1,9-Dihydro-3-phenyl-4H-pyrazolo[3,4-b]quinolin-4-one, a novel fluorescent probe for extreme pH measurement

Article abstract of DOI:10.1039/b101685g

The synthesis of 1,9-dihydro-3-phenyl-4H-pyrazolo[3,4-b]quinolin-4-one and its pH-dependent fluorescent properties for extreme pH measurement are presented.

Full text of DOI:10.1039/b101685g

1,9-Dihydro-3-phenyl-4H-pyrazolo[3,4-b]quinolin-4-one, a novel
fluorescent probe for extreme pH measurement
Meihong Su, Yong Liu, Huimin Ma,* Quanli Ma, Zhihua Wang, Junlin Yang and Meixiang
Wang*
Center for Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100080,
China. E-mail: mhmliang@public.bta.net.cn
Received (in Cambridge, UK) 21st February 2001, Accepted 19th April 2001
First published as an Advance Article on the web 10th May 2001
The synthesis of 1,9-dihydro-3-phenyl-4H-pyrazolo[3,4-
b]quinolin-4-one and its pH-dependent fluorescent proper-
ties for extreme pH measurement are presented.
and three nitrogen atoms and one oxygen atom were set in the
probe 5. To achieve the fluorescent response to extreme pH
values, the environmental difference among the electronegative
atoms in the conjugated molecule should be as large as possible.
The potential binding sites for H+ in 5 were therefore placed in
quite different environments. Particularly following the known
data,11 one nitrogen atom [e.g. N(9)] was arranged in an
electron-deficient position and another one N(1) in an electron-
rich position, expecting the generation of considerably different
pKa values. In addition, for improved selectivity, the arrange-
ment of all the electronegative atoms in the structure should not
provide a suitable cavity or a convenient formation of five- and
six-membered ring complexes for metal ions. As shown in
Scheme 1, the prepared probe 5 does not possess any favorable
complexation sites for metal ions. The benzene ring in the
position 3 of the probe would render a steric hindrance for any
possible complexation of either the adjacent oxygen or nitrogen
atom with metal ions.
The experimental results showed that the probe has a notable
fluorescence quantum yield of obase = 0.14 in basic media (pH
= 13.0) or oacid = 0.12 in acidic media (pH = 1.0) with an
appreciable fluorescence lifetime of 11.12 ± 0.04 ns,12 and its
fluorescence spectra are highly dependent on pH (Fig. 1).
Titration of fluorescence intensity with pH gave two pKa values
of 2.61 and 12.44 (Fig. 1, insets),13 which correspond to the
protonations of the two nitrogen atoms N(9) and N(1) in the
probe, respectively. It is understandable that only two of the
electronegative atoms exhibit pH-sensitivity, because the
basicity of the other two (one nitrogen and one oxygen) is
largely weakened by their lone electron pairs participating in the
system conjugation. As a result, the probe is capable of
measuring the extreme pH values over the two pH ranges of
1.8–3.4 and 11.6–13.3, respectively. Further, the probe was
stable and no obvious changes in fluorescence were observed
within 5 months at rt. It should be pointed out that the probe is
unsuitable for neutral pH measurements, since a nonlinear
response of fluorescence intensity to pH was observed in that
region.
The accurate measurement of pH is very important because it
usually plays a key role in a variety of systems. The most
popular and direct device for pH measurement is the glass pH
electrode. However, the known limitations of the glass pH
electrodes (e.g. its electrical interference or mechanical damage
to small cells, and the presence of acid error and especially
alkaline error1) make them unsuitable for certain applications:
intracellular pH and microscopy studies as well as the
measurements of extreme pH values below 1 or above 9. In
contrast to the electrochemical methods, optical measurements,
based on fluorescent probes that are either protonated or
deprotonated, have no such drawbacks.2 Moreover, fluorescent
measurements are convenient to microscopy studies, and can
reflect the H+ distribution and change within cells.3 It is not
surprising, therefore, that the researches on pH-dependent
fluorescent probes have received a great attention,4 particularly
on the probes which are pH-sensitive to the near neutral pH
value of normal body fluids for the point of biological
application. To the best of our knowledge, however, relatively
less attention was paid on the fluorescent probes which are pH-
sensitive in the lower pH region (pH < 5) or the higher pH
region (pH > 9), though in some cases pH changes (e.g. within
the stomach) can enter the extreme pH ranges.1,5 It is also a
challenge to design a fluorescent probe with linear response
over a broad range, because pH measurement can be accurately
made only over a range of about two pH units, i.e. pKa ± 1.
Although some attempts have been made to broaden the
response range of pH measurement by using a mixture of
multiple pH indicators,6 this makes the system rather complex.
Another feasible approach to the problem is to develop a probe
with multiple steps of H+ binding.7 Unfortunately, the typically
used pH-dependent groups such as –COOH, –OH, etc., have a
high affinity for common metal ions, thus resulting in
complexation and nonspecific response of the probe to H+.
The objective of this research was to design a novel
fluorescent probe which is both pH-selective and pH-sensitive
for extreme pH ranges, by assembling several electronegative
atoms in the different positions of a highly conjugated molecule.
Scheme 1 shows the synthetic route to such a fluorescent probe,
1,9-dihydro-3-phenyl-4H-pyrazolo[3,4-b]quinolin-4-one 5.8
The reaction of 2-benzoylketene dithioacetal 19 with 2 gave
methylthio-substituted quinolone 3. Upon treatment with hydra-
zine, 3 was converted quantitatively to hydrazone 4 which was
then transformed into the designed product 5 by further heating
in pyridine.
To test the selectivity of the probe, the effects of various
diverse ions upon the emission spectra were examined. The
results showed that the selectivity of this probe for H+ over other
In order to increase the selectivity for H+, the use of carboxyl
which is easy to complex with metal ions was avoided. pH-
dependent amino groups, though they are also strong ligands for
metal ions, were mainly chosen as H+ receptors because of their
convenient arrangement in synthesis and the wealth of proton
binding data available.10 The arrangement of multiple H+
receptors is conducive to obtaining a broad pH response range,
Scheme 1 Reagents and conditions: (a) propanoic acid, reflux 48 h; (b)
hydrazine hydrate (30%), ethanol reflux; (c) anhydrous pyridine, argon
protection, reflux 60 h.
960
Chem. Commun., 2001, 960–961
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b101685g

Table 1 The tolerable concentration of foreign ions for the measurement of pHa
2
32
22
22
Mn+
Ca2+
Cu2+
Fe2+
Fe3+
K+
Mg2+
Mn2+
Zn2+
NO3
PO4
CO3
SO4
c/1024 Mb
c/1025 Mc
10
20
10
4.5
0.70
3.6
0.40
3.4
20
20
0.80
0.32
40
0.20
2.6
2.2
0.50
1.0
2.0
1.0
0.20
8.3
4.0
20
a The tolerable concentration was estimated by the criterion at which a species gave a relative error of no more than 5% in the analytical signal (fluorescence
intensity) of the probe (2.0 3 1023 g dm23) in the presence of 0.1 M NaCl. b Measured in acid medium (pH = 2.0). c Measured in basic medium (pH =
12.0).
serve as both H+ receptors and electron donors in the PET
process. In acid media, the probe 5 has a longer fluorescence
emission of 493 nm at the optimal excitation of 385 nm (Fig.
1A), and the H+ binding equilibrium of the N(9) atom yields an
isosbestic point at 462 nm. Further protonation of the probe may
inhibit the related PET, resulting in fluorescence enhancement
(Fig. 1, curve a). On the other hand, the deprotonation of the
N(1) atom in basic media presumably revives the corresponding
PET, causing fluorescence quenching at 433 nm (Fig. 1B).
In summary, a novel fluorescent pH probe 5 has been
prepared by arranging several electronegative atoms in the
different environments of a conjugated molecule. Although this
probe is far from being an ideally broad pH sensor, it greatly
complements glass pH electrodes. Further work in this area
would be beneficial to designing more excellent fluorescent pH
probes.
Financial supports from the NNSF of China and the CMS
fund of CAS are acknowledged.
Notes and references
1 G. Mattock and G. R. Taylor, pH Measurement and Titration,
Heywood, London, 1961.
2 C. Chan, W. Lo and K. Wong, Biosensors Bioelectron., 2000, 15, 7.
3 A. Srivastava and G. Krishnamoorthy, Anal. Biochem., 1997, 249,
140.
4 J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 2nd Edn.,
Kluwer Academic/Plenum Publishers, New York, 1999.
5 S. Ohkuma and B. Poole, Proc. Natl. Acad. Sci. USA, 1978, 75, 3327.
6 J. Lin and D. Liu, Anal. Chim. Acta, 2000, 408, 49.
7 O. S. Wolfbeis and H. Marhold, Fresenius Z. Anal. Chem., 1987, 327,
347.
8 All new compounds 3–5 were fully characterized on the basis of their
spectral (NMR, IR, MS) data and C–H–N analysis. Selected data for 5:
dH (300 MHz, 340 K, DMSO-d6) 11.8 (1H, s, NH), 8.32 (2H, d, J 7.1),
8.19 (1H, d, J 7.9), 7.58 (1H, t, J 7.3), 7.37–7.47 (4H, m), 7.14 (1H, t,
J 7.5); vmax (KBr)/cm21 3132, 3059, 1631, 1573; EI mass spectrum
(relative intensity), m/z 261 (100, M+), 260 (45); Anal. Calcd for
C16H11N3O: C, 73.55; H, 4.24; N, 16.08. Found: C, 73.35; H, 4.40; N,
16.01%.
9 Z. T. Huang and Z. R. Liu, Synth. Commun., 1989, 19, 943.
10 R. A. Bissell, A. P. de Silva, H. Q. N. Gunaratne, P. L. M. Lynch,
Fig. 1 pH-dependence of the fluorescence spectra of 5 (5 3 1023 g dm23
)
G. E. M. Maguire, C. P. McCoy and K. R. A. S. Sandanayake, Top.
Curr. Chem., 1993, 168, 223.
11 D. D. Perrin, B. Dempsey and E. P. Serjeant, pKa Prediction for
Organic Acids and Bases, Chapman and Hall, London and New York,
1981.
in 0.1 M NaCl at various pH values (A): 1.29 (a), 2.59 (b), 2.72 (c), 2.81 (d),
3.09 (e), 3.41 (f); and (B): 11.82 (a), 12.01 (b), 12.17 (c), 12.29 (d), 12.41
(e), 12.69 (f). Different pH values were obtained by adding small amounts
of 0.1 M HCl or NaOH to the solution. The excitation was at 385 nm for
both of the emission spectra (Em) of (A) and (B); the emission was at 493
and 433 nm for the excitation spectra (Ex) of (A) and (B), respectively. Inset
shows the variation of fluorescence emission intensity with pH at 493 (A)
and 433 nm (B), respectively.
12 (a) Fluorescence quantum yield o was determined according to the
equation: o = oR F/FR AR/A, where F is the integrated intensity, A is the
absorbance and the subscript R refers to the reference fluorophore of
quinine sulfate (1.0 3 1026 g dm23 in 0.05 M H2SO4). Fluorescence
measurements were performed on
a HITACHI F-4500 spectro-
fluorimeter; (b) Fluorescence lifetime in neutral medium (pH = 6.6)
was obtained at 25 °C with a single photon-counting apparatus
(HORIBA NAES-1100 time-resolved spectrofluorimeter).
ions (Table 1) is of specificity, making it very useful for
accurate measurement of extreme pH values.
The mechanism of the probe’s fluorescence response to pH is
complex due to the influence of protonation/deprotonation and
the possible presence of enol–amide tautomerism, but it may be
interpreted largely according to photoinduced electron transfer
(PET) principle, which has been widely used to design a variety
of fluorescent ion sensing molecules.14 Nitrogen atoms often
13 pKa was calculated by the equation: pKa = pH ± lg [(Fmax 2 F)/(F 2
F
min)]. See: R. A. Bissell, A. P. de Silva, H. Q. N. Gunaratne, P. L. M.
Lynch, G. E. M. Maguire and K. R. A. S. Sandanayake, Chem. Soc.
Rev., 1992, 21, 187.
14 M. Ayadim, J. L. H. Jiwan, A. P. De Silva and J. Ph. Soumillion,
Tetrahedron Lett., 1996, 37, 7039.
Chem. Commun., 2001, 960–961
961