Fluorogenic transformations based on formation of C-C bonds catalyzed by palladium: An efficient approach for high throughput optimizations and kinetic studies

Article abstract of DOI:10.1002/adsc.200700384

We have developed novel fluorogenic transformations based on formation of C-C bonds catalyzed by palladium using iodocoumarin 1 as a model aryl iodide, where fluorescence is quenched completely due to effects of the heavy, polarizable iodine atom. Substitution of the iodine atom for the carbon using Sonogashira, Suzuki-Miyaura and Heck couplings results in a dramatic fluorescence enhancement. This approach has been used successfully for the optimization of reaction conditions and kinetic studies in high throughput format.

Full text of DOI:10.1002/adsc.200700384

FULL PAPERS
DOI: 10.1002/adsc.200700384
À
Fluorogenic Transformations Based on Formation of C CBonds
Catalyzed by Palladium: An Efficient Approach for High
Throughput Optimizations and Kinetic Studies
Roman V. Rozhkov,a,* V. Jo Davisson,a and Donald E. Bergstroma
a
Bindley Bioscience Center, Birck Nanotechnology Center and Department of Medicinal Chemistry and Molecular
Pharmacology, Purdue University, West Lafayette, IN 47907-2091, USA
Fax : (+1)-765-496-3601; e-mail: romanr@pharmacy.purdue.edu
Received: August 2, 2007; Published online: December 14, 2007
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: We have developed novel fluorogenic Heck couplings results in a dramatic fluorescence en-
À
transformations based on formation of C C bonds hancement. This approach has been used successfully
catalyzed by palladium using iodocoumarin 1 as a for the optimization of reaction conditions and kinet-
model aryl iodide, where fluorescence is quenched ic studies in high throughput format.
completely due to effects of the heavy, polarizable
À
iodine atom. Substitution of the iodine atom for the Keywords: C C coupling; fluorescence; homogene-
carbon using Sonogashira, Suzuki–Miyaura and ous catalysis; palladium
Introduction
tivity at higher conversions, crucial for efficiency
screening, is lower due to fluorolytic (based on disap-
Formation of carbon-carbon bonds catalyzed by palla- pearance of the signal) nature of the method. Here
dium is of a great importance for organic synthesis, we report our studies on novel fluorogenic carbon-
pharmaceutics and material science.[1–5] Development carbon couplings catalyzed by palladium, where the
and study of these transformations are associated fluorescence signal increases during the course of the
with extensive and time-consuming optimizations of reaction, and their utility for high throughput optimi-
numerous variables including catalyst, solvent, base, zations and kinetic studies.
ligand, additives, stoichiometry and temperature. An-
alytical methods, extensively used for screening and
optimization of transition metal-catalyzed transforma- Results and Discussion
tions, such as HPLC,[6,7] electrophoresis,[8] and IR ther-
mography[9] offer quantitative approaches while re- Although the effects of heavy atoms and electron-do-
quiring significant commitment of time and instru- nating/-withdrawing groups on fluorescence are well
mentation.
known,[14] only a few examples of fluorogenic trans-
Fluorescent screening methods, developed recently formations, based on formation of new bonds includ-
by Hartwig and Taran,[10] are based upon immobiliza- ing conversion of phosphine into phosphine oxide at
tion of fluorescent dyes on a solid support,[10] FRET the C-3 position[15] and azide/alkyne into 1,2,3-tri-
quenching of the fluorophore by an incoming quench- azoles at the C-3 and C-7 positions of coumarin,[16,17]
er,[11,12] and antibody-based immunoassays.[13] These have been reported. We envisioned that the presence
methods utilize short fluorescence relaxation times of the iodine at the C-6 position of the coumarin 1
and allow high throughput screening of bases, sol- would quench fluorescence due to the effect of the
vents, ligands and temperatures within the fixed set of heavy iodine atom. Additional quenching is expected
reactants conjugated to dye and quencher.
from steric factors, which may cause deflection of the
However, these methods suffer from several draw- diethylamino group from a planar conformation with
backs: (1) derivatization of both reactants is required, respect to the aromatic ring and reduce the efficiency
thus limiting screening scope; (2) optical properties of of the pull-push chromophore. Substitution of the
À
both FRET partners in various solvents at different iodine atom for carbon via palladium-catalyzed C C
pH may vary and alter results significantly; (3) sensi- coupling is expected to eliminate or significantly
Adv. Synth. Catal. 2008, 350, 71 – 75
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
71

Roman V. Rozhkov et al.
FULL PAPERS
Table 1. Fluorogenic Sonogashira transformations.
reduce these quenching effects and reveal fluores-
cence.
[a]
Dye (R)
Yield [%]
(time, [h])
Fim [nm]
EFi[b] [a.u.]
Our first attempts to synthesize iodocoumarin 1
(Scheme 1) via iodination of the corresponding sali-
cylic aldehyde derivative 2 by iodine, ICl and NIS
(step a) led to a complex, inseparable mixture of
products. Attempts to iodinate coumarin 4 directly
using previously mentioned iodinating reagents in var-
ious solvents were unsuccessful.
7a (Ph)
97 (4)
91 (4)
94 (4)
99 (4)
95 (4)
74(48)
465 nm
469 nm
467 nm
465 nm
464 nm
460 nm
27.4
36.2
3.3
7b (C6H4OMe)
7c (C6H4NMe2)
7d (C6H4F)
7e (C6H4CO2Me)
7f (C5H11)
25.2
26.2
12.0
According to several reports, the C-6 position of
the coumarin ring can be activated for electrophilic
aromatic substitution through opening of the lactone
ring and formation of the phenoxide 5 under basic
conditions.[18–21] When coumarin 4 was dissolved in
5% aqueous sodium hydroxide (step c), followed by
direct iodination with iodine (step d) and lactone ring
closure with hydrochloric acid (step e), nearly quanti-
tative yield of the desired coumarin 1 was obtained.
As predicted, the iodinated coumarin 1 was non-
fluorescent. Furthermore, Sonogashira transformation
of coumarin 1 with phenylacetylene produced a
[a]
F
im =fluorescence maximum.
[b]
EFi =fluorogenic efficiency [Eq. (1)]; Fi(l) and F1(l)=
fluorescence intensity of fluorescent product 7 and fluo-
rogenic dye 1 at wavelength l, respectively.
highly fluorescent product 7a in excellent yield tion proceeded slower giving the desired coumarin 7f
(Figure 1, Table 1). Next, we explored the generality in 74% yield after 48 h. In all studied cases, with the
of this process with respect to various terminal al- exception of 7c where fluorescence is presumably
kynes. Aryl substituted alkynes all gave the desired quenched by internal charge transfer (ICT),[14] Sono-
coumarins 7b–e in nearly quantitative yields within gashira transformation led to the formation of highly
4 h. In case of an alkyl-substituted alkyne, the reac- fluorescent products.
The efficiency of the fluorogenic transformations
was evaluated by integrating the ratio of fluorescence
intensities for dyes 7a–f to that of dye 1 within 420–
550 nm region upon 365 nm excitation [Eq. (1)].
Aryl-substituted products 7a, 7b, 7d, 7e have high flu-
orogenic efficiency ranging from 25.2 to 36.2 a.u.
which is comparable with the “ideal” fluorogenic
transformation from dye 1 to the commercial fluores-
cent dye 4 (51 a.u).
In order to transform this newly developed fluoro-
genic transformation into an efficient and reliable
method for kinetic and high throughput optimization
studies, the effect of various factors on fluorescence
intensity was examined. First, we have evaluated the
impact of starting material and product concentra-
tions on total fluorescence of the model dye 7a using
Scheme 1. Synthesis of fluorogenic dye 1.
365 nm excitation and 465 nm emission wavelengths.
Plots of the dilution series for both 7a and 7a+1 are
linear with a correlation coefficient exceeding 0.99 as
shown in Figure 2, establishing the lack of optical in-
terferences between reaction components and linear
dependence of fluorescence intensity on the concen-
tration of the fluorescent product 7a within the stud-
ied range.
Since solvents have pronounced effects on fluores-
cence due to solvation of the exited states,[22] we ex-
amined the effect of solvents suitable for polystyrene
96-well plates including water, acetonitrile, methanol
and ethanol on fluorescence intensity in 1–10 mM
Figure 1. Fluorogenic Sonogashira transformations.
range. Fluorescence measurements in water were not
72
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 71 – 75

À
Fluorogenic Transformations Based on Formation of C C Bonds Catalyzed by Palladium
FULL PAPERS
&
Figure 2. Calibration curves and inner filter effect; ( ) 7a,
!
~
&
*
^
Figure 3. First-order plots: ( ) 7a, ( ) 7b, ( ) 7d ( ) 7e, ( )
7f.
*
( ) 7a+1.
Table 2. Reactivity of various alkynes.
possible due to the low solubility of the specific dyes.
Since ethanol provides the highest fluorescence inten-
sity among solvents studied, this solvent has been se-
lected for further studies. The concentration of diiso-
propylethylamine in 1–1000 mMrange had virtually
no effect on fluorescence intensity in EtOH. We have
also examined effect of various palladium catalysts
Dye (R)
kobs [hÀ1]
R2
t1/2 [h]
7a (H)
7b (OMe)
7d (F)
7e (CO2Me)
7f (C5H11)
0.531Æ0.087
0.579Æ0.083
0.609Æ0.036
0.551Æ0.0621
0.199Æ0.029
0.962
0.970
0.995
0.982
0.970
1.31
1.20
1.14
1.25
3.49
such
(CH3CN)2PdCl2, Pd
on the fluorescence signal and observed no substantial
as
Pd
A
PdCl2,
(PPh3)2PdCl2,
AHCTREUNG
quenching effects at 5% loading with respect to the version values and occurs due to inhibition of catalyt-
10 mMfluorescent product 7a. Catalyst (dppf)PdCl2 ic palladium intermediates through coordination with
exhibited a modest quenching effect at 5% loading accumulating internal alkynes.[24]
due to presence of iron, but not at 1% loading.
A comparison of our fluorogenic method and
Therefore fluorescence correlates only with the con- HPLC for the transformation of dye 1 and phenylace-
centration of the product and is not effected signifi- tylene into product 7a, demonstrated no significant
cantly by the presence of other reaction components. differences between those methods at the 99.9% con-
Although the Sonogashira transformation is the fidence level. The applicability of the fluorogenic dye
classic example in palladium catalysis, data on the re- as a model aryl iodide has been examined through
activity of alkynes is scattered and mostly based on comparison of Sonogashira coupling between iodo-
reaction yields with variable reaction conditions.[3,23] benzene and coumarin 1 with phenylacetylene by
Since our method allows substrate variations, the next HPLC. No substantial difference in reactivity of these
step was to demonstrate its utility in addressing the substrates was observed (see Supporting Informa-
relative reactivity of alkynes in Sonogashira transfor- tion). There are also no obvious “abnormal” structur-
mations (Figure 3, Table 2).
al components that would distinguish fluorogenic dye
Reaction time courses for these reactions reveal 1 from other aryl iodides in transition metal-catalyzed
pseudo-first-order transformations with respect to the transformations.
fluorogenic coumarin 1, consistent with previously re-
We have also estimated the time required for the
ported results claiming transmetalation or reductive acquisition and analysis of the dataset presented in
elimination as rate-determining steps.[23,24] Although Figure 3 by our fluorogenic method and standard
an alkyl-substituted acetylene is considerably less re- HPLC with UV-visible or fluorescence detectors.
active than an arylacetylene, introduction of electron- Data collection for 5 data points with 5 repetitions for
donating or electron-withdrawing groups into the 5 kinetic curves requires around 100 h when acquired
para-position of the aryl had no pronounced effect on by HPLC, but only 0.2 h by our fluorogenic method.
the reaction rate. A moderate deviation from the Additionally, our method does not require extensive
first-order kinetic model is expected for practical con- calibrations and development of separation programs.
Adv. Synth. Catal. 2008, 350, 71 – 75
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
73

Roman V. Rozhkov et al.
FULL PAPERS
Since virtually no solvent is required, this fluorogenic Miyaura and Heck reactions from reactivity and fluo-
method is environmentally friendly.
rogenic efficiency standpoints.
Next, we have explored the generality of the fluo-
rogenic approach for other classic palladium-cata-
lyzed transformations such as Heck[1] and Suzuki– Conclusions
Miyaura[25] transformations. To our satisfaction, fluo-
rogenic dye 1 reacted with phenylboronic ester and In summary, we have envisioned and developed a
tert-butyl acrylate to form fluorescent products 8 and powerful approach for high throughput screening, op-
9 in 96 and 55% yields, respectively (Scheme 2, timization and kinetic studies of palladium-catalyzed
Figure 4).
carbon-carbon couplings involving aryl iodides. This
Reaction conditions were optimized through varia- method allows one to screen not only standard reac-
tion in temperature, base, catalyst and dilution factors tion variables such as bases, ligands, catalysts, solvents
in high throughput format using the fluorogenic and temperature but also substrates without any addi-
nature of these transformations with dye 1. Our opti- tional modifications or derivatizations. High function-
mized reaction conditions matched recently reported al group tolerance, mild “bioorthogonal” reaction
Suzuki–Miyaura coupling conditions for the synthesis conditions, excellent fluorogenic and chemical effi-
of ethyl ester of coumarin 8 from the corresponding ciency and submicromolar sensitivity enhance the im-
bromocoumarin in 72% yield.[26]
portance of these fluorogenic transformations for the
The fluorogenic efficiency of transformations lead- development and optimization of bioconjugations and
ing to the Sonogashira, Suzuki–Miyaura and Heck surface/nanoparticle derivatization. Further studies on
products 7b, 8 and 9, determined as integrated ratio the synthesis of novel fluorogenic dyes with high
of fluorescences from 420 to 550 nm, was 36.2, 8.0 and chemical stability and variable optical properties as
17.1, respectively. According to our results, the Sono- well as the development of novel fluorogenic transfor-
gashira reaction is more efficient than Suzuki– mations and their bioanalytical applications are un-
derway.
Experimental Section
Procedure for Typical Fluorogenic Sonogashira
Transformation
Coumarin 1 (0.25 mmol, 97 mg), alkyne (0.5 mmol), bis(di-
phenylphosphino)palladium dichloride (5%, 0.0125 mmol,
8.8 mg), CuI (10%, 0.025 mmol, 4.75 mg) and NEt3
(2 mmol, 202 mg) were dissolved in 2 mL of DMF and
stirred for 4 h (unless noted otherwise) at room tempera-
ture. Upon completion of the reaction, the solvent was
evaporated under vacuum and the solid residue was purified
by flash chromatography.
Scheme 2. Fluorogenic Suzuki–Miyaura and Heck couplings.
Fluorescent dye 7a: Obtained in 97% yield as light yellow
1
needles, mp 183–1858C; H NMR (CDCl3): d=1.33 (t, J=
7.0 Hz, 6H), 3.74 (q, J=7.0 Hz, 4H), 6.64 (s, 1H), 7.30–7.40
(m, 3H), 7.43–7.60 (m, 2H), 7.72 (s, 1H), 8.64 (s, 1H), 12.13
(s, broad, 1H); 13C NMR (CDCl3): d=13.1, 46.0, 87.1, 93.7,
100.5, 107.7, 108.9, 109.5, 122.7, 128.5, 128.6, 131.1, 138.4,
149.6, 155.3, 156.3, 163.8, 164.9; HR-MS: m/z=362.1395
(calcd. for C22H19NO4 +H: 362.1392).
Fluorogenic Suzuki Coupling
Coumarin 1 (0.25 mmol, 97 mg), phenylboronic acid pinacol
ester (1 mmol, 204 mg), 1,1’-[bis(diphenylphosphino)ferro-
cene]dichloropalladium(II) (5%, 0.0125 mmol, 10.2 mg),
and CsF (1 mmol, 152 mg) were dissolved in 2 mL of DMF
and stirred for 24 h at 608C. Upon completion of the reac-
tion, the solvent was evaporated under vacuum. The solid
residue was purified by flash chromatography and recrystal-
lized from an ethanol-water mixture.
Figure 4. Fluorescence spectra of 1, 4, 7b, 8 and 9 in EtOH
(10 mM).
74
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 71 – 75

À
Fluorogenic Transformations Based on Formation of C C Bonds Catalyzed by Palladium
FULL PAPERS
Fluorescent dye 8: Obtained in 96% yield as light yellow References
1
crystals, mp 135–1378C; H NMR (CDCl3): d=0.98 (t, J=
7.0 Hz, 6H), 3.10 (q, J=7.0 Hz, 4H), 6.90 (s, 1H), 7.30–7.50
(m, 6H), 8.72 (s, 1H), 12.32 (s, broad, 1H); 13C NM R
(CDCl3): d=12.1, 45.9, 104.9, 108.6, 111.3, 127.7, 127.9,
128.9, 132.1, 133.7, 140.0, 150.5, 156.0, 156.2, 163.7, 165.0;
HR-MS: m/z=337.1320 (calcd. for C20H19NO4: 337.1314).
[1] I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000,
100, 3009.
[2] J. Tsuji, Palladium in OrganicSynthesis , Springer-
Verlag, Berlin Heidelberg, 2002.
[3] E. Negishi, (Ed.), Handbook of Organopalladium
Chemistry for OrganicSynthesis , John Wiley and Sons,
Inc., 2002.
Fluorogenic Heck Coupling
[4] R. Chinchilla, C. Najera, Chem. Rev. 2007, 107, 874.
[5] L. X. Yin, J. Liebscher, Chem. Rev. 2007, 107, 133.
[6] K. Burgess, H. J. Lim, A. M. Porte, G. A. Sulikowski,
Angew. Chem. Int. Ed. 1996, 35, 220.
7-Diethylamino-6-iodocoumarin-3-carboxylic
(0.25 mmol, 97 mg), tert-butyl acrylate (2 mmol, 256 mg),
bis(diphenylphosphino)palladium dichloride
0.0125 mmol, 8.8 mg), and EtNACHTRE(UNG i-Pr)2 (1 mmol, 129 mg)
acid
(5%,
[7] A. M. Porte, J. Reibenspies, K. Burgess, J. Am. Chem.
were dissolved in 20 mL of DMF and stirred for 24 h at
408C. Upon completion of the reaction, the solvent was
evaporated under vacuum. The solid residue was purified by
flash chromatography and recrystallized from an ethanol-
water mixture.
Fluorescent dye 9: Obtained in 55% yield as yellow crys-
tals, mp 200–2028C; 1H NMR (CDCl3): d=1.19 (t, J=
7.0 Hz, 6H), 1.53 (s, 9H), 3.32 (q, J=7.0 Hz, 4H), 6.29 (d,
J=15.5 Hz, 1H), 6.86 (s, 1H), 7.63 (d, J=15.5 Hz, 1H), 7.66
(s, 1H), 8.74 (s, 1H); 13C NMR (CDCl3): d=12.42, 28.1,
46.73, 80.9, 105.0, 109.8, 111.5, 120.9, 126.2, 130.9, 140.2,
150.6, 156.5, 157.2, 163.2, 164.6, 165.7; HR-MS: m/z=
387.1686 (calcd. for C21H25NO6: 387.1682).
Soc. 1998, 120, 9180.
[8] M. T. Reetz, K. M. Kuhling, A. Deege, H. Hinrichs, D.
Belder, Angew. Chem. Int. Ed. 2000, 39, 3891.
[9] S. J. Taylor, J. P. Morken, Science 1998, 280, 267.
[10] K. H. Shaughnessy, P. Kim, J. F. Hartwig, J. Am. Chem.
Soc. 1999, 121, 2123.
[11] S. R. Stauffer, N. A. Beare, J. P. Stambuli, J. F. Hartwig,
J. Am. Chem. Soc. 2001, 123, 4641.
[12] S. R. Stauffer, J. F. Hartwig, J. Am. Chem. Soc. 2003,
125, 6977.
[13] P. Vicennati, N. Bensel, A. Wagner, C. Creminon, F.
Taran, Angew. Chem. Int. Ed. 2005, 44, 6863.
[14] J. R. Lackowitz, Principles of Fluorescence Spectrosco-
py, 3rd edn., Springer Science, New York, 2006.
[15] G. A. Lemieux, C. L. de Graffenried, C. R. Bertozzi, J.
Am. Chem. Soc. 2003, 125, 4708.
[16] Z. Zhou, C. J. Fahmi, J. Am. Chem. Soc. 2004, 126,
8862.
[17] K. Sivakumar, F. Xie, B. M. Cash, S. Long, H. N. Barn-
hill, Q. Wang, Org. Lett. 2004, 6, 4603.
[18] N. H. Cairns, M. Laurence, David P. Astles, Alex Orr,
J. Chem. Soc., Chem. Commun. 1986, 2, 182.
[19] E. L. Zubia, F. Rodriguez, G. M. Massanet, I. G. Colla-
do, Tetrahedron 1992, 48, 4239.
[20] V. F. Traven, R. V. Rozhkov, A. Yu. Tolmachev, N. A.
Kuznezova, N. Ya. Podhaluzina, E. A. Carberry, Het.
Commun. 1997, 3, 339.
Kinetic Experiments
A set of polypropylene PCR tubes was charged with a solu-
tion of coumarin 1 (40 mL, 50 mM), alkyne (40 mL, 200 mM),
(PPh)3PdCl2 (1%, 40 mL, 0.5 mM), copper iodide (1%, 40
mL, 0.50 mM) and ethyldiisopropylamine (40 mL, 200 mM)
in DMF. Final concentrations of the fluorogenic dye, alkyne,
palladium complex, copper iodide and amine were 10, 40,
0.1, 0.1 and 40 mM, respectively.
Next, samples were capped and heated at 508C and 20 mL
aliquots were withdrawn at 0, 0.25, 0.5, 1, 2, 4, 8 and 16 h
time points. Aliquots taken were diluted by a factor of 10
using acetonitrile and by a factor of 1000 using ethanol for
HPLC with UV-visible detection and fluorescence polystyr-
ene 96-well plate experiments, respectively. Samples were
stored at À208C in capped PCR tubes prior to the analysis
to stop further progress of the reaction. Detailed analysis is
presented in Supporting information.
[21] R. V. Rozhkov, A. V. Vasiliev, V. F. Traven, Het.
Commun. 1998, 4, 567.
[22] H. Sunahara, Y. Urano, H. Kojima, T. Nagano, J. Am.
Chem. Soc. 2007, 129, 5597.
[23] E. Negishi, L. Anastasia, Chem. Rev. 2003, 103, 1979.
[24] C. Amatore, S. Bensalem, S. Ghalem, A. Jutand, Y.
Medjour, Eur. J. Org. Chem. 2004, 366.
[25] A. Suzuki, Proc. Jap. Acad. B: Phys. Biol. Sci. 2004, 80,
359.
Acknowledgements
The Authors thank Professor Christoph J. Fahrni for valua-
ble discussions about the fluorescence of synthesized coumar-
ins. This research was supported by the National Institutes of
Health.
[26] T. Hirano, K. Hiromoto, H. Kagechika, Org. Lett. 2007,
9, 1315.
Adv. Synth. Catal. 2008, 350, 71 – 75
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
75