The Design of Boronic Acid Spectroscopic Reporter Compounds by Taking Advantage of the pKa-Lowering Effect of Diol Binding: Nitrophenol-Based Color Reporters for Diols

Article abstract of DOI:10.1021/jo0350357

The complex that forms between a boronic acid and a diol is often much more acidic than the starting boronic acid. In conditions where the solution pH is between the two pK_a values, the boron atom will convert from a neutral trigonal form to an anionic tetrahedral form upon complexation. Such a change is likely to dramatically alter the electron density of neighboring groups. Utilizing this effect, we have designed and synthesized two nitrophenol-based boronic acid reporter compounds that change ionization states and therefore spectroscopic properties upon diol binding. Both compounds show significant UV changes upon addition of saccharides. For example, a blue shift of the absorption max from 373 to 332 nm was observed with the addition of D-fructose to 2-hydroxy-5-nitrophenylboronic acid at neutral pH. Such a reporter compound can be used as a recognition and signaling unit for the construction of polyboronic acid sensors for the selective and specific recognitions of saccharides of biological significance.

Full text of DOI:10.1021/jo0350357

Th e Design of Bor on ic Acid Sp ectr oscop ic Rep or ter Com p ou n d s
by Ta k in g Ad va n ta ge of th e p Ka -Low er in g Effect of Diol Bin d in g:
Nitr op h en ol-Ba sed Color Rep or ter s for Diols
Weijuan Ni,† Hao Fang,‡ Greg Springsteen,† and Binghe Wang*,‡
Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695, and
Department of Chemistry, Georgia State University, 33 Gilmer Street S.E., Atlanta, Georgia 30303
wang@gsu.edu
Received J uly 17, 2003
The complex that forms between a boronic acid and a diol is often much more acidic than the starting
boronic acid. In conditions where the solution pH is between the two pKa values, the boron atom
will convert from a neutral trigonal form to an anionic tetrahedral form upon complexation. Such
a change is likely to dramatically alter the electron density of neighboring groups. Utilizing this
effect, we have designed and synthesized two nitrophenol-based boronic acid reporter compounds
that change ionization states and therefore spectroscopic properties upon diol binding. Both
compounds show significant UV changes upon addition of saccharides. For example, a blue shift of
the absorption max from 373 to 332 nm was observed with the addition of D-fructose to 2-hydroxy-
5-nitrophenylboronic acid at neutral pH. Such a reporter compound can be used as a recognition
and signaling unit for the construction of polyboronic acid sensors for the selective and specific
recognitions of saccharides of biological significance.
In tr od u ction
chromatographic stationary materials.25-30 Recently, our
group for the first time showed that boronic acid-based
fluorescent sensors for cell surface carbohydrate sialyl
Lewis X (sLex) can be used for the fluorescent labeling
of HEPG2 cells that are engineered to overexpress sLex.15
Parallel with our sensor development effort, we have also
examined systematically the binding between boronic
acid with various diols.31,32 Such studies allowed us to
Due to the unique strong interactions between boronic
acid and diols through reversible ester formation, there
has been a great deal of interest in using boronic acid as
the recognition moiety for the development of fluorescent
and color sensors,1-16 carbohydrate transporters,17-24 and
* Address correspondence to this author. Phone: (404) 651-0289.
† North Carolina State University.
‡ Georgia State University.
(16) Yoon, J .; Czarnik, A. W. J . Am. Chem. Soc. 1992, 114, 5874-
5875.
(17) Paugam, M.-F.; Bien, J . T.; Smith, B. D.; Chrisstoffels, L. A.
J .; de J ong, F.; Reinhoudt, D. N. J . Am. Chem. Soc. 1996, 118, 9820-
9825.
(18) Westmark, P. R.; Gardiner, S. J .; Smith, B. J . Am. Chem. Soc.
1996, 118, 11093-11100.s
(19) Riggs, J . A.; Hossler, K. A.; Smith, B. D.; Karpa, M. J .; Griffin,
G.; Duggan, P. J . Tetrahedron Lett. 1996, 37, 6303-6306.
(20) Draffin, S. P.; Duggan, P. J .; Duggan, S. A. M. Org. Lett. 2001,
3, 917-920.
(21) Gardiner, S. J .; Smith, B. D.; Duggan, P. J .; Karpa, M. J .;
Griffin, G. J . Tetrahedron 1999, 55, 2857-2864.
(22) Paugam, M. F.; Bien, J . T.; Smith, B. D.; Chrisstoffels, L. A. J .;
deJ ong, F.; Reinhoudt, D. N. J . Am. Chem. Soc. 1996, 118, 9820-9825.
(23) Smith, B. D.; Gardiner, S. J .; Munro, T. A.; Paugam, M. F.;
Riggs, J . A. J . Inclusion Phenom. Mol. Recognit. Chem. 1998, 32, 121-
131.
(1) J ames, T. D.; Sandanayake, K. R. A. S.; Shinkai, S. Nature
(London) 1995, 374, 345-347.
(2) Eggert, H.; Frederiksen, J .; Morin, C.; Norrild, J . C. J . Org.
Chem. 1999, 64, 3846-3852.
(3) Norrild, J . C.; Eggert, H. J . Am. Chem. Soc. 1995, 117, 1479-
1484.
(4) Sandanayake, K. R. A. S.; Nakashima, K.; Shinkai, S. J . Chem.
Soc., Chem. Commun. 1994, 1621-1622.
(5) Norrild, J . C.; Eggert, H. J . Chem. Soc., Perkin Trans. 2 1996,
2583-2588.
(6) Shinkai, S.; Takeuchi, M. Trend Anal. Chem. 1996, 15, 188-
193.
(7) Wiskur, S. L.; Lavigne, J . L.; Ait-Haddou, H.; Lynch, V.; Chiu,
Y. H.; Canary, J . W.; Anslyn, E. V. Org. Lett. 2001, 3, 1311-1314.
(8) Yang, W.; He, H.; Drueckhammer, D. G. Angew. Chem., Int. Ed.
2001, 40, 1714-1718.
(9) J ames, T. D.; Sandanayake, K. R. A. S.; Shinkai, S. Chem.
Commun. 1994, 477-478.
(24) Westmark, P. R.; Gardiner, S. J .; Smith, B. D. J . Am. Chem.
Soc. 1996, 118, 11093-11100.
(10) (a) Wang, W.; Gao, X.; Wang, B. Curr. Org. Chem. 2002, 6 (14),
1285-1317. (b) Yang, W.; Gao, S.; Wang, B. Med. Res. Rev. 2003, 23,
346-368.
(25) Wulff, G.; Vesper, W. J . Chromatogr. 1978, 167, 171-186.
(26) Wulff, G.; Poll, H.; Minarik, M. J . Liq. Chromatogr. 1986, 9,
385-405.
(11) Lavigne, J . J .; Anslyn, E. V. Angew. Chem., Int. Ed. 1999, 38,
3666-3669.
(27) Liu, X. C.; Hubbard, J . L.; Scouten, W. H. J . Organomet. Chem.
1995, 493, 91-94.
(12) Cabell, L. A.; Monahan, M.-K.; Anslyn, E. V. Tetrahedron Lett.
1999, 40, 7753-7756.
(28) Psotova, J .; J aniczek, O. Chem. Listy 1995, 89, 641-648.
(29) Singhal, R. P.; Ramamurthy, B.; Govindraj, N.; Sarwar, Y. J .
Chromatogr. 1991, 543, 17-38.
(13) Gao, S.; Wang, W.; Wang, B. Bioorg. Chem. 2001, 29, 308-
320.
(14) Wang, W.; Gao, S.; Wang, B. Org. Lett. 1999, 1, 1209-1212.
(15) Yang, W.; Gao, S.; Gao, X.; Karnati, V. R.; Ni, W.; Wang, B.;
Hooks, W. B.; Carson, J .; Weston, B. Bioorg. Med. Chem. Lett 2002,
12, 2175-2177.
(30) Soundararajan, S.; Badawi, M.; Kohlrust, C. M.; Hageman, J .
H. Anal. Biochem. 1989, 178, 125-134.
(31) Springsteen, G.; Wang, B. Chem. Commun. 2001, 1608-1609.
(32) Springsteen, G.; Wang, B. Tetrahedron 2002, 58, 5291-5300.
10.1021/jo0350357 CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/13/2004
J . Org. Chem. 2004, 69, 1999-2007
1999

Ni et al.
correct many literature mistakes33 including the strength
of the binding between phenylboronic acid and various
sugars.
other diol detection.11,12 Herein we report the design,
synthesis, and evaluation of nitrophenol-based boronic
acid reporter compounds that show significant spectro-
scopic changes upon addition of sugars at neutral pH in
aqueous solution. Our design attempts to take advantage
of the ability of a boronic acid functional group to
modulate the pKa and/or the electron density of a
neighboring group upon addition of a diol, through both
proximity effect as well as inductive effect. Such a
spectroscopic reporter compound can be used for the
construction of diboronic acid sensors for the specific
recognition of various carbohydrates.
Critical to the sensor development effort is the avail-
ability of reporter compounds that can generate a detect-
able signal upon binding with the target molecules. Along
this line, Czarnik,16 Shinkai,1,34,35 Lakowicz,36-39 James,40-43
and Heagy44,45 have developed various fluorescent boronic
acid compounds that show fluorescent intensity/wave-
length changes upon binding with diols to varying
degrees. The most prominent one is the Shinkai photo-
electron transfer (PET) system using B-N bond strength
to modulate the fluorescence quenching process and
therefore the fluorescent intensity changes.9,46,47 How-
ever, many aspects of these available reporters need to
be improved. This includes water solubility, the magni-
tude of fluorescence intensity changes, and photostabili-
ties and chemical stabilities of these compounds. Re-
cently, our group has developed a novel type of fluorescent
probe for carbohydrates, 8-quinolineboronic acid. This
boronic acid responds to the binding of a carbohydrate
with over 40-fold increases in fluorescence intensity and
shows optimal fluorescence change at physiological pH
in aqueous solution.48
There has also been a great deal of interest in develop-
ing colorimetric sensors for sugars with use of boronic
acid compounds. Along this line, Strongin,49 J ames,50
Lakowicz,51 and Shinkai52 have reported several com-
pounds that upon addition of sugar change colors. Most
of these designs are based on the modulation of the
electronic properties of the chromophores through B-N
bond formation or inductive effects. The mechanism of
action for the Strongin system, however, is not clear. The
Anslyn group has reported several systems that use a
reporter compound to form an ensemble for sugar and
Resu lts a n d Discu ssion
1. Design a n d Syn th esis. It has long been recognized
that the pKa of a boronic ester is, most of the time, lower
than that of the corresponding boronic acid.32,33 For
example, the pKa of phenylboronic acid is about 8.8 and
the pKa values of its glucose and fructose esters are 6.8
and 4.5, respectively.32 Such significant changes in pKa
upon ester formation have one major implication that can
be taken advantage of in designing boronic acid-based
fluorescent and color sensors. This is particularly impor-
tant for sensors that have a desirable functional pH close
to neutral. We have recently examined a series of about
30 phenylboronic acid compounds with varying substit-
uents and found that the highest pKa was about 9.0 when
the substitute was a methoxy group. If binding to a sugar
lowers its pKa by 2-4 units, this would mean that at pH
7.4 the boronic acid would exist in the neutral form and
addition of a sugar would change it to the anionic
tetrahedral form.
This change in charge states can be used to modulate
the electron density of the neighboring group through
either proximity or inductive effect. There is also the
possibility that the protonation state of a neighboring
group that is either an acid or base can be affected by
the ionization state change of the boronic acid. Both of
these factors are known to affect the spectroscopic
properties of a chromophore. p-Nitrophenol (PNP) is
known to change its ionization states depending on the
pH and such a change causes tremendous shifts in its
spectroscopic properties.53 Therefore, nitrophenol-based
compounds have been widely used for the development
of substrates for enzyme assays among other things.54
We envisioned that if a boronic acid is positioned next to
the phenol group of p-nitrophenol or its analogues, the
ionization state changes of the boronic acid moiety upon
binding with a diol would likely affect the pKa and/or the
electron density of the phenol group through either
proximity or inductive effect. If this change straddles the
neutral pH range, we should be able to develop a
spectroscopic reporter compound functional at physiologi-
cal pH. For this purpose, we designed (Scheme 1)
2-hydroxy-5-nitrophenylboronic acid (1) and, for com-
parison, 4,6-dinitrophenol-2-boronic acid (2).
(33) Lorand, J . P.; Edwards, J . O. J . Org. Chem. 1959, 24, 769.
(34) J ames, T. D.; Linnane, P.; Shinkai, S. Chem. Commun. 1996,
281-288.
(35) Yamamoto, M.; Takeuchi, M.; Shinkai, S. Tetrahedron 1998,
54, 3125-3140.
(36) DiCesare, N.; Lakowicz, J . R. Tetrahedron Lett. 2002, 43, 2615-
2618.
(37) DiCesare, N.; Lakowicz, J . R. Anal. Biochem. 2002, 301, 111-
116.
(38) DiCesare, N.; Lakowicz, J . R. Tetrahedron Lett. 2001, 42, 9105-
9108.
(39) DiCesare, N.; Lakowicz, J . R. Chem. Commun. 2001, 2022-
2023.
(40) Cooper, C. R.; Spencer, N.; J ames, T. D. Chem. Commun. 1998,
1365-1366.
(41) Ward, C. J .; Patel, P.; J ames, T. D. J . Chem. Soc., Perkin Trans.
1 2002, 462-470.
(42) Arimori, S.; Bosch, L. I.; Ward, C. J .; J ames, T. D. Tetrahedron
Lett. 2001, 42, 4553-4555.
(43) Cooper, C. R.; J ames, T. D. Chem. Lett. 1998, 883-884.
(44) Adhikiri, D. P.; Heagy, M. D. Tetrahedron Lett. 1999, 40, 7893-
7896.
(45) Cao, H.; Diaz, D. I.; DiCesare, D.; Lakowicz, J . R.; Heagy, M.
D. Org. Lett. 2002, 4, 1503-1505.
(46) J ames, T. D.; Sandanayake, K. R. A. S.; Iguchi, R.; Shinkai, S.
J . Am. Chem. Soc. 1995, 117, 8982-8987
(47) Sandanayake, K. R. A. S.; Shinkai, S. J . Chem. Soc., Chem.
Commun. 1994, 1083-1084.
(48) Yang, W.; Yan, J .; Springsteen, G.; Deeter, S.; Wang, B. Bioorg.
Med. Chem. Lett. 2003, 13 (6), 1019-1022.
(49) Davis, C. J .; Lewis, P. T.; McCarroll, M. E.; Read, M. W.; Cueto,
R.; Strongin, R. M. Org. Lett. 1999, 1, 331-334.
(50) Ward, C. J .; Patel, P.; J ames, T. D. Org. Lett. 2002, 4, 477-
479.
The synthesis of compound 1, 2-hydroxy-5-nitrophen-
ylboronic acid, started with commercially available 2-bro-
mo-4-nitroanisole. Borylation of bromide 3 with Pd(dppf)-
Cl2 as the catalyst in the presence of potassium acetate
(51) DiCesare, N.; Lakowicz, J . R. Org. Lett. 2001, 3, 3891-3893.
(52) Kimura, T.; Takeuchi, M.; Nagasaki, T.; Shinkai, S. Tetrahedron
Lett. 1994, 36, 559-562.
(53) Dodgson, K. S.; Spencer, B. Biochem. J . 1953, 53, 444-451.
(54) Nistor, C.; Rose, A.; Wollenberger, U.; Pfeiffer, D.; Emneus, J .
Analyst 2002, 127, 1076-1081.
2000 J . Org. Chem., Vol. 69, No. 6, 2004

Nitrophenol-Based Color Reporters for Diols
F IGURE 1. Absorption of compound 1 at 2 × 10-4 M in pH
7.4, 0.1 M phosphate buffer with increasing concentration of
fructose (0, 0.4, 1.0, 2.1, 4.0, 16.0, 100 mM). Inset: Absorption
increase (λ ) 332 nm) and absorption decrease (λ ) 373 nm)
of 1 in the presence of fructose.
F IGURE 2. Absorption intensity changes of 1 (2 × 10-4 M)
at λ ) 373 nm with sugar addition: O, glucose (Ka ) 8.0 M-1);
9, galactose (Ka ) 44.0 M-1); 2, 1,2-cis-cyclopentanediol (Ka
) 34.7 M-1); b, fructose (Ka ) 245 M-1). All studies were in
4% methanol in 0.1 M pH 7.4 phosphate buffer (v/v).
SCHEME 1. Design ed Nitr op h en ol-Ba sed Bor on ic
Acid s 1 a n d 2
fructose-specific sensor. Monoboronic acid compounds are
not expected to have intrinsic preference that is different
from that of phenylboronic acid.32 Such a reporter com-
pound can be used for the preparation of specific sensors
through the construction of di- or multiboronic acid
compounds. The binding studies were carried out at a
concentration of 2 × 10-4 M (4% methanol in pH 7.4, 0.1
M phosphate buffer).
As can be seen from Figure 1, the spectroscopic
properties of 2-hydroxy-5-nitrophenylboronic acid changed
tremendously at pH 7.4 with the addition of fructose.
Millimolar concentrations of fructose were sufficient to
cause a significant blue shift of the absorption λmax (from
373 to 332 nm). With increasing concentrations of
fructose, the UV intensity decreased at 373 nm and
increased at 332 nm. This essentially forms a ratiometric
system with large λmax shifts. The response seems to
reach a plateau at 16 mM. An isosbestic point is observed
at 350 nm showing the equilibrium of two species.
To further examine the detailed effect of sugars on the
spectroscopic properties of 2-hydroxy-5-nitrophenyl-
boronic acid (1), we studied the concentration effect of
fructose, glucose, galactose, and 1,2-cis-cyclopentanediol
on this compound (Figure 2).
furnished pinacolboronate 4 in 20% yield.55 The oxidative
deprotection of pinacolboronate ester 4 by sodium perio-
date56 yielded 70% of methoxy phenylboronic acid 5.
Then, treatment of this substituted phenylboronic 5 with
1 M boron tribromide in methylene chloride deprotected
the methoxyl group to give hydroxyl phenyl boronic acid
1 in 90% yield (Scheme 2).
The synthesis of 4,6-dinitrophenol-2-boronic acid (2)
started with commercially available 2-methoxyphenyl-
boronic acid (6). Nitration of this boronic acid with a
mixture of nitric acid and sulfuric acid at -10 °C gave
the desired dinitro product 7 in 50% yield with 10% of a
side product, 1-methoxy-2,4-dinitrobenzene. There are
literature precedents showing that nitration reactions
can cause the cleavage of the C-B bond.57 Treatment of
boronic acid 7 with 1 M boron tribromide in methylene
chloride yielded compound 2, 4,6-dinitrophenol-2-boronic
acid, in 90% yield (Scheme 3). Boronic acid compounds
sometimes exhibit poor NMR spectra due to the possible
formation of dimers or polymers. Therefore, for the full
characterization of these compounds, boronic acids 7 and
2 were protected by reacting them with 2,2-dimethylpro-
pane-1, 3-diol to give boronate 8 and 9.
Spectroscopic changes were observed with the addition
of all three sugars and cis-1,2-cyclopentanediol. As
expected, compound 1 showed the same type of intrinsic
preference for fructose over others as phenylboornic
acid.32 The binding constants between this boronic acid
and fructose, glucose, galactose, and 1,2-cis-cyclopen-
tanediol were determined as 245, 8.0, 44.0, and 34.7 M-1
respectively.
,
2. Bin d in g Stu d ies w ith Diol-Con ta in in g Com -
p ou n d s. To see whether our design works or not, the
effect of fructose on the absorption of 1 was examined
(Figure 1). Fructose was used because it was known to
bind with monoboronic acid very well.32 It should be noted
that this does not mean that we are trying to develop a
3. Mech a n istic Stu d ies. Since compound 1 works
very well for signaling sugar binding, it is very important
to examine the mechanism. Since the design was based
on the idea that the charge state changes of the boronic
acid group upon ester formation may affect the charge
state or density of a neighboring group, it is important
to examine the possible ionization species existing in the
reaction solution. Compound 1 may exist in different
ionization states (Scheme 4) at neutral pH because of the
presence of two ionizable functional groups, the phenol
(55) Ishiyama, T.; Murata, M.; Miyaura, N. J . Org. Chem. 1995, 60,
7508-7510
(56) Falck, J . R.; Bondlela, M.; Venkataraman, S. K.; Srinivas, D.
J . Org. Chem. 2001, 66, 7148-7150.
(57) Salzbrunn, S.; Simon, J .; Prakash, G. K. S.; Petasis, N. A.; Olah,
G. A. Synlett 2000, 10, 1485-1487.
J . Org. Chem, Vol. 69, No. 6, 2004 2001

Ni et al.
SCHEME 2. Syn th esis of 2-Hyd r oxy-5-n itr op h en ylbor on ic Acid (1)a
a
Reagents and conditions: (a) bis(pinacolato)diboron, Pd(dppf)Cl2, KOAc, anhydrous DMSO, 20%; (b) NaIO4, 2 N HCl, THF/H2O, 70%;
(c) 1 M BBr3/DCM, 90%.
SCHEME 3. Syn th esis of
4,6-Din itr op h en ol-2-bor on ic Acid (2)a
F IGURE 3. pH profile of compound 1 (initial concentration
of 1 was set at 2 × 10-4 M in 4% methanol/96% water, and
the ionic strength was initially fixed at 0.1 M NaCl).
a
Reagents and conditions: (a) H2SO4/HNO3, -10 °C, 50%; (b)
1 M BBr3/DCM, 90%; (c) 2,2-dimethylpropane-1,3-diol, toluene, 115
°C, refluxing, 100%.
hydroxyl group and the boronic acid group. In evaluating
the likelihood for the formation of different ionization
species, one needs to analyze the pKa of each functional
group present (Scheme 4).
At physiological pH, if the hydroxyl group (of 1, R )
H) has a lower pKa than the boronic acid, it would favor
species 10a and if the boronic acid has a lower pKa it
would favor the formation of 10c. If these two groups
have comparable pKa values, both species may exist
simultaneously in different ratios depending on the
conditions. p-Nitrophenol has a pKa of about 7.2. Since
boronic acid is considered an electron-withdrawing func-
tional group, it is most likely that the pKa of the hydroxyl
group of compound 1 is lower than 7.2. 2-Methoxy-5-
nitrophenylboronic acid has a pKa of about 7.1 (see
F IGURE 4. pH profile of compound 1 with 200 mM D-fructose
(initial concentration of 1 was set at 2 × 10-4 M in 4%
methanol/96% water, and the ionic strength was initially fixed
at 0.1 M NaCl).
below). This should be comparable to the boronic acid pKa
of 1. Therefore, it is possible that the pKa values of these
two functional groups are so close that both 10a and 10c
exist at the same time at physiological pH. The formation
of the di-ionized species 10b is unlikely at neutral pH.
Upon ester formation, there are four possible species,
11a -d . Since binding with a diol most of the time lowers
the pKa of the boron by 2-4 pKa units,32 we expect that
species 11c be predominant upon ester formation, which
would lower the ratio of the hydroxyl ionized species 11a .
This shift in the concentration of the species with the
hydroxyl group ionized should result in a change in the
spectroscopic properties of the solution.
solution. Figure 3 shows the pH titration profile of 1 in
the absence of any diols. One can readily see that there
are three absorption maxima at 315, 380, and 427 nm.
These three λmax values represent three different ioniza-
tion stages. At low pH (pH <4), the absorption λmax is at
315 nm corresponding to the nonionized form of 1
(Scheme 4). The absorption peak at 380 nm represents
the first ionization (first pKa), which can be either the
deprotonation of the hydroxyl group (10a ) or the ioniza-
tion of the boron (10c) or a combination of these two since
their pKa values are very close. The third peak at 427
nm represents the fully ionized form with two anions on
the same molecule (10b).
To understand the ionization state changes upon ester
formation, it is also important to examine the spectro-
scopic properties of the different species existing in
Figure 4 shows the pH titration profile of 1 in the
presence of D-fructose (200 mM). In this case, there also
seems to be three ionization stages corresponding to λmax
2002 J . Org. Chem., Vol. 69, No. 6, 2004

Nitrophenol-Based Color Reporters for Diols
SCHEME 4. Equ ilibr iu m a m on g Differ en t Sp ecies of 1 a n d 2 in th e P r esen ce a n d Absen ce of
D-F r u ctose
values of 315, 325, and 427 nm. The first λmax is the same
as that of 1 in the absence of any diol. This is under-
standable since at low pH little binding is expected32 and
the species present should be the free sensor 1 with no
ionization. The last λmax (427 nm) is also similar to that
of 1 in the absence of any diol. It is known that the
binding affinity of boronic acid with diols increases with
increasing pH and at high pH one would expect that
essentially all the boronic acid is converted to its ester.
This would indicate that at pH 10.5, the species is also
the dianion species (11b), which has the same spectro-
scopic properties as that of 10b. This further indicates
that forming an ester alone does not change the electronic
properties of the complex enough to affect their spectro-
scopic properties. The major difference between the
system with a diol (fructose) and without a diol (Scheme
4) is in the neutral pH region, which is ideal (and by
design) for the sensing of diol compounds under near-
physiological conditions. To understand the mechanism
through which this spectroscopic property changes, one
needs to analyze the predominant spectroscopic species
in both situations (with and without fructose).
In the presence of fructose, there are two possible
species at neutral pH, 11a and 11c. To see which the
dominant species is, we also examined the spectroscopic
changes of the methylated version of 1, compound 5, in
the presence (Figure 6) and absence (Figure 5) of fructose.
As expected, the pH titration curve of 5 shows two species
(Figure 5), one with a λmax of 315 nm and the other at
337 nm. It is easy to understand that the peak at 315
nm corresponds to the un-ionized form and the peak at
337 nm corresponds to the anionic tetrahedral boron
species. It needs to be noted that the first λmax is about
the same as that of the un-ionized 1, which indicates that
methylation did not change the electronic properties of
the molecules sufficiently to affect its UV properties. This
would also suggest that 11c should have about the same
λmax as the ionized form of 5. If we compare the UV
F IGURE 5. pH profile of compound 5 (initial concentration
of 5 is set at 2 × 10-4 M in 4% methanol/96% water, and the
ionic strength was initially fixed at 0.1 M NaCl).
spectrum of 1 in the presence of fructose at around
neutral pH, it is easy to see that the λmax is about the
same as that of the ionized (5). This would suggest that
upon addition of fructose to 1, the lowered pKa of the
boron in the ester makes 11c the predominant species.
Such a conclusion is also consistent with the pKa-lowering
effect of fructose on boronic acid.
Now we can go back to analyze the situation of 1 in
the absence of fructose. A detailed examination of the
spectral set with 1 in the absence of any sugar reveals a
“shifting” λmax for the middle peak (380 nm, Figure 3).
The lack of an isosbestic point transitioning from the first
to the second “ionization state” indicates the formation
of more than one ”intermediate”. This could only mean
that both 10a and 10c were formed in the pH range of
5.79 and 8.57. The shifting of the λmax from 339 to 380
nm in this region indicates the formation of a species with
a shorter λmax wavelength at low pH and a species with
a longer λmax wavelength at high pH. Intuitively, it is
reasonable to assign 10a as the species with a longer
J . Org. Chem, Vol. 69, No. 6, 2004 2003

Ni et al.
being the species contributing most to the long-wave-
length UV absorption properties. For 10b, the situation
is similar; there are only two analogous resonance forms.
Therefore, it is reasonable to expect that PNP and 10b
have similar λmax values. Since 11b should behave
similarly to 10b, its resonant forms are not shown. For
10a , the situation is very different. Because the boron
open shell is not occupied, it is resonantly electron-
withdrawing. One can write a third resonance structure,
10a -3. Therefore, the UV absorption of 10a can be
expected to be quite different from that of PNP, 10b, and
11b.
Overall, the pH titration (Figure 3) of 1 in the absence
of any sugar can be interpreted as follows. At low pH,
there is no ionization, and free 1 is the only species
(Scheme 4). With increasing pH, the first ionization
occurs to generate 10c, which has a λmax of around 330
nm. With pH increasing, 10a starts to appear. Since 10a
has a longer λmax, the UV spectra shift to longer wave-
length. Because the transition from free 1 to the species
with one ionized functional group involves two “products”,
no isosbestic point was observed.
F IGURE 6. pH profile of compound 5 in the presence of 200
mM D-fructose (initial concentration of 5 was set at 2 × 10-4
M in 4% methanol/96% water, and the ionic strength was
initially fixed at 0.1 M NaCl).
SCHEME 5. Reson a n ce Str u ctu r es of
Dep r oton a ted p-Nitr op h en ol a n d Som e Sp ecies of
Com p ou n d 1
It is important to note that at pH 7.4, the λmax for the
solution of 1 is at about 373 nm in the absence of any
sugar and 325 nm in the presence of fructose, which
forms the basis for the sugar sensing at physiological pH
in aqueous solution. The mechanism responsible for this
sensing is that the addition of fructose shifted the UV
absorption of the solution from that of 10a to that of 11c.
This was the direct result of the pKa-lowering effect of
fructose addition. Because the ratio of 11a over 11c (and
10a over 10c) is directly related to the relative pKa of
these two species and these two species are mutually
exclusive, the lowered boron pKa of the ester relative to
the hydroxyl group would result in a shift in the acid-
base equilibrium by lowering the percentage of the
hydroxyl-ionized species (11a ). Consequently, addition of
fructose to the solution of 1 results in a shift of the UV
absorption of the solution from that of 10a to that of 11c.
As discussed earlier, 11B NMR is known to indicate the
electronic and ionization state of the boron atom. For
example, the 11B NMR chemical shift for the neutral and
trigonal form (such as 10a ) is about 30 ppm and that for
the anion tetrahedral form (such as 10b) it is about10
ppm.58 At pH 7.4 (phosphate buffer), the 11B NMR of
compound 1 in the presence of excessive cis-1,2-cyclo-
pentanediol was a fairly sharp peak at about 7.7 ppm
(see the Supporting Information), indicative of anionic
species consistent with either 11b or 11c. However, it is
unlikely that at neutral pH, it can be doubly ionized.
Therefore, the NMR results are consistent with the
formation of 11c upon addition of a diol to the solution
of 1 at neutral pH. Cyclopentanediol was used in the
NMR experiments to reduce the complexity of the NMR
spectrum because fructose can bind to a boronic acid in
several forms, which would complicate the interpretation.
When the 11B NMR of 1 in the absence of any diol was
determined at pH 7.4, a broad peak center around 6.3
ppm was observed (see the Supporting Information), an
wavelength, and 10c as the one with a shorter wave-
length. Such an assumption is also consistent with
experimental data. For example, 11c and the ionized
form of 5 all have λmax values around 325 nm, and they
are structurally analogous to 10c. Therefore, it is logical
to assign the species with an approximate λmax of 339 nm
(please note that at this point, it is already a mixture of
more than one species, therefore it may not reflect the
λmax of a single species), not 380 nm. Furthermore, the
ionization of the hydroxyl group in a p-nitrophenol type
of structure is known to significantly increase its λmax
value. For example, the ionized form p-nitrophenol itself
has a λmax of about 425 nm, which is similar to that of
10b and 11b.
One might ask if the deprotonated form of p-nitro-
phenol (PNP) and 10b and 11b all have similar λmax
values, why should the λmax of 10a be very different if it
also has the hydroxyl group ionized. This can be ex-
plained by examining the resonance structures of these
different species (Scheme 5). For PNP, there are two
resonance structures, PNP-1 and PNP-2, with PNP-2
(58) No¨th, H.; Wrackmeyer, B. Nuclear Magnetic Properties of
Boron. In Nuclear Magnetic Resonance Spectroscopy of Boron Com-
pounds; Diehl, P., Fluck, E., Kosfeld, R., Eds.; NMR Basic Principles
and Progress Series 14; Springer-Verlag: Berlin, Germany, 1978; pp
5-14 and 287.
2004 J . Org. Chem., Vol. 69, No. 6, 2004

Nitrophenol-Based Color Reporters for Diols
F IGURE 7. Absorption spectra of 2 (5 × 10-6 M) (4%
methanol in water (v/v), 0.1 M phosphate buffer, pH 7.4) in
the presence of fructose (0, 12, 24, 48, 100, 200, 400 mM).
Inset: Absorption change of 2 at 361 (b) and 272 nm (O) upon
addition of fructose.
indication that the boron is NOT in the trigonal neutral
state. There are two things that need to be noted. First,
the broad peak is an indication that there is most likely
more than one species in the solution, which is consistent
with the earlier proposal that it was a mixture of 10a
and 10c. Second, the chemical shift is consistent with
the boron being more shielded than in the trigonal
neutral form. However, much more work will be needed
to understand exactly what the composition is for all the
different species.
For comparison, we also studied the binding of 4,6-
dinitrophenol-2-boronic acid (compound 2) with fructose,
glucose, and galactose. The studies were carried out at a
concentration of 5 × 10-6 M for 2 (4% methanol/96% 0.1
M phosphate buffer, pH ) 7.4). Spectroscopic changes
were observed with the addition of all three sugars, with
binding constants of 13.5, 1.2, and 0.7 M-1 for fructose,
glucose, and galactose, respectively. It is very interesting
to see that the binding constants for 2 with various
sugars are smaller than that of 1. This contradicts the
conventional notion that boronic acids with lower pKa
values bind more tightly to diols. Our group has also
generated other data that indicate that boronic acid pKa
values are not directly proportional to the binding
constants.59
F IGURE 8. pH profile of 2 (initial concentration: 1 × 10-5
M in 4% methanol/96% water, and the ionic strength was
initially fixed at 0.1 M NaCl).
pH 1.8 to pH 13. From Figure 8, it can be seen that in
the absence of any sugar the first pKa value (the ioniza-
tion of only one functional group that can be either the
boronic acid or the hydroxyl group) of compound 2 is at
about 3.3 and the second pKa value (the ionization of two
functional groups) is at about 9.2. With the addition of
fructose (200 mM) (Figure 9), the first pKa value re-
mained essentially the same and the second one was
lowered to about 7.2.
One can assume a similar mechanism in operation for
2 as for 1 (Scheme 4). In the absence of any sugar, the
peak at 369 nm is attributed to species 12a . However,
upon addition of a sugar and the lowering of the boronic
pKa value,32 the concentration of 12a decreased with
increasing concentration of 13c, which contributed to the
decrease in absorption at 369 nm and increase in absorp-
tion at 323 nm. The major difference between 2 and 1 is
that the second pKa value with the addition of fructose
is much lower (7.2). This means that at physiological pH,
addition of fructose can also result in the formation of
the di-ionized species 13b, which is responsible for the
peak at 425 nm (Scheme 4).
As an example, the spectroscopic changes of 2 with the
addition of fructose are shown in Figure 7. It can be seen
that such spectroscopic changes at physiological pH are
also significant with a pattern similar to that of 1. With
increasing fructose concentrations, the UV intensity
decreased at 361 nm and increased at 272 nm. The
presence of fructose also results in the appearance of a
new peak at 369 nm. Two isosbestic points were observed
at 323 and 423 nm, indicative of two equilibria of three
species (Scheme 4).
Con clu sion s
We have designed and synthesized two nitrophenol-
boronic acids (1 and 2). Binding of sugars to these
compounds resulted in very significant spectroscopic
changes in both intensity and wavelength. The mecha-
nism for the changes is through the manipulation of the
pKa values of the boronic acid, which affect the equilib-
rium among different chromophoric species. Such com-
pounds can be used as colorimetric reporters functioning
as both a recognition and a signaling unit for the
construction of polyboronic acid spectroscopic sensors for
To help understand the mechanism of the spectroscopic
change of the binding of compound 2 with sugar, the
apparent pKa value of compound 2 was determined. This
was achieved by observing the spectroscopic changes from
(59) Yan, J .; Springteen, G.; Wang, B. Unpublished results.
J . Org. Chem, Vol. 69, No. 6, 2004 2005

Ni et al.
d, J ) 3 Hz), 8.30-8.26 (1H, dd, J ) 3, 9 Hz), 6.91 (1H, d, J
) 9 Hz), 3.92 (3H, s), 1.36 (12H, s) ppm. 13C NMR (CDCl3) δ
168.9, 141.3, 132.9, 128.7, 110.3, 84.4, 56.6, 25.0 ppm. Anal.
Calcd for C13H18BNO5‚0.5H2O: C, 54.19; H, 6.65; N, 4.86.
Found: C, 54.52; H, 6.41; N, 4.64. ESI-MS for C13H18BNO5
calcd 279.1, found 279.0.
2-Meth oxy-5-n itr op h en ylbor on ic Acid (5). Sodium pe-
riodate (320 mg, 1.50 mmol) was added to a solution of
pinacolboronate ester 4 (133.7 mg, 0.48 mmol) in THF/H2O
(4:1, 4 mL) at room temperature. The mixture was stirred for
30 min, and then 2 N HCl (0.5 mL) was added and the solution
stirred overnight. Then the reaction mixture was extracted
with ethyl acetate (2 × 30 mL), and the combined organic
extracts were washed with water and brine, dried over MgSO4,
and concentrated in vacuo. Column chromatography of the
residue [silica gel, EtOAc/DCM] afforded the boronic acid
1
(70%). H NMR (CD3OD, 300 MHz) δ 8.29-8.26 (1H, dd, J )
2.8 Hz, 9 Hz), 8.13 (d, 1H, J ) 2.8 Hz), 7.12 (1H, d, J ) 9.2
Hz), 3.95 (3H, s) ppm. 13C NMR (CD3OD) δ 167.7, 142.801,
129.8, 127.9, 111.1, 56.7 ppm. Anal. Calcd for C7H8BNO5: C,
42.69; H, 4.07; N, 7.11. Found: C, 43.07, H, 4.16, N, 6.90. EI-
MS for C7H8BNO5 calcd for (M+) 197, found 197.
2-Hydr oxy-5-n itr oph en ylbor on ic acid (1). Gen er al P r o-
ced u r e for th e Dep r otection of th e Hyd r oxy Gr ou p . To
the solution of methoxy phenylboronic acid (5) in dry DCM
cooled with acetone/dry ice bath was added dropwise 3 equiv
of 1.0 M boron triboromide in methylene choloride solution.
The reaction solution was stirred for 1 h. Then the reaction
mixture was warmed to rt and stirred for another 3 h under
N2. The organic solvent was removed and the residue was
extracted with ether and washed with water. The organic layer
was washed with 2 N sodium hydroxide solution and the
resulting basic solution was neutralized with 2 N HCl. Then
F IGURE 9. pH profile of 2 with 200 mM D-fructose (initial
concentration of 2 was 1 × 10-5 M in 4% methanol/96% water,
and the ionic strength was initially fixed at 0.1 M).
1
the product was obtained by extraction with DCM (50%). H
NMR (CD3OD, 300 MHz) δ 8.31 (1H), 8.14 (1H, dd, J ) 2.7, 9
Hz), 6.86 (1H, d, J ) 9.2 Hz) ppm. 13C NMR (CD3OD) δ 142.0,
132.9, 128.6, 116.5, 103.0 ppm. ESI-MS for C6H6BNO5 calcd
for (2M - H2O) 347, found 347.
monocarbohydrates and complex carbohydrates, many of
which are biomarkers for important biological and patho-
logical events.
2-Meth oxy-3,5-n itr op h en ylbor on ic Acid (7). To the stir-
ring mixture of 1.5 mL of nitric acid and 1.5 mL of sulfuric
acid cooled in acetonitrile/dry ice bath was added slowly
o-methoxyphenyl boronic acid 6 (0.1 g, 6.6 mmol) over 20 min.
The reaction mixture was stirred for 3 h more in salt-ice
solution while the temperature was kept below -10 °C. Brown
solid was seen during the reaction. The reaction mixture was
poured onto 5 g of ice and kept in an ice-water bath for 3 h.
The brown solid precipitated. The solid was filter and washed
with ice-cold water and dried under vacuum. Column chro-
matography of the residue [silica gel, EtOAc/DCM] afforded
the boronic acid 7 (30%). 1H NMR (CD3OD) δ 8.64 (1H, d, J )
3.0 Hz), 8.46 (1H, d, J ) 3.0 Hz), 4.02 (3H, s) ppm. 13C NMR
(CD3OD) δ 160.6, 143.6, 143.2, 133.3, 122.5, 62.4 ppm. Anal.
Calcd for C7H7BN2O7: C, 34.75; H, 2.92; N, 11.58. Found: C,
35.12, H, 2.94, N, 11.35. EI-MS for C7H7BN2O7 calcd for (M+)
242, found 242.
Exp er im en ta l Section
Gen er a l P r oced u r es. Commercially available reagents
were used without additional purification unless otherwise
indicated. All glassware was oven-dried (>120 °C, >4 h) and
cooled under vacuum. Dichloromethane (DCM) was distilled
from CaH2. THF was distilled from sodium and benzophenone.
Analytical thin-layer chromatography (TLC) was performed
with plastic-backed TLC silica gel 60 F hard layer plates. Flash
chromatography was performed with silica gel (flash, 32-63
µm). Mass spectrometry (MS) analyses were performed by the
Mass Spectrometry Laboratories of North Carolina State
University and the University of Kansas. 1H and 13C NMR
spectra were recorded at 300 MHz (for proton). 11B NMR
spectra were measured with Et2O‚BF3 in deuterated chloro-
form as the external standard. Chemical shifts (δ) are given
2-(2-Met h oxy-3,5-d in it r op h en yl)-5,5-d im et h yl[1,3,2]-
d ioxa bor in a n e (8). The mixture of compound 7 (240 mg, 1.22
mmol) and pinacol (145.4 mg, 1.23 mmol) was refluxed in 20
mL of toluene with use of a Dean-Stark to remove water
overnight. Then toluene was evaporated and the solid residue
was extracted with DCM. The combined organic extracts were
washed with water and brine, dried over MgSO4, concentrated,
1
in ppm relative to TMS for H spectra and relative to residual
solvent for 13C spectra. Combustion analyses were performed
by Atlantic Micro labs, Inc., Norcross, GA.
2-(2-Meth oxy-5-n itr op h en yl)-4,4,5,5-tetr a m eth yl[1,3,2]-
d ioxa bor ola n e (4). The mixture of 2-bromo-4-nitroanisole
(360 mg, 1.56 mmol), pinacolboronate (360 mg, 1.41 mmol),
Pd(dppf)2Cl2 (32 mg, 0.04 mmol), and potassium acetate (360
mg, 3.7 mmol) was dried under N2. To this mixture was added
10 mL of anhydrous DMSO. The reaction mixture was stirred
at 85 °C under N2 for 5 h and then cooled to room temperature
and poured into 20 mL of ice-cold water. Then the mixture
was extracted with 2 × 30 mL of ethyl acetate and the
combined organic extracts were washed with water and brine,
dried over MgSO4, and concentrated in vacuo. Column chro-
matography of the residue [silica gel, EtOAc/DCM] afforded
1
and dried in vacuo. H NMR (CDCl3, 300 MHz) δ 8.94 (1H, d,
J ) 2.4 Hz), 8.63 (1H, d, J ) 2.4 Hz), 4.05 (3H, s), 3.85 (4H,
s), 1.09 (6H, s) ppm. Anal. Calcd for C12H15BN2O9: C, 46.48;
H, 4.88; N, 9.03. Found: C, 46.40, H, 4.90, N, 8.89.
2-(5,5-Dim et h yl[1,3,2]d ioxa b or in a n -2-yl)-4,6-d in it r o-
p h en ol (9). The procedure was the same as the preparation
1
of 8 from 7. H NMR (CDCl3, 400 MHz) δ 10.80 (1H, s), 8.95
(1H, d J ) 2.4 Hz), 8.77 (1H, d, J ) 2.4 Hz), 3.89 (4H, s), 1.09
1
the boronate 4 (20%). H NMR (CDCl3, 300 MHz) δ 8.56 (1H,
(6H, s) ppm. 13C NMR (CD3OD) δ 162.2, 139.8, 136.6, 124.6,
2006 J . Org. Chem., Vol. 69, No. 6, 2004

Nitrophenol-Based Color Reporters for Diols
73.1, 32.3, 21.9 ppm. Anal. Calcd for C11H13BN2O7: C, 44.63;
H, 4.43; N, 9.46. Found: C, 44.56; H, 4.47; N, 9.46.
Su p p or tin g In for m a tion Ava ila ble: HPLC analysis for
1, and 11B NMR spectra for 1 in the presence and absence of
cis-cyclopentanediol. This material is available free of charge
via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. Financial support from the Na-
tionalInstitutes ofHealth (NO1-CO-27184 and CA88343),
Georgia Cancer Coalition, and Georgia Research Alli-
ance is greatly acknowledged.
J O0350357
J . Org. Chem, Vol. 69, No. 6, 2004 2007