Second generation calixpyrrole anion sensors [28]

Full text of DOI:10.1021/ja001308t

9350
J. Am. Chem. Soc. 2000, 122, 9350-9351
Scheme 1
Second Generation Calixpyrrole Anion Sensors
Pavel Anzenbacher, Jr., Karolina Jurs´ıkova´, and
Jonathan L. Sessler*
Department of Chemistry and Biochemistry
and Institute for Cellular and Molecular Biology
UniVersity of Texas at Austin, Austin, Texas 78712-1167
ReceiVed April 14, 2000
ReVised Manuscript ReceiVed August 10, 2000
such as aromatic amino acids, it was considered desirable to use
labels that would allow for excitation by visible light (λAbs
>
300 nm). Given these considerations, we decided to use dansyl,
Lissamine-rhodamine B, and fluorescein as fluorescent labels.7
These labels show appreciable fluorescence intensity in aqueous
solutions even at very low concentrations (in this work the
concentrations of the sensors were always kept e5 µM).
The synthesis of sensors 1-3 departs from the general precursor
6. This key intermediate may be prepared in multigram scale in
two steps from Cbz-protected 3-aminoacetophenone 4, 3-pen-
tanone, and pyrrole in the presence of BF3:Et2O (Scheme 1).
Deprotection of the initial product 5 then produces 6 in 21%
overall yield. Sensors 1-3 were then prepared using standard
labeling methodologies8 and were isolated in 92%, 68%, and 93%
yields, respectively.
All three sensors 1-3 proved soluble in a wide range of organic
solvents. Acetonitrile was selected because it is water miscible,
meaning it would allow the sensing of anions added in the form
of aqueous solutions. In the case of sensors 1 and 2, acetonitrile
containing 0.01% water (which corresponds to a water concentra-
tion of ca. 5.6 mM) was used. In the case of sensor 3, a system
designed in such a way that it actually requires the presence of
water to hydrolyze its nonfluorescent precursor (i.e., the corre-
sponding lactone), studies were carried out in solutions of
acetonitrile containing 4% water by volume ([H2O] ) 2.2 M).
Under these conditions, sensor 3 was found to operate at pH 6.5-
8.5, with neutral pH 7.0 ( 0.1 being used for quantitative studies.
Anions tested as potential substrates for sensors 1-3 included
fluoride, chloride, dihydrogenphosphate, and hydrogen pyrophos-
phate. These anions, studied in the form of their tetrabutylam-
monium (TBA+) salts, were chosen because of their biological
importance (especially Cl-, H2PO4-, and HP2O73-).
While traditional methods of anion sensing such as ion selective
electrodes continue to hold their ground, increasing attention is
being devoted to finding alternative ways of effecting anion
detection. Here, sensors based on anion-induced changes in
fluorescence appear particularly attractive. They offer the potential
for high sensitivity at low analyte concentration coupled with
obvious ease of use.1 Unfortunately, few, if any, fluorescent anion
sensors exist that display high phosphate/chloride selectivities,
emit in the visible region, or function in aqueous media over a
wide range of pH. Such attributes, however, would be desirable.
They might allow, among other things, the study of metabolic
processes in biological milieus without interference from endog-
enous substrates such as chloride anion or aromatic amino acids.
Recently we described a new class of fluorescent anion sensors2
that are based on the use of octamethyl calix[4]pyrrole3 as the
anion recognition element.4 In these first generation sensors, an
anthracene derivative was used as the fluorescent signaling device.
Unfortunately, drawbacks, including low phosphate:chloride
selectivity ratios and less-than-ideal generalized affinities for
anions, prompted us to search for improved systems. In this
communication, we report the synthesis of three second generation
calixpyrrole-based fluorescent anion sensors, compounds 1-3.
These systems bind anions with greater affinity than previous
systems while displaying a more efficient fluorescent response.
In the design of sensors 1-3, a rigid aromatic spacer was used
so as to fix the distance between the quencher (anion) and the
signaling moiety. This spacer element contained either a sulfona-
mide5 (compound 1 and 2) or thiourea6 (sensor 3) group. These
linker moieties were introduced with the expectation that they
might provide additional hydrogen bond donor sites that would
act in concert with the calixpyrrole NH protons to enhance the
overall anion binding affinities.
1H NMR spectroscopic analyses were used to establish 1:1
binding stoichiometries.9 They were also used to carry out
qualitative binding titrations. Here, for instance, concerted down-
field shifts were observed for the protons attached to C2 of the
aromatic spacer, the pyrrole, and the sulfonamide nitrogen as
receptors 1 and 2 were exposed to increasing concentrations of
anions. Likewise, the multiplets corresponding to the C4, C5, and
C6 protons of the phenyl spacer were seen to be shifted to higher
field as the concentration of anions was increased. Taken together,
these concerted changes support the contention that all the
The choice of fluorescent label was guided by two consider-
ations. First, to target biological analytes, it was appreciated that
the sensors would have to function in the presence of water (i.e.,
either in water itself or in a solvent in which water is miscible).
Second, to avoid possible interference from fluorescent impurities,
(1) (a) Chemosensors of Ion and Molecular Recognition; Desverne, J.-P.,
Czarnik, A. W., Eds.; NATO ASI Series, Ser. C; Kluwer: Dordrecht, The
Netherlands, 1997; Vol. 492, (b) de Silva, A. P.; Guanarante, H. Q. N.;
Gunnlaugson, T.; Huxley, A. J. M.; McCoy, C. P.; Rademacher, J. T.; Rice,
T. E. Chem. ReV. 1997, 97, 1515-1566. (c) Beer, P. D. Chem. Commun.
1996, 689-696, (d) Dickens, R. S.; Gunnlaugson, T.; Parker, D.; Peacock,
R. D. Chem. Commun. 1998, 1643-1644. (e) Fabbrizzi, L.; Faravelli, I.;
Francese, G.; Licchelli, M.; Perotti, A.; Tagletti, A. Chem. Commun. 1998,
971-972. (f) Black, C. B.; Andrioletti, B.; Try, A. C.; Ruiperez, C.; Sessler,
J. L. J. Am. Chem. Soc. 1999, 121, 10438-10439.
(7) Sensor 1 (fluorescent label: dansyl) has absorption maximum λ ) 350
nm, emission λ ) 520 nm, Φ ) 0.36; sensor 2 (fluorescent label: Lissamine)
has absorption maximum λ ) 550 nm, emission λ ) 575 nm, Φ ) 0.18;
sensor 3 (fluorescent label: fluorescein) has absorption maximum λ ) 510
nm, emission λ ) 525 nm, Φ ) 0.42. Fluorescence quantum yields (Φ) were
measured following the method described by Demas and Crosby (Demas, J.
N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991-1024) with quinine sulfate
as standard (2 µM solution in 0.05 M sulfuric acid, Φ ) 0.546).
(8) Haughland, R. P. Handbook of Fluorescent Probes and Research
Chemicals, 6th ed.; Molecular Probes, Eugene-Leiden, 1996. Lefevre, C.;
Kang, H. C.; Haughland, R. P.; Malekzadeh, N.; Arttamangkul, S.; Haughkand,
R. P. Bioconj. Chem. 1996, 7, 482-489.
(9) Tsukube, H.; Furuta, H.; Odani, A.; Takeda, Y.; Kudo, Y.; Inoue, Y.;
Liu, Y.; Sakamoto, H.; Kimura, K. Determination of Stability Constants, in
ComprehensiVe Supramolecular Chemistry; Atwood, J. L., Davies, J. E. D.,
Macnicol, D. D., Vogtle, F., Eds.; Elsevier Science Ltd.: New York, 1996;
Vol. 8, pp 425-482.
(2) Miyaji, H.; Anzenbacher, P., Jr.; Sessler, J. L.; Bleasdale, E. R.; Gale,
P. A. Chem. Commun. 1999, 1723-1724.
(3) Baeyer, A. Ber. Dtsch. Chem. Ges. 1886, 19, 2184-2185.
(4) Gale, P. A.; Sessler, J. L.; Kra´l, V.; Lynch, V. J. Am. Chem. Soc. 1996,
118, 5140-5141. Gale, P. A.; Sessler, J. L.; Kra´l, V. Chem. Commun. 1998,
1-8.
(5) Kavallieratos, K.; Bertao, C. M.; Crabtree, R. H. J. Org. Chem. 1999,
64, 1675-1683.
(6) Sasaki, S.; Mizuno, M.; Naemura, K.; Tobe, Y. J. Org. Chem. 2000,
65, 275-283. Bu¨hlmann, P.; Nishizawa, S.; Xiao, K. P.; Umezawa, Y.
Tetrahedron 1997, 53, 1647-1654. Scheerder, J.; Engbergsen, J. F. J.; Casnati,
A.; Ungaro, R.; Reinhoudt, D. N. J. Org. Chem. 1995, 60, 6448-6454.
10.1021/ja001308t CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/08/2000

Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 38, 2000 9351
Figure 1. Structures of second generation sensors 1-3. These systems
contain a rigid spacer and rely on the use of dansyl (1), Lissamine-
rhodamine B (2), and fluorescein (3) moieties as the fluorescent elements,
respectively.
Figure 3. Schematic representation of the multiple hydrogen bonding
interactions that are believed to account for the high phosphate and
pyrophosphate affinities observed for sensors 1 and 2 and 3, respectively.
hydrogen bonding interactions involving these two nonspherical
anions. Such effects, illustrated in Figure 3, are likely to be
particularly pronounced in the case of sensor 3 and pyrophosphate
dianion where the coordination of the second anionic center within
the pyrophosphate by the thiourea moiety is believed to be
responsible for the dramatic increase in affinity.10
One of the more interesting findings to emerge from the present
study was that for sensors 1-2, the sensing efficiency11 was
actually improved by the presence of water in acetonitrile (up to
0.01% v/v), presumably as the result of improved solvation of
the charged/ionic moieties present in the fluorescent labels, a
feature that also prevents the formation of sensor aggregates.
Consistent with this proposal is the finding that, in the absence
of water, the relevant anion binding isotherms display biphasic
character. This is rationalized in terms of sensor aggregate
dissociation and anion binding occurring at the same time.
Addition of water, on the other hand, results in binding isotherms
of nearly ideal hyperbolic shape; it also gives rise to higher
numeric values for the affinity constants under consideration. The
presence of water was also found to be beneficial for sensor 3.
This latter species was found to operate as a viable fluorescent
sensor at concentrations of water in acetonitrile of up to 20%
(v/v).
Figure 2. Decrease in fluorescence emission intensity observed when
sensor 2 (0.1 µM in acetonitrile containing 0.01% v/v water) is titrated
with increasing concentrations of tetrabutylammonium fluoride. From top
to lowest trace, [F-] ) 0, 0.30, 0.76, 1.53, 2.30, 3.06, 3.83, 4.60 µM.
Table 1. Affinity Constantsa for Sensors 1-3 and Anionic
Substrates As Determined in Acetonitrile (0.01% v/v water) for
Sensors 1 and 2 and Acetonitrile-Water (96:4, pH 7.0 ( 0.1) for
Sensor 3
association constants (mol-1
determined by emission quenching
)
In summary, sensors 1-3 display the highest anion binding
affinities for anions yet recorded for calixpyrrole-type receptors.
They are also the first to show high phosphate/chloride selectivity
(2 orders of magnitude) and, in the case of sensor 3, the first
such systems to operate successfully in the presence of water at
physiological pH.
sensor 1
sensor 2
sensor 3
(5 × 10-6 M)
(1 × 10-6 M)
(5 × 10-6 M)
F-
222 500
10 500
168 300
131 000
>1 000 000
>0 200 000
<10 000
Cl-
H2PO4
HP2O7
18 200
-
446 000
170 000
682 000
3-
>2 000 000
a For a detailed description of the experimental conditions used for
these titration experiments see the Supporting Information.
Acknowledgment. This work was supported by NSF and NIH (Grants
CHE9725399 and GM58907, respectively, to J.L.S.) as well as the Texas
Advanced Research Program.
hydrogen bond donors within the molecule of sensors 1-3 act in
a cooperative fashion.
Supporting Information Available: Synthetic experimental details
for all 1-6, Scott plots for the fluorescence titrations, and representative
Job plots (PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
More quantitative assessments of the anion affinities came from
observing the extent to which the fluorescence intensity of sensors
1-3 was quenched in the presence of anions (Figure 2).
From the extent of this quenching, appreciable for all three
sensors and all four anions, affinity constants were calculated.1f
The resulting values are listed in Table 1. While, as expected,4
fluoride anion gave rise to the largest response, an inspection of
Table 1 reveals that sensors 1-3 are remarkably selective for
phosphate and pyrophosphate anions relative to Cl-. The high
selectivity for phosphate and pyrophosphate, something not seen
previously for any calixpyrrole-based system,2,4 is potentially
advantageous in biological sensing applications where a high
concentration of Cl- (but not F-) pertains. In the case of sensors
1-3, this selectivity can be explained by the presence of multiple
JA001308T
(10) Related arguments can be advanced to explain the differences in
binding strength between receptors 1, 2, and 3. Sensor 2 contains a sulfonic
acid residue that increases the acidity of the sulfonamide NH and, relative to
1, its presumed efficacy as a hydrogen bond donor. Likewise, the presence of
two ancillary hydrogen bond donors in the thiourea moiety is expected to
increase the general anion binding efficiency of 3.
(11) Sensing efficiency as used here is a qualitative term meant to imply
improvements in a range of desirable features including linear response of
the sensors to the presence of anions and a higher degree of quenching at
lower anion concentrations.