Discovery and characterization of catalysts for azide-alkyne cycloaddition by fluorescence quenching

Article abstract of DOI:10.1021/ja048425z

Copper-based catalysts for the 1,3-dipolar cycloaddition of azides and alkynes were screened in parallel fashion using a fluorescence quenching assay. The method was designed to identify systems able to accelerate the coupling of reactants at micromolar concentrations in aqueous mixtures and to obtain quantitative comparisons of their activities. In addition to the tris(triazolylamines) previously reported, two types of compounds (bipy/phen and 2-pyridyl Schiff bases) were found to exhibit significant ligand-accelerated catalysis, with one complex showing especially dramatic rate enhancements. Preliminary explorations of the dependence of reaction rate on pH, ligand:Cu ratio, and Cu concentration are described. Copyright

Full text of DOI:10.1021/ja048425z

Published on Web 07/08/2004
Discovery and Characterization of Catalysts for Azide-Alkyne Cycloaddition
by Fluorescence Quenching
Warren G. Lewis, Fernando G. Magallon, Valery V. Fokin, and M. G. Finn*
Department of Chemistry and The Skaggs Institute for Chemical Biology, The Scripps Research Institute,
10550 North Torrey Pines Road, La Jolla, California 92037
Received March 19, 2004; E-mail: mgfinn@scripps.edu
The 1,3-dipolar cycloaddition of alkynes and azides1 has emerged
as a highly useful connection method in both uncatalyzed2 and
copper(I)-catalyzed3 forms. The practical importance of the process
derives from the fact that it is the only facile 1,3-dipolar reaction
which uses chemically stable components; others generally employ
at least one reactant that is highly energetic, water-sensitive, or
transient in nature.4 This feature makes the azide-alkyne process
highly tolerant of diverse functional groups and reaction conditions.
Copper(I) catalysis of this process, discovered in 2002,3 has been
adopted by many groups for bioconjugation applications,5 in which
chemoselectivity and high reaction rate are crucial requirements.
To extend these potential applications we have focused on catalysis
of the azide-alkyne cycloaddition with water-soluble copper(I)
systems.3a Here we describe two new families of catalysts discov-
ered by a screening method that allows for measurements of relative
and absolute rates in a convenient microtiter-plate format.
Catalysis of the cycloaddition reaction of dansyl azide fluoro-
phore 1 and dabsyl alkyne 2 in aqueous-organic mixtures induces
a dramatic disappearance of fluorescence, since the triazole product
Figure 1. Fluorescence quenching assay. The microtiter plate was
3 exhibits intramolecular quenching of the dansyl unit by the dabsyl
visualized under long-wavelength ultraviolet irradiation after a reaction time
of 20 min. For all reactions, [Na ascorbate] ) 16 mM, [1] ) [2] ) [4] )
(Figure 1).6 The effect is easily visualized by eye under long-
0.031 mM, [ligand] ) 77 µM, solvent ) 38% DMSO + 62% Tris buffer
(pH 7.5). SDS (sodium dodecyl sulfate) was included to verify that
surfactants do not influence fluorescence intensity.
wavelength ultraviolet illumination or quantitated by fluorescence
spectroscopy. Screening of candidate ligands in parallel format was
done first by simple visual imaging, which offers a rapid readout
with simultaneous evaluation of control reactions. Interesting
candidates were then analyzed more carefully using a 96-well
fluorescence plate reader, which enabled kinetic characterization.7
The use of 1 and 2 at mid-micromolar concentrations in DMSO/
water provided emission intensities in a convenient range for visual
and plate reader detection; these compounds are completely
unreactive in the absence of CuI. The organic cosolvent was used
to improve both the fluorescence intensity of 1 and the solubility
of several classes of ligands tested. Compound 4, with a propyl
group in place of the reactive propargyl of 2, served as an unreactive
quencher to control for false positive readings caused by inter-
molecular quenching, absorbance, or catalyst precipitation, as was
observed for certain highly colored copper complexes. A typical
experiment involved side-by-side comparisons of the reaction of
interest (1 + 2 + catalyst) and a control set of reactions (1 + 4 +
catalyst) at room temperature over the same range of copper and/
or ligand concentrations with constant concentrations of fluorophore
and quenchers. An example is shown in Figure 1, with a positive
indication of catalysis being the disappearance of fluorescence only
in the top half of the plate. Ligand-accelerated catalysis8 was
suspected when the apparent rate of reaction in the presence of
ligand was greater than in its absence. A large excess of copper
relative to ligand was used to boost the rate of the ligand-free
process and thus make the screen for ligand acceleration more
demanding.
Figure 1 shows one example of fluorescence quenching in the
control wells (ligand 6), thereby comprising a case that cannot be
evaluated by this assay. Among the others, 5, 9, 10, and 12 were
ineffective, 7 and 8 marginally so. Tris(triazolyl)amine 11 was
previously identified as a useful ligand for bioconjugation
reactions,5a and the assay identified bipyridine copper complexes
such as those formed by 13 to be comparably potent catalysts. Thus,
catalytic activity persisted upon dilution of copper to 180 µM, a
level at which much slower cycloaddition was observed in the
absence of ligand.9
The same rapid assay was used to compare variations in the
bipyridine theme.11 Under otherwise identical conditions, the
relatively electron-rich ligands 13 and 14, as well as phenanthroline
15, were found to give 2- to 3-fold more rapid rates of cycloaddition
than bipyridine itself. The copper complex of neocuproine 16 is a
less active catalyst than unsubstituted phenanthroline, presumably
because of steric factors. Bathophenanthrolinedisulfonic acid, 17,
used in the specific colorimetric detection of copper(I),10 provided
an excellent and very water-soluble catalyst (see below).
Also screened was a combinatorial library of 88 Schiff bases
prepared from 11 aromatic aldehydes and 8 amines.11 Pyridine-
containing compounds were found to be promising, but several
candidates caused degradation of the fluorophore, and alternative
reducing agents were compromised by side reactions with the diazo
group of the quencher. The detection of side reactions underscores
9
9152
J. AM. CHEM. SOC. 2004, 126, 9152-9153
10.1021/ja048425z CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
In summary, the fluorescence quenching screening method has
led to the discovery and quantitative kinetic characterization of
ligand-accelerated catalysis of the Cu(I) azide-alkyne cycloaddition
reaction under dilute aqueous conditions. Bipyridine-type com-
pounds have been found for the first time to be components of
effective catalysts, and an opposing pH dependence has been
observed for bipy vs triazolylamine compounds. Further studies
directed toward the discovery and application of new ligands are
underway, as are investigations of the reaction mechanism.
Figure 2. Observed second-order rates of reaction at two different ligand
concentrations. The concentrations of azide, alkyne, copper, and ascorbate
were 18 µM, 18 µM, 182µM and 18 mM, respectively.
Acknowledgment. We are grateful to The Skaggs Institute of
Chemical Biology at The Scripps Research Institute for support of
this work; W.G.L. is a Skaggs Predoctoral Fellow. We thank Prof.
K. Barry Sharpless for valuable discussions, Mr. Timothy Chan
for a generous contribution of tris(triazolyl)amines 11, 21, and 22,
Prof. Floyd E. Romesberg for use of the fluorescent plate reader,
and Dr. Sayam Sen Gupta for valuable suggestions and preliminary
work on bioconjugation applications.
the importance of negative controls in ligand screening (here
represented by the use of 4), as well as the potential diversity of
copper(I) reactivity in catalyst libraries. After elimination of false
positives, clear ligand-accelerated catalysis was observed for Schiff
bases such as 18-20, each containing a pyridine group in position
to form a chelating interaction. These systems are being further
refined.
Reactions in microtiter plates were followed quantitatively by
measurement of emission intensity vs time; linear fits for simple
pseudo-second-order kinetics (Cu and ligand in excess with respect
to azide and alkyne) were obtained over the first hour of each
reaction.11 The apparent second-order rate constants thus obtained
allowed quantitative comparisons between ligands at varying pH,
as shown in Figure 2. Rate accelerations of more than 2 orders of
magnitude were observed relative to the rate of the copper-catalyzed
reaction in the absence of ligand.12 The ligand-free process was
accelerated as pH was increased from 7.5 to 8.5, as were many of
the reactions in the presence of ligand. However, tris(triazolyl)-
amine additives13 11, 21, and 22 showed the opposite trend, with
the fastest reactions observed at pH 7.5. The most active catalyst
was that containing bathophenanthroline 17 at pH 8.5.
Kinetics measurements were performed as a function of the 17:
Cu ratio.11 The second-order rate constant was observed to peak at
a 17:Cu value of 2.0 at two different copper concentrations,
suggesting the involvement of two ligands per metal center and
the ability of an extra ligand to inhibit the reaction. Furthermore,
the kinetic rate order in this complex was found to be 2.0 ( 0.1,11
suggesting that two copper centers may be required for catalytic
turnover.
The two most promising new systems identified above were
found to perform well in preparative and bioconjugation applica-
tions. Particularly important was the observation that 17 is highly
effective in promoting the attachment of a dye-alkyne to a virus
particle decorated with azides in analogy to previous results obtained
with 11,5a which is much less soluble in water. A full account of
this type of application will appear elsewhere.
Supporting Information Available: Experimental details and
additional figures. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) (a) Huisgen, R. In 1,3-Dipolar Cycloaddition Chemistry; Padwa, A., Ed.;
Wiley: New York, 1984; Vol. 1, pp 1-176. (b) Huisgen, R. Pure Appl.
Chem. 1989, 61, 613-628.
(2) (a) Mock, W. L.; Irra, T. A.; Wepsiec, J. P.; Manimaran, T. L. J. Org.
Chem. 1983, 48, 3619-3620. (b) Mock, W. L.; Irra, T. A.; Wepsiec, J.
P.; Adhya, M. J. Org. Chem. 1989, 54, 5302-5308. (c) Lewis, W. G.;
Green, L. G.; Grynszpan, F.; Radic, Z.; Carlier, P. R.; Taylor, P.; Finn,
M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2002, 41, 1053-1057.
(d) Wang, Z.-X.; Qin, H.-L. Chem. Commun. 2003, 2450-2451.
(3) (a) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2002, 41, 2596-2599. (b) Tornøe, C. W.; Christensen,
C.; Meldal, M. J. Org. Chem. 2002, 67, 3057-3062.
(4) Carruthers, W. In Cycloaddition Reactions in Organic Synthesis; Pergamon
Press: New York, 1990; pp 270-331.
(5) (a) Wang, Q.; Chan, T. R.; Hilgraf, R.; Fokin, V. V.; Sharpless, K. B.;
Finn, M. G. J. Am. Chem. Soc. 2003, 125, 3192-3193. (b) Speers, A. E.;
Adam, G. C.; Cravatt, B. F. J. Am. Chem. Soc. 2003, 125, 4686-4687.
(c) Fazio, F.; Bryan, M. C.; Blixt, O.; Paulson, J. C.; Wong, C.-H. J. Am.
Chem. Soc. 2002, 124, 14397-14402. (d) Deiters, A.; Cropp, T. A.;
Mukherji, M.; Chin, J. W.; Anderson, J. C.; Schultz, P. G. J. Am. Chem.
Soc. 2003, 125, 11782-11783. (e) Link, A. J.; Tirrell, D. A. J. Am. Chem.
Soc. 2003, 125, 11164-11165.
(6) Zaramella, A.; Conti, N.; Dal Cin, M.; Paio, A.; Seneci, P.; Gehanne, S.
J. Comb. Chem. 2001, 3, 410-420.
(7) A variety of fluorescence methods have been used for catalysis screening.
For example, see: Stambuli, J. P.; Stauffer, S. R.; Shaughnessy, K. H.;
Hartwig, J. F. J. Am. Chem. Soc. 2001, 123, 2677-2678; Copeland, G.
T.; Miller, S. J. J. Am. Chem. Soc. 1999, 121, 4306-4307. Additional
contributions are listed in Supporting Information.
(8) Berrisford, D. J.; Bolm, C.; Sharpless, K. B. Angew. Chem., Int. Ed. Engl.
1995, 34, 1059-1070.
(9) Copper-bipyridine complexes are also well-known catalysts for atom
transfer radical polymerization (ATRP) reactions; see Johnson, R. M.;
Ng, C.; Samson, C. C. M.; Fraser, C. L. Macromolecules 2000, 33, 8618-
8628.
Ligands may assist the azide-alkyne cycloaddition process in
at least two ways: stabilization of the CuI oxidation state and
acceleration of the reaction itself. We did not screen for the former
function, since an excess of ascorbate was maintained throughout.
Thus, active ligands identified here have a direct role in promoting
the reaction, but may not be useful under oxidizing conditions.
Indeed, reactions using 17 without excess reducing agent are
somewhat more air-sensitive than those of 11.
(10) Faizullah, A. T.; Townshend, A. Anal. Chim. Acta 1985, 172, 291-296.
(11) See Supporting Information for details.
(12) If used directly, the values shown in Figure 2 dramatically understate the
magnitude of ligand-accelerated catalysis, since the concentration of
uncomplexed Cu is much higher than that of Cu-ligand species in all
cases.
(13) Chan, T. R.; Hilgraf, R.; Sharpless, K. B.; Fokin, V. V. Org. Lett. 2004,
in press.
JA048425Z
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9153