Published on Web 03/15/2006
`
`An Improved Class of Sugar-Binding Boronic Acids, Soluble and Capable of
`Complexing Glycosides in Neutral Water
`Meenakshi Dowlut and Dennis G. Hall*
`Department of Chemistry, UniVersity of Alberta, Edmonton, AB, T6G 2G2 Canada
`
`Received November 16, 2005; E-mail: dennis.hall@ualberta.ca
`
`Figure1. Ortho-substituted arylboronic acids tested for glycoside binding.
`
`The selective recognition of carbohydrates under physiological
`conditions stands as one of the biggest challenges of chemical
`biology.1 For example, the development of a selective and nonin-
`vasive molecular sensor for monitoring blood glucose has long been
`sought as a key component of insulin-releasing implants for diabetes
`patients.2 While the use of boronic acids is regarded as one of the
`most promising approaches for the recognition of carbohydrates in
`water,3 it is not without severe limitations. No boronic acid unit
`has yet been demonstrated to bind to nonreducing sugars and
`glycosides,4 which account for the large majority of biologically
`important oligosaccharides found in the form of cell-surface
`glycoconjugates. Moreover, the “Wulff-type” ortho-dialkylami-
`nomethyl arylboronic acids,5 currently the established standard for
`the recognition of simple reducing sugars, tend to have limited
`solubility in aqueous solutions.6 Herein, we report that ortho-
`hydroxyalkyl arylboronic acids bind to monosaccharides, such as
`glucose and fructose, with higher affinity than do Wulff-type
`boronic acids in neutral water, and show a better solubility profile.
`Moreover, exciting preliminary evidence reveals the unprecedented
`capability of this new class of boronic acids to complex nonreducing
`glycopyranosides.
`Elegant studies by Norrild and co-workers7 have confirmed that
`glucose binds to boronic acids in water in its weakly populated
`furanose form.8 The work of our group9 and others10 has emphasized
`the existence of similar requirements for disaccharides. This
`behavior is generally ascribed to geometrical preferences in boronate
`formation. Rigid and coplanar vicinal diols, such as the syn 1,2-
`pair of furanoses, are strongly preferred to minimize angle strain
`in the resulting boronic ester. The formation of a coplanar boronate
`with the non-coplanar diols of a glycopyranoside would induce a
`highly unfavorable conformational change to the puckered sugar
`ring.11 It is clear that, if the use of oligoboronic acid receptors is to
`mature into a general approach for oligosaccharide recognition, new
`boronic acids with pyranoside-binding capability are required. Our
`initial approach envisioned the possible formation of a hemi-
`arylboronic ester with cooperative covalent or noncovalent interac-
`tions from an ortho-substituent (Figure 1). To this end, we screened
`a panel of ortho-substituted arylboronic acids using Wang’s
`qualitative colorimetric assay based on the competitive release of
`alizarin red S (ARS).12 From more than a dozen candidates, 1a-
`1m, ortho-hydroxymethyl phenylboronic acid (1m)13 stood out by
`showing strong binding to both glucose and fructose. To our greater
`satisfaction, weak but encouraging binding of the glycosides methyl
`R-D-glucopyranoside and trehalose (a 1,1¢ -glucopyranose dimer)
`was observed. All of the other boronic acids, including Wulff-type
`ortho-dimethylaminomethyl phenylboronic acid (1n) and the highly
`acidic 2,14 failed to provide any visible darkening of the solution,
`even with a large excess of glycosides.15,16 The complexed and
`uncomplexed forms of 1m with methyl glucopyranoside were not
`distinguishable by 1H NMR. Peak broadening of the arylboronate
`protons, however, was observed only in the case of 1m, which
`4226 9 J. AM. CHEM. SOC. 2006, 128, 4226-4227
`
`Table1. Ka Measurements by 1H NMR at Neutral pH16,18
`Ka (M-1)b
`glucose
`fructose
`
`conditionsa
`
`entry
`
`boronic acid
`PhB(OH)2
`1m
`1n
`1n
`1o
`
`1
`2
`3
`4
`5
`
`D2O
`D2O
`33% CD3OD/D2O
`80% CD3OD/D2O
`80% CD3OD/H2O
`a In pH 7.4 sodium phosphate monobasic buffer. b Average of at least
`two measurements. c Not measured. Likely below 5 M-1 according to ARS
`qualitative assay.
`
`0
`17
`c
`c
`c
`
`79
`606
`115
`308
`1960
`
`further supports binding of the model glycosides only with this
`boronic acid.16 The association constant of 1m to methyl R-D-
`glucopyranoside in water (pH 7.4) was best measured by the ARS
`method.12 In agreement with the qualitative assay, complex forma-
`tion with methyl R-D-glucopyranoside was found to be slightly
`weaker than with glucose (Ka ) 22 vs 36 M-1). These affinities
`are comparable or superior to recently reported macrocyclic
`receptors,17 however, with a much simpler compound. A comparison
`of the binding of phenylboronic acid, 1m, and 1n to glucose and
`fructose was done by NMR titrations in neutral aqueous conditions
`(Table 1).18 Although Ka measurements are sensitive to the method
`and conditions employed,19 these values are useful for comparative
`purposes. The data of Table 1 confirm that boronic acid 1m is
`superior to the Wulff-type analogue 1n. Moreover, in contrast to
`1n, 1m does not need an organic cosolvent for solubilization.
`
`10.1021/ja057798c CCC: $33.50 © 2006 American Chemical Society
`
`Downloaded via Patricia Hennessy on August 20, 2018 at 15:13:45 (UTC).
`
`See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
`
`

`

`Table2. Ka Measurementsa of Glycopyranosides Using the ARS
`UV Assay at Neutral pH12,16
`
`pyranoside
`methyl R-D-glucopyranoside
`methyl R-D-galactopyranoside
`methyl (cid:226)-D-glucopyranoside
`methyl (cid:226)-D-galactopyranoside
`
`Ka (M-1)b
`22
`34
`22
`34
`
`a Conditions: pH 7.4 sodium phosphate monobasic buffer (in 100%
`H2O). b Average of at least two measurements.
`
`These results led us to question the contribution of the covalent
`boronate interaction in the binding of Wulff-type boronic acids to
`monosaccharides. The influence of hydrophobic interactions in the
`recognition of carbohydrates by natural (i.e., lectins) and unnatural
`receptors is well-known.20 Here, compared to 1n, we found that
`the hydrophobic nature of the sensing unit of 1o,3a specifically,
`the anthracene group, significantly increases the Ka values (compare
`entries 4 and 5). Thus, when measured in a minimum percentage
`of methanol, the simple Wulff-type boronic acid 1n, which is devoid
`of a hydrophobic unit, is clearly inferior to 1m. This suggests for
`the first time that the saccharide-binding affinity of previously
`reported Wulff-type boronic acid receptors is significantly amplified
`by hydrophobic interactions.
`We looked at the binding requirements of 1m by comparing
`methyl R-D-glucopyranoside with its 6-deoxy derivative.16 The latter
`was found not to bind significantly to 1m, which suggests a key
`in the complexation of 1m to the
`role for the 4,6-diol unit
`glucopyranoside. Although more work is warranted to examine the
`binding selectivity, the ARS assay revealed that complexation of
`1m to methyl R-D-galactopyranoside is even more favorable, and
`the (cid:226)-glycosides display a similar selectivity (Table 2).16
`The precise binding mode of boronic acid 1m to hexopyranosides
`is currently unclear. Certainly, the pKa of 1m is quite low (ca. 7.2)
`as measured by potentiometric and 11B NMR methods,16 but as
`shown with the inefficiency of 1n and 2, acidity alone can hardly
`explain the surprising ability of 1m to bind hexopyranosides. The
`ether derivative 1l is ineffective, so the presence of a coordinating
`oxygen is not sufficient. Boronic acid 1m is believed to exist in its
`cyclic, dehydrated boronophthalide form.13 It is possible that the
`unusually small C-B-O dihedral angle of 1m, as observed by
`X-ray crystallography,21 opens up the cone angle in the resulting
`tetrahedral diol-boronate complex. This distorted geometry may
`better accommodate the 4,6-diol of hexopyranosides compared to
`the usual boronic acids, leading to the proposed complexation model
`of eq 1. This complex may also benefit entropically from the internal
`alkoxy arm of 1m (as compared to a hydroxy ligand with the usual
`boronic acids).
`
`Dimerization has been shown to be a very effective strategy in
`the development of boronic acid based receptors and sensors,
`leading to large increases in binding affinity and often drastic
`changes in selectivity profiles.3 In the current case, the development
`
`C O M M U N I C A T I O N S
`
`of synthetic routes to oligomeric derivatives of 1m would allow
`multivalency effects to be exploited in the recognition of cell-surface
`glycoconjugates.
`In conclusion, a new class of carbohydrate-binding boronic acids
`was characterized. ortho-Hydroxymethyl phenylboronic acid was
`shown to be superior to the well-established dialkylamino (Wulff-
`type) analogues. The most significant finding is the ability of ortho-
`hydroxyalkyl arylboronic acids to complex model glycosides under
`physiologically relevant conditions. This unique boronic acid unit
`appears to complex hexopyranosides mainly using their 4,6-diol,
`and we note that a majority of cell-surface glycoconjugates present
`free 4,6-diols. Conjugatable forms of these boronic acids could be
`used in the design of oligomeric receptors and sensors to exploit
`multivalency effects. Such receptors could dramatically expand the
`potential of boronic acids toward the selective recognition of cell-
`surface glycoconjugates.
`Acknowledgment. This work was funded by the Natural
`Sciences and Engineering Research Council (NSERC) of Canada,
`the Alberta Heritage Foundation for Medical Research (AHFMR),
`and the University of Alberta. We thank Heidi Chau for assistance
`in the qualitative ARS assay.
`Supporting Information Available: Full experimental details. This
`material is available free of charge via the Internet at http://pubs.acs.org.
`
`References
`(1) Davis, A. P.; Wareham, R. S. Angew. Chem., Int. Ed. 1999, 38, 2978-
`2996.
`(2) Rohrscheib, M.; Robinson, R.; Eaton, R. P. Diabetes, Obes. Metab. 2003,
`5, 280-284.
`(3) (a) James, T. D.; Sandanayake, K. R. A. S.; Shinkai, S. Angew. Chem.,
`Int. Ed. Engl. 1996, 35, 1910-1922. (b) James, T. D.; Shinkai, S. Top.
`Curr. Chem. 2002, 218, 159-200. (c) Wang, W.; Gao, X.; Wang, B. Curr.
`Org. Chem. 2002, 6, 1285-1317. (d) Wiskur, S. L.; Ait-Haddou, H.;
`Lavigne, J. J.; Anslyn, E. V. Acc. Chem. Res. 2001, 34, 963-972.
`(4) A single exception would be boronate formation with the sialyl unit of
`sialosides such as Slex: Wang, B.; Yang, W.; Fan, H.; Gao, X.; Gao, S.;
`Karnati, V. V. R.; Ni, W.; Hooks, W. B.; Carson, J.; Weston, B.; Wang,
`B. Chem. Biol. 2004, 11, 439-448.
`(5) Wulff, G. Pure Appl. Chem. 1982, 54, 2093-2102.
`(6) Hoeg-Jensen, T. QSAR Comb. Sci. 2004, 23, 344-351.
`(7) Bielecki, M.; Eggert, H.; Norrild, J. C. J. Chem. Soc., Perkin Trans. 2
`1999, 449-455.
`(8) To the best of our knowledge, only in a single case was a diboronic acid
`receptor successfully designed to bind glucopyranose: Yang, W.; He, H.;
`Drueckhammer, D. G. Angew. Chem., Int. Ed. 2001, 40, 1714-1718.
`(9) Stones, D.; Manku, S.; Lu, X.; Hall, D. G. Chem.sEur. J. 2004, 10, 92-
`100.
`(10) (a) Nagai, Y.; Kobayashi, K.; Toi, H.; Aoyama, Y. Bull. Chem. Soc. Jpn.
`1993, 66, 2965-2971. (b) Arimori, S.; Phillips, M. D.; James, T. D.
`Tetrahedron Lett. 2004, 45, 1539-1542.
`(11) (a) Kuivila, H. G.; Keough, A. H.; Soboczenski, E. J. J. Org. Chem. 1954,
`19, 780-783. (b) Sugihara, J. M.; Bowman, C. M. J. Am. Chem. Soc.
`1958, 80, 2443-2446.
`(12) Springsteen, G.; Wang, B. Chem. Commun. 2001, 1608-1609.
`(13) Lennarz, W. J.; Snyder, H. R. J. Am. Chem. Soc. 1960, 82, 2172-2175.
`(14) Mulla, H. R.; Agard, N. J.; Basu, A. Bioorg. Med. Chem. Lett. 2004, 14,
`25-27. It should be noted that 2 has low solubility in aqueous solvents.
`(15) To validate this result, we cautiously ensured by NMR that the samples
`of glucopyranosides contained no free glucose.
`(16) See Supporting Information for details.
`(17) Klein, E.; Crump, M. P.; Davis, A. P. Angew. Chem., Int. Ed. 2005, 44,
`298-302.
`(18) The complexed (1:1) and uncomplexed forms of fructose are distinguish-
`able by 1H NMR. The isotopic effect of D2O on association constants
`was found to be small.
`(19) Springsteen, G.; Wang, B. Tetrahedron 2002, 58, 5291-5300.
`(20) For example, see: Fernandez-Alonso, M. d. C.; Can˜ada, F. J.; Jime´nez-
`Barbero, J.; Cuevas, G. J. Am. Chem. Soc. 2005, 127, 7379-7386.
`(21) Zhdankin, V. V.; Persichini, P. J., III; Zhang, L.; Fix, S.; Kiprof, P.
`Tetrahedron Lett. 1999, 40, 6705-6708.
`JA057798C
`
`J. AM. CHEM. SOC. 9 VOL. 128, NO. 13, 2006 4227
`
`