Calixarene-mediated liquid membrane transport of choline conjugates 2: Transport of drug-choline conjugates and neurotransmitters

Article abstract of DOI:10.1002/ejoc.201403519

Lower rim carboxylic acid calix[n]arenes and upper rim phosphonic acid functionalized calix[4]arenes effect selective transport of distinct molecular payloads through a liquid membrane. The secret to this success lies in the attachment of a receptor-complementary handle. We find that the trimethylammonium ethylene group present in choline is a general handle for the transport of drug and drug-like species. Furthermore, neurotransmitters possessing ionizable amine termini are also transported. Some limitations to this strategy have been uncovered as payloads become increasingly lipophilic. These developments reveal new approaches to synthetic receptor-mediated selective small molecule transport in vesicular and cellular systems. Ammonium species emerge as general and potent handles for effecting drug transport as mediated by ionizable calixarenes. Trimethylammonium, ammonium and ammonium carboxylates work in conjunction with phosphonic acid and carboxylic acid functionalized calix[n]arenes.

Full text of DOI:10.1002/ejoc.201403519

FULL PAPER
DOI: 10.1002/ejoc.201403519
Calixarene-Mediated Liquid Membrane Transport of Choline Conjugates 2:
Transport of Drug-Choline Conjugates and Neurotransmitters
Birendra Babu Adhikari,[a] Sahar Roshandel,[a] Ayu Fujii,[a] and Michael P. Schramm*[a]
Keywords: Medicinal chemistry / Drug design / Drug delivery / Receptors / Calixarenes / Dopamine / Serotonin
Lower rim carboxylic acid calix[n]arenes and upper rim
phosphonic acid functionalized calix[4]arenes effect selective
transport of distinct molecular payloads through a liquid
membrane. The secret to this success lies in the attachment
of a receptor-complementary handle. We find that the tri-
methylammonium ethylene group present in choline is a ge-
neral handle for the transport of drug and drug-like species.
Furthermore, neurotransmitters possessing ionizable amine
termini are also transported. Some limitations to this strategy
have been uncovered as payloads become increasingly lipo-
philic. These developments reveal new approaches to syn-
thetic receptor-mediated selective small molecule transport
in vesicular and cellular systems.
Introduction
lated to resorcinarene cavitand host binding[5] and cataly-
sis.[6] We have discovered that these systems function in
aqueous lipid systems as hosts,[7] and as platforms for
switching,[8] and we know too that these molecules distrib-
ute themselves evenly in lipid bilayer systems.[9] Our col-
leagues have demonstrated remarkable functions for these
systems as surface bound receptors for protein sensing[10]
and, most interestingly, as facilitators of endocytosis for
guest molecules![11] These properties have been developed
under constraints of the host, as they are largely incompat-
ible with pure water environments. We had hoped that res-
orcinarene cavitands would be a suitable prototype for
transport. However, limitations stemming from their strong
binding of complementary handles such as choline indi-
cated that efficient guest release would be difficult to
achieve.[1] Consequently, calixarene species with ionizable
groups were identified as being capable of extraction and,
ultimately, transport. This work built on a variety of earlier
reports with trimethylammonium guests.[12] In this context,
several important papers using bulk liquid membranes have
resulted in the development of synthetic receptors that
transport metal ions,[13] and can carry out enantioselective
transport of amino acids.[14] Clearly then, the principles of
dynamic combinatorial chemistry have uncovered new
transporters.[15] As mentioned, phosphonic acid 1 was
found to effect transport across a bulk liquid membrane
without a pH gradient, whereas lower-rim functionalized
carboxylates calix[4]arene 2 and calix[6]arene 4 responded
favorably to inverse acid gradients. We hypothesize that
calixarenes provide ion–ion interactions with choline that
overcome two-phase extraction, but are lenient enough to
release a payload into a welcoming receiving phase, whereas
resorcinarene cavitands are unable to accomplish this essen-
tial latter requirement.
We recently reported that calixarenes are promising re-
ceptors for liquid membrane transport of choline-fluoro-
phore conjugates.[1] Our results indicated that the presence
of ionizable, pre-organized functional groups such as those
on a calixarene scaffold enable effective transport of chol-
ine-fluorophore conjugates. Moreover, the tetraphosphonic
acid calix[4]arene 1 was capable of transporting choline
conjugates without a complementary inverse pH gradient.
Our ultimate interest in this work stems from a desire
to overcome drug transport challenges inherent to humans.
Indeed, drug delivery is an important complement to drug
discovery, especially for high-impact (in vitro) lead targets
that (in vivo) the cell membrane resists. Many ingenious
systems have been developed and some have received clin-
ical approval, but challenges remain.[2] The use of calixar-
ene and resorcinarene cavitands is a novel approach relative
to covalent drug attachment to delivery vehicles or emulsifi-
cations in lipid micelles. We aim to develop: 1) a receptor
that localizes in the cell membrane, 2) a receptor that does
not harm the cell or cause non-specific leakage, 3) a com-
plementary receptor handle system that can selectively
transport a variety of handle-payload conjugates and 4) a
receptor that transports more than one equivalent of pay-
load. Valinomycin and nonactin provide the biological in-
spiration for what is possible.[3,4]
Our contribution stems from a rich history of supra-
molecular host–guest chemistry and concepts, primarily re-
[a] Department of Chemistry and Biochemistry, California State
University Long Beach,
1250 Bellflower Blvd., Long Beach, CA, U.S.A.
E-mail: michael.schramm@csulb.edu
http://schrammlab.wordpress.com
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201403519.
Eur. J. Org. Chem. 2015, 2683–2690
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2683

B. B. Adhikari, S. Roshandel, A. Fujii, M. P. Schramm
FULL PAPER
These initial results encouraged us to further explore the competitive partitioning of host into the guest aqueous
range of possible payloads suitable for delivery. Herein our layer appears to be a significant factor.
work with drug and drug-like molecules is reported.
Table 1. Two-phase solvent extraction of tryptophan–choline 5
with different receptors.[a]
Receptor [5] [mM] % Extraction % Extraction Extraction
(0.5 mm)
control
efficiency[b]
Results and Discussion
1
0.1
0.2
0.5
1.0
3.0
0[c]
6.32
3.91
1.85
1.52
1.14
0
0
0
5.02
27.4
0[c]
We employed upper-rim tetraphosphonic acid calix[4]-
arene 1[16] and a homologous series of lower rim carboxylic
acid calix[4] 2,[17] calix[5] 3[18] and calix[6] 4 – arenes (Fig-
ure 1). BOC protected tryptophan–choline 5, acetamino-
phen–choline 6, ibuprofen–choline 7, tamoxifen–choline 8,
dimethylresveratrol–choline 9 and bentiromide–choline 10
were selected as guest–choline conjugates. This collection of
molecules afforded us easy access to a variety of functional
groups, lipophilicities and molecular complexities enabling
an initial study of the limits of selective drug-conjugate
transport. These are some of the central issues to be consid-
ered in solving drug-transport problems.
0[c]
4.03Ϯ0.16
5.70Ϯ0.54
2
4
0.1
0.2
0.5
1.0
3.0
19.92Ϯ0.38
19.37Ϯ1.30
13.83Ϯ1.01
13.21Ϯ0.13
8.00Ϯ0.35
6.32
3.91
1.85
1.52
1.14
2.72
6.18
11.98
23.38
41.2
0.1
0.2
0.5
1.0
3.0
68.42Ϯ2.25
54.60Ϯ2.87
48.14Ϯ2.50
27.27Ϯ0.50
19.20Ϯ0.76
6.32
3.91
1.85
1.52
1.14
12.4
20.28
46.29
51.5
108.4[d]
[a] Receptor concentration 0.5 mm in DCM, extraction% measured
from an aqueous solution of guest after 24 h by UV/Vis measure-
ment at 280 nm against a calibration curve, reported as duplicate
runs with Ϯmaximum deviation from the mean. [b] (Extraction –
control) * ([guest]/[host]). [c] The absorbance was slightly higher
than at time = 0, we believe this is due to low efficiency guest
extraction and some host extraction into the aqueous layer.
[d] "100%" with error.
Two Phase Solvent Extraction of Tryptophan-Choline 5
Extraction, the logical preamble to transport was studied
first. Accordingly, we initially explored receptor-mediated
extraction of tryptophan–choline 5 from an aqueous solu-
tion (Table 1). We exposed a stirring solution of receptor
(0.5 mm, DCM, 160 rpm) to an equal volume of substrate
Carboxylic acid receptors 2 and 4 were more efficient
in water and stirred for 24 h. An aliquot of the aqueous hosts for extracting guest 5 into the organic layer. Appreci-
phase was then removed and absorbance measured to deter- able extents of extraction were measured at all guest con-
mine the net decrease in substrate concentration. We found centrations and significant extraction efficiencies were ob-
that host 1 was ineffective at extracting guests below served at high guest concentrations. For 4 at a ratio of 6:1
0.5 mm, but showed modest extraction efficiencies at ele- guest to host, the extraction efficiency was found to be 100%;
vated guest concentrations. At low guest concentrations the under these conditions each host is occupied by a guest.
Figure 1. Structure of calixarene-based receptors 1–4 and drug conjugate substrates 5–10.
2684
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2683–2690

New Transport of Choline Conjugates and Neurotransmitters
Three-Phase Transport of Tryptophan-Choline 5
(Table 2). The remaining receptors were all less efficient
transporters than the control, a finding that is consistent
with our extraction experiments (Table 1). Indeed hosts 2–
4 proved to be excellent extractors, but failed to deliver the
guest into the receiving phase under neutral conditions.
Notably, transport characteristics changed drastically
when the receiving phase was acidified. Host 1 remained
an efficient transporter vs. control, but calix[4]arene 2 and
calix[5]arene 3 carboxylic acids became exceptional trans-
porters, each facilitating the delivery of over 9 times more
guest vs. control at a guest concentration of 0.5 mm. Calix-
[6]arene hexacarboxylic acid 4 was also efficient at guest
transport when the receiving phase was acidified. Perhaps
the pre-organization of 2 and 3 instills some transpor-
tational advantages that 4 simply cannot offer due to its
dynamic conformation in solution. Indeed, it is not simply
the number of functional groups that matters to the trans-
port process; calix[5]arene may offer the best ratio of fixed
conformation and number of functional groups.
We then explored the three-phase transport of 5 facili-
tated by four different receptors under different substrate
concentrations (low and high) and different receiving phase
conditions (neutral and acidic) (see Supporting Information
for representative UV data and full details). Under neutral
conditions we found that phosphonic acid 1 efficiently en-
hanced transport flux of 5 on the order of 4.5-fold faster
than in the control when guest concentration was low
Table 2. Three-phase transport of tryptophan–choline 5 with
different receptors.[a]
Receptor [5] [mM] Transport
Enhancement[b] Receiving
phase
(0.5 mm)
flux
(µmol cm–2 min–1)
1
0.5
5.0
5.76ϫ10–5
4.5
1.5
neutral
Ϯ 0.09ϫ10–5
4.50ϫ10–4
Ϯ 0.05ϫ10–4
2
3
0.5
5.0
1.18ϫ10–5
0.9
0.8
Ϯ 0.00ϫ10–5
2.32ϫ10–4
Transport of Drug Conjugates
Ϯ 0.11ϫ10–4
Receptors 1 and 2 were selected for further studies to
understand the transport of similar drug–choline conju-
gates acetaminophen–choline 6 and ibuprofen–choline 7.
We found that both receptors can facilitate transport of
0.5
5.0
3.03ϫ10–6
0.2
0.4
Ϯ 0.00ϫ10–6
1.34ϫ10–4
Ϯ 0.37ϫ10–4
4
0.5
5.0
0
0
0.7
2.27ϫ10–4
Ϯ 0.29ϫ10–4
Table 3. Three-phase transport of drug-conjugates acetaminophen–
choline 6 and ibuprofen–choline 7.[a]
Control
(no host) 5.0
0.5
1.27ϫ10–5
3.05ϫ10–4
Receptor
(0.5 mm)
Guest
(0.5 mm) (umol cm–2
min–1
Transport flux
Enhancement[b] Receiving
phase
1
0.5
5.0
1.13ϫ10–4
6.3
1.9
acidic
(0.1 m HCl)
)
Ϯ 0.02ϫ10–4
6.76ϫ10–4
1
2
6
5.42ϫ10–6
2.9
0.8
neutral
acidic
Ϯ 0.09ϫ10–6
1.60ϫ10–6
Ϯ 0.02ϫ10–4
2
0.5
5.0
0.5
5.0
0.5
5.0
0.5
1.66ϫ10–4
9.2
2.8
9.6
3.4
6.6
1.9
Ϯ 0.12ϫ10–6
Ϯ 0.07ϫ10–4
9.62ϫ10–4
6
4.25ϫ10–6
2.2
6.0
neutral
acidic
Ϯ 0.36ϫ10–6
1.17ϫ10–5
Ϯ 0.09ϫ10–4
Ϯ 0.15ϫ10–5
3
1.71ϫ10–4
Ϯ 0.18ϫ10–4
Control
(no host)
6
7
1.89ϫ10–6
1.96ϫ10–6
neutral
acidic
1.20ϫ10–3
Ϯ 0.15ϫ10–3
1
0.00
0
5.6
neutral
acidic
1.32ϫ10–4
Ϯ 0.10ϫ10–4
4
1.18ϫ10–4
Ϯ 0.07ϫ10–4
2
7
7
0.00
0
3.0
neutral
acidic
6.71ϫ10–4
7.08ϫ10–5
Ϯ 0.20ϫ10–5
Ϯ 0.01ϫ10–4
Control
1.79ϫ10–5
3.48ϫ10–4
Control
(no host)
2.37ϫ10–5
2.37ϫ10–5
neutral
acidic
(no host) 5.0
[a] Transport flux measured via U-tube three-phase aparatus:
0.5 mm receptor in 10 mL DCM, the substrate was charged in the
source phase at the indicated concentration in 4 mL of water and
a receiving phase of 4 mL of water or 0.1 m HCl was added. After
72 h of stirring at 400 rpm, an aliquot from the receiving phase was
removed and analyzed by UV/Vis measurement at 280 nm against
a calibration curve, reported as duplicate runs with Ϯmaximum
deviation from the mean. [b] Enhancement: flux/flux of corre-
sponding control.
[a] Transport flux measured via U-tube three-phase aparatus:
0.5 mm receptor in 10 mL DCM, 0.5 mm substrate was charged in
the source phase in 4 mL of water and a receiving phase of 4 mL
of water or 0.1 m HCl was added. After 72 h of stirring at 400 rpm
an aliquot from the receiving phase was removed and analyzed by
UV/Vis measurement at the corresponding absorbance maximum
against a calibration curve, reported as duplicate runs with Ϯmaxi-
mum deviation from the mean. [b] Enhancement: flux/flux of corre-
sponding control.
Eur. J. Org. Chem. 2015, 2683–2690
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2685

B. B. Adhikari, S. Roshandel, A. Fujii, M. P. Schramm
FULL PAPER
both guests, but differential behaviors emerge resulting and combination of phosphonic acid host 1 or carboxylic
from structural differences between guests (Table 3). Guest acid host 2 enhanced the transport of this lipophilic sub-
6 is most effectively transported by calix[4]arene carboxylic strate by 3.4- and 2.6-fold, respectively, vs. control (Table 4).
acid 2 at low pH, whereas the slightly more lipophilic 7 is
Table 4. Three-phase transport of drug-conjugates dimethylresvera-
transported best by calix[4]arene phosphonic acid 1 at low
pH. Although these results are empirical, a complementary
set of hosts continues to demonstrate advantages. Host 1
continues to show broad guest compatibility in neutral envi-
ronments routinely giving enhanced transport. In this ex-
periment 1 clearly demonstrates some advantages over carb-
oxylic acid host 2 for transporting lipophilic guests. Hosts
such as 2, may serve as more potent extractors of lipophilic
guests (see below). Additionally, host 2 shows remarkable
transport enhancements when the receiving phase is acidi-
fied for a variety of guests.
trol–choline 9 and bentiromide–choline 10.[a]
Receptor
(0.5 mm)
Guest
Transport flux
Enhancement[b] Receiving
phase
(0.5 mm) (µmol cm–2 min–1
)
1
9
9
8.32ϫ10–6
Ϯ0.27ϫ10–6
1.40ϫ10–5
Ϯ0.06ϫ10–5
0.4
3.4
neutral
acidic
2
2.38ϫ10–6
0.1
2.6
neutral
acidic
Ϯ0.62ϫ10–6
1.07ϫ10–5
Ϯ0.11ϫ10–5
Control
(no host)
1
9
2.25ϫ10–5
neutral
acidic
4.12ϫ10–6
10
2.27ϫ10–5
7.0
4.0
5.7
neutral
Extraction and Transport of Lipophilic and More Complex
Substrates
Ϯ0.53ϫ10–5 [c]
2.16ϫ10–5
acidic
acidic
Ϯ0.32ϫ10–5
Amphiphilic substrates 5–7 were found to preferentially
reside in the aqueous phase in both extraction and trans-
port experiments. However, in the presence of ionizable
calixarenes 5–7 could be enticed into and through a hydro-
phobic environment. Next, we studied a more lipophilic
choline-drug conjugate, tamoxifen–choline 8. Conjugate 8
was synthesized by single N-methylation of tamoxifen. We
first evaluated the binding of 8 with 1 and 4 in CDCl3 by 1H
NMR (See Supporting Information for complete details).
NMR changes indicated the formation of fast exchange
complexes between 8 and both 1 and 4 in CDCl3. The sig-
nal attributed to the trimethylammonium moiety was
broadened and shifted upfield. We attempted U-tube ex-
periments using a saturated solution of 8 (63 µM)[19] in
milliQ water and found no evidence of transport using
either 0.5 mm of 1 (neutral or acidic receiving phase) or 2
(neutral receiving phase). Instead, we observed significant
2
10
10
3.06ϫ10–5
Ϯ0.80ϫ10–5
Control
(no host)
1[d]
3.23ϫ10–6
5.36ϫ10–6
neutral
acidic
bentiromide 4.70ϫ10–5
Ϯ0.81ϫ10–5
1.4[d]
neutral
Control[d] bentiromide 3.22ϫ10–5
Ϯ0.64ϫ10–5
acidic
[a] Transport flux measured via U-tube three-phase aparatus:
0.5 mm receptor in 10 mL DCM, 0.5 mm substrate was charged in
the source phase in 4 mL of water and a receiving phase of 4 mL
of water or 0.1 m HCl was added. After 72 h of stirring at 400 rpm
an aliquot from the receiving phase was removed and analyzed by
UV/Vis measurement at the corresponding absorbance maximum
against a calibration curve, reported as duplicate runs with Ϯmaxi-
mum deviation from the mean. [b] Enhancement: flux/flux of corre-
sponding control. [c] Reported as triplicate runs with ϮS.D. [d] At
partitioning, as one would expect for a hydrophobic guest – the source – organic interface, a thin layer of precipitate formed
and we measured a significant decrease in concentration of the
source phase (up to 50%) without an equal increase in the receiving
phase.
indeed 46–49% of 8 was extracted into DCM in the absence
of host. Host 1 enabled an additional extraction of
30.14Ϯ1.66% under neutral conditions and 14.86Ϯ1.54%
under acidic receiving phase conditions. Compound 2 en-
Finally, we turned our attention to a large polar payload,
abled an additional 14.39Ϯ0.46% extraction. We then bentiromide–choline 10. We were encouraged to see trans-
turned our attention to the parent compound, tamoxifen, port enhancement in the presence of receptors 1 and 2 un-
which lacks one N-methyl group. We prepared the tamox- der neutral and acidic conditions. We have previously dem-
ifen HCl salt in the hopes of slightly enhancing its solubility onstrated the significance and selectivity of the choline han-
relative to trimethylammonium analogue 8. Unfortunately, dle.[1] However, given the number of ionizable groups on
the HCl salt also displayed limited water solubility (less 10, we decided to confirm that the choline handle is re-
than 0.1 mm). We observed identical behavior – no trans- quired for transport for this polyfunctional molecule. Par-
port and significant partitioning into DCM in the absence ent compound bentiromide[21] presented some challenges
of host, contrasted with enhanced extraction into DCM in associated with its poor solubility. Indeed, precipitation of
the presence of host 1.
bentiromide at the solvent/environment interface was noted.
Next, we examined resveratrol analogue 9,[20] (containing The background transport rate was higher for bentiromide
two aromatic rings) which has enhanced water solubility compared to our modified 10. This is likely due to en-
(0.5 mm). This substrate showed decreased transport vs. hanced solubility in the organic phase relative to 10, which
control under a variety of conditions indicating extraction is characterized by a trimethylammonium group. Very slight
into the organic phase. However an acidic receiving phase enhancement in transport was noted in the presence of 1.
2686
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2683–2690

New Transport of Choline Conjugates and Neurotransmitters
In these measurements we used the receiving phase to calcu-
late flux, but we also measured the decrease of guest con-
centration in the source phase. Comparisons of these num-
bers allowed us to differentiate transport vs. extraction. In
the case of bentiromide, the source phase lost 50% of its
initial concentration due to precipitation and partitioning;
the receiving phase accounted for about 7% of this 50%
loss. Comparisons of transport and extraction between
bentiromide and 10 are complicated by this fact. Neverthe-
less, our results with 10 nicely illustrate the potential of this
host-conjugate + guest transport system. Clearly, enhanced
transport of a compound characterized by both significant
molecular mass and a variety of drug-like features is achiev-
able. Additionally, the trimethylammonium group of 10 was
found to result in enhanced water solubility as well as
differential behaviors in the presence or absence of
host. Despite the imperfect comparisons between 10 and
bentiromide, we think this latter result presents opportuni-
ties for other drug candidates that partition into lipophilic
phases, possess limited water solubility. Altering water solu-
bility of given agents by the addition of trimethylammo-
Figure 2. Structure of calixarene ether 11 and ester 12 receptors
and guests: dopamine 13, serotonin 14.
nium moieties may be used to enhance transport processes. 1H NMR spectroscopy. This was somewhat surprising since
isomeric butylamines showed some affinity for similar
calix[5]arene ethers.[22] It is likely that the larger aromatic
groups of 13 and 14 provide too much of a steric burden.
Calix[5]arene pentaester 12, on the other hand, displayed
Binding of Alkyl Ammonium Ions, Neurotransmitters and
Neurotransmitter Transport
strong binding affinity with both ammonium salts due to
Having demonstrated that anionic calixarenes function formation of inclusion complexes as evidenced by dramatic
as potent transporters of choline conjugates, we became in- upfield shifting of NMR signals representing the ethylene
terested in binding and transport of biologically relevant protons of both amine hydrochlorides (Supporting Infor-
amines such as serotonin and dopamine. These guests in- mation). No NMR evidence of binding of 13 or 14 with
clude an endogenous ammonium group. Indeed calixarenes calix[4]arene tetraphosphonic acid 1 was observed in a vari-
can sequester organic ammoniums via cation-π interac- ety of solvent systems. Nevertheless, we included ionizable
tions,[3] calix[5]arenes and calixarene pentesters with C5v calix[n]arenes alongside pentaether 11 and pentaester 12 in
cone conformations are noteworthy in this regard.[22,23] our transport studies.
Calix[6]arene esters also display this behavior.[24] Hence,
When dopamine 13 was subjected to three-phase trans-
calix[5]arene ether and ester derivatives 11 and 12 were port vs. calixarene receptors, we found many favorable com-
added to this study as potential transporters of neurotrans- binations that exceed transport levels in the absence of host.
mitters 13 and 14 (Figure 2).
Despite having no clear NMR-based evidence of binding,
We first confirmed the pH sensitive complexation and calix[4]arene phosphonic acid 1 transported 22 times more
decomplexation of host 11 with n-octylamine (Supporting dopamine than was observed in control experiments under
Information). The 1H NMR of 11 and n-octylamine did not neutral receiving phase conditions (Table 5). Acidification
show any evidence of guest binding, but dramatic changes of the receiving phase attenuated this effect, yet a 13-fold
were observed in the upfield region upon protonation of enhancement in transport was still achieved. Calix[4]arene
the amine following addition of CF3COOD. Based on inte- carboxylic acid 2 induced a 2-fold enhancement in trans-
gration and a literature report[22] a 1:1 inclusion complex is port, whereas the larger calix[5]arene carboxylic acid 3 en-
evident. Decomplexation occurred upon addition of diiso- abled a 9-fold enhancement under neutral conditions. The
propylamine and this cycle could be repeated (Supporting non-ionizable receptors 11 and 12, (the latter a potent
Information). From these observations it was clear that binder as observed by NMR) failed to effect transport, al-
calix[5]arene derivatives can effectively bind primary though 12 appeared to serve as an extractor of 13 since
ammonium ions and release the bound ammonium ion as transport was 0.8 times that of the appropriate control sys-
a function of pH, in an NMR tube.
tems.
We then turned our attention to the neurotransmitters
Serotonin 14 presented a different story; no background
dopamine 13 and serotonin 14, both possess endogenous transport was measurable and thus comparisons of en-
ammoniums. Unfortunately, calix[5]arene benzyloxy ether hancements were not possible. Nevertheless we found that
11, which was a good receptor of n-octylammonium tri- 1 was the most potent transporter of 14 as measured by
fluoroacetate, did not show any evidence of binding to transport flux alone (7.73ϫ10–5, Table 6) and that 1 is also
dopamine 13 or serotonin 14 hydrochlorides by means of a transporter of dopamine with a similar degree of effi-
Eur. J. Org. Chem. 2015, 2683–2690
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2687

B. B. Adhikari, S. Roshandel, A. Fujii, M. P. Schramm
FULL PAPER
Table 5. Three-phase transport of dopamine 13.[a]
ciency (1.18ϫ10–4, Table 5). Of the carboxylic acids 2 and
3, the latter displayed greater potential. Relative to the low-
est transport levels in this series of hosts and receiving
phases, phosphonic acid 1 under neutral conditions and
calix[5]arene carboxylic acid 3 appear to be superior. The
non-ionizable receptors 11 and 12 again failed to compete
with 1–3.
Receptor
(0.5 mm)
Guest
Transport flux
Enhancement[b] Receiving
phase
(5.0 mm) (µmol cm–2 min–1
)
1
2
13
13
1.18ϫ10–4
Ϯ0.02ϫ10–4
6.38ϫ10–5
Ϯ0.41ϫ10–5
22.1
13.1
neutral
acidic
9.91ϫ10–6
1.9
2.0
9.0
2.9
neutral
acidic
Ϯ1.12ϫ10–6
9.75ϫ10–6
Transport of Ammonium Carboxylate-Conjugates in Water
and Buffer
Ϯ0.25ϫ10–6
3
13
4.80ϫ10–5
neutral
acidic
In addition to the trimethylammonium and ammonium
handles, we identified one final host complementary handle
in our studies. We originally envisioned calcein 15 (Fig-
ure 3) as a control molecule for ongoing work we are doing
in vesicle systems. Indeed, our original work on transport
unveiled very different behaviors with carboxyfluorescein
(no transport) versus FITC–choline (transport).[1] We ob-
served some transport in vesicle systems for calcein. Lack-
ing a trimethylammonium, or ammonium group we carried
out three-phase transport experiments as described above
and found amplified guest molecule transport in response
to the use of host 1. In water no transport of guest 15 was
observed; the addition of 1 afforded a flux of 4.09ϫ10–6
(µM cm–2 min–1). We first noted this result under buffered
conditions and when the source and receiving phase were
changed from water to 10 mm HEPES Buffer (pH 7.4)
transport of 15 in the absence of host was observed
(2.85ϫ10–6); addition of 1 to the organic phase led to a
≈ 4-fold increase in transport (1.11ϫ10–5). This is our first
report using buffered solutions. These buffered systems
more closely approximate those in which we envision ulti-
mate application of the ideas reported herein. Moreover, we
propose that the ammonium dicarboxylate group present in
15 serves as a third handle for receptor 1. Work in both
buffer and with this new handle is underway.
Ϯ0.33ϫ10–5
1.34ϫ10–5
Ϯ0.27ϫ10–5
11
12
13
13
0.00
0
neutral
neutral
4.21ϫ10–6
0.8
Ϯ0.02ϫ10–6
Control
(no host)
13
5.34ϫ10–6
4.87ϫ10–6
neutral
acidic
[a] Transport flux measured via U-tube three-phase aparatus:
0.5 mm receptor in 10 mL DCM, 5.0 mm substrate was charged in
the source phase in 4 mL of water and a receiving phase of 4 mL
of water or 0.1 m HCl was added. After 72 h of stirring at 400 rpm
an aliquot from the receiving phase was removed and analyzed by
UV/Vis measurement at the corresponding absorbance maximum
against a calibration curve, reported as duplicate runs with Ϯmaxi-
mum deviation from the mean. [b] Enhancement: flux/flux of corre-
sponding control.
Table 6. Three-phase transport of serotonin 14.[a]
Receptor
Guest
Transport flux
Relative Receiving
(0.5 mm) (5.0 mm) (µmol cm–2 min–1) transport[b] phase
1
2
3
14
14
14
7.73ϫ10–5
Ϯ0.10ϫ10–5
5.03ϫ10–5
Ϯ0.84ϫ10–5
1.83ϫ10–5
Ϯ0.23ϫ10–5
5.67ϫ10–6
Ϯ0.43ϫ10–6
2.59ϫ10–5
Ϯ0.02ϫ10–5
5.22ϫ10–5
Ϯ0.40ϫ10–5
13.6
8.9
neutral
acidic
3.2
1.0
neutral
acidic
4.6
9.2
neutral
acidic
11
12
14
14
0.00
neutral
neutral
Figure 3. Structure of calcein 15.
6.06ϫ10–7
Ϯ1.38ϫ10–7
Control
(no host)
14
0.00
0.00
neutral
acidic
Conclusions
[a] Transport flux measured via U-tube three-phase aparatus:
0.5 mm receptor in 10 mL DCM, 5.0 mm substrate was charged in
the source phase in 4 mL of water and a receiving phase of 4 mL
of water or 0.1 m HCl was added. After 72 h of stirring at 400 rpm
an aliquot from the receiving phase was removed and analyzed by
UV/Vis measurement at the corresponding absorbance maximum
against a calibration curve, reported as duplicate runs with Ϯmaxi-
mum deviation from the mean. [b] Enhancement could not be de-
termined due to lack of transport in the absence of host; we report
relative transport compared to all ionizable hosts (1–3).
We have successfully demonstrated a variety of calix-
arene-mediated choline-drug conjugate and neurotransmit-
ter transport processes. The key feature is a receptor com-
plementary handle. Choline serves this role well for a vari-
ety of phosphonic acid and carboxylic acid functionalized
receptors. Previously, we had shown that choline-fluores-
cein and choline-NBD dye conjugates were transported ac-
ross a three-phase liquid membrane system with significant
2688
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2683–2690

New Transport of Choline Conjugates and Neurotransmitters
enhancements in the presence of calixarenes 1, 2, and 4.[1]
The parent compounds carboxyfluorescein and NBD chlor-
ide, both lacking receptor complementary choline handles
failed to undergo such transport. We now show that this
strategy extends beyond fluorophores; an amino acid, acet-
aminophen, ibuprofen and bentiromide also are amenable
to such transport processes. The neurotransmitters dopa-
mine and serotonin with ionizable amines work as well. We
had very modest results with dimethylresveratrol–choline
and tamoxifen–choline resisted transport altogether. We
find that these more lipophilic substrates present a con-
siderable challenge to our transport system – they partition
into a lipophilic environment and calixarenes can even en-
hance this extraction event. Can we develop a strategy to
release lipophilic substrates? We hope that a new handle or
host will evolve in answer to this challenge.
We now know that NMR is not powerful as a predictive
tool for transport – indeed non-ionizable calixarene ether
11 and ester 12 serve as hosts for neurotransmitters, but
do not serve as effective transporters. Acids 1–3 serve as
transporters for 13–14, but not as hosts (via NMR). This
realization will, no doubt, be important to the development
of future systems – weak binders should not be ruled out as
possible transporters.
purchased from Cambridge Isotope Laboratories and residual sol-
vent peaks were used as internal standards. SiliaFlash® P60 Silica
Gel, 40–63 μm, 60A was purchased from Silicycle [Quebec City
(Quebec), CA]. Mass analyses were conducted at the University of
California, Riverside High Resolution Mass Spectrometry Facility,
Riverside, CA, USA. See Supporing Information for full synthetic
procedures and spectroscopic data of new compounds.
Extraction Experiments: Two-phase extraction experiments were
carried out in 20 mLglass vials. The organic phase consisted of a
solution of 0.5 mm for calixarene-based receptors in DCM and the
aqueous phase consisted of a solution of 5 in milliQ water. The
substrate concentration was varied from 0.1 mm to 3.0 mm. Equal
volumes (2 mL each) of organic and aqueous phases were mixed in
the vial and the mixture was stirred at 160 rpm for 24 h. After
cessation of stirring and phase separation, samples from aqueous
phase were analyzed with UV/Vis spectrophotometer to determine
the fraction transported in organic phase. The experiments were
conducted in duplicate and the results presented are averages. Con-
trol experiments were performed without the host.
U-Tube Transport Experiments: Three-phase U-tube transport ex-
periments were conducted by using an U-shaped glass tube as
shown below (Figure 4). The organic phase consisted of 10 mL
solution 0.5 mm calixarene-based receptors in DCM. The source
phase consisted of 4 mL aqueous solution of a substrate in MilliQ
water. The receiving phase was either pure milliQ water, 0.1 m HCl
or 10 mm HEPES buffer. The U-tube’s upper ends were sealed and
the experiments were started by stirring the organic phase with
magnetic stirring bar being rotated at 400 rpm. The samples from
source phase as well as receiving phase were drawn after 72 h and
analyzed with UV/Vis spectrophotometer after adequate dilution
with milliQ water against a standard serial dilution curve. Two sets
of experiments were conducted and the data presented are the
average values. Control experiments were performed using pure
DCM as the organic phase.
The correlation between extraction and transport is less
clear-cut but some testable behaviors emerged. Most no-
table was the comparison between acetaminophen 6 and
ibuprofen 7. Compound 6 is more hydrophilic and we found
that calix[4]arene carboxylic acid was the superior trans-
porter for 6 under acidic receiving phase conditions. Con-
versely, the slightly more lipophilic 7 was moderately more
amenable to transport by the calix[4]arene phosphonic acid
1.
Looking forward, this strategy provides new ideas and
approaches to the drug transport problem. We hope to
demonstrate that a drug can be appended to a receptor
complementary handle and that the receptor, once lodged
in a cell membrane, can serve as a selective shuttle. More-
over, we aim to transport multiple complementary guests
per receptor molecule. This strategy diverges from emulsifi-
cation or covalent conjugate strategies yet is not so far-
fetched as to prohibit ultimate applications in cells.[11,25] We
know that supramolecular receptors are compatible with
cells and we can monitor their localization tendencies in
giant unilamellar vesicles.[9] A final report on the behavior
of guests in buffer as well as a comparison of trimethylam-
monium vs. ammonium vs. ammonium dicarboxylate spe-
cies will be reported in due course. Our first report using
vesicles to visualize and quantitate transport will follow
shortly thereafter.
Figure 4. U-tube dimensions.
Acknowledgments
The project described was supported by the National Cancer Insti-
tute, grant number SC2CA167636. The content is solely the respon-
sibility of the authors and does not necessarily represent the views
of the National Cancer Institute or the National Institutes of
Health (NIH). NMR instrumentation was provided by the
National Science Foundation (NSF) (MRI CHE-1337559). Any
opinions, findings and conclusions expressed are those of the au-
thor(s) and do not necessarily reflect the views of NSF.
Experimental Section
General: All reagents were of reagent grade or better and used with-
out further purification unless otherwise noted. NMR were ac-
quired with Bruker Fourier 300-MHz or Bruker Avance III 400-
MHz spectrometers. Spectra were processed (iNMR) using a Fou-
rier transform with exponential weighting. NMR solvents were
Eur. J. Org. Chem. 2015, 2683–2690
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2689

B. B. Adhikari, S. Roshandel, A. Fujii, M. P. Schramm
FULL PAPER
[14] P. Breccia, M. Van Gool, R. Perez-Fernandez, S. Martin-San-
tamaria, F. Gago, P. Prados, J. de Mendoza, J. Am. Chem. Soc.
2003, 125, 8270–8284.
[15] R. Perez-Fernandez, M. Pittelkow, A. M. Belenguer, L. A.
Lane, C. V. Robinson, J. K. M. Sanders, Chem. Commun. 2009,
3708–3710.
[1]
[2]
B. B. Adhikari, A. Fujii, M. P. Schramm, Eur. J. Org. Chem.
2014, 2972–2979.
a) R. Langer, Science 1990, 249, 1527–1533; b) R. Haag, F.
Kratz, Angew. Chem. Int. Ed. 2006, 45, 1198–1215; Angew.
Chem. 2006, 118, 1218; c) W. M. Pardridge, Drug Discovery
Today 2007, 12, 54–61.
A. Späth, B. König, Beilstein J. Org. Chem. 2010, 6, 1.
K. Neupert-Laves, M. Dobler, Helv. Chim. Acta 1975, 58, 432–
442.
S. M. Biros, J. J. Rebek, Chem. Soc. Rev. 2007, 36, 93–104.
a) T. Iwasawa, R. J. Hooley, J. Rebek, Science 2007, 317, 493–
496; b) F. R. Pinacho Crisóstomo, A. Lledó, S. R. Shenoy, T.
Iwasawa, J. Rebek, J. Am. Chem. Soc. 2009, 131, 7402–7410.
[16] K. Ohto, T. Yamasaki, K. Inoue, Ars Separatoria Acta 2006,
4, 96–106.
[3]
[4]
[17] A. Arduini, A. Pochini, S. Raverberi, R. Ungaro, J. Chem. Soc.,
Chem. Commun. 1984, 981–982.
[18] B. B. Adhikari, M. Gurung, H. Kawakita, K. Ohto, Analyst
2011, 136, 3758–3769.
[19] Compound 8 was insoluble at 0.1 mm in water. A saturated
solution was prepared, filtered and we estimated the concentra-
tion using a standard curve for 8 that was obtained using
chloroform. No significant differences were observed in the UV
spectra of compound 8 in water vs. chloroform. Nevertheless
0.063 mm remains an estimate for the solubility of compound
8 in water.
[20] M. P. Schramm, Vol. Serial No.: 62/077,780, our Reference
No.: 195893 7.00002 (Ed.: U. S. P. P. Application), USA, 2014.
[21]
[5]
[6]
[7] M. P. Schramm, R. J. Hooley, J. Rebek, J. Am. Chem. Soc.
2007, 129, 9773–9779.
[8] Y. J. Kim, M. T. Lek, M. P. Schramm, Chem. Commun. 2011,
47, 9636–9638.
[9] K. M. Feher, H. Hoang, M. P. Schramm, New J. Chem. 2012,
36, 874–876.
[10] Y. Liu, P. Liao, Q. Cheng, R. J. Hooley, J. Am. Chem. Soc.
2010, 132, 10383–10390.
[11] Y.-J. Ghang, M. P. Schramm, F. Zhang, R. A. Acey, C. N. Da-
vid, E. H. Wilson, Y. Wang, Q. Cheng, R. J. Hooley, J. Am.
Chem. Soc. 2013, 135, 7090–7093.
[12] a) H. Tao, D. Cao, L. Liu, Y. Kou, L. Wang, H. Meier, Sci.
China Chem. 2012, 55, 223–228; b) K. D. Daze, T. Pinter, C. S.
Beshara, A. Ibraheem, S. A. Minaker, M. C. F. Ma, R. J. M.
Courtemanche, R. E. Campbell, F. Hof, Chem. Sci. 2012, 3,
2695–2699; c) T. Ogoshi, S. Kanai, S. Fujinami, T.-a. Yamagi-
shi, Y. Nakamoto, J. Am. Chem. Soc. 2008, 130, 5022–5023; d)
R. Zadmard, T. Schrader, J. Am. Chem. Soc. 2004, 127, 904–
915; e) K. N. Koh, K. Araki, A. Ikeda, H. Otsuka, S. Shinkai,
J. Am. Chem. Soc. 1996, 118, 755–758; f) S. Shinkai, K. Araki,
T. Matsuda, O. Manabe, Bull. Chem. Soc. Jpn. 1989, 62, 3856–
3862; g) S. Shinkai, K. Araki, O. Manabe, J. Am. Chem. Soc.
1988, 110, 7214–7215.
[22] F. Arnaud-Neu, S. Fuangswasdi, A. Notti, S. Pappalardo,
M. F. Parisi, Angew. Chem. Int. Ed. 1998, 37, 112–114; Angew.
Chem. 1998, 110, 120.
[23] S. Pappalardo, M. F. Parisi, J. Org. Chem. 1996, 61, 8724–8725.
[24] S.-Y. Han, M.-H. Kang, Y. Jung, S.-K. Chang, J. Chem. Soc.
Perkin Trans. 2 1994, 835–839.
[25] R. Lalor, H. Baillie-Johnson, C. Redshaw, S. E. Matthews, A.
Mueller, J. Am. Chem. Soc. 2008, 130, 2892–2893.
Received: November 24, 2014
[13] a) Y. Nakatsuji, M. Okahara, Pure Appl. Chem. 1993, 65, 557–
561; b) A. J. Wright, S. E. Matthews, W. B. Fischer, P. D. Beer,
Chem. Eur. J. 2001, 7, 3474–3481.
Published Online: March 9, 2015
2690
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 2683–2690