt effect was consistent with sensitivity to base because both
increasing ion pairing and deprotonation decrease internal charge
repulsion. Higher sensitivity of pore 2 to ionic strength compared
to pore 1 was thus in support of the comparably poor internal
charge repulsion. Ongoing attempts to treat the dependence of
pore activity on ionic strength quantitatively76,78 were so far not as
successful as with pH profiles.74,75
The Hill coefficients of pore 2 decreased with increasing ionic
strength. This trend could originate from increasing incompatibil-
ity with Hill analysis as the monomer concentration decreases
with increasing thermodynamic pore stability in response to
decreasing internal charge repulsion.55,74,76 This interpretation was
in agreement with the suspicion that a closed, inactive barrel
without internal charge repulsion is more stable than an internally
charged, open and active barrel. The stabilization of internal space
by intermediate internal charge repulsion may thus necessarily
destabilize the pore itself and require counter pressure from the
surrounding membrane to exist.76
Nucleotide recognition
Nucleotide recognition is an ideal topic for p-clamping because
their central role in biology promises many applications, from
multicomponent sensing in complex matrixes18–20 to gene sequenc-
ing with pores.82–84 Previous approaches to nucleotide clamping
include Schneider’s porphyrin clefts5 and Waters’ elegant b-
hairpins.15,16 Interestingly, previous examples of nucleotide recog-
nition by synthetic20 and bioengineered84 pores on the one hand
and by NDIs on the other hand85 do not operate on p-clamping.
Planar bilayer conductance
The addition of rod 4 to planar EYPC membranes caused the
appearance of rather heterogeneous single-pore currents (Fig. 7).
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1369–1380 | 1375
©

View Article Online
Table 1 Nucleotide recognition by pores 1 and 2
Blockera Ionic strengthb IC50 (2)/lMc IC50 (1)/lMc
Pd
1
2
3
4
5
6
7
8
9
ATP
ATP
+
++
+
14
21
83
1062
938
22
22e
131e
224e
8100
9000
73
1.6
6.2
2.7
7.6
9.6
3.3
ADP
AMP
AMP
GTP
GDP
GMP
GMP
+
++
++
+
+
++
+
f
77
—
—
—
—
f
1055
967
505
488
f
f
10 CDP
11 UDP
f
+
—
Fig. 9 Dose response curves for the blockage of pore 2 (200 nM monomer
4) by ATP (᭜), ADP (᭹), AMP (ꢀ), GDP (᭺) and GMP (ꢁ). Conditions
as in Fig. 8, ionic strength, 107 mM NaCl (Table 1, +).
a Nucleotides (three-letter abbreviations). b Outside: 107 mM NaCl (+) or
200 mM NaCl (++), 10 mM HEPES, pH 6.5; inside: 10 mM NaCl (+)
or 100 mM NaCl (++), 50 mM CF, 10 mM HEPES, pH 6.5. c Nucleotide
concentration required for 50% blockage, determined from fluorogenic CF
significant P = 6.2 at higher ionic strength (Table 1, entry 2).
An increasing clamping factor with increasing ionic strength was
consistent with an increasing dominance of p-clamping over ion
pairing under these conditions.
efflux from EYPC-LUVs CF (see Fig. 8 and 9). d Clamping factor P =
⊃
IC50 (1)/IC50 (2). e Data from ref. 20. f Not determined.
The discrimination of ATP and ADP is important for sensing
applications because it assures detectability of all ATP-dependent
processes.20 Compared to the ADP/ATP discrimination factor
D = IC50(ADP)/IC50(ATP) = 10.2 for pore 1, the introduction of
p-clamps within pore 2 reduced D to 5.9 (Table 1, entry 1 vs. 3;
Fig. 9, ᭜ vs. ᭹). This result was as expected for interference of
constant contributions from p-clamping with sensitivity toward
changes in ion pairing. However, an excellent D = 11.8 was
reported for Waters’ b-hairpin p-clamps at low ionic strength,15
and pore 1 gave up to D = 18.0 at low ionic strength.20
Nucleotide recognition by pore 2 with internal NDI p-clamps
was monitored with the CF assay. An original trace is shown in
Fig. 8, where addition of pore 2 at t = 30 seconds is followed
by a dramatic increase in CF emission. To determine molecular
recognition with this assay on molecular translocation of CF, the
same experiment was repeated in the presence of the analyte of
interest, e.g., ATP (Fig. 8). The observed increasing hindrance of
CF efflux with increasing concentration of ATP is exactly what is
expected from p-clamping of ATP within pore 2. Hill analysis of
the dose response curves for ATP gave IC50 = 14 lM in 107 mM
NaCl and IC50 = 21 lM in 200 mM NaCl (Table 1, entries 1
and 2, Fig. 9᭜). ATP recognition by pore 2 was more than three
orders of magnitude better than ion-pair assisted ATP recognition
Moving from ADP to AMP, IC50s further increased for both
pores 1 and 2 (Table 1, entries 3–5; Fig. 9, ᭹ vs. ꢁ). Contributions
from p-clamping were, however, most relevant at reduced ion
pairing with AMP. Clamping factors P = 7.6 at lower and
P = 9.6 at higher ionic strength were clearly better than those
with ATP. Clamping factors for AMP thus still increased with
ionic strength, although the trend was less pronounced than with
ATP. As a result, IC50s for AMP recognition by pore 2 decreased
with ionic strength, whereas IC50s for AMP recognition by p-
clamp-free pore 1 increased with ionic strength. Although a small
effect, this increasing recognition with increasing ionic strength
provided quite remarkable direct experimental evidence for the
existence and power of p-clamping within pore 2. Identical results
for GMP confirmed the general relevance of this finding (Table 1,
entries 8 and 9).
by tryptophan clamps on one side of a synthetic b-sheet (KD
=
48 000 lM, 200 mM KCl).15
The discrimination of purines and pyrimidines by p-clamp pore
2 was as unproblematic as expected. Already at low ionic strength,
molecular recognition of both ADP and GDP was about 6-times
better than that of CDP and UDP (Table 1, entries 3, 7, 10 and
11). This difference was in the range of that reported for b-hairpin
p-clamps.15 Naturally, this result was in support of effective p-
clamping within pore 2.
Discrimination between different purines by e.g., contributions
from aromatic donor–acceptor interactions to p-clamping within
pore 2, was less obvious. The IC50s for ANPs and GNPs at low
ionic strengths were overall almost the same (Table 1; Fig. 9, ᭹ꢁ
vs. ᭺ꢀ). The difference of the redox potentials is possibly too
small (G: 1.29 V, A: 1.42 V vs. NHE at pH 7)87 for substantial
contributions from aromatic donor–acceptor interactions with
the p-acidic NDI clamps. The higher cooperativity found in the
Fig. 8 Blockage of pore 2 by ATP. Fractional change in CF emission Y
(kex 492 nm, kem 517 nm) as a function of time after addition of ATP (top
down: 0, 1, 3, 5, 10, 20, 30, 50, 100 and 500 lM) and monomer 4 (200 nM
⊃
(at t ∼ 30 s) to EYPC-LUVs CF (∼125 lM EYPC, 10 mM HEPES;
inside: 50 mM CF, pH 7.4, 10 mM NaCl; outside: pH 6.5, 107 mM NaCl),
calibrated with final addition of excess triton X-100.
The clamping factor P = IC50 (1)/IC50 (2) was introduced to
quantify isolate contributions from p-clamping within pore 2 by
comparison with the p-clamp-free pore 1. The clamping factor
P = 1.6 for ATP recognition at lower ionic strength was poor
(Table 1, entry 1), perhaps also because of emerging interference
from stoichiometric binding at low IC50s.86 It increased to a more
1376 | Org. Biomol. Chem., 2007, 5, 1369–1380
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
dose response curves for GNP compared to ANP recognition
suggested that purine recognition by p-clamp pore 2 is dominated
by phenomena beyond charge-transfer complexation (Fig. 9, ᭹ꢁ
vs. ᭺ꢀ; G: n ≈ 1.8, A: n ≈ 1.1). We are currently exploring the
possibility of contributions from G-quartet29,88,89 formation to this
poorly understood effect. The often more complex trends found
for nucleobase discrimination by related systems such as porphyrin
tweezers5 or synthetic b-hairpins15,16 are in agreement with these
observations.
representative analytes of diagnostic interest. The hydrazones 3,
26, and 28–30 obtained by in situ reaction with hydrazides 31–33
under mildly acidic conditions were expected to be stable under
neutral and basic conditions.90–95
DAN hydrazide 31 was prepared from the commercially avail-
able naphthalene 36 by Williamson ether synthesis with methyl
bromoacetate 37 followed by the reaction of ester 3896 with
hydrazine. Williamson ether synthesis of naphthalene 36 with t-
butyl bromoacetate 7 for mild hydrolysis of ester 39 with TFA was
used to prepare DAN diacid 3.96
NDI hydrazide 32 was synthesized from dianhydride 11 and Z-
protected b-alanine 40. Procedures for the conversion of the latter
into the Boc-protected hydrazide 41 followed by hydrogenolytic
amine deprotection are available in the literature.97 Reaction of
amine 42 with dianhydride 11 afforded diimide 43, which was
readily deprotected with TFA. Reaction of dianhydride 11 with
b-alanine 44 afforded NDI diacid 27 in one step.98
The efficiency of this series of DAN and NDI intercalators to
block pore 2 and p-clamp-free control pore 1 was determined
from the inhibition of the fluorogenic efflux of CF from EYPC-
Adhesive p-clamping
The terms “adhesive “ or “sticky” are for p-clamps that can
use aromatic electron donor–acceptor interactions (AEDA) to
catch analytes. To systematically explore the existence of adhesive
p-clamping within the p-acidic pore 2, the p-basic dialkoxy-
naphthalene (DAN)33–37,43 intercalators 3, 25 and 26, p-acidic
NDI intercalators 27–29 and the non-aromatic control 30 were
synthesized (Scheme 2).1 A modular approach for rapid blocker
screening with DAN hydrazide 31, NDI hydrazide 32 and adipic
dihydrazide 33 was conceived for future use in sensing applications.
Reaction of these p-acidic, p-basic and non-aromatic scaffolds
with ketones of various size and charge90–95 was expected to provide
facile access to quantitative insights on adhesive p-clamping within
pore 2. Pyruvate 34 and a-ketoglutarate 35 were selected as
⊃
LUVs CF as described above for nucleotides (compare Fig. 8
and 9). The results provided remarkably consistent evidence for the
occurrence and significance of adhesive p-clamping within pore 1
(Table 2). As for nucleotides, contributions from ion pairing were
the easiest to identify: IC50s decreased with increasing negative
Scheme 2 Synthesis of NDI and DAN modules for in situ screening of pore blockers. (a) Cs2CO3, acetone, reflux. (b) TFA, quant.96 (c) Cs2CO3, acetone,
reflux, 77%.96 (d) NH2NH2, EtOH, reflux, 12 h, quant. (e) 135 ◦C, 3 d, 93%.98 (f) NH2NHBoc, HBTU, CH2Cl2, 0 ◦C–rt, 1 h, 63%.97 (g) Pd(OH)2/C, H2,
MeOH, quant.97 (h) DMF, 90 ◦C, 12 h, 81%. (i) TFA, 89%. (j) Ketone–hydrazide = 1 : 10, DMSO, 2 h, 50 ◦C.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1369–1380 | 1377
©

View Article Online
Table 2 Adhesive p-clamping within pore 2a,b,1
Blockerc
Formal charge
IC50 (2)/lMd
IC50 (1)/lMd
Pe
PDA
f
1
2
3
4
5
6
7
25
26
3
27
28
29
30
−2
−2
−4
−2
−2
−4
−4
94.7
4.5
0.24
32.7
25.7
2.5
2043
136
9.8
21.6
30.3
40.8
4.5
7.7
10.8
4.8
3.9
3.8
147
197
26.9
g
>100
−
a Data from ref. 1. Determined from dose response curves for fluorogenic CF efflux from EYPC-LUVs CF, compare Fig. 8 and 9. c See Scheme 2
for structures. d Blocker concentration required for 50% blockage. Data are the average value of at least three independent measurements, standard error
mostly <5%, always <10%. e Clamping factor P = IC50 (1)/IC50 (2). f Donor–acceptor factor PDA = PD/PA; e.g., PDA (entry 1) = P (25)/P (27). g Not
determined.
b
⊃
charge of aromatic blockers. Remarkably, the aliphatic hydrazone
30 did not block pore 2 at all despite its high charge. Significant
contributions from p-clamping were revealed by clamping factors
reaching from P = 4.5 for the small NDI dianion 27 up to a
remarkable P = 40.8 for the large DAN tetraanion 3 (Table 2,
entries 3 and 4). The “assistance” of ion pairing to maximize p-
clamping was confirmed by increasing clamping factors P with
increasing blocker charge. The adhesiveness of p-clamping within
pore 2, that is, the existence of aromatic electron donor–acceptor
interactions, was revealed by comparison of clamping factors P
for p-basic DANs and p-acidic NDIs. Clamping factors up to
P = 10 for the p-acidic NDI intercalators 27–29 were in the range
found for nucleotides. The clamping factors between P = 20 and
P = 40 for the p-basic DAN intercalators 3, 25 and 26 exceeded
this range by far. The observed donor–acceptor factors PDA ≈ 4
were roughly independent of intercalator size and charge (Table 2,
entries 1–3).
The identified donor–acceptor factors PDA ≈ 4 provided
surprisingly clear-cut experimental evidence for the existence of
significant contributions from aromatic electron donor–acceptor
interactions to molecular recognition within pore 2. Most impor-
tantly, the findings demonstrate that adhesive p-clamping provides
access to molecular recognition not only with high sensitivity but
also with high selectivity. This evidence for operational sticky p-
clamps was in agreement with results from molecular dynamics
simulations (Fig. 2 and 3). Experimental insights concerning
cooperativity were less convincing. The observed Hill coefficients
for blockage of pore 2 by pertinent DAN and NDI blockers 3, 26,
28 and 29 were in the range of n = 1.1–1.5. Although not very
convincing, this result suggested that more than one blocker may
be needed to block pore 2. With minimal internal pore diameters
of d = 0.7 nm with one and d = 0.4 nm with two DAN blockers 3,
results from molecular dynamics simulations were less ambiguous
with regard to cooperativity (Fig. 3). Consistently higher Hill
coefficients n = 1.4–2.2 were found for blockage of the less crowded
pore 1 with the same blockers.
high selectivity. This breakthrough is important for the application
of synthetic pores as sensors.
As far as the characteristics of the new pore as such are
concerned, insensitivity to acid but closing at high pH and
high ionic strength were identified. Hill plot and single pore
conductance reveal the dynamic characteristics of an unstable,
tri- to tetrameric pore that are of interest for practical sensing
applications.
Molecular dynamics simulations revealed that empty naph-
thalenediimide (NDI) clamps within tetrameric pores can open
up sideward to cover the rigid-rod staves and leave the internal
space free for unhindered translocation of molecules as large as
carboxyfluorescein. Molecular modeling further suggested that
these “open” NDI clamps could flip inward to firmly catch
matching analytes passing by.
p-Clamping within the new pore was explored with nucleotides,
p-basic and p-acidic naphthalenes and aliphatic controls. Nu-
cleotide clamping was characterized by excellent sensitivity (orders
of magnitude beyond similar systems), excellent discrimination of
purines over pyrimidines, modest discrimination between different
charges (ATP vs. ADP) and poor discrimination between different
purines (ATP vs. GTP).
The introduction of p-basic DAN, p-acidic NDI and non-
aromatic dihydrazide scaffolds provided access to a conve-
nient screening approach of blocker libraries. Pyruvate and a-
ketoglutarate hydrazones as representative analytes of biological
significance were already sufficient to secure experimental evidence
for all aspects of adhesive p-clamping, including assistance from
ion pairing and relevance of aromatic electron donor–acceptor
interactions.
The availability of refined pores for molecular recognition
beyond ion pairing on one hand and hydrazone scaffolds for
rapid analyte screening on the other are exceptionally promising
with regard to practical applications of synthetic pores as sensors.
Studies on multicomponent sensing in complex matrixes with
sticky p-clamps within synthetic pores and the synthesis of pores
with either more slippery or more flexible p-clamps are ongoing.
Conclusions
Acknowledgements
The substantial synthetic efforts required to introduce artificial
amino acids at the inner surface of rigid-rod b-barrels to create
pores with sticky p-clamps were fully rewarded. The reported
findings demonstrate that adhesive p-clamping provides access to
molecular recognition not only with high sensitivity but also with
We thank D. Jeannerat, A. Pinto and S. Grass for NMR spec-
troscopy measurements, P. Perrottet and the group of F. Gu¨lac¸ar
for MS measurements, the Swiss National Supercomputing Center
in Manno for CPU time on their CRAY-XT3 computer, the
1378 | Org. Biomol. Chem., 2007, 5, 1369–1380
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
referees for their assistance in improving the quality of the
manuscript, and the Swiss NSF for financial support.
38 V. Gorteau, G. Bollot, J. Mareda, A. Perez-Velasco and S. Matile, J. Am.
Chem. Soc., 2006, 128, 14788–14789.
39 L. L. Miller and K. R. Mann, Acc. Chem. Res., 1996, 29, 417–423.
40 H. E. Katz, A. J. Lovinger, J. Johnson, C. Kloc, T. Siegrist, W. Li, Y. Y.
Lin and A. Dodabalapur, Nature, 2000, 404, 478–481.
41 J. Lee, V. Guelev, S. Sorey, D. W. Hoffman and B. L. Iverson, J. Am.
Chem. Soc., 2004, 126, 14036–14042.
References
42 S. Bhosale, A. L. Sisson, P. Talukdar, A. Fu¨rstenberg, N. Banerji, E.
Vauthey, G. Bollot, J. Mareda, C. Ro¨ger, F. Wu¨rthner, N. Sakai and S.
Matile, Science, 2006, 313, 84–86.
1 H. Tanaka, S. Litvinchuk, G. Bollot, J. Mareda, D.-H. Tran, N. Sakai
and S. Matile, J. Am. Chem. Soc., 2006, 128, 16000–16001.
2 C.-W. Chen and H. W. Whitlock, J. Am. Chem. Soc., 1978, 100, 4921–
4922.
3 A. E. Rowan, J. A. A. W. Elemans and R. J. M. Nolte, Acc. Chem. Res.,
1999, 32, 995–1006.
4 F. G. Kla¨rner and B. Kahlert, Acc. Chem. Res., 2003, 36, 919–932.
5 H.-J. Schneider, L. Tianjun, M. Sirish and V. Malinovski, Tetrahedron,
2002, 58, 779–786.
6 D. Sun, F. S. Tham, C. A. Reed, L. Chaker and P. D. W. Boyd, J. Am.
Chem. Soc., 2002, 124, 6604–6612.
7 A. T. Charlton, A. L. Davis, D. P. Jones, J. R. Lewis, A. P. Davis, E.
Haslam and M. P. Williamson, J. Chem. Soc., Perkin Trans. 2, 2000,
317–322.
43 S. Bhosale, A. L. Sisson, N. Sakai and S. Matile, Org. Biomol. Chem.,
2006, 4, 3031–3039.
44 S. H. White and G. von Heijne, Curr. Opin. Struct. Biol., 2005, 15,
378–386.
45 W. M. Yau, W. C. Wimley, K. Gawrisch and S. H. White, Biochemistry,
1998, 37, 14713–14718.
46 M. R. R. De Planque, B. B. Bonev, J. A. A. Demmers, D. V. Greathouse,
R. E. Koeppe, II, F. Separovic, A. Watts and J. A. Killian, Biochemistry,
2003, 42, 5341–5348.
47 J. M. Sanderson, Org. Biomol. Chem., 2005, 3, 201–212.
48 B. A. Krantz, R. A. Melnyk, S. Zhang, S. J. Juris, D. B. Lacy, Z. Wu,
A. Finkelstein and R. J. Collier, Science, 2005, 309, 777–781.
49 D. A. Doyle, J. M. Cabral, R. A. Pfuetzner, A. Kuo, J. M. Gulbis, S. L.
Cohen, B. T. Chait and R. MacKinnon, Science, 1998, 280, 69–77.
50 S. Khademi, J. O’Connell, III, J. Remis, Y. Robles-Colmenares, L. J. W.
Miercke and R. M. Stroud, Science, 2004, 305, 1587–1594.
51 A. Okada, T. Miura and H. Takeuchi, Biochemistry, 2001, 40, 6053–
6060.
8 S. C. Zimmerman and K. W. Saionz, J. Am. Chem. Soc., 1995, 117,
1175–1176.
9 H. M. Colquhoun, Z. Zhu and D. J. Williams, Org. Lett., 2003, 5,
4353–4356.
10 T. Korenaga, Y. Kawauchi, T. Kosaki, T. Ema and T. Sakai, Bull. Chem.
Soc. Jpn., 2005, 78, 2175–2179.
11 S. Yagi, M. Ezoe, I. Yonekura, T. Takagishi and H. Nakazumi, J. Am.
Chem. Soc., 2003, 125, 4068–4069.
52 X. Guan, L.-Q. Gu, S. Cheley, O. Braha and H. Bayley, ChemBioChem,
2005, 6, 1875–1881.
12 X. Huang, K. Nakanishi and N. Berova, Chirality, 2000, 12, 237–
255.
53 S. Litvinchuk, G. Bollot, J. Mareda, A. Som, D. Ronan, M. R. Shah,
P. Perrottet, N. Sakai and S. Matile, J. Am. Chem. Soc., 2004, 126,
10067–10075.
13 Y. Shoji, K. Tashiro and T. Aida, J. Am. Chem. Soc., 2006, 128, 10690–
10691.
54 J. Kumaki, E. Yashima, G. Bollot, J. Mareda, S. Litvinchuk and S.
Matile, Angew. Chem., Int. Ed., 2005, 44, 6154–6157.
55 S. Bhosale and S. Matile, Chirality, 2006, 18, 849–856.
56 B. Baumeister and S. Matile, Chem. Commun., 2000, 35, 913–914.
57 V. Gorteau, G. Bollot, J. Mareda, D. Pasini, D.-H. Tran, A. N. Lazar,
A. W. Coleman, N. Sakai and S. Matile, Bioorg. Med. Chem., 2005, 13,
5171–5180.
58 MacroModel 7.0, Schro¨dinger Inc., Portland, OR, 1999.
59 F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M. Lipton,
C. Caufield, G. Chang, T. Hendrickson and W. C. Still, J. Comput.
Chem., 1990, 11, 440–467.
60 Maestro 4.1, Schro¨dinger Inc., Portland, OR, 2001.
61 T. A. Halgren, J. Comput. Chem., 1999, 20, 720–748.
62 D. A. Case, T. A. Darden, T. E. Cheatham, III, C. L. Simmerling, J.
Wang, R. E. Duke, R. Luo, K. M. Merz, B. Wang, D. A. Pearlman,
M. Crowley, S. Brozell, V. Tsui, H. Gohlke, J. Mongan, V. Hornak, G.
Cui, P. Beroza, C. Schafmeister, J. W. Caldwell, W. S. Ross and P. A.
Kollman, AMBER 8, University of California, San Francisco, 2004.
63 B. Baumeister, N. Sakai and S. Matile, Org. Lett., 2001, 3, 4229–4232.
64 T. Wagner, W. B. Davis, K. B. Lorenz, M. E. Michel-Beyerle and U.
Diederichsen, Eur. J. Org. Chem., 2003, 3673–3679.
65 A. A. Mokhir and R. Kraemer, Bioconjugate Chem., 2003, 14, 877–883.
66 N. Ashkenasy, W. S. Horne and M. R. Ghadiri, Small, 2006, 1, 99–102.
67 L. Zhang, G. S. Kauffman, J. A. Pesti and J. Yin, J. Org. Chem., 1997,
62, 6918–6920.
14 J. M. Kerckhoffs, M. G. ten Cate, M. A. Mateos-Timoneda, F. W.
van Leeuwen, B. Snellink-Ruel, A. L. Spek, H. Kooijman, M. Crego-
Calama and D. N. Reinhoudt, J. Am. Chem. Soc., 2005, 127, 12697–
12708.
15 S. M. Butterfield and M. L. Waters, J. Am. Chem. Soc., 2003, 125,
9580–9581.
16 S. M. Butterfield, M. M. Sweeney and M. L. Waters, J. Org. Chem.,
2005, 70, 1105–1114.
17 N. Sakai, J. Mareda and S. Matile, Acc. Chem. Res., 2005, 38, 79–87.
18 G. Das, P. Talukdar and S. Matile, Science, 2002, 298, 1600–1602.
19 G. Das and S. Matile, Chem.–Eur. J., 2006, 12, 2936–2944.
20 S. Litvinchuk, N. Sorde´ and S. Matile, J. Am. Chem. Soc., 2005, 127,
9316–9317.
21 Y. Baudry, D. Pasini, M. Nishihara, N. Sakai and S. Matile, Chem.
Commun., 2005, 40, 4798–4800.
22 P. Scrimin and P. Tecilla, Curr. Opin. Chem. Biol., 1999, 3, 730–735.
23 G. W. Gokel and A. Mukhopadhyay, Chem. Soc. Rev., 2001, 30, 274–
286.
24 U. Koert, L. Al-Momani and J. R. Pfeifer, Synthesis, 2004, 8, 1129–
1146.
25 S. Matile, A. Som and N. Sorde´, Tetrahedron, 2004, 60, 6405–6435.
26 A. L. Sisson, M. R. Shah, S. Bhosale and S. Matile, Chem. Soc. Rev.,
2006, 35, 1269–1286.
27 T. M. Fyles, Chem. Soc. Rev., 2007, 36, 335–347.
28 A. P. Davis, D. N. Sheppard and B. D. Smith, Chem. Soc. Rev., 2007,
36, 348–357.
29 J. T. Davis and G. P. Spada, Chem. Soc. Rev., 2007, 36, 296–313.
30 Y. Kobuke, K. Ueda and M. Sokabe, J. Am. Chem. Soc., 1992, 114,
7618–7622.
68 S. Hanessian and P. LaVallee, Can. J. Chem., 1975, 53, 2975–2977.
69 D. Wenninger and H. Seliger, Nucleosides Nucleotides, 1997, 16, 977–
980.
70 M. T. Tierney and M. W. Grinstaff, J. Org. Chem., 2000, 65, 5355–5359.
71 B. Baumeister, N. Sakai and S. Matile, Angew. Chem., Int. Ed., 2000,
39, 1955–1958.
31 W.-H. Chen, X.-B. Shao and S. L. Regen, J. Am. Chem. Soc., 2005, 127,
12727–12735.
72 B. Baumeister, Thesis 3302, University of Geneva, 2001.
73 N. Sakai, N. Majumdar and S. Matile, J. Am. Chem. Soc., 1999, 121,
4294–4295.
74 B. Baumeister, A. Som, G. Das, N. Sakai, F. Vilbois, D. Gerard, S. P.
Shahi and S. Matile, Helv. Chim. Acta, 2002, 85, 2740–2753.
75 V. Borisenko, M. S. P. Sansom and G. A. Woolley, Biophys. J., 2000,
78, 1335–1348.
76 A. Som and S. Matile, Chem. Biodiversity, 2005, 2, 717–729.
77 A. V. Hill, Biochem. J., 1913, 7, 471–480.
78 C. Park and R. T. Raines, J. Am. Chem. Soc., 2001, 123, 11472–11479.
79 J. Popot and D. Engelman, Biochemistry, 1990, 29, 4031–4037.
32 F. Otis, N. Voyer, A. Polidori and B. Pucci, New J. Chem., 2006, 30,
185–190.
33 R. S. Lokey and B. L. Iverson, Nature, 1995, 375, 303–305.
34 G. J. Gabriel and B. L. Iverson, J. Am. Chem. Soc., 2002, 124, 15174–
15175.
35 S. A. Vignon, T. Jarrosson, T. Iijima, H. R. Tseng, J. K. M. Sanders
and J. F. Stoddart, J. Am. Chem. Soc., 2004, 126, 9884–9885.
36 P. Mukhopadhyay, Y. Iwashita, M. Shirakawa, S. Kawano, N. Fujita
and S. Shinkai, Angew. Chem., Int. Ed., 2006, 45, 1592–2595.
37 P. Talukdar, G. Bollot, J. Mareda, N. Sakai and S. Matile, J. Am. Chem.
Soc., 2005, 127, 6528–6529.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1369–1380 | 1379
©

View Article Online
80 N. Sakai, D. Houdebert and S. Matile, Chem.–Eur. J., 2003, 9, 223–232.
81 P. Talukdar, N. Sakai, N. Sorde´, D. Gerard, V. M. F. Cardona and S.
Matile, Bioorg. Med. Chem., 2004, 12, 1325–1336.
91 E. C. Rodriguez, K. A. Winans, D. S. King and C. R. Bertozzi, J. Am.
Chem. Soc., 1997, 119, 9905–9906.
92 V. W. Cornish, K. M. Hahn and P. G. Schultz, J. Am. Chem. Soc., 1996,
118, 8150–8151.
93 P. T. Corbett, J. Leclaire, L. Vial, K. R. West, J.-L. Wietor, J. K. M.
Sanders and S. Otto, Chem. Rev., 2006, 106, 3652–3711.
94 B. Levrand, Y. Ruff, J.-M. Lehn and A. Herrmann, Chem. Commun.,
2006, 41, 2965–2967.
82 D. W. Deamer and D. Branton, Acc. Chem. Res., 2002, 35, 817–825.
83 N. Sorde´, G. Das and S. Matile, Proc. Natl. Acad. Sci. U. S. A., 2003,
100, 11964–11969.
84 Y. Astier, O. Braha and H. Bayley, J. Am. Chem. Soc., 2006, 128, 1705–
1710.
85 J. M. Lavin and K. D. Shimizu, Org. Lett., 2006, 8, 2389–2392.
86 O. H. Straus and A. Goldstein, J. Gen. Physiol., 1943, 26, 559–585.
87 S. Steenken and S. Jovanovic, J. Am. Chem. Soc., 1997, 119, 617–618.
88 S. L. Forman, J. C. Fettinger, S. Pieraccini, G. Gottarelli and J. T. Davis,
J. Am. Chem. Soc., 2000, 122, 4060–4067.
89 N. V. Hud, P. Schultze, V. Sklenar and J. Feigon, J. Mol. Biol., 1999,
285, 233–243.
90 J. Shao and J. P. Tam, J. Am. Chem. Soc., 1995, 117, 3893–3899.
95 A. Som and S. Matile, Eur. J. Org. Chem., 2002, 3874–3883.
96 P. M. Kramer, M.-P. Marco and B. D. Hammock, J. Agric. Food Chem.,
1994, 42, 934–943.
97 H. D. King, D. Yurgaitis, D. Willner, R. A. Firestone, M. B. Yang, S. J.
Lasch, K. E. Hellstrom and P. A. Trail, Bioconjugate Chem., 1999, 10,
279–288.
98 B. Abraham, S. McMasters, M. A. Mullan and L. A. Kelly, J. Am.
Chem. Soc., 2004, 126, 4293–4300.
1380 | Org. Biomol. Chem., 2007, 5, 1369–1380
This journal is
The Royal Society of Chemistry 2007
©