Efficient transport of aromatic amino acids by sapphyrin-lasalocid conjugates

Article abstract of DOI:10.1002/(sici)1521-3765(199801)4:1<159::aid-chem159>3.0.co;2-n

The synthesis and characterization of several sapphyrin-lasalocid conjugates is reported. This family of receptors is capable of acting as efficient and selective carriers for aromatic α-amino acids, as judged from both U-tube and W-tube through-model-mem

Full text of DOI:10.1002/(sici)1521-3765(199801)4:1<159::aid-chem159>3.0.co;2-n

FULL PAPER
Efficient Transport of Aromatic Amino Acids by
Sapphyrin ± Lasalocid Conjugates
Jonathan L. Sessler* and Andrei Andrievsky
Abstract: The synthesis and character-
ization of several sapphyrin ± lasalocid
conjugates is reported. This family of
receptors is capable of acting as efficient
and selective carriers for aromatic a-
amino acids, as judged from both U-tube
and W-tube through-model-membrane
transport experiments. The first member
of this family, system 6, was found to
display an inherent preference for phen-
ylalanine > tryptophan > tyrosine. Fur-
ther, l-amino acids were shown to be
transported with greater efficiency than
the corresponding d-enantiomers by this
particular carrier. The high level of
amino acid carrier capability displayed
by receptor 6 in dichloromethane solu-
tions correlates well with the results of
equilibrium binding studies carried out
using visible-spectroscopic titrations.
These latter studies revealed that system
6 does display significant affinity for
zwitterionic amino acids in this organic
solvent. These binding studies, as well as
a number of control experiments involv-
ing, inter alia, porphyrin ± lasalocid con-
jugate 7, showed the importance of
having both the sapphyrin and lasalocid
subunits contained within the same
overall receptor framework. The four
other second-generation sapphyrin ± la-
salocid conjugates reported here (11 ±
14) were also tested as carriers for the
transport of Phe, Trp, and Tyr. It was
found that the esterified systems 11 and
12 functioned well as amino acid carri-
ers, while the free-acid compounds 13
and 14 did not. These latter conjugates,
containing both carboxylic acid and
sapphyrin subunits, presumably undergo
self-assembly in organic solutions, a
process that hampers their ability to act
as effective carriers. In the case of the
functioning systems 11 and 12, the con-
figuration of the stereogenic phenylala-
nine appendages could be varied such
that either the l- or d-antipodes of the
aromatic amino acid substrates being
studied were transported at a greater
rate.
Keywords: amino acids
phores molecular recognition
sapphyrin ´ amino acid transport
´
iono-
´
´
analysis,[7] and from a desire to extend what is learned with
amino acids into the arena of peptide recognition.[8]
Introduction
As a part of a program concerned with the generation of
efficient methods for effecting the trans-membrane transport
of important biological species,[9] we became interested in
developing carriers capable of facilitating the transport of a-
amino acids across lipophilic barriers. At present, only a
minute amount of information is available concerning the
dynamics and regulation of amino acid transport in nature.[10]
Hence, our approach to the construction of abiotic amino acid
carriers has been guided as much by intuition as by an analysis
of natural amino acid binding systems. The very structure of
amino acids, however, invites us to explore a variety of
possible host ± guest interactions. In particular, the zwitter-
ionic form of this species contains a negatively charged
carboxylate group, a positively charged ammonium moiety,
and a side chain. Therefore, we came to appreciate that an
ideal carrier for amino acids would a) have different, nonself-
complementary binding sites capable of recognizing all three
parts of a putative amino acid substrate, b) have enough
extraction power to compensate for the energetic desolvation
costs associated with bringing the heavily hydrated carbox-
Nature uses amino acids as the building blocks for construc-
tion of proteins,[1] and as molecular messengers capable of
transmitting information in living organisms.[2] Therefore,
elaborate procedures for synthesizing, manipulating, and
transporting these key biological molecules have developed.
In all cases, the critical processes rely on molecular recog-
nition of amino acids, which provides the basis for a precise,
selective read out of the chemical information inherent in
their structure and functionality.[3] Given the importance of
the common amino acids, it is not surprising that supra-
molecular chemists have long been interested in designing
abiotic receptors and carriers for this group of compounds.[4±6]
Here, some of the motivating factors have come from a need
to develop efficient methods of amino acid sensing and
[*] Prof. J. L. Sessler, A. Andrievsky
Department of Chemistry and Biochemistry
The University of Texas at Austin
Austin, Texas 78712 (USA)
Fax: Int. code ‡ 1512471-7550
e-mail: sessler@mail.utexas.edu
Chem. Eur. J. 1998, 4, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0159 $ 17.50+.25/0
159

J. L. Sessler and A. Andrievsky
FULL PAPER
ylate and ammonium groups out of the aqueous environment,
c) be able to release the substrates into an aqueous receiving
phase rapidly, and, ideally, d) display stereochemically de-
pendent interactions with the substrates, thereby allowing for
their enantioselective transport.[11]
The above design requirements are, of course, very severe.
Compounding the problem is the fact that the ammonium and
carboxylate functionalities of a-amino acids are tied up in the
form of a zwitterion. This makes each of these functional
groups more difficult to chelate.[12] In view of this, much work
in the area has been concerned with the problem of
recognizing derivatized amino acids (i.e., those in which
either the carboxylate or ammonium group is protected).[5,6]
For the same reason, many of the amino acid transport studies
have been carried out under conditions of either high or low
pH.[13]
As our ammonium binding group, we have chosen the
naturally occurring polyether ionophore lasalocid (cf. parent
structure 3, Scheme 1).[18±20] Lasalocid itself is composed of a
conformationally constrained cyclic polyether chain termi-
nated at one end by a carboxylic acid and at the other end by a
hydroxyl group. These structural features allow lasalocid to
form intramolecular head-to-tail hydrogen bonds, thereby
generating a pseudocyclic cavity capable of binding cations.
Indeed, lasalocid is known to bind and transport a range of
cations across both biological and artificial membranes.[19,20]
Importantly, lasalocid has been shown to effect the enantio-
selective recognition of chiral ammonium cations.[20a±c] On this
basis it was thought that a covalently connected sapphyrin ±
lasalocid conjugate might indeed function as a bona fide
amino acid receptor.[21] In particular, it was expected that the
protonated sapphyrin portion of the conjugate would bind the
carboxylate portion of an a-amino acid,[22] while the lasalocid
subunit would simultaneously chelate the corresponding
protonated amino group. We expected that such a receptor,
featuring non-self-complementary binding sites, would be free
of problems associated with internal collapse. Further, since it
would contain an asymmetric cavity, this kind of system might
function as an enantioselective amino acid binding agent. We
have thus prepared and recently reported in communication
form the synthesis of a prototypical sapphyrin ± lasalocid
conjugate 6.[15e] In this paper full details as to the recognition
and transport properties of this system are given. Also
reported are several second-generation sapphyrin-lasalocid
conjugates (11 ± 14) which bear extra, chiral, phenylalanine-
based covalent appendages. As expected, these systems do in
fact bind and transport zwitterionic a-amino acids through
model membrane barriers with good selectivity.
In this paper we report the synthesis of carriers designed to
effect the selective transport of zwitterionic a-amino acids.
We have shown previously that sapphyrins (cf. parent
structure 1), when protonated,[14] unlike their simple porphy-
rin congeners (e.g., 2), act as excellent receptors for a variety
of anions,[9b,d±h,15] including carboxylates.[16] Comprehensive
solution and solid-state data[17] show that the bound carbox-
ylate anion is located above the protonated sapphyrin plane,
held in place by a combination of electrostatic interactions
and NH-to-O hydrogen bonds. Given this, it was surmised that
a sapphyrin subunit, once conjugated to an appropriate
receptor for ammonium groups, would function as a receptor
for zwitterionic amino acids.
Results and Discussion
Synthesis of the sapphyrin ± lasalocid conjugate 6 and por-
phyrin ± lasalocid conjugate 7: The sapphyrin ± lasalocid con-
jugate 6 was prepared from its constituents in a stepwise
fashion as illustrated in Scheme 1. First, lasalocid (3) was
coupled with mono-tert-butyloxycarbonylethylenediamine,[23]
yielding the amide 4. After deprotection with TFA, the
resulting lasalocid amino derivative 5 was treated with the
Scheme 1. Synthesis of sapphyrin ± lasalocid conjugate 6, porphyrin ± lasalocid conjugate 7, and lasalocid methyl ester 8. Reagents and conditions:
i) H2N(CH2)2NH(t-Boc), EDC, HOBt, Py, DMF, 72 h; ii) TFA, CH2Cl2, 2 h; iii) 1a, EDC, HOBt, Py, DMF, 48 h; iv) 2, EDC, HOBt, Py, DMF, 48 h;
v) EDC, HOBt, MeOH, 48 h.
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0160 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 1
160

Sapphyrin ± Lasalocid
159±167
sapphyrin monoacid 1a[9d] in DMF, using EDC as a coupling
agent. Compound 6 was isolated in approximately 73% yield
after careful chromatographic purification. By means of the
same approach, the structurally cognate porphyrin ± lasalocid
conjugate 7 was synthesized from the porphyrin monoacid 2[24]
as a control for compound 6. To the best of our knowledge,
these systems constitute the first examples of porphyrin and
expanded-porphyrin appended lasalocids.[15e]
U-tube amino acid transport studies with conjugates 6 and 7:
The aromatic a-amino acids phenylalanine, tryptophan, and
tyrosine were selected for the transport studies, because
a) the side chains of these amino acids could provide for
additional, stereogenic van der Waals-type interactions in-
volving both the aromatic surfaces of the sapphyrin macrocycle
and the salicylate portion of the lasalocid subunit (thereby
favoring selective transport), and b) the transport rates of
these aromatic substrates can be monitored easily by HPLC
analyses (by means of a UV detector). The transport experi-
ments themselves were then run using a Pressman-type U-tube
model membrane system.[25] Here, the initial phase, Aq. I, was
charged with either a solution of the particular zwitterionic a-
amino acid under study, or a mixture of all three amino acids
being considered (Phe, Trp, Tyr), while the receiving phase,
Aq. II, was made basic to increase the off-rates.[26]
It was found that at neutral pH compound 6 acts as a very
efficient carrier for phenylalanine and tryptophan, but not
tyrosine (Table 1, Figure 1). When this same system was tested
under conditions of competitive transport, it was found that l-
phenylalanine was transported four times faster than l-
tryptophan and a thousand times faster than l-tyrosine
(Table 1, entry 1). Interestingly, l-amino acids were trans-
ported with greater efficiency by this carrier than the
corresponding d-antipodes. Taken together, these results are
consistent with a high level of amino-acid-based selectivity,
and provide clear evidence that the amino acid side chains
actively participate in the binding and transport process.
Figure 1. Relative rates of l-Phe and l-Trp transport as determined from
U-tube model membrane experiments.
To help us understand more fully the above results, a
number of control experiments were then carried out. These
were designed to elucidate the roles the constituent parts of 6
could be playing in terms of mediating the overall transport
phenomenon. For instance, in order to probe the role of
carboxylate binding, the porphyrin ± lasalocid conjugate 7 was
tested. This system proved inefficacious for amino acid
transport (entry 2). Likewise, lasalocid itself (3) proved
incapable of effecting amino acid transport (entry 4). Taken
together, these results support the contention that sapphyrin-
based carboxylate binding plays a critical role in mediating
the amino acid transport effected by conjugate 6.
Interestingly, only modest transport was achieved by use of
sapphyrin 1b[27] either alone (entry 3), or as a mixture with 3
(entry 5).[28] In the latter case one could envision the
formation of putative supramolecular complexes between
protonated sapphyrin 1b and lasalocid 3 as a result of
lasalocid carboxylate chelation by sapphyrin. Formation of
Table 1. Amino acid transport rates measured in a U-tube (kt) [a, b].
1
Carrier
kt (10 5 molcm 2 h
)
kF/kW/kY (l) [c]
kF/kW/kY (d) [c]
l-Phe
d-Phe
l-Trp
d-Trp
l-Tyr
d-Tyr
1
2
3
4
5
6
7
8
6 [d]
20.0
0.7
6.9
0.5
3.2
7.8
0.8
6.4
10.5
1.9
12.7
0.9
5.0
0.3
1.4
0.2
0.9
1.4
0.2
1.0
2.5
0.8
0.7
0.01
4.2
0.2
NA [e]
0.2
0.9
1.4
0.2
2.3
1.1
0.7
0.7
0.01
0.02
< 0.001
< 0.001
< 0.001
0.2
0.02
< 0.001
NA [e]
< 0.001
0.2
1000/250/1
2.3/1
4.9/1
635/210/1
4.5/1
7 [d]
1b [d]
3 [d]
NA [e]
0.4
NA [e]
2/1
2.5/1
1b ‡ 3 [f]
1b ‡ 8 [g]
8 [d]
3.1
7.2
0.8
8.2
6.7
1.4
0.7
0.05
16/4.5/1
15.6/2.8/1
4/1
6.4/1
4.2/1
2.4/1
1.3/1
NA [e]
15.5/4.5/1
12/2.3/1
4/1
3.6/1
6.1/1
2/1
1/1
NA [e]
0.5
0.6
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
11 [d]
12 [d]
13 [d]
14 [d]
none
9
10
11
12
0.9
0.05
[a] Transport experiments were performed as described in ref. [9d] under competitive conditions (with respect to amino acid substrates): the initial rate
values given for l-Phe, l-Trp, and l-Tyr transport are thus obtained from experiments involving mixtures of l-Phe (50mm), l-Trp (50mm), and l-Tyr (5 mm)
in the initial aqueous phase, Aq. I. Likewise, those given for d-Phe, d-Trp and d-Tyr are derived from studies involving analogous mixtures of d-Phe (50mm),
d-Trp (50mm) and d-Tyr (5 mm) in Aq. I. In all runs 10mm NaOH was used as the receiving phase, Aq. II. [b] Values are averages of two or three separate
experimental runs; estimated errors are Æ 15%. [c] Single-letter symbols for amino acids are used: F ˆ Phe, Wˆ Trp, and Yˆ Tyr. When Tyr initial
transport rate was less than 10 8 molcm 2 h 1, the kF/kW ratio was used instead of kF/kW/kY. [d] 1  10 4 m in dichloromethane. [e] NA ˆ not measured due to
achirality of the carrier. [f] Organic phase containing a mixture of 1b (1 Â 10 4 m) and 3 (1 Â 10 4 m) in dichloromethane was used. [g] A mixture of 1b (1 Â
10 4 m) and 8 (1 Â 10 4 m) in dichloromethane was used as the organic phase.
Chem. Eur. J. 1998, 4, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0161 $ 17.50+.25/0
161

J. L. Sessler and A. Andrievsky
FULL PAPER
such complexes would inevitably reduce the rate of amino
acid transport by the mixed 1b ‡ 3 system. To exclude this
possibility, a control experiment involving a mixture of
sapphyrin 1b and lasalocid methyl ester 8 (synthesized as
shown in Scheme 1) was carried out (entry 6). Even in this
case, however, no significant increase in transport rate was
observed compared to the 1b ‡ 3 mixture.
Amino acid binding efficacy in dichloromethane was
assessed quantitatively by visible-light spectroscopic titra-
tions. Upon addition of aromatic amino acids, the Soret
maximum of conjugate 6 underwent an approximate 5 nm
blue shift from 452 nm to 447 nm that was accompanied by the
growing-in of a new band also at 447 nm. By following the
increase in absorbance at 447 nm as a function of substrate-to-
receptor ratio, we determined both the relative association
constants (Ka) and the stoichiometry of binding.[29] For
instance, it was found that compound 6 forms 1:1 complexes
with phenylalanine and tryptophan with association constants
Figure 2. Proposed mechanism for amino acid transport catalyzed by the
sapphyrin ± lasalocid conjugate 6. Pentagon, sapphyrin; horseshoe, lasalo-
cid.
1
(Ka) as follows: Ka(l-Phe) ˆ 4.86  105 m , Ka(d-Phe) ˆ
1
1
5.35  105 m , Ka(l-Trp) ˆ 0.83  105 m , Ka(d-Trp) ˆ 0.94 Â
part of carrier 6 chelates a sodium cation and carries it back to
‡
1 [15e]
105 m .
the Aq. I/CH2Cl2 interface. Release of Na and protonation of
the sapphyrin macrocycle then occurs at this locus. This, in
turn, regenerates carrier 6 in a form ready to effect amino acid
transport again. Consistent with this mechanism is the fact
that addition of anions (e.g., Cl , F )[32] to the organic phase
did not result in a significant change in the transport rates
(Table 2, entries 3, 4). This is most likely because the putative
complex of the conjugate 6 with amino acids is a neutral,
tightly bound entity, and does not require any extra anionic
species for its efficient through-membrane transport. Also in
accord with the proposed mechanism was the detection of
The high levels of absolute receptor-to-substrate binding
affinity and amino acid binding selectivity are in accord with
the transport results. However, the finding that the l-
enantiomers are bound with the same affinity (within error)
as the d-congeners may reflect the fact that not only binding
affinities, but also the substrate release rates, are important in
terms of mediating transport efficiency.[3b] Consistent with this
contention is the fact that when water was used as a receiving
phase (as opposed to a sodium hydroxide solution), the l/d
ratio of the transport rates was found to be still further
amplified (Table 2, entries 1, 2).
‡
countertransported Na cations in Aq. I (see Experimental
Section), and the observation of an increase in the pH of Aq. I
( ꢀ 0.3 pH units over 24 h) over the course of the transport
experiments.[33,34]
Table 2. Amino acid transport rates (kt) determined from U-tube experi-
ments using carrier 6 [a].
Taken together, these results lead us to conclude that
system 6 is an efficient carrier for aromatic a-amino acids and
one that shows inherent selectivity within this limited
substrate subset. This receptor and its amino acid substrates
react to form lipophilic, well-defined supramolecular com-
plexes that are stabilized by stereogenic multiple point
interactions. The stable, neutral character of these complexes
provides the basis for the fast, enantiomerically dependent
amino acid transport observed in the model U-tube mem-
branes.
1
Carrier
Aq. II
kt (10 5 molcm 2 h
l-Phe
)
d-Phe
1
2
3
4
6 [b]
6 [b]
10mm NaOH
H2O
10mm NaOH
10mm NaOH
23.9
21.7
19.4
19.3
15.4
10.2
14.3
13.8
[c]
6 ‡ Cl
[d]
6 ‡ F
[a] In these experiments, the initial phase, Aq. I, contained a single amino
acid: either l-Phe (50mm) or d-Phe (50mm). [b] 1 Â 10 4 m in dichloro-
methane. [c] A mixture of 6 (1 Â 10 4 m) and Bu4NCl (1 Â 10 4 m) in
dichloromethane was used as the organic phase. [d] A mixture of 6 (1 Â
10 4 m) and Bu4NF (1 Â 10 4 m) in dichloromethane was used as the organic
phase.
Synthesis and U-tube amino acid transport of the second-
generation sapphyrin ± lasalocid conjugates 11 ± 14: Encour-
aged by the successful transport results obtained using 6 as a
carrier, we decided to take this methodology a step further
and to prepare two new, second-generation sapphyrin ± lasa-
locid conjugates 13 and 14, featuring phenylalanine-derived
carboxylate appendages. The reasons for this were threefold:
first, we were interested to see whether the use of a
zwitterionic carrier would be advantageous for the transport
of zwitterionic amino acids. Secondly, we wanted to examine
how the chirality of phenylalanine adjuvants affects the amino
acid transport selectivity. Finally, on a more practical level, we
considered that the carboxylate tail of 13 and 14 could provide
Taken together, the above results lead us to propose the
transport mechanism outlined in Figure 2. In particular, it is
suggested that carrier 6, featuring a monoprotonated sap-
phyrin macrocycle[22] and a deprotonated lasalocid,[30] binds
zwitterionic amino acid at the Aq. I/CH2Cl2 interface, forming
a neutral complex.[31] This complex then diffuses through the
membrane and is released in the more basic receiving phase,
Aq. II. The amino acid thus transported then reacts with
NaOH to form an amino acid sodium salt. This transformation
and sapphyrin deprotonation at the CH2Cl2/Aq. II interface
facilitates product release. Once at this interface, the lasalocid
162
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0162 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 1

Sapphyrin ± Lasalocid
159±167
a handle allowing these receptors to be attached to a solid
support. These latter materials could act as chromatographic
media for the separation and analysis of amino acid mix-
tures.[35] Here we realized, of course, that the protonated
sapphyrin macrocycle and the anionic carboxylate of 13 are
mutually complementary. This feature could result in the
formation of the higher-order self-assembled aggregates of
13 in solution (e.g., non-covalent dimers), as has been
To test this idea, compounds 12 and 14 were prepared; these
incorporate a d-phenylalanine residue instead of the l-phenyl-
alanine subunit found in 11 and 13 (Scheme 2). This inversion
of chirality in the phenylalanine appendage led, as expected, to
a reversal (compared to compound 11) of the observed amino
acid transport selectivity. In particular, when the U-tube trans-
port experiments were carried out with conjugate 12 (conditions
as described in Table 1), l-Phe was found to be transported
1.6 times faster than d-Phe. Likewise, l-Trp was found to be
transported 2.3 times faster than d-Trp (Table 1, entry 9).[39,40]
shown previously with the zwitterionic sapphyrin monoacid
36]
1a[16b,
and with the other systems.[37] Therefore, we were
interested to see which process (self-assembly versus amino
acid binding) is dominant under conditions of our transport
studies.
W-Tube transport experiments: To compare aromatic a-
amino acid transport rates under conditions where two
carriers were competing for two different amino acids (e.g.
phenylalanine and tryptophan), transport experiments were
carried out using a W-tube model membrane.[41] These W-tube
experiments involve Aq. II/organic/Aq. I/organic/Aq. III ar-
rangements and provide a better way of comparing efficien-
cies and selectivities under competitive conditions. They could
also provide a better model for the chemical conditions of
nature wherein a variety of ion-transporting entities are found
to compete simultaneously for the same pool of ions. In the
present instance, we chose to compare 6, the most efficient
carrier, with each of the second-best systems, 11 and 12. The
results obtained (Table 3) are consistent with what one would
The synthesis of compound 13 is depicted in Scheme 2.
The sapphyrin bis-acid 1c was coupled with one equivalent of
l-phenylalanine tert-butyl ester to yield monoacid mono-
amide 9 in 67% yield after careful chromatographic purifi-
cation. Compound 9 was further conjugated with the lasalocid
amino derivative 5 to yield the protected product 11. It was
then deprotected with TFA to produce the desired compound
13. Transport studies using 13 showed, however, that it is a
very poor carrier for aromatic zwitterionic a-amino acids
(Table 1, entry 10, and Figure 1). This is because it self-
assembles in organic media and this prevents it from binding
an amino acid and effecting the through-membrane transport
of these substrates.[38] Not so with its precursor 11! This
compound, when used in our standard U-tube model mem-
brane transport studies, proved to be a much better carrier
than 13 for both phenylalanine and tryptophan (Table 1,
entry 8).
Table 3. Amino acid transport rates (kt) determined from W-tube experi-
ments carried out under competitive conditions.
Carrier
left arm right arm
Amino acids
Left arm [a]
Phe Trp
Right arm [a]
Phe
Trp
Interestingly, the enantiomeric selectivity of transport with
carrier 11 was reversed compared to 6. In particular, d-
enantiomers of amino acid substrates were transported some-
what faster by 11 than the l-ones. This effect is clearly
attributable to the presence of the l-phenylalanine appendage
in 11. Specifically, it was considered likely that the chirality of
the lasalocid component and the l-phenylalanine tert-butyl
ester moiety act to exert a contradictory effect on amino acid
transport selectivity. It was thus thought that replacing the l-
phenylalanine tert-butyl ester with its d-analogue would lead
to a more synergistic effect.
1
2
3
4
6
6
6
6
11
11
12
12
l-Phe, l-Trp [b]
10.3 2.1
1.8
11.5 2.3
d-Phe, d-Trp [c] 7.7
1.8
3.0
4.9
8.7
4.2
1.1
2.3
1.9
1.0
d-Phe, d-Trp [c] 7.6
l-Phe, l-Trp [b]
[a] Initial amino acid transport rates (kt, 10 5 molcm 2 h 1) were measured
by HPLC (UV detection). Two separated organic phases containing
dichloromethane solutions of competing carrier compounds (1 Â 10 4 m in
each) were put in contact with the same initial phase, Aq. I. In all runs
10mm NaOH was used as the two receiving phases, Aq. II and Aq. III.
[b] The initial phase, Aq. I, contained mixtures of l-Phe (50mm) and l-Trp
(50mm). [c] The initial phase, Aq. I, contained mixtures of d-Phe (50mm)
and d-Trp (50mm).
Scheme 2. Synthesis of sapphyrin ± lasalocid conjugates 11 ± 14. Reagents and conditions: i) l- or d-phenylalanine tert-butyl ester hydrochloride, EDC,
HOBt, Py, DMF, 72 h; ii) 5, EDC, HOBt, Py, DMF, 48 h; iii) TFA, CH2Cl2, 4 h.
Chem. Eur. J. 1998, 4, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0163 $ 17.50+.25/0
163

J. L. Sessler and A. Andrievsky
FULL PAPER
purchased from Aldrich. Phenylalanine, tryptophan, tyrosine, and l-
phenylalanine tert-butyl ester hydrochloride were purchased from Sigma.
d-Phenylalanine tert-butyl ester hydrochloride was purchased from Ad-
vanced Chemtech. Transport experiments were performed using a standard
glass U-tube, or modified W-tube at 293 K.[5a,25] The release of the amino
acid substrates into the receiving phase was monitored as a function of time
by HPLC product analysis (Varian, equipped with a UV detector; l ˆ
209 nm). In all cases, control experiments were performed in the absence of
the carrier. Errors are less than 15%. The presence of countertransported
expect based on the predicative U-tube studies (Table 1). In
particular, phenylalanine was found to be transported faster
than tryptophan by all carriers under investigation. Likewise,
the enantiomeric selectivity of all carriers was found to
correlate well with what was concluded from the U-tube
experiments. In certain cases, however, the actual k(L)/k(D)
transport rate ratio was found to be enhanced in the W-tube
experiments as compared to the U-tube ones.[42]
‡
Na ions in the Aq. I phase was confirmed by observation of the intense
‡
peaks corresponding to the [2.2.1] Na cryptate in the FAB mass spectra of
samples prepared from 4,7,13,16,21-pentaoxa-1,10-diazabicyclo[8.8.5]trico-
sane ([2.2.1] cryptand) mixed with solutions taken from Aq. I; these
measurements were made 24 h after the transport experiments were
commenced. Binding studies were effected by means of visible spectro-
scopic titrations (Beckman DU640) and were carried out at 293 K in
dichloromethane. All receptors were prepared by washing organic
solutions (dichloromethane containing 10% methanol) twice with 10%
NaOHaq and three times with water. They were then taken to dryness on
the rotorary evaporator and dried in vacuo.
Conclusions
The present results have served to confirm that sapphyrin ±
lasalocid conjugates which feature binding sites for both
carboxylate anion complexation and ammonium group rec-
ognition, can act as effective carriers for zwitterionic aromatic
a-amino acids as judged from simple Pressman-type U-tube
model membrane experiments. In the case of carrier 6,
selectivity for phenylalanine over tryptophan was observed. It
was also found that tyrosine is not transported to any
significant extent. Also, l-amino acids were found to be
transported faster than their d-congeners using this carrier. In
analogy to what is often observed in nature,[10c] the mechanism
is thought to involve sodium-cation antiport. Specifically, the
amino acid substrates from the initial aqueous phase are
transported through the organic phase and into the receiving
aqueous phase at the expense of sodium being carried back
from Aq. II to Aq. I. The transport process itself is thought to
be predicated upon formation of a tight supramolecular
complex between 6 and its amino acid substrates as illustrated
in Figure 2.
In the case of the second-generation conjugates 11 ± 14 with
chiral phenylalanine adjuvants (in addition to the sapphyrin
and lasalocid moieties present in 6), a clear difference was
observed between the free acid (13 and 14) and esterified (11
and 12) forms. The former did not effect amino acid transport,
presumably as the result of the carrier undergoing self-
assembly in the organic phase. By contrast, the esterified
systems 11 and 12 showed utility as amino acid transport
enhancing agents. In fact, depending on the chirality of the
phenylalanine appendage used, either l- or d-enantiomers of
amino acid substrates were transported faster. This leads us to
suggest that the approach illustrated here could provide a
generalized basis for the design of other, module-based
carriers capable of binding and transporting selectively a full
range of chiral substrates.
Sapphyrin 1b: Sapphyrin monoacid 1a (189 mg, 0.3 mmol) was dissolved
under argon in anhydrous DMF (10 mL) containing dry pyridine (Py)
(0.1 mL). A solution of 1,1'-carbonyldiimidazole (97 mg, 0.6 mmol) in
anhydrous DMF (2 mL) was then added, followed by HOBt (13.5 mg,
0.1 mmol). The resulting mixture was stirred at room temperature under
argon for 2 h. A solution of mono-tert-butyloxycarbonylethylenediamine
(96 mg, 0.6 mmol) in anhydrous DMF (3 mL) was then added, and the
resulting reaction mixture stirred at room temperature under argon for
24 h. The protected amine 1b that resulted was then concentrated by means
of a rotorary evaporator and purified by column chromatography on silica
gel as the solid support and with a gradient of 2 ± 7% methanol in
dichloromethane as the eluent. Solvents were then evaporated off, and the
residue redissolved in dichloromethane containing 20% methanol
(100 mL). This solution was washed consecutively twice with 1m NaOH
(30 mL), and three times with water. The organic phase was evaporated
and the product dried in vacuo. The yield of compound 1b was 185 mg (free
base, 80%). 1H NMR (300 MHz, CDCl3): d ˆ 3.39 (brs, 3H, pyrrole NH),
0.88 (s, 9H, CH3), 1.81 ± 1.97 (m, 12H, CH3), 2.63 (brs, 2H, CH2), 2.88 (brs,
2H, CH2), 3.03 (brs, 2H, CH2), 3.73 (s, 3H, CH3), 3.74 (s, 3H, CH3), 3.84 (s,
3H, CH3), 3.86 (s, 6H, CH3), 4.18 (m, 4H, CH2), 4.35 (m, 4H, CH2), 4.59 (t,
3J (H,H) ˆ 8 Hz, 2H, CH2), 5.80 (brs, 1H, NH), 10.49 (s, 1H, CH), 10.56 (s,
1H, CH), 10.61 (s, 1H, CH), 10.63 (s, 1H, CH); 13C NMR (75 MHz, CDCl3,
with 5% CD3OD): d ˆ 12.12, 12.50, 15.91, 17.59, 20.34, 20.45, 20.56, 23.24,
27.74, 39.14, 39.42, 39.51, 77.21, 78.87, 90.07, 90.34, 96.16, 126.92, 132.81,
132.97, 133.61, 133.94, 134.04, 134.17, 135.80, 135.92, 137.14, 138.38, 139.28,
139.44, 140.02, 140.35, 142.32, 156.23, 173.16; HRMS (FAB) calcd for
‡
C47H62N7O3 [M‡H] 772.4914, observed 772.4922.
t-Boc-protected aminolasalocid 4: The sodium salt of lasalocid 3 (123 mg,
0.2 mmol),
mono
tert-butyloxycarbonylethylenediamine
(64 mg,
0.4 mmol), and dry Py (0.1 mL) were dissolved in anhydrous DMF
(10 mL) under argon. A solution of EDC (76 mg, 0.4 mmol) in anhydrous
DMF (2 mL) was then added, followed by a catalytic amount of HOBt. The
resulting reaction mixture was stirred at room temperature under argon for
72 h. The solvents were then evaporated, the solids dried in vacuo, and the
product was purified by column chromatography on silica gel as the solid
support and with a gradient of 2 ± 5% methanol in dichloromethane as the
‡
eluent. The yield of compound 4 was 107 mg (Na salt, 71%). 1H NMR
(300 MHz, CD2Cl2): d ˆ 0.74 ± 0.86 (m, 12H, CH3), 0.93 (d, 3J (H,H) ˆ 7 Hz,
3H, CH3), 1.02 (d, 3J (H,H) ˆ 6.7 Hz, 3H, CH3), 1.12 (d, 3J (H,H) ˆ 7.7 Hz,
3H, CH3), 1.20 (m, 3H), 1.39 (s, 9H, CH3), 1.30 ± 1.90 (m, 12H, CH and
CH2), 2.16 (s, 3H, CH3), 2.20 (m, 1H, CH), 2.58 ± 2.74 (m, 2H, CH2), 2.89 ±
2.95 (m, 2H, CH2), 3.38 (m, 4H, CH and CH2), 3.66 (m, 2H, CH), 3.91 (m,
1H, CH), 4.12 (m, 1H, CH), 5.60 (m, 1H, NH), 6.65 (d, 3J (H,H) ˆ 8.3 Hz,
Experimental Section
1
General materials and methods: H and 13C NMR spectra were recorded
on General Electric QE-300 (300 MHz), GE GN-500 (500 MHz), and
Bruker AM-500 (500 MHz) instruments. Visible spectra were recorded on
Beckman DU640 instrument with cuvettes of 1 cm path length.
Sapphyrin mono- and bisacids 1a and 1c were prepared according to
procedures described previously.[9d] Porphyrin monoacid 2 was synthesized
as previously described.[24] Mono-tert-butyloxycarbonylethylenediamine
was prepared according to a literature procedure.[23] The sodium salt of
lasalocid, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride
3
1H, CH), 6.86 (s, 1H, NH), 7.07 (d, J (H,H) ˆ 8.6 Hz, 1H, CH), 10.34 (s,
a
1H, OH); 13C NMR (75 MHz, CD2Cl2): d ˆ 6.47, 8.98, 12.85, 13.72, 14.39,
15.56, 15.91, 17.07, 21.06, 28.53, 30.14, 30.81, 31.02, 31.60, 34.07, 35.04, 36.82,
38.78, 40.50, 40.83, 48.43, 54.67, 70.59, 71.26, 72.88, 77.74, 79.25, 85.15, 87.15,
118.19, 121.33, 123.77, 133.00, 138.84, 156.88, 157.08, 171.25, 213.62; HRMS
‡
(FAB) calcd for C41H69N2O9 [M‡H] 733.5003, observed 733.4987.
(EDC),
1,1'-carbonyldiimidazole,
1-hydroxybenzotriazole
hydrate
Aminolasalocid 5: t-Boc-protected 4 (90 mg, 0.12 mmol) was deprotected
(HOBt), pyridine, trifluoroacetic acid, and dimethylformamide were
by treatment with 3:1 dichloromethane/TFA mixture (5 mL) for ca. 2 h
164
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0164 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 1

Sapphyrin ± Lasalocid
159±167
3
(TLC control). Solvents were then evaporated off, and the residue purified
chromatographically on silica gel as the solid support and 10% methanol
(saturated with NH3 gas) in dichloromethane as the eluent. Fractions with
Rf ˆ 0.2 ± 0.5 were collected. The yield of compound 5 was 63 mg (84%). 1H
NMR (300 MHz, CD2Cl2): d ˆ 0.74 ± 0.89 (m, 15H, CH3), 1.01 (d, 3J
(H,H) ˆ 7 Hz, 3H, CH3), 1.13 (d, 3J (H,H) ˆ 7.7 Hz, 3H, CH3), 1.20 ± 1.90
(m, 15H, CH and CH2), 2.14 (s, 3H, CH3), 2.20 (m, 1H, CH), 2.45 ± 2.60 (m,
2H, CH2), 2.75 ± 2.94 (m, 4H, CH and CH2), 3.33 ± 3.42 (m, 2H, CH), 3.65 ±
3.73 (m, 2H, CH), 3.87 (m, 1H, CH), 3.98 (m, 1H, CH), 6.59 (d, 3J (H,H) ˆ
8.7 Hz, 1H, CH), 7.01 (d, 3J (H,H) ˆ 8.3 Hz, 1H, CH); 13C NMR (75 MHz,
CD2Cl2): d ˆ 6.54, 9.00, 12.84, 13.59, 14.40, 15.73, 16.15, 16.85, 20.61, 30.09,
30.50, 31.26, 31.31, 34.12, 35.36, 36.65, 39.05, 41.28, 48.52, 55.10, 70.82, 71.76,
72.15, 77.65, 84.80, 86.51, 120.41, 121.09, 123.73, 132.25, 138.72, 155.89,
8.3 Hz, 1H, CH), 7.17 (d, J (H,H) ˆ 8.3 Hz, 1H, CH), 11.45 (s, 1H, OH);
13C NMR (75 MHz, CD2Cl2): d ˆ 6.60, 8.78, 12.36, 12.86, 13.85, 14.37, 15.79,
15.95, 17.53, 21.20, 29.93, 30.74, 31.16, 34.71, 35.04, 35.30, 37.14, 39.19, 48.95,
52.51, 54.96, 70.46, 71.69, 74.41, 77.78, 85.08, 86.51, 111.57, 122.11, 124.35,
135.37, 144.06, 161.13, 172.85, 214.35; HRMS (FAB) calcd for C35H57O8
‡
[M‡H] 605.4053, observed 605.4063.
Sapphyrin mono acids 9 and 10: The sapphyrin bis acid 1c (205 mg,
0.31 mmol) and phenylalanine tert-butyl ester hydrochloride (80 mg,
0.31 mmol) (l- or d- for the synthesis of 9 and 10, respectively) were
dissolved under argon in anhydrous DMF (10 mL) containing dry Py
(0.1 mL). A solution of EDC (89 mg, 0.47 mmol) in anhydrous DMF
(2 mL) was then added, followed by HOBt (13.5 mg, 0.1 mmol). The
resulting reaction mixture was stirred at room temperature under argon for
72 h. The progress of the reaction was followed by TLC. Three major
sapphyrin-containing spots were observed: that of a bisamide (highest Rf),
followed by the monoamide monoacid (9 or 10, depending on antipode
used), and the starting material, 1c (lowest Rf). When the reaction was
deemed complete, solvents were evaporated off, the solids dried in vacuo,
and the product (middle fraction) was carefully isolated by column
chromatography on silica gel as the solid support and with a gradient of 2 ±
15% methanol in dichloromethane as the eluent. The yield of compound 9
‡
171.26, 214.80; HRMS (FAB) calcd for C36H61N2O7 [M‡H] 633.4479,
observed 633.4466.
Sapphyrin ± lasalocid conjugate 6: The sapphyrin mono acid 1a (189 mg,
0.3 mmol) and aminolasalocid 5 (379 mg, 0.6 mmol) were dissolved under
argon in anhydrous DMF (10 mL) containing dry Py (0.1 mL). A solution
of EDC (115 mg, 0.6 mmol) in anhydrous DMF (4 mL) was then added,
followed by HOBt (13.5 mg, 0.1 mmol). The resulting reaction mixture was
stirred at room temperature under argon for 48 h. The solvents were
evaporated off, and the resulting solids were then dried in vacuo. The
product was then purified chromatographically on silica gel as the solid
support and with a gradient of 2 ± 15% methanol in dichloromethane as the
1
or 10 was found to be of the order of 180 mg (67%). H NMR (300 MHz,
CDCl3 with 5% CD3OD): d ˆ 1.29 (brs, 9H, CH3), 2.24 (m, 12H, CH3),
2.4 ± 3.9 (m, 4H), 4.13 (s, 3H, CH3), 4.25 (s, 3H, CH3), 4.38 (s, 3H, CH3),
3.95 ± 5.12 (m, 14H), 5.63 (m, 1H, NH), 6.60 ± 7.10 (m, 5H, phenyl CH),
10.88 (s, 1H, CH), 11.16 (s, 1H, CH), 11.75 (s, 2H, CH); 13C NMR (75 MHz,
CDCl3, with 5% CD3OD): d ˆ 11.92, 14.76, 14.97, 16.02, 16.59, 17.66, 18.20,
19.85, 20.16, 20.26, 26.79, 33.39, 36.87, 54.27, 81.42, 86.83, 93.24, 97.14, 99.21,
126.04, 127.20, 127.59, 128.46, 128.79, 129.78, 132.18, 133.25, 134.79, 136.13,
139.64, 141.95, 142.70, 142.80, 143.34, 170.28; HRMS (FAB) calcd for
C53H63N6O5 [M‡H] 863.4860, observed 863.4849.
Sapphyrin ± lasalocid conjugates 11 and 12: The sapphyrin monoamide
monoacid 9 or 10, for the synthesis of compounds 11 and 12 respectively,
(60 mg, 0.07 mmol) and aminolasalocid
dissolved under argon in anhydrous DMF (10 mL) containing dry Py
(0.1 mL). A solution of EDC (27 mg, 0.14 mmol) in anhydrous DMF
(2 mL) was then added, followed by HOBt (6.8 mg, 0.05 mmol). The
resulting reaction mixture was stirred at room temperature under argon for
48 h. The solvents were then evaporated off. After the resulting solids were
dried in vacuo, the products 11 and 12, as appropriate, were purified
chromatographically on silica gel as the solid support and a solution of
methanol (12% v/v) in dichloromethane as the eluent. The yield of
compound 11 was 94 mg (92%), and the yield of compound 12 was 88 mg
(86%). 1H NMR (11, 300 MHz, CDCl3): d ˆ 0.35 (t, 3J (H,H) ˆ 7.3 Hz, 1H),
0.55 ± 0.90 (m, 15H), 1.05 (s, 9H, CH3), 1.1 ± 2.2 (m, 37H), 2.3 ± 3.2 (m,
10H), 3.84 (s, 3H, CH3), 3.86 (s, 3H, CH3), 3.91 (s, 3H, CH3), 4.16 (s, 3H,
CH3), 3.3 ± 4.7 (m, 19H), 5.37 (q, 3J (H,H) ˆ 18.7 Hz, 2H), 5.62 (s, 1H), 6.02
(t, 3J (H,H) ˆ 8 Hz, 2H, CH), 6.17 (d, 3J (H,H) ˆ 8.3 Hz, 1H, CH), 6.25 (m,
3H, CH), 6.78 (d, 3J (H,H) ˆ 8.3 Hz, 1H, CH), 10.43 (s, 1H, CH), 10.46 (s,
1H, CH), 10.84 (s, 1H, CH), 10.89 (s, 1H, CH); 13C NMR (75 MHz, CDCl3
with 5% CD3OD): d ˆ 5.83, 8.39, 12.01, 12.25, 12.43, 12.70, 12.75, 13.21,
15.43, 16.15, 16.43, 16.89, 17.67, 17.74, 18.31, 18.41, 19.65, 20.13, 20.48, 27.36,
28.91, 28.96, 30.46, 32.19, 33.65, 34.78, 34.89, 36.27, 37.27, 38.08, 38.79, 40.10,
49.29, 53.58, 54.91, 69.63, 70.37, 72.83, 76.19, 77.22, 81.98, 83.66, 86.03, 90.00,
90.68, 96.35, 96.47, 117.28, 119.03, 125.30, 125.91, 127.37, 128.40, 130.70,
130.96, 131.20, 132.48, 132.85, 133.09, 135.16, 135.30, 135.67, 135.85, 137.81,
138.75, 140.44, 142.41, 142.64, 143.06, 143.15, 163.34, 170.00, 170.55, 171.88,
1
eluent. The yield of compound 6 was 272 mg (73%). H NMR (300 MHz,
CD2Cl2): d ˆ 0.49 ± 1.70 (m, 33H), 1.82 (brs, 2H), 1.91 ± 2.15 (m, 13H), 2.13
(s, 3H, CH3), 2.14 (m, 1H, CH), 2.47 (m, 2H), 2.69 (m, 2H), 3.16 (m, 4H),
3.13 (m, 2H), 3.45 (m, 2H), 3.64 (m, 4H), 3.84 (s, 3H, CH3), 3.87 (s, 3H,
CH3), 4.01 (m, 3H, CH3), 4.06 (m, 3H, CH3), 4.08 (m, 3H, CH3), 4.35 (m,
4H, CH2), 4.57 (m, 4H, CH2), 4.87 (m, 2H, CH2), 6.37 (d, 3J (H,H) ˆ 8.7 Hz,
1H, CH), 6.72 (d, 3J (H,H) ˆ 8.3 Hz, 1H, CH), 10.51 (s, 1H, CH), 10.60 (s,
1H, CH), 10.64 (s, 1H, CH), 10.72 (s, 1H, CH); 13C NMR (126 MHz,
CD2Cl2): d ˆ 6.33, 6.44, 8.44, 8.51, 11.96, 12.43, 12.60, 13.05, 13.14, 13.59,
13.83, 13.99, 14.10, 15.59, 15.67, 15.81, 16.42, 16.45, 17.45, 17.78, 18.08, 18.15,
20.61, 20.86, 20.97, 21.21, 23.57, 29.37, 29.51, 29.74, 29.96, 30.13, 30.71, 30.95,
31.88, 33.76, 34.29, 34.39, 35.29, 35.53, 35.86, 36.75, 38.82, 39.29, 39.98, 40.44,
48.42, 48.68, 55.06, 70.27, 70.43, 71.15, 71.71, 73.76, 74.23, 77.13, 77.35, 84.49,
84.66, 85.99, 86.07, 90.23, 90.34, 96.64, 122.10, 124.45, 127.51, 128.71, 130.84,
134.04, 135.49, 135.66, 136.27, 137.06, 140.00, 140.21, 141.09, 141.89, 144.19,
160.95, 169.66, 172.81, 214.28, 215.05; HRMS (FAB) calcd for C76H106N7O8
‡
5 (88 mg, 0.14 mmol) were
‡
[M‡H] 1244.8103, observed 1244.8122.
Porphyrin ± lasalocid conjugate 7: This material was made by the procedure
used for the synthesis of conjugate 6, with the exception that the starting
material was the porphyrin monoacid 2. The yield of compound 7 obtained
this way was 86%. 1H NMR (300 MHz, CDCl3): d ˆ 3.91 (brs, 2H,
pyrrole NH), 0.20 ± 2.24 (m, 49H), 2.13 (s, 3H, CH3), 2.40 ± 4.10 (m, 31H),
4.37 (brs, 2H, CH2), 5.30 (s, 1H, NH), 6.43 (d, 3J (H,H) ˆ 7.3 Hz, 1H, CH),
6.53 (s, 1H, NH), 6.95 (d, 3J (H,H) ˆ 7.3 Hz, 1H, CH), 10.05 (m, 4H, CH);
13C NMR (126 MHz, CDCl3): d ˆ 6.91, 6.14, 6.31, 8.16, 8.18, 8.19, 9.01, 11.39,
11.43, 11.56, 11.99, 12.29, 12.41, 12.53, 12.96, 12.99, 13.14, 13.26, 13.35, 13.81,
15.08, 15.33, 15.53, 15.95, 16.07, 17.57, 17.59, 17.64, 18.51, 18.53, 18.66, 28.37,
28.99, 29.04, 29.33, 30.13, 30.43, 32.75, 33.47, 33.75, 35.39, 37.04, 37.40, 39.50,
48.30, 54.31, 55.21, 67.65, 69.99, 70.20, 70.58, 76.31, 82.36, 86.01, 86.76, 96.08,
96.22, 96.37, 119.32, 121.96, 122.39, 122.83, 124.27, 131.22, 131.57, 136.99,
143.02, 154.92, 160.69, 170.48, 173.45, 175.92, 218.39; HRMS (FAB) calcd
‡
for C70H99N6O8 [M‡H] 1151.7524, observed 1151.7518.
Lasalocid methyl ester 8: The sodium salt of lasalocid 3 (95 mg, 0.16 mmol)
and dry Py (0.1 mL) were dissolved in anhydrous methanol (10 mL). EDC
(59 mg, 0.31 mmol) and a catalytic amount of HOBt were then added. The
resulting reaction mixture was stirred at room temperature under argon for
48 h. The solvents were then evaporated off, and the resulting solids dried
in vacuo. The resulting crude product was then purified by column
chromatography using silica gel as the solid support and 3% methanol in
dichloromethane as the eluent. Compound 8 eluted off the column first,
and was followed by trace quantities of various unidentified by-products.
‡
172.20, 216.56; HRMS (FAB) calcd for C89H121N8O11 [M‡H] 1477.9155,
observed 1477.9118. Compound 12 showed similar NMR characteristics.
Sapphyrin ± lasalocid conjugates 13 and 14: tert-Butyl-protected 11 or 12
for the synthesis of compound 14 (59 mg, 0.04 mmol) was deprotected by
treatment with 3:1 dichloromethane/TFA mixture (2 mL) for ca. 4 h at
room temperature (TLC control). Solvents were then evaporated off, and
the resulting solids were purified on a short chromatographic column, on
silica gel as the solid support and dichloromethane ± methanol (6:1 v/v) as
the eluent. Small amounts of macrocyclic by-products (presumably, a
mixture of esters obtained from the reaction of terminal carboxylic acid
portion of the phenylalanine appendage with one or more of the free
hydroxy groups present in the lasalocid appendage) were separated out
during column chromatography. The yield of compound 13 obtained in this
way was 40 mg (71%), while that of compound 14 was 43 mg (75%). 1H
‡
The yield of compound 8 was 72 mg (Na salt, 74%). 1H NMR (300 MHz,
CD2Cl2): d ˆ 0.78 ± 0.86 (m, 9H, CH3), 0.92 ± 0.94 (m, 6H, CH3), 1.01 (d, 3J
(H,H) ˆ 7 Hz, 3H, CH3), 1.15 (d, 3J (H,H) ˆ 7.6 Hz, 3H, CH3), 1.20 ± 1.90
(m, 15H, CH and CH2), 2.19 (s, 3H, CH3), 2.20 (m, 1H, CH), 2.72 (m, 1H,
CH), 2.88 ± 3.00 (m, 3H, CH), 3.25 (brs, 1H, CH), 3.42 (m, 1H, CH), 3.71
3
(m, 1H, CH), 3.87 (m, 1H, CH), 3.96 (s, 3H, OCH3), 6.68 (d, J (H,H)ˆ
Chem. Eur. J. 1998, 4, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0165 $ 17.50+.25/0
165

J. L. Sessler and A. Andrievsky
FULL PAPER
NMR (13, 500 MHz, CDCl3): d ˆ 0.60 ± 2.10 (m, 52H), 2.3 ± 3.7 (m, 10H),
3.99 (s, 3H, CH3), 4.03 (s, 3H, CH3), 4.05 (s, 3H, CH3), 4.11 (s, 3H, CH3),
4.3 ± 4.7 (m, 16H), 4.8 (s, 1H), 5.46 (s, 2H), 5.56 (s, 1H), 6.40 ± 7.00 (m, 7H),
11.35 ± 11.65 (m, 4H, CH); 13C NMR (126 MHz, CDCl3 with 5% CD3OD):
d ˆ 5.91, 6.03, 7.95, 8.43, 12.26, 12.29, 12.82, 13.12, 13.69, 13.93, 15.68, 16.13,
16.60, 17.31, 17.37, 17.89, 17.99, 18.24, 18.32, 19.92, 20.05, 20.19, 20.36, 20.55,
20.73, 20.77, 23.04, 23.18, 26.30, 28.46, 29.24, 29.46, 30.31, 30.71, 33.40, 35.04,
35.67, 36.26, 37.10, 38.68, 38.92, 45.17, 50.53, 54.25, 55.00, 70.46, 70.78, 72.38,
73.02, 83.59, 84.34, 85.70, 86.13, 91.11, 91.78, 92.13, 95.86, 120.26, 123.36,
125.58, 127.44, 127.79, 128.35, 128.68, 129.09, 129.28, 129.71, 131.61, 132.39,
132.53, 133.84, 134.37, 134.79, 137.73, 138.77, 140.74, 143.67, 150.88, 152.60,
169.33, 170.48, 170.60, 170.98, 215.00, 215.13; HRMS (FAB) calcd for
[6] For studies involving the recognition of amino acids in their anionic
forms and of N-protected amino acids, see: a) V. Alcazar, F. Die-
derich, Angew. Chem. 1992, 104, 1503 ± 1505; Angew. Chem. Int. Ed.
Engl. 1992, 31, 1521 ± 1523; b) R. J. Pieters, F. Diederich, Chem.
Commun.. 1996, 2255 ± 2256; c) K. Maruyama, H. Tsukube, T. Araki,
J. Am. Chem. Soc. 1982, 104, 5197 ± 5203; d) H. Tsukube, Tetrahedron
Lett. 1981, 22, 3981 ± 3984; e) K. Maruyama, H. Tsukube, T. Araki,
ibid. 1981, 22, 2001 ± 2004; f) G. J. Pernia, J. D. Kilburn, M. Rowley, J.
Chem. Soc. Chem. Commun. 1995, 305 ± 306; g) H. Kataoka, T.
Katagi, Tetrahedron 1987, 43, 4519 ± 4530; h) K. Konishi, K. Yahara,
H. Toshishige, T. Aida, S. Inoue, J. Am. Chem. Soc. 1994, 116, 1337 ±
1344; i) M. Zinic, L. Frkanec, V. Skaric, J. Trafton, G. W. Gokel, J.
Chem. Soc. Chem. Commun. 1990, 1726 ± 1728.
‡
C85H113N8O11 [M‡H] 1421.8529, observed 1421.8500. Compound 14
showed similar NMR characteristics.
[7] Issues relevant to the analytical chemistry of amino acids are discussed
in Amino Acid Analysis (Ed.: J. M. Rattenbury), Ellis Horwood,
Chichester, 1981.
Acknowledgments: This work was supported by NIH grant A.I. 33577 to
J.L.S., and, in part, by Pharmacyclics Inc.
[8] a) For a short review, see: H.-J. Schneider, Angew. Chem. 1993, 105,
890 ± 892; Angew. Chem. Int. Ed. Engl. 1993, 32, 848 ± 850. A number
of excellent peptide receptors have been prepared in the Still
laboratory: b) G. Li, W. C. Still, J. Org. Chem. 1991, 56, 6964 ± 6966;
c) J.-I. Hong, S. K. Namgoong, A. Bernardi, W. C. Still, J. Am. Chem.
Soc. 1991, 113, 5111 ± 5112; d) S. S. Yoon, W. C. Still, ibid. 1993, 115,
823 ± 824. For other peptide recognition work, see: e) J. S. Albert,
M. S. Goodman, A. D. Hamilton, ibid. 1995, 117, 1143 ± 1144; f) K.
Konishi, S. Kimata, K. Yoshida, M. Tanaka, T. Aida, Angew. Chem.
1996, 108, 3001 ± 3003; Angew. Chem. Int. Ed. Engl. 1996, 35, 2823 ±
2825; g) C. P. Waymark, J. D. Kilburn, I. Gillies, Teterahedron Lett.
1995, 36, 3051 ± 3054; h) J. Dowden, P. D. Edwards, J. D. Kilburn,
Tetrahedron Lett. 1997, 38, 1095 ± 1098.
Received: June 3, 1997 [F713]
[1] a) D. Voet, J. G. Voet, Biochemistry, 2nd ed., Wiley, New York, 1995;
b) C. Branden, J. Tooze, Introduction to Protein Structure, Garland,
New York, 1991; c) A. Fersht, Enzyme Structure and Mechanism, 2nd
ed., Freeman, New York, 1985.
[2] Excitatory Amino Acid Receptors (Eds.: P. Krogsgaard-Larsen, J. J.
Hansen), Ellis Horwood, Chichester, 1992.
[3] a) For an excellent collection of reviews on molecular recognition,
see: Comprehensive Supramolecular Chemistry (Eds.: J. L. Atwood,
J. E. D. Davies, D. D. Macnicol, F. Vögtle), Elsevier, Exeter, 1996.
Also see: b) J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim,
1995; c) F. Vögtle, Supramolecular Chemistry, Wiley, Chichester, 1991.
[9] a) H. Furuta, K. Furuta, J. L. Sessler, J. Am. Chem. Soc. 1991, 113,
4706 ± 4707; b) J. L. Sessler, D. A. Ford, M. J. Cyr, H. Furuta, J. Chem.
Soc. Chem. Commun. 1991, 1733 ± 1735; c) J. L. Sessler, T. D. Mody,
D. A. Ford, V. Lynch, Angew. Chem. 1992, 104, 461 ± 464; Angew.
Â
[4] For a recent review see: a) C. Seel, A. Galan, J. de Mendoza, Top.
Curr. Chem. 1995, 175, 101 ± 132, and references therein. For
zwitterionic a-amino acid recognition, see: b) J. Rebek, Jr., B. Askew,
D. Nemeth, K. Parris, J. Am. Chem. Soc. 1987, 109, 2432 ± 2434; c) A.
Â
Chem. Int. Ed. Engl. 1992, 31, 452 ± 455; d) V. Kral, J. L. Sessler, H.
Furuta, J. Am. Chem. Soc. 1992, 114, 8704 ± 8705; e) J. L. Sessler, H.
 Â
Furuta, V. Kral, Supramol. Chem. 1993, 1, 209 ± 220; f) V. Kral, J. L.
Â
Galan, D. Andreu, A. M. Echavarren, P. Prados, J. de Mendoza, ibid.
Â
Sessler, Tetrahedron 1995, 51, 539 ± 554; g) V. Kral, A. Andrievsky, J. L.
1992, 114, 1511 ± 1512; d) A. Metzger, K. Gloe, H. Stephan, F. P.
Schmidtchen, J. Org. Chem. 1996, 61, 2051 ± 2055; e) I. Tabushi, Y.
Kuroda, T. Mizutani, J. Am. Chem. Soc. 1986, 108, 4514 ± 4518; f) Y.
Aoyama, M. Asakawa, A. Yamagishi, H. Toi, H. Ogoshi, ibid. 1990, 112,
3145 ± 3151; g) J. Sunamoto, K. Iwamoto, Y. Mohri, T. Kominato, ibid.
1982, 104, 5502 ± 5504; h) S. Marx-Tibbon, I. Willner, J. Chem. Soc.
Chem. Commun. 1994, 1261 ± 1262; i) R. P. Bonomo, V. Cucinotta, F.
DꢁAllessandro, G. Impellizzeri, G. Maccarrone, E. Rizzarelli, G. Vecchio,
J. Inclusion Phenom. Mol. Recognit. Chem. 1993, 15, 167 ± 180; j) R.
Corradini, A. Dossena, G. Impellizzeri, G. Maccarrone, R. Marchelli,
E. Rizzarelli, G. Sartor, G. Vecchio, J. Am. Chem. Soc. 1994, 116,
10267 ± 10274; k) P. Scrimin, P. Tecilla, U. Tonellato, Tetrahedron
1995, 51, 217 ± 230; l) L. K. Mohler, A. W. Czarnik, J. Am. Chem. Soc.
1993, 115, 7037 ± 7038; m) M. T. Reetz, J. Huff, J. Rudolph, K. Töllner,
A. Deege, R. Goddard, ibid. 1994, 116, 11588 ± 11589; n) H. C. Chen,
S. Ogo, R. H. Fish, ibid. 1996, 118, 4993 ± 5001; o) H. Tsukube, J.
Uenishi, T. Kanatani, H. Itoh, O. Yonemitsu, Chem. Commun.. 1996,
477 ± 478; p) N. Higashi, M. Saitou, T. Mihara, M. Niwa, ibid. 1995,
2119 ± 2120; q) K. B. Lipkowitz, S. Raghothama, J. Yang, J. Am.
Chem. Soc. 1992, 114, 1554 ± 1562.
Â
Sessler, J. Chem. Soc. Chem. Commun. 1995, 2349 ± 2351; h) V. Kral,
A. Andrievsky, J. L. Sessler, J. Am. Chem. Soc. 1995, 117, 2953 ± 2954.
[10] For reviews of biological amino acid transport systems, see: a) Mam-
malian Amino Acid Transport. Mechanisms and Control (Ed.: M. S.
Kilberg, D. Häussinger), Plenum, New York, 1992; b) M. S. Kilberg,
B. R. Stevens, D. A. Novak, Annu. Rev. Nutr. 1993, 13, 137 ± 165; c) K.
Ring, Angew. Chem. 1970, 82, 343 ± 355; Angew. Chem. Int. Ed. Engl.
1970, 9, 345 ± 356.
[11] For a review about stereochemical aspects of molecular recognition,
see: T. H. Webb, C. S. Wilcox, Chem. Soc. Rev. 1993, 383 ± 395.
[12] a) C. J. Pederson, J. Am. Chem. Soc. 1967, 89, 7017 ± 7036. b) For a
discussion see also refs. [4a,b], and [5c].
[13] a) J.-P. Behr, J.-M. Lehn, J. Am. Chem. Soc. 1973, 95, 6108 ± 6110.
b) See also refs. [5a,e,g,6c ± e,g,i]. However, several groups have
succeeded in effecting the trans-membrane transport of zwitterionic
amino acids under neutral conditions; see refs. [4a ± d,f ± h,k ± m].
[14] For reviews about sapphyrins and other expanded porphyrins, see:
a) J. L. Sessler, A. K. Burrell, Top. Curr. Chem. 1991, 161, 176 ± 273;
b) J. L. Sessler, A. K. Burrell, H. Furuta, G. W. Hemmi, B. L. Iverson,
Â
V. Kral, D. J. Magda, T. D. Mody, K. Shreder, D. Smith, S. J. Weghorn,
[5] For efforts directed at the recognition of amino acids in their cationic
form and of cationic amino acid esters, see: a) M. Newcomb, J. L.
Toner, R. C. Helgeson, D. J. Cram, J. Am. Chem. Soc. 1979, 101, 4941 ±
4947; b) G. D. Y. Sogah, D. J. Cram, J. Am. Chem. Soc. 1979, 101,
3035 ± 3042; c) J.-P. Behr, J.-M. Lehn, P. Vierling, Helv. Chim. Acta
1982, 65, 1853 ± 1867; d) M. Sawada, Y. Takai, H. Yamada, T. Kaneda,
K. Kamada, T. Mizooku, K. Hirose, Y. Tobe, K. Naemura, J. Chem.
Soc. Chem. Commun. 1994, 2497 ± 2498; e) S.-K. Chang, H.-S. Hwang,
H. Son, J. Youk, Y. S. Kang, ibid. 1991, 217 ± 218; f) K. Maruyama, H.
Sohmiya, H. Tsukube, ibid. 1989, 864 ± 865; g) H. Miyake, T.
Yamashita, Y. Kojima, H. Tsukube, Tetrahedron Lett. 1995, 36,
7669 ± 7672. For work on the recognition of neutral amino acid esters,
see: h) Y. Kuroda, Y. Kato, T. Higashioji, J. Hasegawa, S. Kawanami,
M. Takahashi, N. Shiraishi, K. Tanabe, H. Ogoshi, J. Am. Chem. Soc.
1995, 117, 10950±10958, and references therein; i) M. J. Crossley, L. G.
Mackay, A. C. Try, J. Chem. Soc. Chem. Commun. 1995, 1925 ± 1927.
in Transition Metals in Supramolecular Chemistry, NATO ASI Series
(Eds.: L. Fabbrizzi, A. Poggi), Kluwer, Amsterdam, 1994, 391 ± 408;
c) J. L. Sessler, S. J. Weghorn, Expanded, Contracted, and Isomeric
Porphyrins, Elsevier, London, 1997, in press; d) J. L. Sessler, M. Cyr,
H. Furuta, V. Kral, T. Mody, T. Morishima, M. Shionoya, S. J. Weghorn,
Pure Appl. Chem. 1993, 65, 393 ± 398; e) B. L. Iverson, K. Shreder, V.
Â
Â
Kral, D. A. Smith, J. Smith, J. L. Sessler, ibid. 1994, 66, 845 ± 850.
Â
[15] a) V. Kral, H. Furuta, K. Shreder, V. Lynch, J. L. Sessler, J. Am. Chem. Soc.
Â
1996, 118, 1595 ± 1607; b) B. L. Iverson, K. Shreder, V. Kral, P. I. San-
som, V. Lynch, J. L. Sessler, ibid. 1996, 118, 1608 ± 1616; c) J. L. Sess-
ler, M. J. Cyr, V. Lynch, E. McGhee, J. A. Ibers, ibid. 1990, 112, 2810 ±
2813; d) M. Shionoya, H. Furuta, V. Lynch, A. Harriman, J. L. Sessler,
ibid. 1992, 114, 5714 ± 5722; e) J. L. Sessler, A. Andrievsky, Chem.
Commun.. 1996, 1119 ± 1120; f) J. L. Sessler, P. I. Sansom, V. Kral, D.
OꢁConnor, B. L. Iverson, J. Am. Chem. Soc. 1996, 118, 12323 ± 12330.
Â
166
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0166 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 1

Sapphyrin ± Lasalocid
159±167
Â
[16] a) V. Kral, S. L. Springs, J. L. Sessler, J. Am. Chem. Soc. 1995, 117,
methanol ˆ 99:1) being added to a solution of receptor 6 at fixed
(10 6 m) concentration. Association constants and the stoichiometric
compositions of the complexes were determined by Benesi ± Hilde-
brand plots and by standard nonlinear curve-fitting protocols: K. A.
Connors, Binding Constants, Wiley, New York, 1987. For a complete
discussion, see supplementary material to ref. [9h]. Also, in the case of
l-Phe, 1:1 stoichiometry was confirmed by a Job plot.
8881 ± 8882; b) J. L. Sessler, A. Andrievsky, P. A. Gale, V. Lynch,
Angew. Chem. 1996, 108, 2954 ± 2957; Angew. Chem. Int. Ed. Engl.
1996, 35, 2782 ± 2785. c) See also ref. [9h].
[17] Two different carboxylate anion complexes of protonated pentaaza-
sapphyrin macrocycles have been structurally characterized in the
solid state: firstly, the sapphyrin bisacid 3,12,13,22-tetraethyl-8,17-
bis(carboxyethyl)-2,7,18,23-tetramethylsapphyrin has been found to
form a supramolecular dimer in the solid state (see ref. [16b]). In this
case, a carboxylate hook from one molecule is chelated to the
protonated pyrrolic core of the second macrocycle. Simultaneously,
this second macrocycle shares its carboxylate tail with the first
sapphyrin subunit. A trifluoroacetate ion is coordinated to each of
unchelated faces in the dimer and prevents these sites from being
involved in further binding. The second example consists of a 2:1
complex formed between trifluoroacetic acid and the diprotonated
form of 3,8,12,13,17,22-hexaethyl-2,7,18,23-tetramethylsapphyrin. In
this case, the TFA carboxylate anion is chelated above the plane of the
protonated sapphyrin macrocycle by a combination of electrostatic
attractions and hydrogen bonding involving the ligated oxygen atom
and the pyrrolic hydrogens: J. L. Sessler, A. Andrievsky, V. Lynch,
unpublished results. The available solution data is consistent with this
kind of binding occuring in organic media (ref. [16]).
[30] The salicylic hydroxy group of lasalocid is acidic enough to be
deprotonated by a strong base, which, in this instance, could be either
sapphyrin or NaOH.
[31] Charge neutralization is required for efficient carrier-mediated trans-
port (a corollary of Fickꢁs first law, see ref. [25]). While we favor a
model wherein the net neutralization process takes place as shown in
Figure 2, a referee has suggested that it could occur in part as the
result of proton transfer between oppositely charged species taking
place within the lipophilic medium of the model membrane (e.g., from
the primary ammonium group of an amino acid zwitterion to the
salicylamide moiety of the lasalocid subunit). While at present it is
difficult to distinguish between these interpretive extremes of what
may well be a chemical continuum, it is important to point out that
functionally (i.e., in terms of experimental prediction) they are the
same.
[32] Fluoride anions have a much higher affinity for protonated sapphyrin
than chloride (ref. [15d]).
[33] Importantly, the pH increase in question is not sufficient to change the
predominant form of amino acid species in Aq. I from zwitterionic to
anionic.
[34] Finally, in support of proposed mechanism are the results of
two negative control experiments: When either a) the pH of
the initial phase, Aq. I, was made basic by means of NaOH
addition (pH ˆ 13 and the other conditions as in entry 1 of Table 1)
or b) the pH was lowered by adding HCl (pH ˆ 1, with other
conditions as in entry 1 of Table 1), no amino acid transport by carrier
6 was observed.
[18] a) M. Dobler, Ionophores and Their Structures, Wiley, New York,
1981; b) R. Hilgenfeld, W. Saenger, Top. Curr. Chem. 1982, 101, 1 ± 82;
c) G. R. Painter, B. C. Pressman, ibid. 1982, 101, 83 ± 110.
[19] For examples of metal cation complexation by lasalocid, see
a) ref. [18]; b) R. Lyazghi, Y. Pointud, G. Dauphin, J. Juillard, J.
Chem. Soc. Perkin Trans. 2 1993, 1681 ± 1686, and references therein;
c) H. Tsukube, K. Takagi, T. Higashiyama, T. Iwachido, N. Hayama,
Inorg. Chem. 1994, 33, 2984 ± 2987, and references therein. For
examples of adducts of lasalocid with metal complexes, see:
d) P. S. K. Chia, L. F. Lindoy, G. W. Walker, G. W. Everett, Pure
Appl. Chem. 1993, 65, 521 ± 526, and references therein; e) R.
Ballardini, M. T. Gandolfi, M. L. Moya, L. Prodi, V. Balzani, Isr. J.
Chem. 1992, 32, 47 ± 51.
Â
[35] a) B. L. Iverson, R. E. Thomas, V. Kral, J. L. Sessler, J. Am. Chem.
Â
Soc. 1994, 116, 2663 ± 2664; b) J. L. Sessler, J. W. Genge, V. Kral, B. L.
[20] For examples of ammonium cation binding by lasalocid and its
derivatives, see: a) J. W. Westley, R. H. Evans, Jr., J. F. Blount, J. Am.
Chem. Soc. 1977, 99, 6057 ± 6061; b) H. Tsukube, H. Sohmiya, J. Org.
Chem. 1991, 56, 875 ± 878; c) H. Tsukube, H. Sohmiya, Supramol.
Chem. 1993, 1, 297 ± 304, and references therein; d) R. C. R. Gueco,
G. W. Everett, Tetrahedron 1985, 41, 4437 ± 4442; e) J. F. Kinsel, E. I.
Melnik, S. Lindenbaum, L. A. Sternson, Yu. A. Ovchinnikov, Bio-
chim. Biophys. Acta 1982, 684, 233 ± 240; f) J. F. Kinsel, E. I. Melnik,
L. A. Sternson, S. Lindenbaum, Yu. A. Ovchinnikov, ibid. 1982, 692,
377 ± 383; g) C. Shen, D. J. Patel, Proc. Natl. Acad. Sci. USA 1977, 74,
4734 ± 4738.
[21] Importantly, it was shown by Tsukube that chemically modified
natural polyether ionophores (i.e., as their ester or amide derivatives)
retain the ability to chelate ammonium cations. In most cases
improved enantioselectivity is actually observed (see ref. [20b]).
[22] The central core of sapphyrin macrocycle is monoprotonated at
neutral pH (ref. [14]).
Iverson, Supramol. Chem. 1996, 8, 45 ± 52.
[36] Analysis of CPK space-filling models indicated that the linker
between the sapphyrin core and the carboxylate group in 13 is not
long enough to allow for efficient, within-monomer carboxylate-to-
sapphyrin binding. This precludes internal collapse.
[37] M. Berger, F. P. Schmidtchen, J. Am. Chem. Soc. 1996, 118, 8947 ±
8948.
[38] The evidence for compound 13 self-dimerization was obtained from
FAB mass-spectrometric analysis. An intense dimer peak at twice the
molecular weight was found when carboxylate 13 was analyzed in this
way. Such a peak, however, was not seen when the control tert-butyl
ester 11 was subject to similar mass spectrometric analyses. Although
these analyses were carried out in the gas phase, they can be safely
extrapolated to solution conditions, as shown previously (see
ref. [16b]). Further, in the case of 13, the proposed aggregation effect
was confirmed by 1H NMR dilution experiments carried out in
[D2]dichloromethane.
[23] A. P. Krapcho, C. S. Kuell, Synth. Commun. 1990, 20, 2559 ± 2564.
[24] J. L. Sessler, G. W. Genge, A. Urbach, P. I. Sansom, Synlett 1996, 187 ±
188.
[25] H. Tsukube, Liquid Membranes: Chemical Applications (Eds.: T.
Araki, H. Tsukube), CRC, Boca Raton, FL, 1990.
[26] In particular, increasing the pH of the receiving phase leads to
deprotonation of the sapphyrin macrocycle. This, in turn should lead
to weakening of the sapphyrin ± amino acid carboxylate non-covalent
interactions and thus result in a release of the amino acid substrate
into Aq. II.
[27] Compound 1b was chosen as a control because it has a large
hydrophobic tail attached to the sapphyrin macrocycle. It was
prepared by coupling the sapphyrin mono acid 1a with mono-tert-
butyloxycarbonylethylenediamine, using 1,1'-carbonyldiimidazole as
the coupling agent (see Experimental Section).
[39] The deprotected, presumably zwitterionic substance 14 displayed
poor transport capabilities, just as did its congener compound 13.
[40] These results, not fully understood from a mechanistic point of view,
somewhat resemble those of chiral amplification, a nonlinear phe-
nomenon observed in asymmetric catalysis: R. Noyori, Asymmetric
Catalysis in Organic Synthesis, Wiley, New York, 1994.
[41] The W-tube was constructed in a fashion similar to Cramꢁs Catalytic
Resolving Machine (ref. [5a]). It has two organic phases (each
containing competing carrier compounds) which share the same
initial phase, Aq. I, (containing a mixture of the amino acids in
question). Each organic phase was put in contact with its own
receiving phase (Aq. II and Aq. III). The initial Aq. I phase and both
of the organic phases were continuously stirred at equal rates. The
increase in concentration of amino acids in phases Aq. II and Aq. III
was monitored by HPLC analysis.
[28] It appears, however, that the sapphyrin part of the conjugate plays
greater role in enhancing amino acid transport than the lasalocid part.
[29] These spectroscopic titrations were carried out in dichloromethane
with aliquots of the substrates in question (10 5 m in dichloromethane:
[42] For instance, the kt(l-Phe)/kt(d-Phe) ratio for compound 12 went up
from 1.6 when measured in the U-tube to 2.1 when studied by means
of the W-tube. The more complex equilibria relevant to the W-tube
conditions are probably responsible for this effect.
Chem. Eur. J. 1998, 4, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0401-0167 $ 17.50+.25/0
167