Computer-guided design in molecular recognition: Design and synthesis of a glucopyranose receptor

Article abstract of DOI:10.1002/1521-3773(20010504)40:9<1714::AID-ANIE17140>3.0.CO;2-F

Greater than 400-fold selectivity for glucose over galactose, mannose, and fructose has been achieved with a receptor designed by the computer program CAVEAT. The receptor has precisely positioned boronic acid groups of α-D-glucopyranose (see picture) and

Full text of DOI:10.1002/1521-3773(20010504)40:9<1714::AID-ANIE17140>3.0.CO;2-F

COMMUNICATIONS
[6] a) K. Ichimura, S.-K. Oh, M. Nakagawa, Science 2000, 288, 1624;
b) J. W. Weener, E. W. Meijer, Adv. Mater. 2000, 12, 741.
mixture generally requires the interaction of multiple func-
tional groups of the receptor with complementary function-
ality of the guest compound.[3, 4] Good affinity and selectivity
requires the precise placement of the recognition elements of
the receptor in the proper position and orientation for optimal
complementarity to the guest. Nature uses proteins (enzymes,
antibodies, and other protein-based receptors) as a basis for
specific receptors. Efforts to achieve specificity and affinity in
smaller synthetic molecules has relied primarily on intuition
for the identification of molecular scaffolds that would permit
proper orientation of the functionality for molecular recog-
nition. We envisioned that the computer program CAVEAT,
developed by Bartlett and co-workers and previously utilized
for the design of enzyme inhibitors and conformationally
constrained peptides, could serve as a valuable tool for the
discovery of molecular backbones for the orientation of
functional groups for molecular recognition.[5, 6] Described
here is the demonstration of a CAVEAT-based design
approach in the development of a glucopyranose receptor
incorporating precisely positioned arylboronic acid groups as
recognition elements.
Â
[7] A. P. H. J. Schenning, C. Elissen-Roman, J. W. Weener, M. W. P. L.
Baars, S. J. van der Gaast, E. W. Meijer, J. Am. Chem. Soc. 1998, 120,
8199.
[8] T. Gensch, J. Hofkens, J. van Stam, H. Feas, S. Creutz, K. Tsuda, R.
Jerome, H. Masuhara, F. C. De Schryver, J. Phys. Chem. B 1998, 102,
8440.
[9] Procedures in Scanning Probe Microscopies, Wiley, Chichester, 2000.
[10] B. M. Discher, Y.-Y. Won, D. S. Ege, J. C.-M. Lee, F. S. Bates, D. E.
Discher, D. A. Hammer, Science 1999, 284, 1143.
[11] H. Hub, B. Hupfer, H. Koch, H. Ringsdorf, Angew. Chem. 1980, 92,
962; Angew. Chem. Int. Ed. Engl. 1980, 19, 938.
[12] K. Tsuda, G. Dol, T. Gensch, J. Hofkens, L. Latterini, J. W. Weener,
E. W. Meijer, F. C. De Schryver, J. Am. Chem. Soc. 2000, 122, 3445.
[13] M. Shimomura, R. Ando, T. Kunitake, Ber. Bunsenges.Phys. Chem.
1983, 87, 1134.
[14] M. Shimomura, T. Kunitake, J. Am. Chem. Soc. 1987, 109, 5175.
[15] S. R. McAlpine, S. L. Schreiber, Chem. Eur. J. 1999, 5, 3528.
[16] F. M. Menger, A. Archut, Acc. Chem. Res. 1998, 31, 789.
[17] F. M. Menger, J. S. Keiper, Adv. Mater. 1998, 10, 888.
[18] A. Ashkin, Phys. Rev. Lett. 1970, 24, 156.
[19] A. Ashkin, J. M. Dziedzic, J. E. Bjorkholm, S. Chu, Opt. Lett. 1986, 11,
288.
[20] K. Svoboda, S. M. Block, Annu. Rev. Biophys. Biomol. Struct. 1994, 23,
247.
Arylboronic acids have long been known to form stable
complexes with sugars and other diols in aqueous solution.[7]
Numerous arylboronic acids have been prepared and studied
as sugar receptors, and those incorporating fluorescent aryl
groups have been explored as fluorescence-based sugar
sensors.[8±15] Of particular interest are glucose sensors for
potential application in the maintenance of blood glucose
levels in persons with diabetes.[16, 17] However, the simple
arylboronic acids form stable complexes with a variety of
sugars and thus are not useful as specific receptors or sensors
for a single sugar. Compounds have been prepared that
contain a pair of arylboronic acid groups in somewhat flexible
structures which form two cyclic boronates with a single sugar
molecule.[11±14] These bis-boronic acids, while not designed for
complexation with a specific sugar, have demonstrated some-
what enhanced, though still modest, selectivity. A glucopyr-
anose receptor 1 containing a pair of precisely positioned
phenylboronic acid groups was chosen as an initial target for
receptor design using CAVEAT (Scheme 1). The large oval
structure in 1 represents a polycyclic organic framework to be
identified using CAVEAT. Since glucopyranose 2 can form
cyclic boronates between the a-1,2 and 4,6-hydroxy
groups,[14, 18] the receptor was designed to incorporate aryl-
boronic acid groups in the proper relative position to form a
[21] L. I. McCann, M. Dykman, B. Golding, Nature 1999, 785.
[22] C. DꢁHelon, E. W. Dearden, H. Rubinsztein-Dunlop, N. R. Hecken-
berg, J. Mod. Opt. 1994, 41, 595.
[23] M. E. J. Friese, H. Rubinsztein-Dunlop, N. R. Heckenberg, E. W.
Dearden, Appl. Opt. 1996, 35, 7112.
[24] R. M. Simmons, J. T. Finer, S. Chu, J. A. Spudich, Biophys. J. 1996, 70,
1813.
[25] J. D. Moroz, P. Nelson, R. Bar-Ziv, E. Moses, Phys. Rev. Lett. 1997, 78,
386.
[26] P. Borowicz, J. Hotta, K. Sasaki, H. Masuhara, J. Phys. Chem. B 1998,
102, 1896.
[27] J. Hofkens, J. Hotta, K. Sasaki, H. Masuhara, T. Taniguchi, T.
Miyashita, J. Am. Chem. Soc. 1997, 119, 2741.
[28] J. Hofkens, J. Hotta, K. Sasaki, H. Masuhara, K. Iwai, Langmuir 1997,
13, 414.
Computer-Guided Design in Molecular
Recognition: Design and Synthesis of a
Glucopyranose Receptor**
Wei Yang, Huan He, and Dale G. Drueckhammer*
The design of compounds which are capable of specific
recognition of molecules and ions is a long-standing challenge
in organic chemistry. In addition to its fundamental interest,
this work is of increasing practical value in the development of
chemosensors for compounds of biological and environmental
importance[1] and is a key element of biomimetic catalysis.[2]
The specific binding of one compound from a complex
HO
HO
HO
O
O
HO
-
HO
O
O
OH
B
2
HO
B(OH)2
–2 H3O+
O
- B
O
B(OH)2
OH
[*] Prof. D. G. Drueckhammer, W. Yang, H. He
Department of Chemistry
State University at Stony Brook
Stony Brook, NY 11794-3400 (USA)
Fax : (‡1)631-632-7960
E-mail: dale.drueckhammer@sunysb.edu
3
1
[**] This work was supported by the National Institutes of Health (grant
DK5523402).
Scheme 1. Complex formation between glucose and a bis-arylboronic acid.
1433-7851/01/4009-1714 $ 17.50+.50/0 Angew. Chem. Int. Ed. 2001, 40, No. 9
1714
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

COMMUNICATIONS
complex represented by structure 3. This is in contrast to the
simple arylboronic acids and the bis-arylboronic acids which
form complexes in aqueous solution with the a-furanose form
of glucose.[14, 15]
The design approach for the receptor 1 is illustrated in
Scheme 2. Initially a conformational search of the bis-p-
tolylboronate derivative of glucose, namely 4, was performed
and synthetically accessible but isosteric furan ring. The other
major modification is the introduction of a bridging oxygen
atom to avoid the out of plane twisting caused by the
interaction of the two hydrogen atoms for which the oxygen
was substituted. More minor changes are the introduction of a
hydroxymethylene group on the furan ring for increasing
water solubility and for possible derivatization for further
solubility enhancement, and the introduction of a methyl
group on the pyran ring, a substitution which was dictated by
the availability of the commercially available starting material
9 (see Scheme 3) bearing this extra methyl group. Geometry
optimization of 7 using HF//6-31G* showed that its lowest
energy conformation matched very well with the bis-phenyl-
boronate glucose complex 4. Likewise, geometry optimization
of the glucose complex 8 by the same method indicated almost
no change in the conformation of the receptor 7 and no
accompanying increase in strain upon glucose binding. The
superimposed structures of 7 and 8 (Figure 1) give a root
mean square deviation (RMSD) of 0.279 Š for common
atoms. On the basis of these positive results, the synthesis of
the designed glucose receptor 7 was undertaken.
O
HO
-
O
O
B
HO
computer
modeling
O
- B
O
CAVEAT
OH
CH3
vector pair
CH3
4
B(OH)2
H
B(OH)2
H
H
5
H
6
Scheme 2. Computer-based design of a glucopyranose receptor.
using AM1-SM2, a semiemperical method incorporating an
aqueous solvation model.[19] This search was followed by
geometry optimization using HF//6-31G* to predict the lowest
energy conformation of 4. The resulting computer-generated
structure was used to define a pair of vectors corresponding to
the methyl ± aryl bonds (in bold) of 4. The program CAVEAT
was then used to identify polycyclic organic structures having
substituent bonds to the polycyclic framework matching the
vector pair. TRIAD, a computer-generated collection of
4 Â 105 tricylic hydrocarbons, was used as the database for
the search.[5] This search identified about 300 structures
having substituent bonds closely matching the vector pair
defined by the computer-generated model of 4. After
elimination of most of the structures because of the presence
of several chiral centers and/or highly strained and unusual
ring systems, compound 5 was chosen as the lead structure,
primarily because of its apparent ease of synthesis. The C H
bonds of 5 in bold correspond to the vectors defined by 4.
Simple replacement of these hydrogen atoms with phenyl-
boronic acid groups leads to the prospective glucose receptor
6. The structure was further modified to arrive at the structure
7. One significant modification in arriving at structure 7 from
6 is the replacement of the fulvene ring with the more stable
Figure 1. Superposition of the energy-minimized structures of the glucose
receptor 7 and the glucose complex 8.
The right side of 7 was derived from compound 10, which
was prepared as shown in Scheme 3. Commercially available
7-hydroxy-4-methylcoumarin (9) was converted into the
triflate derivative followed by reaction with paraformalde-
hyde and HCl to introduce the chloromethylene group at C3
and give 10.[20, 21] The synthesis of 7 is shown in Scheme 4.
O
HO
B(OH)2
O
O
B
-
HO
O
O
OTf
O
1. triflic anhydride
pyridine
O
OH
O
O
B(OH)2
B
-
Cl
OH
2. (CH2O)n, HCl,
ZnCl2
O
O
O
O
CH3
CH3
10
9
OH
CH3
OH
CH3
H
Scheme 3. Synthesis of the right side of the glucopyranose receptor. Tf ˆ
H
7
8
trifluoromethanesulfonyl.
Angew. Chem. Int. Ed. 2001, 40, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4009-1715 $ 17.50+.50/0
1715

COMMUNICATIONS
The formation of a complex with glucose was initially
analyzed by H NMR spectroscopy and mass spectrometry.
O
O
O
SmI3
1
O
O
Cl
+
The receptor 7 was combined with 1.8 equivalents of NaOD
and 0.5 equivalents of d-glucose in CD3OD. Only 0.5 equiv-
alents of glucose were used in an effort to observe complex
formation with only the preferred enantiomer of 7 and thus
avoid the spectral complexity of diasteromeric complexes.
The resulting sample was used for both mass spectral and
1H NMR analysis. The electrospray ionization (ESI) mass
spectrum showed large peaks at m/z 301.12 for the dianionic
tetradeuterated complex 8 (C31H24B2O11D4, Mr ˆ 602) and at
m/z 603.25 for the corresponding protonated monoanion.
Smaller peaks were observed at m/z 249.01 for the dianionic
heptadeuterated boronate form of 7 (C25H17B2O9D7, Mr ˆ
497) and at m/z: 498.72 for the corresponding protonated
monoanion. Importantly, no peaks were observed at m/z 321
or 643, which correspond to the dianionic and protonated
monoanionic forms, respectively, of a complex of glucose with
only one of the boronic acid groups of 7. Such a complex
would be 40 mass units (two equivalents of D2O) larger than
8. Interestingly, no peaks were observed that corresponded to
complete or partial exchange of the boronic acid OH groups
in complex 8 or the free receptor 7 with OCD3 from the
solvent.
I
I
13
12
11
O
O
1. TosN3, Et3N
2. Rh(OAc)2
1. 2 equiv / LDA
2.
I
OTBS
Br
14
TBSO
15
I
I
OTf
O
1. LDA, THF
O
O
O
2. 10
O
O
CH3
OTBS
16
17
OTBS
O
O
O
O
O
O
B
B
B
B O
O
O
18
The 300 MHz 1H NMR spectra obtained within a few
minutes of mixing 7 and glucose showed complete disappear-
ance of the peaks corresponding to free glucose, with
substantial shifts of these protons mostly downfield, which is
indicative of complex formation. No further changes were
observed in the spectra taken 4 and 16 h after mixing, which
indicated that complex formation was complete within a few
minutes and that no isomerization of the initially formed
complex occurred.[14] The protons of the glucose moiety in the
complex were assigned using chemical shift predictions
generated using ACDLabs software. Coupling constants were
measured and compared to the coupling constants for 1,2;4,6-
di-O-benzylidine glucopyranose and 1,2;5,6-di-O-isopropyl-
idine glucofuranose (Table 1). The coupling constants be-
tween H5 and the protons on C6 are not included, as they are
dependent upon the conformation of the C5 C6 bond and not
necessarily indicative of the ring form. A similar comparison
of coupling constants has been used in the assignment of the
structures of other boronate complexes of glucose.[14] The
coupling constants, most notably J2,3 and J3,4, are clearly
consistent with complex formation with the pyranose form of
glucose, as expected on the basis of the receptor design and
computer modeling studies.[30] The NMR and mass spectral
results confirm the formation of complex 8 as predicted by
computer modeling studies.
O
O
[PdCl2(dppf)]
KOAc/DMSO
O
CH3
19
OTBS
O
O
B
O
B
O
1. LiAlH4
2. H3O+
TiCl3
LiAlH4
7
O
O
OTBS
CH3
20
Scheme 4. Synthesis of the glucopyranose receptor 7. LDA ˆ lithium
diisopropylamide; TBS ˆ tert-butyldimethylsilyl; Tos ˆ toluene-4-sulfonyl;
Ac ˆ acetyl; dppf ˆ 1,1'-bis(diphenylphosphanyl)ferrocene; DMSO ˆ di-
methyl sulfoxide.
Reaction of p-iodobenzoyl chloride (11) with 2,4-pentane-
dione (12) and samarium iodide gave the diketone 13.[22]
Formation of the dienolate of 13 followed by reaction with
the bromide 14, which was prepared in two steps from
2-butyn-1,4-diol,[23] gave 15. Conversion of 15 into the diazo
compound by reaction with tosyl azide followed by a
rhodium-catalyzed cyclization gave 16.[24] Deprotonation of
16 adjacent to the carbonyl group followed by alkylation of
the resulting enolate with 10 resulted in the formation of 17.
Reaction of 17 with the diboron diester 18 using a palladium
catalyst gave the protected bis-boronic acid 19.[25, 26] An
intramolecular McMurry reaction between the ketone and
lactone carbonyl groups produced 20.[27, 28] Deprotection of
the boronic acids and the primary alcohol gave the glucose
receptor 7 in racemic form.[29]
Table 1. 1H-1H Coupling constants for the glucose moiety of complex 8 and
related reference compounds.
Compound
J1,2
J2,3
J3,4
J4,5
8
6.3
4.9
3.6
7.8
6.4
0
8.0
9.4
2.8
8.8
9.2
6.8
pyranose[a]
furanose[b]
[a] 1,2;4,6-Di-O-benzylidine-a-d-glucopyranose. [b] 1,2;5,6-Di-O-isoproyl-
idine-a-d-glucofuranose.
1716
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4009-1716 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 9

COMMUNICATIONS
Fluorescence experiments were conducted to evaluate the
affinity of 7 for glucose and its selectivity for glucose versus
other common sugars. Profiles of the changes in fluorescence
versus concentrations of glucose, galactose, mannose, and
fructose at pH 7.5 are shown in Figure 2.[31] Glucose exhibited
find use in the design of receptors for the recognition of a wide
range of molecules and ions.
Received: October 13, 2000
Revised: February 26, 2001 [Z15951]
[1] A. W. Czarnik, Chem. Biol. 1995, 2, 423 ± 428.
[2] R. Breslow, Acc. Chem. Res. 1995, 28, 146 ± 153; P. Tecilla, V. Jubian,
A. D. Hamilton, Tetrahedron 1995, 51, 435 ± 438.
[3] E. V. Anslyn, A. Metzger, Angew. Chem. 1998, 110, 682 ± 684; Angew.
Chem. Int. Ed. 1998, 37, 649 ± 652; C.-Y. Huang, L. A. Cabell, E. V.
Anslyn, J. Am. Chem. Soc. 1994, 116, 2778 ± 2792.
[4] G. Das, A. D. Hamilton, J. Am. Chem. Soc. 1994, 116, 11139 ± 11140;
T. W. Bell, Z. Hou, Y. Luo, M. G. B. Drew, E. Chapoteau, B. P. Czech,
A. Kumar, Science 1995, 269, 671.
[5] G. Lauri, P. A. Bartlett, J. Comput. Aided Mol. Des. 1994, 8, 51 ± 66.
[6] P. A. Bartlett, G. T. Shea, S. J. Telfer, S. Waterman in Molecular
Recognition: Chemical and Biological Problems (Ed.: S. M. Roberts),
Royal Society of Chemistry, London, 1989, pp. 182 ± 196; P. A.
Bartlett, F. A. Etzkorn, T. Guo, G. Lauri, K. Liu, M. Lipton, B. P.
Morgan, G. T. Shea in Chemistry at the Frontiers of Medicine,
Proceedings of the Robert A. Welch Foundation Conference on
Chemical Research XXXV (Houston TX), 1992, pp. 45 ± 68.
[7] J. P. Lorand, J. O. Edwards, J. Org. Chem. 1959, 24, 769 ± 774.
[8] J. Yoon, A. W. Czarnik, J. Am. Chem. Soc. 1992, 114, 5874 ± 5875; T. D.
James, K. R. A. S. Sandanayake, S. Shinkai, J. Chem. Soc. Chem.
Commun. 1994, 477 ± 478.
Figure 2. Relative fluorescence intensity Irel of 7 as a function of the
saccharide concentrations lgc at 258C with 1.0 Â 10 5 m of 7 in 30% MeOH/
&
aqueous phosphate buffer at pH 7.5. lex ˆ 375 nm, lem ˆ 447 nm.
d-
&
*
*
glucose, d-galactose, d-mannose, d-fructose.
[9] H. Suenaga, M. Mikami, K. R. A. S. Sandanayake, S. Shinkai,
Tetrahedron Lett. 1995, 36, 4825 ± 4828.
a 400-fold greater affinity than any of the other sugars for 7,
with an apparent dissociation constant for the complex of
2.5 Â 10 5 m.[32] The dissociation constants for the galactose
[10] P. Linnane, T. D. James, S. Imazu, S. Shinkai, Tetrahedron Lett. 1995,
36, 8833 ± 8834; K. R. A. S. Sandanayake, T. D. James, S. Shinkai,
Chem. Lett. 1995, 503 ± 504; T. D. James, P. Linnane, S. Shinkai, Chem.
Commun. 1996, 281 ± 288. T. D. James, K. R. A. S. Sandanayake, S.
Shinkai, Nature 1995, 374, 345 ± 347.
[11] T. D. James, K. R. A. S. Sandanayake, S. Shinkai, Angew. Chem. 1996,
108, 2038 ± 2050; Angew. Chem. Int. Ed. Engl. 1996, 35, 1910 ± 1922.
[12] T. D. James, K. R. A. S. Sandanayake, S. Shinkai, Angew. Chem. 1994,
106, 2287 ± 2289; Angew. Chem. Int. Ed. Engl. 1994, 33, 2207 ± 2209;
T. D. James, K. R. A. S. Sandanayake, R. Iguchi, S. Shinkai, J. Am.
Chem. Soc. 1995, 117, 8982 ± 8987.
[13] K. Tsukagoshi, S. Shinkai, J. Org. Chem. 1991, 56, 4089 ± 4091.
[14] M. Bielecki, H. Eggert, J. C. Norrild, J. Chem. Soc. P2 1999, 449 ± 455.
[15] H. Eggert, J. Frederiksen, C. Morin, J. C. Norrild, J. Org. Chem. 1999,
64, 3846 ± 3852.
[16] C. R. Cooper, T. D. James, Chem. Lett. 1998, 883 ± 884.
[17] D. A. Gough, J. C. Armour, Diabetes 1995, 44, 1005 ± 1009. R. J.
McNichols, G. L. Cote, J. Biomed. Opt. 2000, 5, 15 ± 16.
2
and mannose complexes were 1.0 Â 10 and 1.6 Â 10 2 m,
respectively. The dissociation constant for the fructose com-
plex appears similar to the latter two, although the change in
fluorescence intensity was too small for an accurate determi-
nation. The selectivity is attributed to the fact that only
glucose is believed to form a bidentate complex involving
both of the boronic acid groups of the receptor. The degree of
selectivity observed is unprecedented in receptors based on
boronic acid; the apparent previously most selective glucose
sensor exhibits a 12-fold and 25-fold selectivity versus fructose
and mannose, respectively.[12] The maximum change in the
fluorescence intensity was also much greater with glucose,
with a 50% decrease in the intensity observed at saturating
glucose concentrations, while a change of less than a 30% was
observed for the other sugars.
[18] G. T. Morin, M.-F. Paugam, M. P. Hughes, B. D. Smith, J. Org. Chem.
1994, 59, 2724 ± 2728.
[19] C. J. Cramer, D. G. Truhlar, Science 1992, 26, 213 ± 217; all calculations
were carried out using the Gaussian 98 suite of programs: Gaussian98
(RevisionA.7), M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A.
Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M.
Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,
S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K.
Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.
Foresman, J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J.
Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C.
Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen,
M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle, J. A.
Pople, Gaussian, Inc., Pittsburgh, PA, 1998.
Apparently both enantiomers of 7 form glucose complexes
with similar binding constants: the fluorescence curve of
Figure 2 shows no evidence for two discernable binding
constants and the 50% decrease in fluorescence is too large to
be attributed to complex formation with only half of the
racemic 7. This proposal is supported by computational
evaluation of the glucose complexes with both enantiomers
of 7 and the fact that the two phenyl boronic acids and their
point of attachment to the rigid structure are almost coplanar
and parallel.
Work is underway to further explore the potential applica-
tion of 7 and derivatives thereof as fluorescence-based glucose
sensors. In addition to the practical application of 7, this work
demonstrates a novel computer-based design approach for
molecular recognition that uses the computer program
CAVEAT. It is anticipated that this general approach may
[20] S. S. Lele, N. G. Savant, S. Sethna, J. Org. Chem. 1960, 25, 1713 ± 1716.
[21] D. I. Ivanova, S. V. Eremin, V. I. Shvets, Tetrahedron 1996, 52, 9581 ±
9588.
[22] W. Hao, Y. Zhang, T. Ying, P. Lu, Synth. Commun. 1996, 26, 2421 ±
2427.
Angew. Chem. Int. Ed. 2001, 40, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4009-1717 $ 17.50+.50/0
1717

COMMUNICATIONS
[23] 14 was prepared by reaction of 2-butyn-1,4-diol with 1.0 equiv tert-
butyldimethylsilyl chloride and imidazole to form the mono-TBS
protected derivative followed by reaction with triphenylphosphane
and carbon tetrabromide.
[24] A. Padwa, F. R. Kinder, J. Org. Chem. 1993, 58, 21 ± 28; T. R. Hoye,
C. J. Dinsmore, D. S. Johnson, P. F. Korkowski, J. Org. Chem. 1990, 55,
4518 ± 4520.
hollow framework which is expected to open up a new area of
porphyrin chemistry based on host ± guest interactions. Our
strategy employed here is the molecular paneling[5] of a
pyridine-functionalized porphyrin by metal coordination. A
prismlike hollow structure is quantitatively assembled from
three porphyrin ligands and six [PdII(en)]2‡ (en ˆ ethylene-
diamine) building blocks (Scheme 1). The spatially fixed
[25] T. Ishiyama, M. Murata, N. Miyaura, J. Org. Chem. 1995, 60, 7508 ±
7510.
[26] A. Giroux, Y. Han, P. Prasit, Tetrahedron Lett. 1997, 38, 3841 ± 3844; T.
Ishiyama, Y. Itoh, T. Kitano, N. Miyaura, Tetrahedron Lett. 1997, 38,
3447 ± 3450.
[27] J. E. McMurry, D. D. Miller, J. Am. Chem. Soc. 1983, 105, 1660 ± 1661.
[28] J. E. McMurry, T. Lectka, J. G. Rico, J. Org. Chem. 1989, 54, 3748 ±
3749.
[29] Compounds 20 and 7 were characterized by high- and low-resolution
1
mass spectra, respectively, in addition to H NMR analysis.
[30] Computer-based modeling also predicts that glucofuranose should not
form a stable complex with both boronic acid groups of 7.
[31] Values of Irel versus c were fitted to the equation c ˆ [(Irel 1)/De]/
K{[7]init. (Irel 1)/De} ‡ (Irel 1)/De using Origin 5.0. The curves of
Figure 2 were generated using B-spline fitting.
[32] The reported values are apparent dissociation constants as the
reactions involve uptake or release of hydroxide ions on complex
formation and dissociation, respectively, and the apparent equilibrium
constants are thus pH dependent.[8]
Scheme 1. Self-assembly of porphyrin prism 312‡
.
A Porphyrin Prism: Structural Switching
Triggered by Guest Inclusion**
porphyrin ligands surround a large hydrophobic cavity which
can accommodate neutral organic molecules, such as pyrene
and perylene, in an aqueous solution. Though metal-linked
porphyrinic square arrays have been reported by several
groups,[6] the porphyrin cores in the arrays seem to adopt
cofacial rather than perpendicular conformations and provide
no distinct hollow structures capable of guest inclusion. Quite
recently, a PdII-linked porphyrin cage which bound diamines
was reported by Shinkai and co-workers.[7]
Norifumi Fujita, Kumar Biradha, Makoto Fujita,*
Shigeru Sakamoto, and Kentaro Yamaguchi
Porphyrin assemblies play an essential role in biological
systems for oxygen transport,[1] electron transfer, and energy
migration and conversion.[2] In this regard, designing molec-
ular assemblies with porphyrin units is a promising approach
for constructing artificial functional systems. Both covalent[3]
and noncovalent[4] syntheses of polyporphyrins have offered
some novel classes of compounds with respect to their
structures and functions. Here, we report a porphyrin-based
The ligand designed here is the (3-pyridyl)-functionalized
porphyrin 1. This square panel-like ligand can be hinged by
[PdII(en)] units at two opposite ridges of the square, and
hence is anticipated to give a discrete porphyrin box
structure.[8] Thus, the treatment of porphyrin ligand 1 with 2
(2 molequiv) in H2O/CH3CN at 808C gave a purple homoge-
neous solution after one day. 1H NMR spectroscopic analysis
showed the formation of a single product, which was isolated
as a violet precipitate in 96% yield by adding acetone to the
solution. After exchange of NO3 with PF6 ions, the structure
of the product was assigned as 312‡ by coldspray-ionization
mass spectrometry (CSI-MS)[9] and NMR analyses. The CSI
mass spectrum displayed prominent peaks corresponding to
[M n(PF6 )‡m(CH3CN)]n‡,(n ˆ 3 ± 9, m ˆ 0 ± 16)[10] (see
Supporting Information). Interestingly, one NO3 ion was
not exchanged by an PF6 ion even when the complex was
treated with a large excess of NH4PF6. The unexchanged
NO3 ion was strongly trapped by the prism 312‡. This was
clearly evidenced by exact mass measurement.[11] For exam-
[*] Prof. Dr. M. Fujita, Dr. K. Biradha
Department of Applied Chemistry
Graduate School of Engineering
Nagoya University
Furocho, Chikusaku, Nagoya 464-8603 (Japan)
Fax : (‡81)52-789-3199
E-mail: mfujita@apchem.nagoya-u.ac.jp
N. Fujita
The Graduate University for Advanced Studies
Myodaiji, Okazaki 444-8585 (Japan)
S. Sakamoto, Prof. Dr. K. Yamaguchi
Chemical Analysis Center, Chiba University
Yayoicho, Inageku, Chiba 263-8522 (Japan)
[**] We thank the Noyori COE project at Nagoya University for the
800 MHz NMR measurement. N.F. acknowledges the JSPS for a
research fellowship. M.F. is responsible for the CREST project of the
Japan Science and Technology Corporation (JST).
‡
ple, the highest peak at m/z 567.7613 (7 charge) corresponds
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
to [M 6PF6
NO3 ‡7CH3CN]7‡ (calcd 567.7561) and not
1718
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4009-1718 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 9