Fluoride protects boronic acids in the copper(i)-mediated click reaction

Article abstract of DOI:10.1039/b909575f

Fluoride has been found to protect boronic acids from copper(i)-mediated decomposition; such findings should be very useful for the preparation of boronic acid-based carbohydrate sensors and boronic acid conjugates using the copper(i)-mediated click react

Full text of DOI:10.1039/b909575f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Fluoride protects boronic acids in the copper(I)-mediated click reactionw
Shan Jin,a Gaurav Choudhary,a Yunfeng Cheng,a Chaofeng Dai,a Minyong Li*ab
and Binghe Wang*a
Received (in Austin, TX, USA) 14th May 2009, Accepted 30th June 2009
First published as an Advance Article on the web 6th August 2009
DOI: 10.1039/b909575f
Fluoride has been found to protect boronic acids from copper(I)-
mediated decomposition; such findings should be very useful for
the preparation of boronic acid-based carbohydrate sensors and
boronic acid conjugates using the copper(I)-mediated click reaction.
a way to minimize copper(I)-mediated decompositions.
Herein, we report our findings that fluoride can protect
boronic acids from copper(I)-mediated decomposition either
alone or in a cycloaddition reaction mixture. Such results
should be very useful for the future development of boronic
acid-based carbohydrate sensors and boronic acid conjugates
for various other applications such as aptamer selections.
Again, our initial interest was for the preparation of boronic
acid–nucleotide conjugates for incorporation into DNA
aptamers.9 The very first boronic acid used was 8-quinoline-
boronic acid (8-QBA, 1, Fig. 1), which led to the relatively
uneventful synthesis of the desired conjugates. This turned out
to be fortuitous because later efforts in incorporating other
boronic acids in a similar fashion led to decomposition,
presumably due to copper insertion. In looking for a way to
minimize this side reaction, we thought about the ability for
fluoride to convert the boron atom in a boronic acid to its
anionic tetrahedral form.12 We reasoned that the presence of a
negative charge on boron would change the overall electronic
properties of the B–C bond and negatively affect copper
insertion, thus leading to a reduction in side reactions.
Therefore, we first studied the stabilities of a series of eight
boronic acids (Fig. 1) for a period of 5 h in the presence of
Cu(I). This time course was chosen because that was the
typical length of time for our click reaction for conjugation.
In doing so, HPLC was used to monitor the progression of the
decomposition reaction. The conditions for the stability studies
were based on the optimized conditions in our lab (Fig. 2).
Among the eight boronic acid studied, very different
stabilities in the presence of Cu(I) were observed. For example,
8-QBA (1) was essentially stable with 95% remaining after
being exposed to 100 mM of Cu(I) for 5 h. On the other hand,
when boronic acid 3 was subjected to the same conditions,
35% degradation was observed. Fig. 2 shows a set of repre-
sentative HPLC chromatograms using boronic acids 1, 3 and 5
as examples. One can see that at the 5 h point (Fig. 2c),
multiple peaks appeared for compounds 3 and 5. However, the
It is well known that boronic acids are very important in
synthetic organic chemistry, medicinal chemistry and molecular
recognition. Specifically, in synthetic chemistry, boronic
acids are also very important intermediates in reactions such
as the Suzuki coupling;1,2 boronic acid compounds have
shown diverse biological activities and could be useful pharma-
ceutical agents,3 and in molecular recognition, boronic acids
are known to form tight complexes with compounds containing
the diol moiety, which can be used for chemosensor
design.2,4–6 Along this line, the copper(I)-mediated Huisgen
cycloaddition7,8 (click reaction, Scheme 1) could be a very
powerful method for the preparation of boronic acids with
diverse structural features for synthetic and medicinal
chemistry applications, and bisboronic acids and boronic acid
conjugates for biosensing applications. The ability to modify
boronic acids using click chemistry should allow for a drastic
increase in the structural diversity of available boronic acids.
Recently, our lab has prepared boronic acid–nucleic acid
conjugates using the copper-mediated click reaction for the
selection of DNA aptamers for glycoproteins.9 In expanding
our work to include a large set of boronic acids for conjuga-
tion with nucleotides, we found that copper(I)-mediated
degradation was a problem with some boronic acids. This is
understandable since copper is known to insert between the
carbon–boron bond, which leads to useful coupling reactions
when desired.10,11 However, in applying this click reaction to
the preparation of boronic acid conjugates, we desired to find
Scheme 1 The click reaction between alkyne (i) and azide (ii) in the
presence of copper(I) leading to triazole (iii).
a Department of Chemistry and Center for Biotechnology and Drug
Design, Georgia State University, Atlanta, GA 30302-4098, USA.
E-mail: wang@gsu.edu; Fax: +1 404-413-5543;
Tel: +1 404-413-5544
b Department of Medicinal Chemistry, School of Pharmacy,
Shandong University, Jinan, Shandong 250012, China.
E-mail: mli@sdu.edu.cn; Fax: +86 531-8838-2076;
Tel: +86 531-8838-2076
w Electronic supplementary information (ESI) available: HPLC
conditions for analytical studies, procedures for model click reactions,
and all the NMR and MS spectra, and HPLC chromatograms of the
final products. See DOI: 10.1039/b909575f
Fig. 1 Structures of compounds 1–8 for the HPLC stability studies
and the percentage of remaining boronic acids after being treated with
Cu(I) for 5 h without any protection.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 5251–5253 | 5251

View Article Online
Fig. 2 HPLC chromatograms of compounds 5, 3, and 1 (50 mM for each)
in the presence of 20 mM Cu(I) in (H2O–DMF–t-BuOH = 1 : 3 : 1)
solution, monitored at 260 nm. (a) Pure boronic acids; (b) immediately
after addition of Cu(I); (c) 5 h after mixing with Cu(I); (d) immediately
after addition of Cu(I) and fluoride; (e) 5 h after mixing with Cu(I) and
fluoride.
Fig. 3 HPLC chromatograms of model reaction (C) in the presence
of 20 mM Cu(I) in (H2O–DMF–t-BuOH = 1 : 3 : 1) solution,
monitored at 260 nm. (a) pure compound C-ii, (namely compound 5);
(b) pure compound C-i, namely the alkyne (i) in the reaction (C);
(c) pure compound C-iii, namely the triazole product (iii) in the
reaction (C); (d) 5 h reaction mixture with Cu(I); (e) 5 h reaction
mixture with Cu(I) and fluoride.
8-QBA (1) solution gave essentially one peak (starting material)
indicating minimal degradation.
To test the ability for fluoride to protect boronic acids from
Cu(I)-mediated degradation, we studied the stabilities of
boronic acids 2–8 in the presence of 100 mM of fluoride
(CsF). Fig. 2d–e shows a set of representative chromatograms
using boronic acid 5 as an example. As can be seen, 5 under-
went extensive degradation at the 5 h point after mixing with
Cu(I), giving four additional peaks (Fig. 2c). In the meantime,
the intensity of the boronic acid peak decreased by 21%. In
contrast, if fluoride was added together with Cu(I), degrada-
tion diminished to a negligible level. Similar results were
observed with boronic acids 5–8. To our surprise, the
degradation of boronic acids 2–4 was not prevented with the
addition of fluoride. It is not readily clear as to what structural
features led to the variation in fluoride protection.
results clearly indicate that fluoride played a protective role
and can help increase the yield of such cycloaddition reactions.
In an effort to achieve a quantitative understanding of the
beneficial role of fluoride addition, reaction yields were
calculated from the ratio of the product concentrations to
starting material concentrations. Standard curves were used
for the quantitative studies. First, we studied the reaction
(A, Table 1) that did not involve a boronic acid. Essentially
the same yields (80%) were obtained with and without fluoride
Table 1 Reaction yields of model click reactionsa
Yield (%)a Yield (%)a
Entry Alkyne (i)
Azide (ii)
without FÀ with FÀ
Though experimental results from boronic acids alone seem
to suggest that fluoride is able to protect at least some boronic
acids from Cu(I)-mediated degradations, there is more than
one way to interpret these results. For example, if the observed
protective effect was due to interactions between fluoride and
Cu(I), then fluoride addition may not help in increasing the
yield of the desired cycloaddition reaction, which is our
ultimate goal. To further examine this, we also studied the
effect of added fluoride on the copper(I)-mediated click
reaction of several pairs of boronic acid-containing alkynes/
azides. All studies clearly indicated that fluoride addition
indeed increases the reaction yields. Fig. 3 shows an example
set of chromatograms (reaction C, Table 1) of these HPLC
studies. The yield studies were conducted on an analytical
column using the same conditions as the stability studies
(see the ESIw for details). In this case, both starting materials
(Fig. 3; C-i and C-ii) disappeared after 5 h of reaction with a
concomitant increase in product peak. In the absence of
fluoride, multiple peaks appeared after 5 h of reaction
(Fig. 3d). On the other hand, in the presence of fluoride, the
desired product is the predominant component (Fig. 3e). Such
(A)
(B)
79 Æ 6
57 Æ 5
81 Æ 3
96 Æ 1
(C)
(D)
(E)
49 Æ 1
60 Æ 5
79 Æ 1
89 Æ 1
71 Æ 2
94 Æ 2
(F)
(G)
(H)
43 Æ 3
47 Æ 6
57 Æ 4
43 Æ 5
77 Æ 9
83 Æ 4
86 Æ 11
86 Æ 11
(I)
a
All the yields were calculated from standard curves (HPLC peak
integrals over concentrations).
ꢀc
This journal is The Royal Society of Chemistry 2009
5252 | Chem. Commun., 2009, 5251–5253

View Article Online
Scheme 2 A possible mechanism of boronic acid protection by
fluoride from copper-mediated degradation.
Fig. 4 Energy profile for the fluoride-mediated stability studies of
boronic acid.
addition. Such results indicate that fluoride has no intrinsic
effect on the click reaction yield (Table 1). For the other
reactions that do involve boronic acids, the reaction yields
increased by 11–43% with the addition of fluoride (Table 1).
For example, in the two cases (B and I) where the boronic acid
moiety is part of an arylalkyne, yield improvements from 57 to
96% (about 70% increase) (B) and 43 to 86% (one-fold
increase) (I) were observed, respectively. In cases where the
boronic acid compounds contain an azido group, 20–75%
improvements in yield were observed.
In conclusion, fluoride can protect some boronic acids from
copper(I)-mediated degradation and thus increase the yields of
click reactions. Reaction yield improvements of 20–100% were
observed. The mechanism of fluoride protection is proposed to
be the prevention of copper insertion by the formation of
anionic tetrahedral form of boron. This work should be very
useful for the preparation of boronic acid-based carbohydrate
sensors, boronic acid-modified aptamers, and structurally
diverse boronic acids for synthetic and medicinal chemistry
applications.
It is well known that metals can insert into the C–B bond of
boronic acids in general coupling reactions.13,14 It is reason-
able to suspect that this insertion is a prerequisite step in the
Cu(I)-mediated decomposition of boronic acids (Scheme 2).
To understand how fluoride addition could affect the relative
stabilities of the different species involved, we have also
conducted DFT calculations based on the BP86 functional
using the standard 6-31G* basis.15 The PCM solvation
model16 was used in single-point energy calculations
(PCM(sp)), and for geometry optimizations and frequency
calculations (PCM(opt)). All the calculations using the PCM
solvent model employed the UAHF atomic radii when
constructing the solvent cavity, as recommended in the
Gaussian 03 user’s reference. All geometries were fully
optimized, and the characters of the stationary points found
were confirmed by a harmonic frequency calculation at the
same level of theory to ensure a minimum was located. As
shown in Fig. 4 of the relative energy diagram, the insertion of
Cu(I) into phenylboronic acid (9) without fluoride protection
would give compound 10d with release of 13.2 kcal molÀ1 of
energy. The addition of fluoride would generate tetrahedral
anionic species 11, which would decrease the energy level by
5.4 kcal molÀ1 for monofluoride form 11a, 9.5 kcal molÀ1 for
difluoride form 11b, and 14.3 kcal molÀ1 for the trifluoride
form 11c. The energy of 11c is even lower than that of the
copper insertion forms 10d,c. Based on the calculated energies
of 10a–c, Cu(I) insertion would release less energy from
fluoride-protected tetrahedral forms 11a–c to 10a–c than from
phenylboronic acid 9 to 10d. The transition of the trifluoride
form 11c (À14.3 kcal molÀ1) to 10c (À11.2 kcal molÀ1) is an
energy gaining process. Therefore, fluoride addition decreases
(or sometimes reverses) the energy drives for copper insertion
into a B–C bond.
Financial support from the Molecular Basis of Disease
program at Georgia State University, Georgia Cancer
Coalition, Georgia Research Alliance and the National
Institutes of Health (CA123329, CA113917, GM086925 and
GM084933) are gratefully acknowledged. We also thank
Frontier Scientific, Inc. for providing boronic acids to us.
Notes and references
1 N. Miyaura and A. Suzuki, J. Chem. Soc., Chem. Commun., 1979,
866.
2 D. G. Hall, Boronic Acids, Wiley-VCH, Weinheim, 2004.
3 W. Yang, X. Gao and B. Wang, Med. Res. Rev., 2003, 23, 346–368.
4 J. Yan, H. Fang and B. Wang, Med. Res. Rev., 2005, 25, 490–520.
5 T. D. James and S. Shinkai, Top. Curr. Chem., 2002, 218, 159–200.
6 S. Jin, Y. F. Cheng, S. Reid, M. Li and B. Wang, Med. Res. Rev.,
2009, DOI: 10.1002/med.20155.
7 C. W. Tornøe, C. Christensen and M. Meldal, J. Org. Chem., 2002,
67, 3057–3064.
8 W. G. Lewis, L. G. Green, F. Grynszpan, Z. Radic, P. R. Carlier,
P. Taylor, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed.,
2002, 41, 1053–1057.
9 M. Li, N. Lin, Z. Huang, L. Du, H. Fang, C. Altier and B. Wang,
J. Am. Chem. Soc., 2008, 130, 12636–12638.
10 D. M. T. Chan, K. L. Monaco, R. Li, D. Bonne, C. G. Clark and
P. Y. S. Lam, Tetrahedron Lett., 2003, 44, 3863–3865.
11 P. Y. S. Lam, C. G. Clark, S. Saubern, J. Adams, M. P. Winters,
D. M. T. Chan and A. Combs, Tetrahedron Lett., 1998, 39,
2941–2944.
12 C. R. Cooper, N. Spencer and T. D. James, Chem. Commun., 1998,
1365–1366.
13 H. Zhao, L. Dang, T. B. Marder and Z. Lin, J. Am. Chem. Soc.,
2008, 130, 5586–5594.
14 H. K. Prokopcov and C. O., Angew. Chem., Int. Ed., 2008, 47,
3674–3676.
15 L. J. Goossen, D. Koley, H. L. Hermann and W. Thiel,
Organometallics, 2006, 25, 54–67.
16 E. Cances, B. Mennucci and J. Tomasi, J. Chem. Phys., 1997, 107,
3032–3041.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 5251–5253 | 5253