Carbohydrate recognition through H-bonding and CH-π interactions by porphyrin-based receptors

Article abstract of DOI:10.1021/jo101384u

Porphyrin-based synthetic receptors containing urea, carbamate, or amide groups were designed and synthesized for carbohydrate recognition. The receptors have equatorially and convergently directed hydrogen-bonding sites into which the urea, carbamate, or

Full text of DOI:10.1021/jo101384u

pubs.acs.org/joc
Carbohydrate Recognition through H-Bonding and CH-π Interactions
by Porphyrin-Based Receptors
Jung-Deog Lee, Yeon-Hwan Kim, and Jong-In Hong*
Department of Chemistry, College of Natural Sciences, Seoul National University, Seoul 151-747, Korea
jihong@snu.ac.kr
Received July 14, 2010
Porphyrin-based synthetic receptors containing urea, carbamate, or amide groups were designed and
synthesized for carbohydrate recognition. The receptors have equatorially and convergently directed
hydrogen-bonding sites into which the urea, carbamate, or amide groups were introduced above the
porphyrin plane. The receptors exhibited remarkable affinities to pyranoside/furanoside derivatives
in organic media, demonstrating not only the importance of multiple hydrogen-bonding interactions
but also CH-π interactions in carbohydrate recognition. Among the three hydrogen-bonding
groups, urea NHs were used as the strongest H-bonding donors for sugar hydroxyl oxygens, and
the porphyrin plane was used for mimicking CH-π interactions with sugar CHs, which are found in
sugar-binding proteins. The binding interactions between the artificial receptors and carbohydrates
were elucidated by various spectroscopic analyses such as UV-vis titration, 1H NMR titration, CD
measurement, and computer-assisted modeling.
Introduction
found that the multiple hydrogen bonds in the hydrophobic
pocket of a protein play an important role in protein-
carbohydrate interactions.4 Numerous attempts have been
made to develop artificial carbohydrate receptors.5,6 How-
ever, most effective receptors working in aqueous media
still depend on the formation of a boronic acid ester between
the boronic acid of the receptors and the diol of the
carbohydrates.5b,e,f,p,q,s,t On the other hand, a few outstand-
ing studies have been reported on carbohydrate recognition
through hydrogen-bonding interaction.5c,h,n,u,w,6b,6j,6n How-
ever, these studies have shown a fairly reduced recognition
ability in an aqueous solution because carbohydrates are
hydrophilic and highly solvated in water.5g,j,k,n,u,v Therefore,
Carbohydrate recognition by artificial receptors has been
a challenging subject for synthetic chemists in the field of
supramolecular chemistry.1 This is partly due to the various
cellular recognition processes of oligosaccharides on cell
surfaces2 and partly due to the 3D complexity that is present
even in monosaccharide structures.3 From the X-ray crystal-
lographic structures of sugar-protein complexes, it has been
(1) Reviews: (a) Davis, A. P.; Wareham, R. S. Angew. Chem., Int. Ed.
1999, 38, 2978–2996. (b) Mazik, M. Chem. Soc. Rev. 2009, 38, 935–956.
Book: (c) Aoyama, Y. In Comprehensive Supramolecular Chemistry;
€
Atwood, J. L., Davis, J. E. D., MacNicol, D. D., Vogtle, F., Eds.; Pergamon:
Oxford, 1996; Vol. 2, pp 279-307.
(2) (a) Hokomori, S. Pure Appl. Chem. 1991, 63, 473–482. (b) Kobata, A.
Acc. Chem. Res. 1993, 26, 319–324.
(3) (a) Sharon, N.; Lis, H. Science 1989, 246, 227–234. (b) Sharon, N.; Lis,
H. Sci. Am. 1993, 268 (1), 74–81.
(4) (a) Lemieux, R. U. Chem. Soc. Rev. 1989, 18, 347–374. (b) Quiocho,
F. A. Pure Appl. Chem. 1989, 61, 1293–1306. (c) Weiss, W. I.; Drickamer, K.
Annu. Rev. Biochem. 1996, 65, 441–473.
7588 J. Org. Chem. 2010, 75, 7588–7595
Published on Web 10/22/2010
DOI: 10.1021/jo101384u
r
2010 American Chemical Society

Lee et al.
JOCArticle
an effective method for mimicking the carbohydrate recog-
nition in nature is still needed to develop artificial carbohy-
drate receptors.
Results and Discussion
Design and Synthesis of Porphyrin-Based Receptors. As
envisioned from the X-ray structures of sugar-protein
complexes, an effective approach to carbohydrate recogni-
tion is to surround the polar hydroxyl groups of carbohy-
drates with complementary hydrogen-bonding groups and
introduce aromatic surfaces into the receptor against carbo-
hydrate CH moieties. This can be effectively realized in
porphyrin-based carbohydrate receptors with a preorga-
nized complementary and convergent arrangement of hydro-
gen-bond donor and acceptor functionalities in the RRRR-
positions of the porphyrin, along with a rigid platform
(porphyrin skeleton) providing aromatic surfaces for CH-π
interaction. Eight porphyrin-based receptors were prepared
for carbohydrate recognition, as illustrated in Chart 1. In
a series of urea-appended porphyrins, benzyl groups were
introduced in 1a and 1b, phenyl groups in 1c and 1d, and
methyl benzoate groups in 1e and 1f. By varying the attached
groups in this way, we could investigate how the binding
properties of porphyrins were affected by the attached groups.
Carbamate-appended porphyrin 1g and amide-appended
porphyrin 1h were also prepared to compare the hydrogen-
bonding ability of a urea group with that of carbamate and
amide groups toward carbohydrates.
The receptors were synthesized from meso-RRRR-tetrakis-
(o-aminophenyl)porphyrin, which was prepared by Collman’s
method,9a followed by Lindsey’s atropisomerization method.9b
Urea-appended porphyrins were synthesized from the reaction
between meso-RRRR-tetrakis(o-aminophenyl)porphyrin and
appropriate isocyanates that were commercially available or
generated in situ.10 Commercially available carbobenzoxy
chloride and phenylpropionyl chloride were utilized for the
preparation of 1g and 1h, respectively. Zinc porphyrins were
prepared in quantitative yields from free-base porphyrins and
Zn(OAc)2 in a mixture of chloroform and methanol. All of the
compounds were characterized by 1H NMR, 13C NMR, high-
resolution mass spectrometry, and elemental analysis.
Binding Affinities of the Receptors for Carbohydrates.
UV-vis absorption experiments in chloroform showed that
the Soret band of the receptors underwent a slight shift or was
changed in intensity as the carbohydrates were bound in the
inner space of the receptors. Without adding a large excess of
carbohydrate, clear isosbestic points were observed, which
indicatesthe existenceoftwo states throughthe formationof a
1:1 complex. This was also confirmed by a Job’s plot in the 1H
NMR experiments. The apparent association constants for
the formation of the complexes of receptors (1a-h) and various
carbohydrates (2-6) were estimated by fitting the curve of the
absorbance at λmax as a function of the change in carbo-
hydrate concentration to a 1:1 binding isotherm (Table 1).
As shown in Table 1, the receptors showed high affinities
for carbohydrates, and a general trend was seen in their
affinities for guests in the order 4 ≈ 2 > 3 > 5 > 6. Uridine
and thymidine derivatives 5 and 6 having a ribose unit
Recently, we reported the synthesis of aspartate urea-
appended porphyrins that displayed the highest current
levels of binding for octyl pyranosides in chloroform.7 In
addition, Bonar-Law and co-workers found that analogous
urea-appended porphyrins exhibited excellent affinity for
various carbohydrates.8 In the present study, we have con-
ducted systematic studies to develop a method that can be
used to control the binding affinities by varying the func-
tional groups on the periphery of the porphyrin and eluci-
dated the binding modes using various spectroscopic
techniques and computer-assisted modeling. In this paper,
we describe the syntheses and binding properties of porphyrin-
based receptors 1a-h with regard to monosaccharides
and the binding modes between the selected receptors and
monosaccharides.
(5) For recent studies in aqueous or partial aqueous media, see:
(a) Morales, J. C.; Penades, S. Angew. Chem., Int. Ed. 1998, 37, 654–657.
(b) Sugasaki, A.; Ikeda, M.; Takeuchi, M.; Shinkai, S. Angew. Chem., Int. Ed.
2000, 39, 3839–3842. (c) Hamachi, I.; Nagase, T.; Shinkai, S. J. Am. Chem.
Soc. 2000, 122, 12065–12066. (d) Bielecki, M.; Eggert, H.; Norrild, J. C. J.
Chem. Soc., Perkin Trans. 2 1999, 449–455. (e) Davis, C. J.; Lewis, P. T.;
McCarroll, M. E.; Read, M. W.; Cueto, R.; Strongin, R. M. Org. Lett. 1999,
1, 331–334. (f) Lewis, P. T.; Davis, C. J.; Cabell, L. A.; He, M.; Read, M. W.;
McCarroll, M. E.; Strongin, R. M. Org. Lett. 2000, 2, 589–592. (g) Rusin, O.;
ꢀ
ꢀ
Kral, V. Chem. Commun. 1999, 2367–2368. (h) Kral, V.; Rusin, O.;
Charvatova, J.; Anzenbacher, P., Jr.; Fogl, J. Tetrahedron Lett. 2000, 41,
ꢀ
ꢀ
ꢀ
10147–10151. (i) Rusin, O.; Lang, K.; Kral, V. Chem.;Eur. J. 2002, 8, 655–
663. (j) Sugimoto, N.; Miyoshi, D.; Zou, J. Chem. Commun. 2000, 2295–2296.
(k) Klein, E.; Crump, M. P.; Davis, A. P. Angew. Chem., Int. Ed. 2005, 44,
298–302. (l) Tong, A.-J.; Yamauchi, A.; Hayashita, T.; Zhang, Z.-Y.; Smith,
B. D.; Teramae, N. Anal. Chem. 2001, 73, 1530–1536. (m) Striegler, S.; Dittel,
M. J. Am. Chem. Soc. 2003, 125, 11518–11524. (n) Hou, J.-L.; Shao, X.-B.;
Chen, G.-J.; Zhou, Y.-X.; Jiang, X.-K.; Li, Z.-T. J. Am. Chem. Soc. 2004,
126, 12386–12394. (o) Goto, H.; Furusho, Y.; Yashima, E. J. Am. Chem. Soc.
2007, 129, 9168–9174. (p) Zhao, J.; Davidson, M. G.; Mahon, M. F.; Kociok-
€
Kohn, G.; James, T. D. J. Am. Chem. Soc. 2004, 126, 16179–16186. (q) Zhao,
J.; Fyles, T. M.; James, T. D. Angew. Chem., Int. Ed. 2004, 43, 3461–3464. (r)
Ferrand, Y.; Crump, M. P.; Davis, A. P. Science 2007, 318, 619–622. (s) Cao,
H.; McGill, T.; Heagy, M. D. J. Org. Chem. 2004, 69, 2959–2966. (t) Wang,
Z.; Zhang, D.; Zhu, D. J. Org. Chem. 2005, 70, 5729–5732. (u) Schmuck, C.;
Schwegmann, M. Org. Lett. 2005, 7, 3517–3520. (v) Mazik, M.; Cavga, H. J.
Org. Chem. 2006, 71, 2957–2963. (w) Mazik, M.; Cavga, H. Eur. J. Org.
Chem. 2007, 3633–3638. (x) Pal, A.; Berube, M.; Hall, D. G. Angew. Chem.,
Int. Ed. 2010, 49, 1492–1495. (y) Ferrand, Y.; Klein, E.; Barwell, N. P.;
ꢀ
Crump, M. P.; Jimenez-Barbero, J.; Vicent, C.; Boons, G. J.; Ingale, S.;
Davis, A. P. Angew. Chem., Int. Ed. 2009, 48, 1775–1779. (z) Barwell, N. P.;
Crump, M. P.; Davis, A. P. Angew. Chem., Int. Ed. 2009, 48, 7673–7676.
(6) For recent studies in polar or apolar organic media, see: (a) Davis,
A. P.; Wareham, R. S. Angew. Chem., Int. Ed. 1998, 37, 2270–2273. (b)
Lecollinet, G.; Dominey, A. P.; Velasco, T.; Davis, A. P. Angew. Chem., Int.
Ed. 2002, 41, 4093–4096. (c) Velasco, T.; Lecollinet, G.; Ryan, T.; Davis,
€
A. P. Org. Biomol. Chem. 2004, 2, 645–647. (d) Bahr, A.; Droz, A. S.;
Puntener, M.; Neidlein, U.; Anderson, S.; Seiler, P.; Diederich, F. Helv.
€
Chim. Acta 1998, 81, 1931–1963. (e) Smith, D. K.; Zingg, A.; Diederich, F.
€
Helv. Chim. Acta 1999, 82, 1225–1241. (f) Bahr, A.; Felber, B.; Schneider, K.;
Diederich, F. Helv. Chim. Acta 2000, 83, 1346–1376. (g) Welti, R.; Diederich,
F. Helv. Chim. Acta 2003, 86, 494–503. (h) Inouye, M.; Takahashi, K.;
Nakazumi, H. J. Am. Chem. Soc. 1999, 121, 341–345. (i) Inouye, M.; Chiba,
K. J.; Nakazumi, H. J. Org. Chem. 1999, 64, 8170–8178. (j) Abe, H.; Aoyagi,
Y.; Inouye, M. Org. Lett. 2005, 7, 59–61. (k) Mazik, M.; Bandmann, H.;
Sicking, W. Angew. Chem., Int. Ed. 2000, 39, 551–554. (l) Mazik, M.; Sicking,
W. Chem.;Eur. J. 2001, 7, 664–670. (m) Mazik, M.; Radunz, W.; Sicking,
W. Org. Lett. 2002, 4, 4579–4582. (n) Mazik, M.; Radunz, W.; Boese, R. J.
Org. Chem. 2004, 69, 7448–7462. (o) Mazik, M.; Sicking, W. Tetrahedron
Lett. 2004, 45, 3117–3121. (p) Mazik, M.; Cavga, H.; Jones, P. G. J. Am.
Chem. Soc. 2005, 127, 9045–9052. (q) Ishi-i, T.; Mateos-Timoneda, M. A.;
Timmerman, P.; Crego-Calama, M.; Reinhoudt, D. N.; Shinkai, S. Angew.
Chem., Int. Ed. 2003, 42, 2300–2305. (r) Kim, H. -J.; Kim, Y. -H.; Hong, J. -I.
Tetrahedron Lett. 2001, 42, 5049–5052. (s) Cho, H.-K.; Kim, H.-J.; Lee,
K.-H.; Hong, J.-I. Bull. Korean Chem. Soc. 2004, 25, 1714–1716. (t) Lee,
D.-H.; Kim, H.-J.; Hong, J.-I. Supramol. Chem. 2007, 19, 251–256.
(9) (a) Collman, J. P.; Gagne, R. R.; Reed, C. A.; Halbert, T. R.; Long,
G.; Robinson, W. T. J. Am. Chem. Soc. 1975, 97, 1424–1439. (b) Lindsey,
J. S. J. Org. Chem. 1980, 45, 5215.
(10) Compounds 1c and 1d were previously synthesized and used as anion
receptors by Burns’ group: Jagessar, R. C.; Shang, M.; Scheidt, W. R.;
Burns, D. H. J. Am. Chem. Soc. 1998, 120, 11684–11692. 1c: HRMS calcd for
(7) Kim, Y.-H.; Hong, J.-I. Angew. Chem., Int. Ed. 2002, 41, 2947–2950.
(8) Ladomenou, K.; Bonar-Law, R. P. Chem. Commun. 2002, 2108–2109.
C72H55N12O4 m/z 1151.4469, found 1151.4447. 1d: HRMS calcd for
C72H52N12O4Zn m/z 1212.3526, found 1212.3536.
J. Org. Chem. Vol. 75, No. 22, 2010 7589

Lee et al.
JOCArticle
CHART 1
showed less affinity than the 6-membered carbohydrates
possessing four hydroxyl groups. As expected, the inner
space of the receptors had appropriately arranged hydro-
gen-bonding sites, which were able to recognize the hydroxyl
groups of the carbohydrates through complementary inter-
action. A comparison of the free base porphyrins (1c and 1e)
with the corresponding zinc porphyrins (1d and 1f) indicated
that metal insertion played an ambiguous role in the binding
of guests. This was presumably because two factors compete
with each other in the binding of guests. The first is the fact that
zinc porphyrins show the favorable effect of Lewis acid-
the order 1b > 1d > 1f. Therefore, a possible explanation is
that steric hindrance in the binding of guests increased in
the order 1b < 1d < 1f because of the different functional
groups attached at the urea and prohibited guest sugars
from properly interacting with H-bonding groups within
the binding cavity, although the o-carbomethoxy group of
1f would have provided more H-bonding sites. As for 1f, the
ester groups were considered to have formed either inter-
molecular hydrogen bonds with the hydroxyl groups of the
guest or intramolecular hydrogen bonds with the NH of
urea, competing with the intermolecular hydrogen-bonding
interaction between the urea and guest. A systematic binding
study with various receptors revealed that the ester group did
not have a favorable effect on the hydrogen bonding with
guests. The same trend was observed when compared with 1c
and 1e. We were also able to investigate the hydrogen-bonding
ability of urea, carbamate, and amide toward monosacchar-
ides. Comparing 1b, 1g, and 1h, it can be clearly seen that urea
was the most effective functional group forming hydrogen
bonds with guests, and more insight about the binding was
obtained from the molecular modeling (vide infra).
Lewis base interaction (Zn O coordination). The other
3 3 3
is the fact that metal insertion caused less conformational
freedom in the binding site and had more difficulty in
adjusting to different guest structures, which resulted in
decreased or similar affinities for guests in comparison with
metal free receptors. Although several zinc porphyrins were
reported to have better binding affinities,11c urea-appended
porphyrins did not show significant improvements in their
binding properties.7,8 The binding affinities of 1b, 1d, and
1f toward various carbohydrates were found to decrease in
7590 J. Org. Chem. Vol. 75, No. 22, 2010

Lee et al.
JOCArticle
TABLE 1. Binding Constants (Ka, M-1) and Free Energy Changes (ΔGo, kcal mol-1) from UV-vis Titrations of 1a-h with n-Octyl Pyranosides/
3
Furanosides (2-6) in CHCl3 at 298 Ka
2
3
4
5
6
Ka
-ΔGo
Ka
-ΔGo
Ka
-ΔGo
Ka
-ΔGo
Ka
-ΔGo
1a NDb
NDb
NDb
NDb
NDb
1b 5.7((0.68c) ꢀ 105
1c 9.7((2.0) ꢀ 104
1d 1.7((0.16) ꢀ 104
1e 3.2((0.16) ꢀ 103
1f 4.4((0.43) ꢀ 103
1g 6.2((0.52) ꢀ 102
1h 3.8((0.46) ꢀ 102
7.8
6.8
5.8
4.8
5.0
3.8
3.5
4.8((0.29) ꢀ 104
3.1((0.49) ꢀ 104
6.9((1.2) ꢀ 103
8.9 ((1.6) ꢀ 102
2.0((0.14) ꢀ 103
4.7((0.32) ꢀ 102
1.9((0.19) ꢀ 102
6.4
6.1
5.2
4.0
4.5
3.6
3.1
6.8((0.48) ꢀ 105
2.2((0.25) ꢀ 105
3.8((0.37) ꢀ 104
1.2((0.19) ꢀ 103
3.2((0.38) ꢀ 103
1.2((0.042) ꢀ 103
1.3((0.015) ꢀ 103
8.0
7.3
6.2
4.2
4.8
4.2
4.2
2.5((0.45) ꢀ 104
2.7((0.20) ꢀ 104
1.0((0.046) ꢀ 103
9.8((1.9) ꢀ 102
NDb
6.0
6.0
4.1
4.1
3.2((0.22) ꢀ 103
3.5((0.33) ꢀ 103
1.2((0.21) ꢀ 103
2.2((0.28) ꢀ 102
7.1((0.58) ꢀ 10
5.8((0.68) ꢀ 10
4.9((0.20) ꢀ 10
4.8
4.8
4.2
3.2
2.5
2.4
2.3
3.9((0.16) ꢀ 102
3.5
2.6
8.4((0.30) ꢀ 10
aExperimental conditions: [1b]=2.4 μM for all guests, [guest]=0.00037-0.029 mM for 2, 0.0068-0.38 mM for 3, 0.00046-0.052 mM for 4,
0.011-0.2 mM for 5, and 0.12-2.9 mM for 6; [1c] = 2.1 μM for 2, 3, and 6, 2.0 μM for 4 and 5, [guest] = 0.0003-0.078 mM for 2, 0.0013-0.33 mM for 3,
0.00073-0.19 mM for 4, 0.0034-0.87 mM for 5, and 0.005-1.3 mM for 6; [1d] = 1.7 μM for all guests, [guest] = 0.00077-0.2 mM for 2, 0.0065-1.7 mM
for 3, 0.0018-0.46 mM for 4, 0.031-8.0 mM for 5, and 0.24-6.2 mM for 6; [1e] = 2.0 μM for all guests, [guest] = 0.019-4.9 mM for 2, 0.03-7.8 mM for
3, 0.024-6.1 mM for 4, 0.017-4.3 mM for 5, and 0.038-9.8 mM for 6; [1f ] = 1.75 μM for 2 and 4, 1.74 μM for 3, 1.76 μM for 6, [guest] = 0.019-4.8 mM
for 2, 0.014-3.6 mM for 3, 0.019-4.9 mM for 4, and 0.060-15 mM for 6; [1g] = 1.9 μM for 2 and 4, 2.0 μM for 3, 5, and 6, [guest] = 0.32-5.9 mM for 2,
1.0-17 mM for 3, 0.066-6.1 mM for 4, 0.65-11 mM for 5, and 0.94-17 mM for 6; [1h] = 2.1 μM for 2, 3, and 4, 2.0 μM for 5 and 6, [guest] = 0.29-
12 mM for 2, 1.1-19 mM for 3, 0.21-9.1 mM for 4, 0.88-16 mM for 5, and 1.4-25 mM for 6. Each host concentration was determined from the
standard curve using the Beer-Lambert law. bND=not determined due to little change of the absorbance. cStandard deviations in Ka.
FIGURE 1. 1H NMR spectra of a CDCl3 solution of n-octyl β-D-galactopyranoside (4) in the presence of 1e (0.04 equiv to 0.33 equiv from
bottom to top) in CDCl3 at 298 K.
1H NMR Study of Binding and Computer-Assisted Con-
formational Searches. In order to elucidate the intermolecular
binding mode, we performed 1H NMR titrations of the recep-
tors with various guests in CDCl3. As the equivalent of
carbohydrate increased, the chemical shift of the urea NHs
moved toward the downfield region (see Figure S1 (Supporting
Information) for the 1H NMR titration of 1f with 2 in CDCl3
and the 1H NMR titration of 1a with 3 in CDCl3). This implies
that the urea NH protons of the receptors participate in
hydrogen bonding with the hydroxyl oxygens of carbohydrates.
To investigate the binding mode in more detail, the com-
plexation-induced shifts (CIS) of the guests were estimated
mined by extrapolation to 100% complexation are listed in
Table 2. CIS values are known to provide valuable informa-
tion on the binding mode.11 Two opposing anisotropic
effects should be taken into account to interpret the CIS
values for various carbohydrates upon complexation with
the receptors: upfield shifts due to the ring current of the
porphyrin when carbohydrate protons are located above the
center of the porphyrin plane and potential downfield shifts
of the OH protons if they participate in hydrogen bonds with
urea groups. Table 2 indicates that the resonance of the 1-H
proton of the guest underwent the largest upfield shift for
all of the carbohydrates, which is ascribed only to the effect
of the ring current. Therefore, the 1-H proton is expected
to exist above the porphyrin plane. On the other hand, most
of the OH protons were shifted only slightly upfield, perhaps
because the hydrogen bonding compensated for the ring
current effect. However, the 6-OH proton experienced a
downfield shift upon the addition of 1e, which indicates
that the hydroxyl proton should be involved in hydrogen-
bonding interaction with 1e. Further, the CH protons in the
ring of 4 also moved upfield, indicating that 4 is positioned
1
by the inverse H NMR titrations of various guests with
receptors 1e and 1f in CDCl3 (Figure 1 and Table 2). The CIS
values increased linearly as the fraction of the carbohydrate
bound by the receptor was increased, and the values deter-
(11) (a) Imada, T.; Kijima, H.; Takeuchi, M.; Shinkai, S. Tetrahedron
Lett. 1995, 36, 2093–2096. (b) Imada, T.; Kijima, H.; Takeuchi, M.; Shinkai,
S. Tetrahedron 1996, 52, 2817–2826. (c) Mizutani, T.; Kurahashi, T.;
Murakami, T.; Matsumi, N.; Ogoshi, H. J. Am. Chem. Soc. 1997, 119,
8991–9001.
J. Org. Chem. Vol. 75, No. 22, 2010 7591

Lee et al.
JOCArticle
TABLE 2. 1H NMR Complexation-Induced Shifts (CIS, Δδ/ppm) of Guests upon the Addition of 1e and 1f in CDCl3 at 298 Ka
Δδsat (Δδmax obs) by 1e/1f
2
3
4
5
6
1-H
-3.4 (-0.68)/
-0.98 (-0.45)
NDb/ND
-2.9 (-0.84)/
-1.2 (-0.27)
-3.6 (-0.38)/-1.8 (-0.41)
-2.9 (-0.93)/UP
-3.0 (-0.68)/
-2.7 (-0.60)
2-OH
3-OH
4-OH
-0.43 (-0.13)/
þ0.69 (þ0.16)
-1.8 (-0.53)/
-1.4 (-0.43)
-0.51 (-0.16)/-0.70 (-0.15)
-1.4 (-0.29)/-3.1 (-0.59)
-1.2 (-0.39)/-1.6 (-0.56)
-1.4 (-0.29)/UP
-2.2 (-0.34)/ND
-2.5 (-0.39)/ND
-1.9 (-0.61)/-1.3 (-0.40)
-0.68 (-0.16)/DN
þ0.016 (þ0.0086)/
-1.3 (-0.28)
5-OH
6-OH
þ0.12 (þ0.037)/þ0.17 (þ0.056)
DN/0
ND/þ0.27 (þ0.13)
-1.4 (-0.32)/UPc
þ0.33 (þ0.061)/DNc
aΔδsat was obtained by extrapolating to 100% complexation from the CIS values at 0-30% complexation (Δδmax obs). Signal assignments were made
on the basis of the 1H-1H COSY experiments and reference. bND = not determined due to the broadening or overlap of peaks. cUP = upfield shift,
DN = downfield shift, CIS values could not be calculated because of the broadening or overlap of peaks, but the tendency of the shift could be estimated.
above the porphyrin plane of 1e. This was supported by the
molecular modeling study (see Figure 2). When 4 was bound
by 1e/1f, the patterns of the CIS values were similar to 2,
except that the CIS value of axial 4-OH was less than that of
equatorial 3-OH. These results suggest that 2 and 4 have
similar binding modes, but the axial 4-OH of 4 is expected to
be farther from the porphyrin plane than the equatorial
4-OH of 2. In the case of 3, a different trend was seen, i.e.,
the 4-OH for 1e and 2-OH for 1f were shifted downfield,
while 6-OH was shifted upfield in a manner opposite to 2 and
4. Clearly, the R-anomer 3 is likely to have a different binding
mode from 2 and 4. In addition, the CIS values suggest that
the carbohydrate portion of 5 and 6, rather than the base
portion, is located near and above the porphyrin plane,
hydrogen bonded with urea NHs.
The calculated structures13 revealed the multiple hydrogen
bonds between the hydroxyl groups of the carbohydrate and
the urea NH or ester CdO of the receptor (Figure 2). When 2
was bound by 1e, the OH groups of 2 were near the center of
the porphyrin plane, forming hydrogen bonds with the urea
NH and/or ester CdO of 1e (Figure 2). As mentioned
FIGURE 2. Baseline conformation resulting from a 1000-step
Monte Carlo molecular mechanics study of 1e plus 2 modeled as
the analogous methyl β-D-glucoside. The five intermolecular and
three intramolecular hydrogen bonds are shown as broken lines.
suggests that the number of hydrogen-bonding groups does
not always determine the strength of the binding affinity.
Instead, the proper orientation of the hydrogen-bonding
groups of the host would be more desirable for strong
binding to the guest.
previously, the 6-OH resonance of 2 was shifted downfield
as bound by 1f (Table 2), presumably because the 6-OH was
hydrogen bonded to the ester CdO of 1f at a distance from
the porphyrin plane, which was revealed by the molecular
modeling study. The carbohydrate is located above the
porphyrin plane in the inner space of the receptor, which is
almost in accordance with the CIS values obtained from the
We also performed computer-assisted conformational
searches to investigate how urea groups make up the effective
binding sites for carbohydrates on the porphyrin plane.
Three porphyrins were chosen for the systematic study: 1a
and the free base porphyrins of 1g and 1h.12 As shown in
Figure S4 (Supporting Information), the calculated struc-
tures reveal the multiple hydrogen bonds between the hydroxyl
groups of the carbohydrate and functional groups of the
receptors.13 The urea-appended porphyrin binds a carbohy-
drate with eight intermolecular hydrogen bonds, six of which
are contributed by three bifurcated hydrogen bonds (Figure
S4-aSupporting Information). On the other hand, the complex
between the carbamate-appended porphyrin and carbohy-
drate could be potentially stabilized by five intermolec-
ular hydrogen bonds, and NH OH and OH OCONH
1
inverse H NMR titration experiments. From the binding
model and the unusual chemical shift (about 10 ppm) of one
of the two urea NHs in the complexes in CDCl3, it is likely
that the ester CdO of 1e/1f, which was introduced as a
hydrogen-bonding acceptor to the ortho position of the
urea NH, preferred an intramolecular hydrogen bond with
the urea NH to an intermolecular hydrogen bond with
the hydroxyl groups of the carbohydrate. This preference
reduced the ability of the hydrogen bond donor of the urea
NH. In Table 1, 1e and 1f show weaker affinities to various
carbohydrates than 1c and 1d without the ester group. This
(12) Although zinc porphyrins were used for UV-vis titration, free base
porphyrins were selected for this study because a general optimized force
field for Zn(II) does not exist.
(13) Conformational searches were performed with MacroModel 7.0,
Amber* force field in chloroform solvent, Monte Carlo conformational
searches, 1000 steps: Mohamadi, F.; Richards, N. G. J.; Guida, W. C.;
Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still,
W. C. J. Comput. Chem. 1990, 11, 440.
3 3 3
3 3 3
were shown as possible binding interactions (Figure S4-b,
Supporting Information). Finally, only three hydrogen bonds
were shown in the complex between the amide-appended
porphyrin and carbohydrate (Figure S4-c, Supporting
Information). From the calculated structures, it could be
7592 J. Org. Chem. Vol. 75, No. 22, 2010

Lee et al.
JOCArticle
inferred that the urea-appended porphyrin is able to form the
most stabilized structure with the highest number of hydrogen
bonds contributed by bifurcated hydrogen bonding. In addi-
tion, this study is considered to indirectly support the result
from UV-vis titration that 1b is much superior to 1g and 1h
for carbohydrate recognition.
Induced Circular Dichroism (ICD) Study of Binding. The
binding event was also confirmed by circular dichroism (CD).
When excess n-octyl β-D-galactoside (4) was added to the CD
silent receptor 1f, the complex showed distinct biphasic CD in
the Soret band with a positive Cotton effect at a short wave-
length and negative Cotton effect at a long wavelength (Figure 3).
This result indicates that the relative orientations of the
C-O and O-H moieties of the carbohydrate guest, which
are located near the porphyrin plane, induced distinct CD
spectra, as similar results were reported by Shinkai and
co-workers and Mizutani et al., respectively.7,11
Carbohydrate Recognition in More Polar Solvent. Keeping
in mind that 1b was the strongest binder among the receptors
1
from UV-vis titration, we performed H NMR titration
with 1a and 1b in a polar solvent system. The addition of
n-octyl β-D-glucoside (2) to 1a in CDCl3/CD3OD (v/v, 10:1)
showed distinct shifts of the other signals of 1a despite the
1
disappearance of the urea NHs (Figure 4). The H NMR
spectrum of 1a was independent of the concentration within
the range in which titration was conducted. The association
constants could be calculated by 1:1 nonlinear least-squares
curve fitting, and the apparent binding constants of 1a and
1b for n-octyl β-D-glucoside (2) were 950 and 470 M-1
,
respectively. This implies that the interaction between the
pyranoside and the receptors still remained effective in the
presence of competing methanol, although the binding affi-
nity decreased by a significant amount. The binding constants
for the other guests are reported in Table 3. The selectivity
trend (2>4>3) was the same as previously reported for urea-
appended receptors, but the affinity increased 3-fold and
furanosides 5 and 6 were also bound.
Conclusion
FIGURE 3. Circular dichroism induced in receptor 1f by n-octyl
β-D-galactoside (4) in CHCl3 at 288 K: [1f] = 5.0 ꢀ 10-6 M and [4] =
1.0 ꢀ 10-3 M.
Porphyrin-based receptors containing urea, carbamate, or
amide groups exhibited effective recognition behavior for
1
FIGURE 4. Titration of receptor 1a (1.3 mM) with n-octyl β-D-glucoside (2). (a) H NMR spectra (CDCl3/CD3OD (v/v, 10:1), 298 K) of
receptor 1a after addition of 0.00-3.95 equiv (from bottom to top) of 2. (b) 1H NMR titration curve from the change in the chemical shift of
benzyl CH2.
TABLE 3. Binding Constants (Ka, M-1) and Free Energy Change (ΔGo, kcal/mol) from 1H NMR Titrations of 1a and 1b with Octyl Pyranosides and
Furanosides 2-6 in CD3OD/CDCl3 (1/10) at 298 Ka
2
3
4
5
6
Ka
950(11b)
470(7)
-ΔGo
Ka
-ΔGo
Ka
-ΔGo
Ka
-ΔGo
Ka
-ΔGo
1a
1b
4.06
3.64
330(11)
150(2)
3.43
2.97
510(5)
250(4)
3.69
3.27
77(2)
51(4)
2.57
2.33
68(4)
34(5)
2.50
2.09
a[1a] or [1b] is in the range 2.0-1.3 mM for titration with octyl pyranosides and furanosides. The 1H NMR spectra of 1a and 1b were independent of
the concentrations within this range. Guest concentrations: 0.4-16 mM for 2, 0.8-20 mM for 3 and 4, 1.2-18 mM for 5, and 1.6-16 mM for 6.
bStandard deviations.
J. Org. Chem. Vol. 75, No. 22, 2010 7593

Lee et al.
JOCArticle
monosaccharides through multiple hydrogen bonds and
CH-π interactions, and generally, the binding affinities
of these receptors toward monosaccharides decreased in
the following sequence: n-octyl β-D-galactoside (4) > n-octyl
β-D-glucoside (2) >n-octyl R-D-glucoside (3) >n-octyluridine
(5) > n-octyldeoxythymidine (6). UV-vis titration of receptors
1b, 1g, and 1h demonstrated that urea was an excellent func-
tional group to construct an efficient binding site for carbohy-
drates. In addition, receptors 1a and 1b interacted with n-octyl
β-D-glucoside (2) in hydroxylic organic media, CDCl3/CD3OD
(v/v, 10:1). An effective approach to carbohydrate recognition is
not only to surround the polar hydroxyl groups of the carbohy-
drates with complementary hydrogen-bonding groups of recep-
tors but also to introduce aromatic surfaces into the receptor
against carbohydrate CH moieties. This study showed that urea
NHs can be used as strong H-bonding donors for sugar
hydroxyl oxygens and that the porphyrin plane can be used
for mimicking the CH-π interactions with sugar CHs, which
are found in sugar-binding proteins. In addition, the four
flexible phenyl groups of 1a and 1b could freely wrap the
carbohydrate guest and, therefore, allowed for hydrophobic
interaction with carbohydrate CHs, especially in hydroxylic
cosolvents. We believe that this design principle can be applied
to the development of artificial receptors for natural carbohy-
drates in water.
reduced pressure to give the crude product, which was purified by
column chromatography on silica gel eluting with CH2Cl2/metha-
nol = 20:1 (Rf = 0.35) and CH2Cl2/methanol =40:1 to yield 1a
(97 mg, 80% yield): UV-vis (CHCl3) λmax (log ε) 423 (5.37), 517
(4.14), 553 (3.75), 591 nm (3.72); 1H NMR (300 MHz, DMSO-d6)
δ -2.67 (s, 2H), 3.87 (d, J = 5.4 Hz, 8H), 6.34 (t, J = 5.3 Hz, 4H),
6.86 (m, 8H), 7.02 (m, 12H), 7.23 (s, 4H), 7.35 (t, 4H), 7.67 (d, J =
7.5 Hz, 4H), 7.76 (t, 4H), 8.46 (d, J = 8.5 Hz, 4H), 8.76 (s, 8H); 13C
NMR (75 MHz, DMSO-d6) δ 43.3, 115.9, 122.4, 123.0, 127.1,
127.2, 128.4, 130.1, 132.4, 135.3, 138.8, 139.6, 155.5; HRMS calcd
for C76H63N12O4 m/z 1207.5095, found 1207.5092. Anal. Calcd for
C76H62N12O4: C, 75.59; H, 5.18; N, 13.93. Found: C, 75.50; H, 5.31;
N, 13.91.
Compound 1b. To a solution of free-base porphyrin 1a in
CHCl3 was added a solution of zinc acetate dihydrate dissolved
in methanol. The resulting solution was stirred under N2 for 24
h. Water was added, and the mixture was stirred for 30 min. The
organic layer was separated, dried over Na2SO4, concentrated,
and purified by flash chromatography (CH2Cl2/methanol =
40:1) to give rise to 1b (quantitative yield): UV-vis (CHCl3)
λmax (log ε) 428 (5.41), 553.5 (4.19); 1H NMR (300 MHz,
DMSO-d6) δ 3.79 (d, J=5.5 Hz, 8H), 6.15 (t, J=5.5 Hz, 4H,),
6.76 (s, 4H), 6.79 (m, 8H), 6.97 (m, 12H), 7.39 (t, J = 7.5 Hz,
4H), 7.76 (t, J=8.2 Hz, 4H), 7.82 (dd, J=7.5, 1.0 Hz, 4H), 8.44
(d, J=8.2 Hz, 4H), 8.74 (s, 8H); 13C NMR (75 MHz, DMSO-d6)
δ 42.4, 115.5, 120.8, 120.9, 126.4, 126.7, 127.8, 128.6, 131.6,
131.8, 135.1, 139.4, 139.4, 149.6, 154.9; HRMS calcd for
C76H61N12O4Zn m/z 1269.4230, found 1269.4248. Anal. Calcd
for C80H60N12O12Zn 2H2O: C, 64.85; H, 4.36; N, 11.35. Found:
C, 64.62; H, 4.28; N, 11.50.
3
Experimental Section
Compound 1e. Phosgene in toluene (20% solution, >10 equiv,
∼5 mL) was added to 10 mL of a dry tetrahydrofuran solution of
2-aminobenzoic acid methyl ester (150 mg, 1 mmol) and triethyla-
mine (2 mL) under N2. The mixture was stirred for 1 h at room
temperature and then was brought to dryness on a rotary evap-
orator. The resulting solid was dissolved in 10 mL of dry dichlor-
omethane. The reaction mixture was treated with R,R,R,R-
5,10,15,20-tetrakis(2-aminophenyl)-21H,23H-porphine (67 mg,
0.1 mmol) in 10 mL of dry tetrahydrofuran. After being stirred at
room temperature for 24 h, the reaction mixture was washed with
1 N HCl, 5% aq NaHCO3, and then brine. The extracted organic
phase was dried over Na2SO4 and concentrated to give the crude
product, which was purified by column chromatography on silica
gel, eluting with CH2Cl2/methanol = 20:1 (Rf = 0.4) and CH2Cl2/
methanol = 40:1 to furnish 1e (55 mg, 40% yield): UV-vis
(CHCl3) λmax (log ε) 423 (5.53), 517 (4.33), 550.5 (3.87), 590 nm
(3.78); 1H NMR (300 MHz, CDCl3) δ -2.53 (s, 2H), 2.86 (s, 12H),
6.40 (t, J = 7.8 Hz, 4H), 6.43 (s, 4H), 6.89 (t, J = 7.7 Hz, 4H), 7.16
(d, J = 7.8 Hz, 4H), 7.47 (t, J = 7.2 Hz, 4H), 7.66 (d, J = 8.6 Hz,
4H), 7.75 (t, J = 7.2 Hz, 4H), 7.99 (d, J = 7.2 Hz, 4H), 8.16 (d, J =
8.1 Hz, 4H), 8.82 (s, 8H), 9.22 (s, 4H); 13C NMR (75 MHz, CDCl3)
δ 51.4, 114.4, 115.5, 119.7, 120.8, 122.6, 122.9, 129.8, 130.1, 131.4,
132.5, 133.6, 135.0, 138.8, 141.3, 152.5, 167.7; HRMS calcd for
C80H63N12O12 m/z 1383.4688, found 1383.4690. Anal. Calcd for
C80H62N12O12: C, 69.44; H, 4.52; N, 12.15. Found: C, 69.31; H,
4.57; N, 12.18.
UV-vis Titrations. Binding constants were determined through
UV-vis titrations. Chloroform was used immediately after
deacidification. No atropisomerization was observed in the solid
state and during UV-vis titration. The prepared receptor solu-
tion was used as a solvent in the preparation of a 1-20 mM stock
solution of various carbohydrates. Therefore, in spite of the
volume change due to successive addition of the guest stock
solution, the concentration of the receptor was maintained
constant throughout the duration of UV-vis titration. The
details concerning the determination procedure of the binding
constant are as follows. The above-mentioned stock solution of
various carbohydrates was successively added to the host solu-
tion in a 1-cm quartz UV-vis cell. The changes in absorbance in
the Soret band at λmax were monitored for different concentra-
tions of carbohydrate. Assuming a 1:1 complexation, the binding
constants were evaluated by least-squares parameter estimation.
The calculated curves were well fitted to absorbance changes.
1H NMR Titrations. CDCl3 was completely deacidified by
keeping it in anhydrous potassium carbonate for 1 day. Titration
of 0.5 mL of a known concentration of CDCl3/CD3OD (10:1, v/v)
solution of the receptor (1a and 1b) in an NMR tube by stepwise
addition of an appropriate volume of carbohydrate solution was
followed by 1H NMR measurements; the spectrum was recorded
after each addition. The chemical shifts of protons were monitored.
The complex-induced chemical shifts were plotted against the molar
percentage of added carbohydrate to obtain titration curves. If the
calculated curve assuming 1:1 complexation was fitted to the
experimental titration curve, the binding constants could be eval-
uated by least-squares parameter estimation. In reverse 1H NMR
titration (Figure 1 and Table 2), the signals of carbohydrate protons
(2 mM solution) were monitored through stepwise addition of an
appropriate volume of the receptor (1e and 1f) to CDCl3.
Compound 1f. 1f was obtained quantitatively from 1e through
the same procedure used for the preparation of 1b: UV-vis
1
(CHCl3) λmax (log ε) 428 (5.68), 557 (4.38), 595 nm (3.71); H
NMR (300 MHz, CDCl3) δ 3.12 (s, 12H), 6.26 (s, 4H), 6.60 (t,
J=7.6 Hz, 4H), 7.09 (t, J=7.4 Hz, 4H), 7.32 (d, J = 7.4 Hz,
4H), 7.58 (t, J=7.2 Hz, 4H), 7.73 (d, J=8.5 Hz, 4H), 7.83 (t,
J=7.6 Hz, 4H), 8.19 (d, J = 6.6 Hz, 4H), 8.26 (d, J = 8.1 Hz,
4H), 8.81 (s, 8H), 9.46 (s, 4H); 13C NMR (75 MHz, CDCl3) δ
51.7, 114.0, 115.2, 119.3, 120.9, 122.1, 123.2, 129.5, 130.2, 131.8,
133.9, 134.2, 134.6, 139.0, 141.5, 150.2, 152.7, 168.3; HRMS
calcd for C80H60N12O12Zn m/z 1444.3745, found 1444.3765.
Compound 1a. To R,R,R,R-5,10,15,20-tetrakis(2-aminophenyl)-
21H,23H-porphine (67 mg, 0.1 mmol) dissolved in 30 mL of
distilled dichloromethane was slowly added benzyl isocyanate
(250 μL, 2 mmol). After the reaction mixture was stirred at room
temperature overnight under N2, the solvent was evaporated under
7594 J. Org. Chem. Vol. 75, No. 22, 2010

Lee et al.
JOCArticle
Anal. Calcd for C80H60N12O12Zn 2H2O: C, 64.85; H, 4.36; N,
11.35. Found: C, 64.62; H, 4.28; N, 11.50.
water. After the organic phase was dried under reduced vacuum, the
crude product was dissolved in 15 mL of chloroform, followed by
the addition of a solution of zinc acetate dihydrate dissolved in
methanol. The resulting solution was stirred under nitrogen for 24 h.
Ethylenediamine (5 mL) was added to the solution to deactivate the
remaining acid chloride, and then the solution was stirred for 30
min. Water was added to the reaction mixture, which was then
stirred for 30 min and an organic layer was separated. The organic
layer was dried over Na2SO4, evaporated, and purified by flash
chromatography (CH2Cl2/methanol = 100:1) to yield 1h (81 mg,
64% overall): UV-vis (CHCl3) λmax (log ε) 421 (5.68), 547 nm
(4.36); 1H NMR (300 MHz, DMSO-d6) δ 1.51 (br, 8H), 2.19 (br,
8H), 6.53 (m, 8H), 6.80 (m, 12H), 7.55 (t, J = 7.3 Hz, 4H), 7.81 (t,
J = 7.7 Hz, 4H), 7.93 (d, J = 7.3 Hz, 4H), 8.17 (s, 4H), 8.24 (d, J =
7.2 Hz, 4H), 8.66 (s, 8H); 13C NMR (75 MHz, DMSO-d6) δ 30.4,
36.9, 115.9, 123.6, 124.0, 126.0, 128.1, 128.3, 129.1, 131.9, 135.3,
136.1, 138.5, 140.8, 149.9, 170.6; HRMS calcd for C80H64N8O4Zn
3
Compound 1g. Carbobenzoxy chloride (30-35% in toluene,
2 mL) and 2,6-lutidine (1 mL) were added to a tetrahydrofuran
solution of R,R,R,R-5,10,15,20-tetrakis(2-aminophenyl)-21H,23H-
porphine (67 mg, 0.1 mmol). The mixture was stirred at room
temperature for 2 h, and then 5 mL of 0.5 N aqueous NaOH
solution was added. After 10 min, the reaction mixture was brought
to dryness on a rotary evaporator. The resulting mixture was
dissolved in CHCl3 and then washed with water. After the organic
phase was dried under reduced vacuum, the crude product was
dissolved in 15 mL of chloroform, followed by the addition of a
solution of zinc acetate dihydrate dissolved in methanol. The
resulting solution was stirred under nitrogen for 24 h. Water was
added to the reaction mixture, the mixture was stirred for 30 min,
and an organic layer was separated. The organic layer was dried
over Na2SO4, evaporated, and purified by flash chromatography
(hexane/ethyl acetate = 200:1) to yield 1g (95 mg, 75% overall):
UV-vis (CHCl3) λmax (log ε) 420 (5.70), 546 nm (4.36); 1H NMR
(300 MHz, CDCl3) δ 4.60 (s, 8H), 6.35 (s, 4H), 6.80 (m, 8H), 7.00
(m, 12H), 7.46 (t, J = 7.4 Hz, 4H), 7.84 (t, J = 8.0 Hz, 4H), 7.93 (d,
J = 6.9 Hz, 4H), 8.48 (d, J = 6.9 Hz, 4H), 8.84 (s, 8H); 13C NMR
(75 MHz, CDCl3) δ 66.6, 115.3, 119.5, 122.2, 127.5, 127.7, 127.9,
128.1, 129.7, 132.4, 134.8, 135.7, 138.3, 150.5, 153.2; HRMS calcd
for C76H56N8O8Zn m/z 1272.3513, found 1272.3508. Anal. Calcd
m/z 1264.4342, found 1264.4366. Anal. Calcd for C80H64N8O4Zn
3
H2O: C, 74.79; H, 5.18; N, 8.72. Found: C, 74.52; H, 5.10; N, 8.37.
Spectroscopic Measurements. The CD scanning conditions
were as follows: scanning rate =100 nm per min, bandwidth =
0.5 nm, response time = 2 s, accumulations = one1 time.
Acknowledgment. This work was supported by an NRF grant
funded by the MEST (Grant No. 2009-0080734). J.-D.L and
Y.-H.K. thank the Ministry of Education for the BK 21 fellowship.
for C76H56N8O8Zn H2O: C, 70.61; H, 4.52; N, 8.67. Found: C,
70.29; H, 4.39; N, 8.58.
3
Compound 1h. 3-Phenylpropionyl chloride (2 mL) and 2,
6-lutidine (1 mL) were added to a tetrahydrofuran solution of
R,R,R,R-5,10,15,20-tetrakis(2-aminophenyl)-21H,23H-porphine
(67 mg, 0.1 mmol). The mixture was stirred for 2 h at room
temperature and then brought to dryness on a rotary evaporator.
The resulting mixture was dissolved in CHCl3 andthenwashed with
1
Supporting Information Available: Job’s plot, selected H
NMR titration spectra, UV-vis titration curves, and 1H NMR
spectra of 1a in CDCl3/CD3OD (v/v, 10:1) upon dilution. This
material is available free of charge via the Internet at http://
pubs.acs.org.
J. Org. Chem. Vol. 75, No. 22, 2010 7595