You are on page 1of 11

Journal of Inorganic Biochemistry 152 (2015) 8292

Contents lists available at ScienceDirect

Journal of Inorganic Biochemistry


journal homepage: www.elsevier.com/locate/jinorgbio

Molar absorption coefcients and stability constants of metal complexes


of 4-(2-pyridylazo)resorcinol (PAR): Revisiting common chelating probe
for the study of metalloproteins
Anna Kocya, Adam Pomorski, Artur Krel
Laboratory of Chemical Biology, University of Wrocaw, ul. Joliot-Curie 14a, 50-383 Wrocaw, Poland

a r t i c l e

i n f o

Article history:
Received 19 May 2015
Received in revised form 25 August 2015
Accepted 26 August 2015
Available online 5 September 2015
Keywords:
4-(2-Pyridylazo)resorcinol (PAR)
Molar absorption coefcient
Dissociation constant
Stability constant
Zinc protein

a b s t r a c t
4-(2-Pyridylazo)resorcinol (PAR) is one of the most popular chromogenic chelator used in the determination of
the concentrations of various metal ions from the d, p and f blocks and their afnities for metal ion-binding
biomolecules. The most important characteristics of such a sensor are the molar absorption coefcient and the
metalligand complex dissociation constant. However, it must be remembered that these values are dependent
on the specic experimental conditions (e.g. pH, solvent components, and reactant ratios). If one uses these
values to process data obtained in different conditions, the nal result can be under- or overestimated. We
aimed to establish the spectral properties and the stability of PAR and its complexes accurately with Zn2+,
Cd2+, Hg2+, Co2+, Ni2+, Cu2+, Mn2+ and Pb2+ at a multiple pH values. The obtained results account for the presence of different species of metalPAR complexes in the physiological pH range of 5 to 8 and have been frequently
neglected in previous studies. The effective molar absorption coefcient at 492 nm for the ZnHx(PAR)2 complex
at pH 7.4 in buffered water solution is 71,500 M1 cm1, and the dissociation constant of the complex in these
conditions is 7.08 1013 M2. To conrm these values and estimate the range of the dissociation constants of
zinc-binding biomolecules that can be measured using PAR, we performed several titrations of zinc nger peptides and zinc chelators. Taken together, our results provide the updated parameters that are applicable to any
experiment conducted using inexpensive and commercially available PAR.
2015 Elsevier Inc. All rights reserved.

1. Introduction
Proteins utilize a large number of cofactors to achieve a variety of
functions and structures. Critical among these cofactors are metal ions,
that differ substantially from organic cofactors. For example, bioinformatic analyses of the human genome suggest that up to 3000 proteins
may participate in Zn2+ binding [1]. This number corresponds to ~10%
of all encoded proteins, which may additionally contain various
numbers of zinc domains and motifs with different metal afnities [2].
One of the critical steps in the metalloprotein studies is monitoring of
the metallic cofactors association or dissociation with proteins. Moreover, metalloproteins, particularly those that bind metal ions through
sulfur donors of cysteine residues, might be oxidized or chemically
modied depending on the number of biological reactive species that
typically decrease metal ion-to-protein afnity [3,4]. The best examples
of such cysteine-containing proteins are the metallothioneins and zinc
nger domains [511].
The release of metal ions from proteins and signicant decreases in
metal ion-to-protein afnities are difcult to observe in in vitro conditions due to the typically low concentrations of the metalloproteins
Corresponding author.
E-mail address: artur.krezel@uwr.edu.pl (A. Krel).

http://dx.doi.org/10.1016/j.jinorgbio.2015.08.024
0162-0134/ 2015 Elsevier Inc. All rights reserved.

used in such experiments. Weak spectroscopic properties of some metal


ions, especially Zn2+, do not aid the spectrophotometric determination
of metal ions concentrations. To measure the micro- or submicromolar
metal ion concentrations precisely, specialized chelating chromophores
(also termed metallochromic indicators), that change their spectral properties upon metal ion binding and have appropriate afnities toward
metal ions, are strictly required [12]. In analytical chemistry, one of the
best known (from the historical perspective) metallochromic indicators
for the Hg2+, Pb2+, Cu2+, Zn2+, Co2+ and Ni2+ detection is dithizone
(Fig. 1). It requires hydrophobic solvents and thus is not suitable for
metalloprotein-based applications [1314]. One of the most popular
chelating chromophores for the bioinorganic analysis of Zn2+ is watersoluble 4-(2-pyridylazo)resorcinol (PAR) [15]; however, a number
of other compounds have been applied to measure Zn 2 + , such as
Zincon, Eriochrome Blue SE, Eriochrome red B, Naphtylazoxine 6S
and SNAZOXS [16,17] (Fig. 1, Fig. S1, ESI). Among these compounds,
PAR and Zincon are the most popular in bioanalytics. The former one
has a higher Zn2+ afnity, and its complex has a 3-fold higher molar
absorption coefcient than the latter one [18,19].
The last two decades have produced a number of useful uorogenic
chelating chromophores that are frequently termed zinc probes such as
TSQ, Fura, FluoZin, ZnAF, Zinpyr families that are used in similar applications [2024] (Fig. 1, Fig. S1, ESI). Fluorogenic probes with uorescence

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

83

Fig. 1. Examples of chelating chromophores and uorophore: Zincon (a), Eriochrome Blue SE (b), Dithizone (c), Naphtylazoxine 6S (d), SNAZOXS (e), and TSQ (f).

modulated by metal ion binding are used in most applications at


much lower concentrations than classical metallochromic indicators
due to the greater sensitivities of uorescent techniques and the
unique chemical properties of uorogenic probes. The important
drawbacks of the majority of uorescent chromophores are limited
availabilities from commercial suppliers and high costs. Additionally,
many uorescent chromophores are soluble only in water at the low
micromolar concentrations and precipitate when used above that
range. Despite the many advantages of uorescent chelating probes,
particularly in molecular biology studies, classical chromophores,
including PAR, are the most popular for in vitro studies due to their
protein stability and Zn2 + release characteristics. Classical chromophores can be easily used at submillimolar levels and are inexpensive, which makes them the rst choice probes for measuring metal
ion binding and release from metalloproteins or other metalbinding molecules [2535].
The most important physicochemical parameters of a metal chelating chromophore are its molar absorption coefcient and the stability
constant of the formed complex. In general, the higher molar absorption
coefcients are associated with the lower metal ion detection limits.
Molar absorption coefcients are used in metalloprotein studies dually:
to establish the metal ion concentration and to quantitatively monitor
competition with macromolecules to determine the metal binding
afnities of the macromolecules. In both cases, anyone using the
molar absorption coefcient and dissociation constant values available in the literature should know the conditions for which those
values are valid and employ precisely the same setup in their experiments. The data from the literature, listed in Table 1, indicate that
these values have been determined in various conditions [3649].
Application of the molar absorption coefcients determined in signicantly different conditions and the use of inappropriately determined reference values obviously results in major experimental
errors.
Here, we aimed to re-determine the molar extinction coefcients of
the commonly used chelating chromophore PAR and its complexes with
eight biogenic and toxic metal ions (i.e., Zn2+, Cd2+, Hg2+, Co2+, Ni2+,
Cu2+, Mn2+, and Pb2+) due to the signicant differences in the literature data or lack of it (Table 1) [3649]. Because the application of
PAR for determination of the metalloproteins afnity constants requires
well-dened stability constants, we performed potentiometric and
spectroscopic studies across a wide range of pH to re-determine the
dissociation constants. These values were then applied to determine
the zinc afnities of several previously characterized zinc ngers and
zinc chelators. In this study, we provide re-determined values of the
molar absorption coefcients and stability constants that are valid for
various experimental conditions.

2. Experimental
2.1. Materials
ZnSO47H2O, 4-(2-pyridylazo)resorcinol (PAR), CdCl2, NiCl26H2O,
HgCl2, MnSO4H2O, Pb(NO3)2, CuCl22H2O, Na2HPO4, Na3PO46H2O,
sodium acetate, potassium hydrogen phthalate (KHP), 4-(2hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 2-(Nmorpholino)ethanesulfonic acid (MES), 1,2-ethanedithiol (EDT),
special quality HNO3, 1,4,8,11-tetraazacyclotetradecane (cyclam),
ethylene-bis(oxyethylenenitrilo)tetraacetic acid (EGTA), tris(2carboxyethyl)phosphine hydrochloride (TCEP), perchloric acid, acetic
anhydride, thioanisole, anisole, and the standard solution of 0.1 M

Table 1
Molar absorption coefcients of PARmetal ion complexes reported in the literature.
Metal
ion

Wavelength
(nm)

(M1 cm1)

Experimental conditions

References

Zn2+

492
500

84 100
80 000

[36]
[37]

495
495
495
493
485
485

63 400
77 400
81 000
83 000
38 750
43 300

497

46 700

500
494
485
485

66 000
84 400
21 666
32 000

Hg2+
Co2+

540
508
513

20 059
58 600
33 300

Ni2+
Cu2+

494
496
515

72 200
67 000
68 700

Mn2+

496
490
496

78 000
38 300
86 500

500

78 000

518

35 900

50 mM ammonia solution
50 mM MOPS pH 7.3,
100 NaCl
Ammonia solution, pH 8
Borate, pH 9
Borate, pH 8
Carbonate, pH 9.7
5 mM TrisHCl, pH 8.6
5 mM TrisHCl, pH 7.4,
100 mM NaClO4
50 mM HEPES, pH 7.4, 4 M
GdnHCl
40 mM HEPES, pH 7.0
50 mM ammonia solution
5 mM TrisHCl, pH 8.6
5 mM TrisHCl,
pH 7.4, 0.1 M NaClO4
pH 3.13.3
50 mM ammonia solution
50 mM HEPES, pH 7.4, 4 M
GdnHCl
50 mM ammonia solution
50 mM ammonia solution
Borate, pH 7.969.40,
0.1 M NaClO4
50 mM ammonia solution
NaOH solution, pH 10.0
100 mM phosphate buffer,
pH 11.2
Ammonia solution,
pH 9.710.7
50 mM ammonia solution

Cd2+

Pb2+

[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[36]
[45]
[43]
[46]
[36]
[44]
[36]
[36]
[47]
[36]
[48]
[49]
[40]
[36]

84

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

NaOH were purchased from Sigma-Aldrich. The metal-chelating resin


Chelex 100 was from BioRad. Acetonitrile (ACN) and Co(NO3)26H2O
were from Merck Millipore. NaCl, NaOH, HCl, acetic acid, and ethyl ether
were from Avantor Performance Materials Poland (Gliwice, Poland).
N,N-dimethylformamide (DMF), dichloromethane (DCM), 1-methyl2-pyrrolidinone (NMP), N,N,N,N-tetramethyl-O-(1H-benzotriazol-1yl)uronium hexauorophosphate (HBTU), triuoroacetic acid (TFA),
N,N-diidopropylethylamine (DIEA), piperidine, TentaGel R Ram and all
Fmoc-protected amino acids were obtained from Iris Biotech GmbH
(Marktredwitz, Germany). The concentration of the stock solutions of
the metal ion salts was 0.05 M, and the exact concentrations were
conrmed by representative series of ICP-MS measurements [50]. All
of the pH buffers used in this study were treated with Chelex-100
resin to eliminate trace metal ion contamination [5153]. A PAR
solution in DMSO was prepared freshly before each experiment. It
should be noted that DMSO solutions should not be stored at a room
temperature longer than one week due to degradation. PAR water
solutions are stable for hours and were prepared immediately before
the measurements from DMSO stock solutions.
2.2. Peptide synthesis
The tetracysteine motif (TC), MTF1-1 zinc nger and C@E mutant of
the ZF133-11 zinc nger peptides (Table S1, ESI) were synthesized via
solid phase synthesis using a Fmoc microwave-assisted synthesizer
(CEM) [54]. The reagent excess, cleavage and purication were
performed essentially as previously described [5557]. The peptides
were acetylated on the N-terminus using acetic anhydride in the
presence of DIEA and cleaved from the resin with a mixture of TFA/
thioanisole/EDT/anisol (90/5/3/2 v/v/v/v) over a period of 2 h followed
by precipitation in cold ( 20 C) diethyl ether. The crude peptide
pellets were collected by centrifugation, dried and puried via HPLC
(Dionex Ultimate 3000) on Phenomenex Aeris Peptide 3.6 m 100
C18 columns using a gradient of ACN in 0.1% TFA/water from 0 to 50%
over 30 min. The puried peptides were identied by ESI mass
spectrometry with an API 2000 Applied Biosystems instrument. The
identied/calculated monoisotopic molecular masses were 3137.3/
3137.5 (ZF133-11 C@E), 3600.4/3600.7 (MTF1-1), and 1356.4/1356.5
(TC motif).
2.3. Potentiometry
The protonation constants of PAR and the stability constants were
measured at 0.1 M ionic strength (from 96 mM KNO3 in 4 mM HNO3)
at 25 C using pH-metric titration over the pH range of 2.5 to 11.0
(Molspin automatic titrator, Molspin) with 0.1 M NaOH as the titrant.
The exact concentration of the titrant was established via previous titrations of 3 mM KHP. Due to the limited solubility of PAR in water in acidic
conditions, 1% DMSO was added to all media used in the experiments.
Changes in pH were monitored using a combined glass-Ag/AgCl
electrode (Biotrode, Methrom) that was calibrated daily in hydrogen
concentrations achieved with HNO3 titrations [58]. Sample volumes of
2.0 ml and PAR concentrations of ca. 600 M were used. The data
were analyzed using the Superquad program [59]. Value of 13.83 was
used in the data analysis to account for the ionic product of water. It
represents a 0.1 M ionic strength in 1% of DMSO [58,60].

(Pb2+ and Cu2+) concentrations were performed at pH of 7.0, 7.4, 8.0,


9.0 and 9.9 using 50 mM HEPES or 50 mM Tris buffers with I = 0.1 M
from NaCl. The volume of each sample was 2.0 ml, and the increase in
volume during the titrations was included in the data processing. The
samples were equilibrated for 12 min after the addition of each portion
of 1.0 mM metal ion solution; however, the equilibrations with Ni2+
and Co2+ required 45 h of incubation prior to the measurements.
2.5. Molar absorption coefcients
The molar absorption coefcients of the PAR species that were
present in highly basic pH were calculated based on the spectroscopic
pH-titration data of 20 M PAR and the molar fraction plot from the
potentiometric data using multiple linear regression. The effective
molar absorption coefcients were determined as the slopes of increases of the linear absorption observed after the addition of the appropriate metal ion in the range of 010 M to 100 M PAR at pH 7.4 and
25 C (50 mM HEPES buffer, I = 0.1 M) and 11.0 (50 mM phosphate
buffer, I = 0.1 M). Additional measurements of the effective molar absorption coefcients were performed for Zn2+ across a wide pH range
using 50 mM acetate (pH 4.0, 4.5, and 5.0), 50 mM MES (pH 5.5, 6.0,
and 6.5), 50 mM HEPES (pH 7.0, 7.4, and 8.0), 50 mM Tris (pH 9.0,
and 9.9) with I = 0.1 M from NaCl. The molar absorption coefcients
of the individual ZnHx(PAR)y species were calculated by combining
the potentiometric and spectroscopic data using multiple linear
regression.
2.6. Competition studies
The competition studies of 100 M PAR with the zinc binding peptides (zinc ngers and the TC motif) and the zinc chelators (cyclam
and EGTA) were conducted in 50 mM HEPES buffer at pH 7.4 and 25
C, I = 0.1 M from NaCl. TCEP, which is non-metal-binding protein disulde reducing agent, was added to the peptides containing cysteines
to protect them from oxidation to its nal concentration of 100 M
[62]. PAR was partially saturated by the addition of Zn2+ to its nal concentration of 5 M. The 2.0 mM stock solutions of the zinc nger peptides were prepared in 5 mM HCl to avoid oxidation and were then
used for the competition assays in the range of 012 M of the peptide
[51,53]. The zinc chelators were prepared similarly; however the stock
solutions were prepared in 50 mM HEPES buffer. Each sample was incubated for 10 h after the addition of the specied amounts of the Zn2+
binding ligand. The samples with EGTA and cyclam were incubated for
2 h without the addition of TCEP. All incubations were performed in
glass vials to prevent the adhesion of the metal ion PAR complexes to
plastic tubes. The exact concentrations of the ZnHx(PAR)2 complexes
present in each sample after equilibration were calculated based on
the absorbances at 492 nm using the appropriate effective molar coefcients determined in this study. All calculations of the dissociation
constants of the zinc ngers and zinc complexes with chelators were
performed using the effective dissociation constant of ZnHx(PAR)2
at pH 7.4 that was determined in this study. The results of all of the
calculations are presented in Tables S2-S6 (ESI).
3. Results and discussion
3.1. Acidbase properties of PAR

2.4. Spectroscopic titrations


The absorbance spectra were recorded on a Jasco V-650 spectrophotometer at 25 C over the spectral range of 300650 nm in 1.0-cm quartz
cuvettes. The spectroscopic pH titrations of 20 M PAR were performed
in 0.1 M NaClO4 over the pH range of 2 to 13 using Mettler Toledo InLab
Semi Micro glass electrodes to capture all deprotonation events [61].
The spectroscopic titrations of 20 M PAR with metal ions to their
nal 10 M (Zn2 +, Co2 +, Cd2 +, Ni2 +, Hg2 +, and Mn2 +) or 20 M

The PAR molecule contains three protonating groups, i.e., two


chemically inequivalent hydroxyl groups of resorcinol moieties and
one pyridyl functional group (Fig. 2a). The dissociation constants of
these groups were measured independently with the two techniques,
i.e., spectrophotometry and potentiometry. These methods differ
in terms of numerical accuracy. Potentiometry is superior in this respect; however, due to the limitations of its application at basic pH,
potentiometry cannot be used to determine pK a values above

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

85

Fig. 2. Structures of the fully protonated PAR molecule: H3L+ (a), MHL+ (b), and MH2L2 (c) metal complex species. M refers to any bivalent metal ions examined in this study. Green color
2
represents protons that dissociate at basic pH transforming complexes to ML, MHL
species with the same coordination mode.
2 and ML2

10.511.0 with high accuracy. Moreover, PAR has limited solubility at


acidic pH required in potentiometry. To avoid the limitation of water
solubility 1% DMSO was added to all media used in the potentiometric
experiments. Spectrophotometric titration of 20 M PAR over the
wide range of pH from 1.5 to 13.4 allowed us to characterize all of the
deprotonation events due to the signicant changes in the absorption
spectra that are associated with all acidbase equilibria (Fig. S2a, ESI).
The pyridyl group exhibited the highest acidity with a pKa1 = 2.92
(Table 2), which is comparable to other pyridyl derivatives with pKa
values that vary between 2 and 4 [63,64].
The most basic deprotonation occurred above pH 10.5 with a
pKa3 = 12.1, and this deprotonation was associated with the formation
of a new characteristic band at 489 nm (Fig. S2b, ESI). Due to the inability to determine this pKa value with potentiometry, it was obtained
spectrophotometrically and then xed as a constant value in the data
processing using Superquad program [59]. All pKa values obtained by
both techniques are provided in Table 2. These data are highly convergent, which demonstrates the lack of concentration-dependence of
the deprotonation processes. Comparisons of these data with the values
that were previously determined in 50%50% dioxane-water revealed
signicant differences, especially regarding the second deprotonation,
which differed by ~1.5 log units (Table 2) [65,66]. Similar observations
were made regarding methanolwater mixtures in which the pKa2
values varied by 1.3 log units as the concentration of methanol was
increased from 0 to 90% [67]. This difference was due to the exposition
of the 1-hydroxyl group to the solvent and the lack of additional interactions of this group. In contrast, the second 3-hydroxyl group participates in the formation of a hydrogen bond with a nitrogen atom as
presented in Fig. 2a. This interaction increases the basicity of the
hydroxyl group compared with the pKa values of the chemically equivalent groups of resorcinol, which are 9.2 and 10.9, respectively [68].
The dissociation constants of PAR, presented in Table 2, allowed us to
plot the distributions of the molar fractions of all protonated and
deprotonated species across a wide range of pH (Fig. 3). The absorbance

changes measured at 342, 413, 458 and 489 nm (which corresponded to


the absorbance maxima of the given species) were well correlated with
the molar fraction distributions. The known molar fractions of the individual species in combination with the absorbance values at 489 nm
allowed us to calculate the molar absorption coefcient () of 32
300 300 M1 cm1 for the fully deprotonated L2 species (Fig. S2b,
ESI) using multiple linear regression.

3.2. Complex formation


3.2.1. Metal binding properties of PAR
The relatively exible PAR molecule containing two nitrogen donors and one oxygen donor serves as an attractive ligand for many
metal ions with various radii, oxidation states and geometries.
Early literature indicates that PAR is able to associate with large
numbers of metal ions from the d, p and f blocks of the periodic
table [6974]. The binding of metal ions to the PAR molecule is associated with the deprotonation of the phenolic group (pKa3), which
has major implication for the electronic spectra of PAR (Fig. 2a). All
metal complexes exhibit the formation of a new band between 480
to 540 nm at a neutral or basic pH (Figs. S3S12, ESI). The basicity of
the phenolic group decreases depending on the metal ion and obviously
affects the stabilities of the formed complexes. Because the PAR molecule provides only three donors, ML and ML2 complexes are commonly
formed. Moreover, both complexes might exist in water solutions as

Table 2
Dissociation constants of PAR determined by potentiometric and spectrophotometric
(UVvis) titration at 25 C and I = 0.1 M and the values reported in the literature.
UVvis
pKa1
pKa2
pKa3
Ref.

2.92 0.03
5.43 0.02
12.10 0.02
This study

Potentiometry
a

2.86 0.03
5.45 0.03a
12.10b
This study

Potentiometryc

UV-visd

2.3
6.9
12.4
[65,66]

3.072.45
5.506.28
12.0412.77
[67]

a
log H2L = 17.55 0.01, log H3L+ = 20.41 0.02 with xed log HL = 12.1. The
values were determined in 1% DMSO water solution (see Experimental section).
b
The value taken from UVvis titration.
c
Values determined in 50% dioxanewater mixture.
d
Values determined in methanolwater mixtures (from 0 to 90% methanol).

Fig. 3. Comparison of the spectrophotometric pH titration of 20 M PAR in 0.1 NaClO4 at 25


C with the distribution of the molar fractions of PAR based on potentiometry. Blue, orange, red and green correspond to the absorptions measured at 342, 413, 458 and
489 nm, respectively, which were the maximal absorbances of each species.

86

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

differently protonated forms, such as MHL+, ML, MH2L2, MHL


2 and
ML22 , which additionally complicates the equilibria. MHL+ and
MH2L2 complexes have hydroxyl groups at position 1 (Fig. 2b, c) that
are deprotonated at basic pH (ML and ML2 species, respectively).
MHL
2 is a chemical form with one fully deprotonated PAR molecule.
It should be noted that hydroxyl group deprotonation in those complexes does not change the coordination mode of the metal ion. Because
all of the species present in water solution exhibit spectroscopic properties, it is vitally important to know the stabilities and molar absorption
coefcients of the species that are formed upon metal binding. The PAR
molecule is frequently treated as a simple ligand that forms only the
M(PAR)2 complex when it is used in excess over a metal ion. Due to
the acidbase properties of PAR, several species with various spectral
properties and protonation states are simultaneously present in a
wide range of pH. The use of molar absorption coefcients determined
in different conditions results in the over- or underestimation of metal
ion concentrations.

3.2.2. Potentiometric and spectroscopic studies


The complexation of metal ions to chromophoric chelating compounds can be studied either by potentiometry or with pH-dependent
and pH-constant spectroscopic studies. Potentiometric techniques require relatively high concentrations of the compound. The addition of
1% DMSO to a 4 mM HNO3 water solution is required for potentiometric
studies and provides a solubility of PAR in the range of ~230 M, which
is acceptable for this technique. It must be remembered that PAR in a
DMSO solution is not stable for more than a week and thus should be
freshly prepared. The protonation constants agree well with the spectroscopic measurements (Table 2). Unfortunately, the addition of all of
the metal ions studied here at weakly acidic pH, results in the major precipitation of the complexes, most likely as neutral MH2L2 complexes.
The usage of a high 7:1 excess of PAR over the metal, which is the
limit for this technique, also results in some precipitation in weakly
acidic conditions. This eliminates the possibility of the potentiometry
use to study metal ion complexation by PAR. In a very early study,

Fig. 4. Spectrophotometric titrations of 20 M PAR with Zn2+, Co2+, Cd2+, Ni2+, Hg2+, Mn2+, Pb2+, and Cu2+ at pH 7.0 (green), 7.4 (red), 8.0 (yellow), 9.0 (blue), and 9.9 (black). The
titrations were performed in acetate, MES, HEPES, and Tris buffers at 25 C, I = 0.1 M from NaCl. The dashed lines denote the stoichiometric points.

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

Corsini et al. applied a 50% water solution of dioxane for potentiometric


studies [65,66]. In these conditions, metal complexes are easily soluble
in a wide range of pH. However, such conditions are not attractive for
other researchers due to the high organic solvent content, which affect
both, the stability constants and the spectroscopic properties of PAR
and its complexes.
Decreasing the PAR concentration to 20 M is ideal for the spectroscopic studies and prevents the precipitation of complexes in a wide
range of pH. Fig. 4 presents titrations of 20 M PAR with Zn2 +, Cd2+,
Hg2 +, Co2 +, Ni2+, and Mn2 + to nal metal:ligand molar ratio of 1:2
and 1:3 with Pb2+ and Cu2+ at pH 7, 7.4, 8, 8.5, and 9.0. The absorption
values were measured at the characteristic wavelength of each metal
ion as listed in Table 3. All of the metal ions exhibited a tendency to
bind to PAR; however, the complexations were strictly dependent on
the pH and metal ion. In the case of Co2+ and Ni2+ the titration curve
shapes were almost the same between the pH 7 and 9, which indicates
formation of highly stable MHx(PAR)2 complexes. Here, the long-term
incubation with PAR was required to reach equilibrium. The Zn2+ titration plot also reveals the formation of ZnHx(PAR)2 complexes; with
characteristic inection point above pH 8.0. In contrast, the Cd2 +,
Hg2 + and Mn2 + titration plots were parabolic, and full saturations
were not observed even at pH 9, which might be attributable to the
signicantly lower metal-to-ligand afnities. However, Pb2+ exhibited
a tendency to primarily form 1:1 complexes, which was easily observable in both, the pH-constant titrations (Fig. 4) and the pH-dependent
titrations (Figs. S9 and S10, ESI). Absorbance increased linearly up to
the molar ratio of 1:1, which might be explained by the size of the
metal ion and its strong tendency to from complexes with coordination
number of 4. Cu2+ exhibits the dual properties of forming both stoichiometries, i.e., CuHx(PAR)2 complexes at a neutral pH and mixed
CuHx(PAR)2 and CuHxPAR complexes at a basic pH.
Figs. S3S12 (see ESI) demonstrate the electronic spectra of 20 M
PAR with 10 M and 20 M metal ions (for Pb2+ and Cu2+) recorded
from pH 2.5 to 11. Similar to the experiment above, the full complexations of Co2+, Ni2+, Zn2+, and Cd2+ were reached above the pH of 6,
7, 9 and 11, respectively. In the case of Mn2+, no absorbance plateau
was observed. Above pH 10, the absorption at 498 nm decreased
dramatically, which most likely demonstrates the dissociation of
Mn2+ due to the formation of hydroxide-complexes [75]. The titrations
of Pb2+ with PAR at reactant ratios of 1:2 and 1:1 demonstrated that
the absorbances at the saturation points were ~ 0.3 and ~ 0.6, respectively, which conrms the formation of PbHxPAR complexes at pH
above 77.5. Similar to the Mn2+ titrations, slight decreases in absorbance were observed above pH 9 for the 2:1 ratios due to the partial dissociation of Pb2+ and the formation of hydroxide-complexes [76]. The
stoichiometries of the Cu2+ complexes strictly depend on the reagent
molar ratios. At the ratio of 1:1, the formation of CuHxPAR complexes
was observed above pH 7. At a metal-to-ligand ratio of 1:2, two

Table 3
Effective molar absorption coefcients of PAR complexes with metal ions determined at
the indicated wavelengths, which correspond to the maximum absorbances. The values
were obtained from the slopes of the titrations of 100 M PAR with the appropriate metal
ions over the of 010 M at pH 7.4 (50 mM HEPES buffer) or 11.0 (50 mM phosphate
buffer) at 25 C, I = 0.1 M from NaCl. The slope the SD corresponds to the eff.
Metal ion

max (nm)

Zn2+
Cd2+
Hg2+
Co2+
Ni2+
Cu2+
Mn2+
Pb2+

492
495
505
508
494
508
498
520

7.4
eff
(103 M1 cm1)

11.0
eff
(103 M1 cm1)

71.5 0.4
49.3 0.3
15.19 0.09
51.3 0.3
61.1 0.4
44.6 0.2
7.71 0.06
30.5 0.3

81.5 0.2
74.6 0.4
48.9 0.3
51.4 0.2
61.2 0.5
59.3 0.1
67.2 0.3
36.0 0.1a

a
The value was determined in non buffered conditions due to the precipitation of
Pb3(PO4)2 in phosphate buffer. The pH value was adjusted manually using 2 M NaOH.

87

equilibria events were observed over the wide range of pH. Similar to
the 1:1 ratio, the rst saturation was observed at pH ~ 7, and the
second one occurred at a basic pH. These ndings conrm the equilibria
between CuHxPAR and CuHx(PAR)2 complexes in these conditions.
The spectroscopic results presented above allow us to state that
MHx(PAR)2 complexes of Zn2 +, Cd2 +, Hg2 +, Co2 +, Mn2 + and Ni2 +
are formed, depending on the metal ion, at a neutral and basic pH.
However, 1:1 complexes might also be present in neutral or slightly
acidic conditions. Cu2+ clearly forms both 1:1 and 1:2 complexes in a
reagent-ratio dependent manner. Pb2 + formed only 1:1 complexes
independent of the reagent ratio. The use of PAR at a higher metal-ion
ratios increases the concentration of MHx(PAR)2 complexes (with the
exception of Pb2+) and is frequently used to prevent the formation of
1:1 complexes.
3.2.3. Molar absorption coefcients
The absorbance of a metal-chromophoric chelator complex at its
characteristic wavelength depends on both, the molar fraction of the
complex and its molar absorption coefcient. An accurate molar absorption coefcient () value can be determined at specic conditions only
when the fraction of the metal complex is known. Frequently, multiple
metal complex species with various spectroscopic properties are simultaneously present or incomplete complexation of the metal ions occurs
in certain experimental conditions. In such cases, the molar absorption
coefcient should refer to the total amount of metal (variously
complexed and free) present in the solution, and this quantity is termed
as the effective molar absorption coefcient (eff) in this article. Using
this value, one is able to precisely calculate the total concentration of
the metal ion for any condition required by an experiment. If incomplete complexation occurs, e.g., in acidic pH, the application of the effective molar absorption coefcient does not provide information about
the absorption species concentration but rather about the total amount
of metal. However, in most cases, chelating chromophores are used in
high excess over metal ions, ensuring that 100% of the metal ions are
complexed. Notably, a difference in the probe concentration or in the
pH changes the eff value. To avoid major errors, the effective molar
coefcient should be used for the same experimental conditions.
Taking these facts into consideration, we measured the absorbances
at the maximum wavelengths in buffered solutions at pH of 7.4 and 11.0
(Fig. 5). To avoid the formation of MHxPAR complexes in the cases of
Zn2+, Cd2+, Hg2+, Co2+, Ni2+, and Mn2+, high molar excesses of PAR
over the metal ions were used. PAR at a concentration of 100 M was
titrated at a constant pH with the specied metal ion from 0 to 10 M.
Linear ts representing the effective molar absorption coefcients
were obtained in all cases and are presented in Table 3 along with the
standard deviations. With the exceptions of Ni2+ and Co2+, the absorption coefcients were lower at pH 7.4 than at pH 11.0 due to either
fractional saturation of PAR or the presence of differentially protonated
complexes with various absorptivities. As discussed above, the complexes of Co2 + and Ni2 + with PAR are extremely stable, and those
metal ions were fully complexed at pH 7.4 and 11.0 to form ML2 species
(Figs. S4 and S6, ESI). Therefore, the effective molar coefcients were
identical for these metal ions in both conditions.
Table 1 illustrates the molar absorption coefcients obtained in different experimental conditions, methods and determined at various
wavelengths. A prime example is the Zn2+-PAR system; the effective
molar coefcients that have been determined for this system at various
pH are in the absorption range of 485 to 500 nm. In the literature, the
most widely cited molar extinction coefcient of the Zn2+-PAR system
is 66 000 M1 cm1, which was determined at pH 7.0 at 500 nm by
Hunt et al. [45]. Because the fractions of particular species of the
Zn2+-PAR system differ signicantly at neutral pH, the effective coefcients determined at experimental pH values should be used to avoid
errors. To highlight this issue, we determined the effective molar
absorption coefcients of Zn2+ complexes across the very wide range
of pH value from 4.0 to 11.0 (Table 4, Fig. 6). The effective molar

88

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

Fig. 5. Determination of effective molar absorption coefcients of the Zn2+, Cd2+, Hg2+, Co2+, Ni2+, Cu2+, Mn2+ and Pb2+ complexes with PAR at pH 7.4 (red circles, 50 mM HEPES, I = 0.1 M)
or pH 11.0 (black circles, 50 mM phosphate buffer, I = 0.1 M).

coefcients (at 492 nm) for the pH of 7.0, 7.4, and 8.0 were 60 800, 71
500, and 76 800 M 1 cm1, respectively. These values differ from
those determined at 500 nm (Table 4). Plotting the determined effective
molar coefcients as a function of pH resulted in a Zn2+ complexation
isotherm curve similar to that obtained from spectrophotometric
pH-titration (Fig. 7 and the inset of Fig. 6). This nding conrms the
statement that the changes in the molar fractions of ZnH2L2, ZnHL
2
and ZnL2
(when PAR is used in excess over the metal ion) from pH 5
2
to 9 critically affect the average absorbance at 492 nm and consequently
alter the effective molar coefcients. Therefore, it is imperative to use
the appropriate coefcient values that are determined in the same
conditions as the experiment of interest.

3.2.4. Stability constants of Zn2+ complexes with PAR


PAR is widely used for determinations of Zn2+ concentrations, its release from metalloproteins or determinations of the stability constants
of proteins or other colorless zinc complexes. For the rst two applications, the molar absorption coefcient is required; additionally, for the
latter application, the stability constants of the PAR complexes are also
critical. There are two reports in the literature from Tanaka et al. and
Pollak et al. on the stability constants of Zn2+PAR complexes [18,77].
The protonation and stability constant values were determined spectrophotometrically and are presented in Table 5 as cumulative constants

Table 4
Effective molar absorption coefcients of the Zn2+PAR system determined at 492
and 500 nm across a wide range of pH. The values were obtained from the slopes of the
titrations of 100 M PAR with 010 M Zn2+ in 50 mM acetate, MES, HEPES, Tris or
phosphate buffers at 25 C, I = 0.1 M from NaCl. The slope the SD corresponds to the eff
(103 M1 cm1).
pH

4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.4
8.0
9.0
9.9
11.0

eff (103 M1 cm1)


492 nm

500 nm

2.80 0.03
8.05 0.04
16.1 0.1
24.7 0.1
37. 7 0.1
49.9 0.4
60.8 0.5
71.5 0.4
76.8 0.4
81.1 0.4
81.4 0.3
81.5 0.2

3.40 + 0.03
9.41 0.04
17.7 0.1
26.3 0.1
37.9 0.2
49.8 0.4
59.8 0.3
70.1 0.4
75.0 0.3
78.5 0.2
78.9 0.3
79.0 0.2

Fig. 6. Determination of effective molar absorption coefcients of the Zn2+PAR system.


PAR (100 M) was titrated with Zn2+ over the range of 010 M at various pH values.
The inset demonstrates the pH-dependence of the effective molar absorption coefcients.

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

89

Table 6
Effective dissociation constants (Kdeff)a of the ZnHx(PAR)2 species at various pH values from
7 to 9. The values were calculated based on the protonation and corrected Zn2+ stability
constants reported in this study.b
pH

log Keff
d

7.0
7.2
7.3
7.4
7.5
7.6
7.8
8.0
8.2
8.4
8.6
8.8
9.0

11.67
11.89
12.02
12.15
12.29
12.43
12.75
13.08
13.44
13.81
14.19
14.57
14.96

a
Kdeff = [Zn2+][L]2 / [Zn(L)2], where [L] is the concentration of unbound PAR at different
protonation states (i.e., the of the H3L+, H2L, HL and L2 species); [Zn(L)2] refers to the
sum of all protonated and deprotonated Zn(PAR)2 complexes (ZnH2L2, ZnHL2 and ZnL2).
b
To avoid the presence of ZnHxPAR complexes, a twenty-fold excess of PAR (100
M) over Zn2+ (5 M) was used in the calculations of Kdeff.

Fig. 7. Speciation plots of the Zn2+PAR complexes calculated using the corrected protonation and Zn2+ stability constants presented in this study. The molar fraction distributions
are plotted for the molar ratios of (a) 10 M Zn2+:20 M PAR and (b) 10 M Zn2+:100 M.
The circles included in the top plot demonstrate the absorptions of the Zn2+-PAR complexes at 407 (green) and 492 nm (cyan) recorded with the same reactant concentrations
as a function of pH. The dotted line indicates the pH of 7.4.

(ijk). Using our protonation constants that were determined potentiometrically and spectrophotometrically (Fig. 7), we were able to recalculate these values. The results are comparable with those reported by
Tanaka et al. and Pollak et al.; however, the values differ signicantly
for the ML2 complexes (Table 5).
Because the ZnHx(PAR)y system includes ZnHL+, ZnL, ZnH2L2,
2
ZnHL
complexes that are present across a wide range of
2 and ZnL2
pH, it was not possible to create experimental conditions that could
maintain only one species with the exception of conditions that involved high pH and high molar excesses over Zn2+. The most frequently
used Kd value for the Zn(PAR)2 complex at pH 7.0 presented in the
report of Hunt et al. was calculated by an unknown method based on
the numerical data from the reports of Tanaka et al. and Pollak et al.
[18,45,77]. Fig. 7a demonstrates the molar fraction distributions of the
ZnHx(PAR)y complexes (10 M Zn2 + and 20 M PAR) over a wide
range of pH. This gure shows that at neutral pH, there are several species that differ not only in stoichiometry but also in spectral properties
and stability. The overlap of the absorbance increases at 492 nm,
2
which demonstrates that ZnL, ZnHL
2 and ZnL2 signicantly contribute
to the average absorbance at this wavelength. For the 100 M:10 M

ratio of PAR over Zn2+ (Fig. 7b) at pH 7.4, the percentages of the ZnL,
2
ZnH2L2, ZnHL
complexes are 5.2, 5.8, 52.1, and 36.9%, re2 , and ZnL2
spectively. For this ratio or higher ones, the amount of ZnL complex
can be neglected, which signicantly simplies all calculations without
a loss in accuracy (data not shown). Providing the simple effective Kd of
a particular complex is not useful because all of the differently protonated ZnHx(PAR)2 complexes are simultaneously present across a wide
range of pH. Therefore, the most convenient method for describing the
stability of the system is to provide the effective dissociation constant
of ZnHx(PAR)2, which is the sum of all of the 1:2 complexes at a given
pH. Table 6 includes the effective dissociation constants that were calculated for several pH values from 7.0 to 9.0. These values are based on the
corrected stability values and molar fraction speciations with high PAR
excesses over Zn2+ presented in this study. The effective dissociation
constants can be used directly for any competition experiment at a
specied pH value when PAR is used in high excess (at least 10) over
the total Zn2+.

3.3. Competitive studies: determinations of metalloprotein afnities


3.3.1. General considerations
The determination of metal ion-to-protein afnities is broadly interesting to many researchers who specialize in chemistry, biochemistry,
biophysics and molecular biology. Many methods and techniques
can be applied for direct determination of the stability constants of
metalloprotein complexes [2,2535,54]. If the dissociation constant
is below the concentration of the metal-binding protein used in a particular technique, indirect methods, such as competition with other metal
ions or other ligands, should be applied [51,54,78,79]. For indirect method, the exact stability of the metal ion relative to the competitor must be
known along with the spectral properties at a given experimental
conditions.

Table 5
Stability constants and molar absorption coefcients of the Zn-PAR complexes.
Species

ijka
Tanaka et al.

ijka
Pollak et al.

ijk
This study

() (103 M1 cm1)
Tanaka et al.

() (103 M1 cm1)
Pollak et al.

() (103 M1 cm1)
This study

ZnHL+
ZnL
ZnH2L2
ZnHL
2
ZnL2
2

17.8
11.9
36.2
29.75
22.2

16.7
11.5
31.99
26.29
20.5

17.6
11.7
35.15
28.7
21.15

10.7(490)
31.8(490)
17.5(495)
67.4(495)
95.8(495)

10.7(490)
28.5(495)
16.8(495)
70.9(490)
92.9(490)

14.0 0.4(492)
48.0 0.6(492)
19.0 0.3(492)
75.5 0.5(492)
81.3 0.2(492)

ZnHxLy = [ZnHxLy] / ([Zn2+][H]x[L]y), where [L] is the concentration of fully deprotonated PAR (L2).

90

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

PAR is relatively often used for indirect measurements of Zn2+-toprotein afnities [51,53]. These measurements can be performed either
by metalloprotein equilibration with PAR (Zn2+ transfer from protein to
PAR) or by the titration of partially Zn2+-saturated PAR with apoprotein
(Zn2+ transfer from PAR to protein). In both approaches, the knowledge
about the Zn2+-to-PAR afnity and the spectral properties is essential.
However, one needs to remember that application of PAR for studying
Zn2 + transfer from zinc protein can be a problematic issue, mostly
due to time required for the equilibration. In many cases PAR causes
dissociation of Zn2+ from metalloproteins but the amount of metal released is lower than one expected based on thermodynamic stability
of protein and PAR. This is because of the slow dissociation of Zn2 +
(low koff) from many zinc proteins. Long equilibration of cysteinecontaining proteins with PAR may result in protein oxidation and its
subsequent denaturation, when non-reducing conditions are used.
The application of disulde reductants, e.g. DTT (dithiothreitol) additionally complicates equilibria due to their high afnity for metal ions
[80]. Use of PAR at a high concentration, usually above 300 M, may
result in non-specic interactions with protein, precipitation of PAR or
its neutral metal complexes during experiments.
Alternatively, the application of partially saturated PAR with Zn2+
for competition with apoproteins avoids problematic issue of kinetics.
Relatively high koff of ZnHx(PAR)2 complexes and high kon of most zinc
proteins result in a reasonable timeframe equilibration. To avoid
apoprotein oxidation during its incubation with PAR, TCEP, known as
a non-metal binding reducing agent, should be used [62]. Its negligible
afnity toward transition metal ions makes it a great reducing agent
that does not interfere with PAR-Zn2+-protein equilibria.
It is known that stability constants of metal complexes with low and
high molecular weight ligands are dependent on many factors including
solvent, pH value, ionic strength, temperature, etc. It is obvious, but
frequently overlooked, that afnity constant determined in one set of
conditions cannot be used for the interpretation of experiments that
are run in other conditions. Besides solvent composition, pH and ionic
strength are the most neglected factors in choosing the right competitor
and experimental conditions. It has been shown that a 0.5 difference in
pH value may result in one or even more orders of magnitude difference
in stability constants [78,81]. Similarly, differences in ionic strength
result in major difference in stability constants. Dissociation constant
of theoretical zinc complex determined at zero ionic strength and
0.2 M differs by ~ 0.5 logarithmic scale [82]. Composition of buffer is
another factor. It is well known that some buffer components, such as
phosphates, have signicant afnities toward metal ions that cannot
be neglected, especially when reference complex is not very stable.
Some frequently used buffers, e.g. Tris buffer also forms complexes
with metal ions, such as Cu2+ [83].
3.3.2. Competition between ZnHx(PAR)2 complexes with apoproteins
Our main goal was to provide spectral and stability data across
a wide range of pH for accurate use in any experiment of interest.
Table 4 shows the effective molar absorption coefcients from pH 4 to
11 as determined for conditions involving a high excess of PAR over
Zn2 + to simplify the system to include only the ZnHx(PAR)2 species.
Table 6 shows the effective dissociation constants of ZnHx(PAR)2 from
pH 7 to 9. Table 5 presents pH-independent cumulative constants
that may be used to calculate speciation in any condition of interest
(i.e., any pH and reactant ratio condition).
To examine numerical data obtained in this study we aimed to use
PAR to determine the stability constants of several chemically different
metal ion binding molecules, including zinc nger peptides (MTF1-1,
C@E ZF133-1, Table S1), zinc binding motifs used for uorescent protein
labeling (TC) and the known zinc ion chelators, including cyclam
and EGTA. These compounds were chosen because of their wellestablished Zn2 +-to-ligand afnities [53,54,8486]. The previously
uncharacterized C@E mutant of ZF133-1 was chosen to close the stability and afnity gap between EGTA and the MTF1-1 zinc nger. All

Fig. 8. Competition between ZnHx(PAR)2 complexes with ligands that form ZnL complex
stoichiometries with various Zn2+ stabilities. a) Titration of 100 M PAR partially saturated
by Zn2+ (5 M) with ligands over a concentration range of 011.9 M, 25 C, and I = 0.1 M.
The gray, blue, dark green, green and red circles correspond to cyclam, ZF133-11 C@E,
EGTA, TC motif, and MTF1-1 ZF, respectively b) Simulated titration curves of ZnHx(PAR)2
complexes with ligands with different afnities for Zn2+.

targets were equilibrated with partially Zn2+-saturated PAR, and the


absorbances at 492 nm were measured after equilibration in glass
vials (Fig. 8a). The changes in absorbance were used to calculate the
total and equilibrated concentrations of the reagents at all of the experimental points. Because all of the experiments were performed at
pH 7.4, the concentrations of the ZnHx(PAR)2 complex were calculated
using 71 500 M1 cm1 as the effective molar absorption coefcient.
The exchange constants (Kex) of the PAR and ligand competitions for
Zn2+ (Eq. (1)) were calculated according to the Eq. (2) and are presented in Tables S2S6 (ESI).
ZnHx PAR2 L 2Hx PAR ZnL




K ex Hx PAR2  ZnL= ZnHx PAR2  L

The dissociation constants of the zinc complexes with the examined


peptides and other ligands (KLd) were calculated using the effective
13
dissociation constant (Keff
M2) of ZnHx(PAR)2 from
d = 7.08 10
Table 6 based on Eq. (3).
K d L K d eff  1=K ex

All of the determined dissociation constants are presented in Table 7


and were convergent well with the previously published values. Relative to the dissociation constants and molar absorption coefcients
from Hunt et al., we obtained either different data or were unable to determine the values due to negative concentration values obtained in the
calculations. Moreover, the standard deviations obtained using our set
of data were much lower than those obtained by examining the
previously published values [45].

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292

91

Table 7
Dissociation constants (KdL) of Zn2+-binding ligands used in this work determined at pH 7.4 (50 mM HEPES, I = 0.1 M from NaCl) based on the competition with ZnHx(PAR)2 complexes.
The values were calculated using the effective molar absorption coefcient of 71 500 M1 cm1 and the effective dissociation constant of ZnHx(PAR)2, logKdeff = 12.15 reported here or
in the data reported by Hunt et al. [45].
Ligand

logKLd of ZnL complex calculated based on


stability constants presented in this work

logKLd of ZnL complex calculated based


on stability constants from refs [13]

logKLd of ZnL complex


reported in the literature

Reference

Cyclam
ZF133-11 C@E
EGTA
TC motif
MTF1-1 ZF

8.03 0.08
8.44 0.04
9.15 0.06
10.17 0.13
11.40 0.25

No dataa
8.06 0.34
9.06 0.27
9.98 0.35
11.14 0.61

7.97
No datab
9.20
10.11
11.44, 11.62

[84,85]

[86]
[53]
[54]

a
b

It was not possible to determine the constant due to the negative concentrations obtained during the calculation.
The stability constant of the Zn2+ complex with ZF133-11 C@E is reported here for the rst time.

The zinc-binding models used in the PAR competitions exhibited


afnities toward Zn2 + that ranged widely from high micromolar
(cyclam, pKLd = 8.03) to subpicomolar (MTF1-1, pKLd = 11.40) values
as also indicated in the theoretical curves presented in Fig. 7b, which
were generated using the data obtained in this study. These results
conrm that the PAR probe is a good choice for determinations of
Zn2+ afnities for metalloproteins and other Zn2+-binding compounds
in competition studies. Depending on the reactant concentrations, the
range of the dissociation constants that can be determined using PAR
varies between 108 and 1012 M assuming a 1:1 stoichiometry of
the zincprotein complex.
4. Conclusions
We demonstrated that at physiological pH values, complexes of
4-(2-pyridylazo)resorcinol (PAR) with various metal ions exhibit
different protonation states and stoichiometries (i.e., MHx(PAR)y).
Particular species have different molar absorption coefcients and
contribute to the average absorbance value. Therefore, measurements
of metal ion concentrations require the use of an effective molar absorption coefcient that is determined in the same conditions as those of
planned experiment. Because PAR is commonly used in Zn2+-related
biochemical studies, we have provided ready-to-use effective molar
absorption coefcients for pH ranging from 4 to 11. Based on these ndings, we have also re-determined the values of the dissociation
constants of ZnHx(PAR)2 complexes across a wide range of pH. We
have also established the range of zinc-to-proteins afnities within
which PAR can easily be used in competition studies. Using six different
models of zinc-binding molecules, we conrmed that the dissociation
constants of the resultant Zn2+ complexes convergent well with the
previously publish stability data. Taken together, the values of eff and
Kd presented here allow for the precise determination of metal concentrations and can be easily applied to metallome studies.
Acknowledgments
This research was supported by the National Science Centre (NCN)
under OPUS grant No. 2011/01/B/ST5/00830 and in part by a
PRELUDIUM grant (No. 2012/07/N/NZ1/03079). The instrumentation
was supported by the Polish Foundation for Science under Focus
projects F1/2010/P/2013, and FG1/2010. The authors thank Wojciech
migiel, ukasz Winkler, and Tomasz Kochaczyk for help with the
peptide synthesis.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.jinorgbio.2015.08.024.
References
[1] C. Andreini, L. Banci, I. Bertini, A. Rosato, J. Proteome Res. 5 (2006) 196201.

[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]

[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]

T. Kochaczyk, A. Drozd, A. Krel, Metallomics 7 (2015) 244257.


J.B. Bae, J.H. Park, M.Y. Hahn, M.S. Kim, J.H. Roe, J. Mol. Biol. 335 (2004) 425435.
W. Maret, J. Nutr. 130 (2000) 1455S1458S.
W. Maret, Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 237241.
A. Krel, W. Maret, Biochem. J. 402 (2007) 551558.
L. Khatai, W. Goessler, H. Lorencova, K. Zangger, Eur. J. Biochem. 271 (2004)
24082416.
A. Krel, W. Maret, J. Am. Chem. Soc. 129 (2007) 1091110921.
A. Witkiewicz-Kucharczyk, W. Bal, Toxicol. Lett. 162 (2006) 2942.
E. Bourls, M. Isaac, C. Lebrun, J.M. Latour, O. Snque, Chemistry 17 (2011)
1376213772.
C.F. Shaw III, L. He, A.A. Muoz, M. Savas, S. Chi, C.L. Fink, T. Gan, D.H. Petering, J. Biol.
Inorg. Chem. 2 (1997) 6573.
R.W. Sabnis, Handook of Biological Dyes and Stains. Synthesis and Industrial
ApplicationsISBN: 9780470407530 John Wiley and Sons, 2010.
L.J. Synder, Anal. Chem. 19 (1947) 684687.
R.S. Young, A. Leibowitz, Analyst 71 (1946) 477479.
A.E. Chichibabin, Zh. Russ. Fiz.-Khim. Obshchestva 50 (1920) 512513.
G.K. Singhal, K.N. Tandon, Talanta 14 (1967) 13511353.
K. Vytras, J. Vytrasov, S. Kotrly, Talanta 22 (1975) 529534.
M. Tanaka, S. Funahashi, K. Shirai, Inorg. Chem. 7 (1968) 573578.
F.S. Sadek, R.W. Schmid, C.N. Reilly, Talanta 2 (1959) 3851.
M.S. Nasir, C.J. Fahrni, D.A. Suhy, K.J. Kolodsick, C.P. Singer, T.V. O'Halloran, J. Biol.
Inorg. Chem. 4 (1999) 775783.
K.R. Gee, Z.L. Zhou, W.J. Qian, R. Kennedy, J. Am. Chem. Soc. 124 (2002) 776778.
T. Hirano, K. Kikuchi, Y. Urano, T. Nagano, J. Am. Chem. Soc. 12 (2002) 65556562.
S.C. Burdette, G.K. Walkup, B. Spingler, R.Y. Tsien, S.J. Lippard, J. Am. Chem. Soc. 123
(2001) 78317841.
K. Kikuchi, Chem. Soc. Rev. 39 (2010) 20482053.
E.A. Peroza, A. dos Santos Cabral, X. Wan, E. Freisinger, Metallomics 5 (2013)
12041214.
X. Sun, X. Zhou, L. Du, W. Liu, Y. Liu, L.G. Hudson, K.J. Liu, Toxicol. Appl. Pharmacol.
274 (2014) 313318.
V.S. Javalkote, P.A. Zawar, P.R. Puranik, Appl. Microbiol. Biotechnol. (2015), http://
dx.doi.org/10.1007/s00253-014-6371-6.
P.W. Thomas, M. Zheng, S. Wu, H. Guo, D. Liu, D. Xu, W. Fast, Biochemistry 50 (2011)
1010210113.
J.B. Bae, J.H. Park, M.Y. Hahn, M.S. Kim, J.H. Roe, J. Mol. Biol. 335 (2004) 425435.
J. Smirnova, L. Zhukova, A. Witkiewicz-Kucharczyk, E. Kopera, J. Oldzki, A.
Wysouch-Cieszyska, P. Palumaa, A. Hartwig, W. Bal, Chem. Res. Toxicol. 21
(2008) 386392.
M. Huang, D. Krepkiy, W. Hu, D.H. Petering, J. Inorg. Biochem. 98 (2004)
775785.
J.B. Bae, J.H. Park, M.Y. Hahn, M.S. Kim, J.H. Roe, J. Mol. Biol. 335 (2004) 425435.
W. Maret, B.L. Vallee, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 34783482.
A. Daiber, D. Frein, D. Namgaladze, V. Ullrich, J. Biol. Chem. 277 (2002) 1188211888.
Q. Hao, W. Maret, FEBS J. 273 (2006) 43004310.
E. Gomez, J.M. Estela, V. Cerda, M. Blanco, Fresenius J. Anal. Chem. 342 (1992)
318321.
M. Zimmermann, Z. Xiao, C. Cobbett, A.G. Wedd, Chem. Commun. (Camb.) (2009)
63646366.
M. Hnillickova, L. Sommer, Collect. Czechoslov. Chem. Commun. 26 (1961) 21892205.
M. Tanaka, S. Funahashi, K. Shirai, Anal. Chim. Acta 39 (1967) 437445.
S. Ahrland, R.G. Herman, Anal. Chem. 47 (1975) 24222426.
M. Kitano, J. Ueda, Nippon Kagaku Zasshi 91 (1970) 983;
Chem. Abstr. 74 (1971) 71352e.
J.E. Laib, C.F. Shaw III, D.H. Petering, Biochemistry 24 (1985) 19771986.
C.F. Shaw III, J.E. Laib, M.M. Savas, D.H. Petering, Inorg. Chem. 29 (1990) 403408.
C.E. Sbel, J.L. Shepherd, S. Siemann, Anal. Biochem. 391 (2009) 7476.
J.B. Hunt, S.H. Neece, A. Ginsburg, Anal. Biochem. 146 (1985) 150157.
M.C. Eshwar, S.G. Nagarkar, Fresenius' Z. Anal. Chem. 260 (1972) 289.
S. Funahashi, S. Yamada, M. Tanaka, Inorg. Chem. 10 (1971) 257263.
O.A. Tataev, L.G. Anisimova, 19661968, J. Anal. Chem. USSR (1971) 26.
D. Nonova, B. Evtimova, Talanta 20 (1973) 13471351.
Y. Ha, O.G. Tsay, D.G. Churchil, Monatsh. Chem. 142 (2011) 385398.
T. Kochaczyk, P. Jakimowicz, A. Krel, Chem. Commun. (Camb.) 49 (2013)
13121314.
A. Pomorski, T. Kochaczyk, A. Mioch, A. Krel, Anal. Chem. 85 (2013) 1147911486.

92
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]

A. Kocya et al. / Journal of Inorganic Biochemistry 152 (2015) 8292


A. Pomorski, J. Otlewski, A. Krel, Chembiochem 11 (2010) 12141218.
A. Mioch, A. Krel, Metallomics 6 (2014) 20152024.
A. Krel, W. Bal, Org. Biomol. Chem. 1 (2003) 38853890.
V. Dork, A. Krel, Dalton Trans. (2003) 22532259.
A. Krel, J. Wjcik, M. Maciejczyk, W. Bal, Inorg. Chem. 50 (2011) 7285.
H. Irving, M.G. Miles, L.D. Pettit, Anal. Chim. Acta 38 (1967) 475488.
P. Gans, A. Sabatini, A. Vacca, J. Chem. Soc. Dalton Trans. (1985) 11951199.
E.P. Serjeant, Potentiometry and Potentiometric Titrations, John Wiley and Sons,
New York, 1984.
J. Adamczyk, W. Bal, A. Krel, Inorg. Chem. 54 (2015) 596606.
A. Krel, R. Latajka, G.D. Bujacz, W. Bal, Inorg. Chem. 42 (2003) 19942003.
P. Doungdee, S. Sarel, N. Wongvisetsirikul, S. Avramovici-Grisaru, J. Chem. Soc.
Perkin Trans. 2 (1995) 319323.
I. Gaubeur, M.V. Rossi, K. Iha, M.E.V. Suarez-Iha, Ecletica. Qum. 25 (2000) 6376
(online).
A. Corsini, I.M. Yih, Q. Fernando, H. Freiser, Anal. Chem. 34 (1962) 10901093.
A. Corsini, Q. Fernando, H. Freiser, Inorg. Chem. 2 (1963) 224226.
J. Ghasemi, A. Niazi, M. Kubista, A. Elbergali, Anal. Chim. Acta 455 (2002) 335342.
S.E. Blanco, M.C. Almandoz, F.H. Ferretti, Spectrochim. Acta A Mol. Biomol. Spectrosc.
6d1 (2005) 93102.
R.W. Stanley, G.E. Cheney, Talanta 13 (1966) 16191629.
A.A. Ensa, M. Fouladgar, J. Anal. Chem. 69 (2014) 143148.

[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]

T.M. Florence, Y. Farrar, Anal. Chem. 35 (1963) 16131616.


S. Yamada, N. Taki, T. Katoh, A. Murata, Anal. Sci. 6 (1990) 567572.
M. Tabata, M. Tanaka, Inorg. Chem. 17 (1978) 27792784.
S. Funahashi, M. Tabata, M. Tanaka, Bull. Chem. Soc. Jpn. 44 (1971) 15861590.
F. Scheffer, P. Schachtschabel, Lehrbuch der Bodenkunde, Enke Verlag, Stuttgart,
1989.
R. Escudero, E. Espinoza, F.J. Tavera, Res. J. Recent Sci. 2 (2013) 14.
M. Pollk, V. Kub, Collect. Czechoslov. Chem. Commun. 44 (1979) 725741.
O. Snque, J.M. Latour, J. Am. Chem. Soc. 132 (2010) 1776017774.
M. Hohl, T. Kochaczyk, C. Tous, A. Aguilera, A. Krel, J.H. Petrini, Mol. Cell 57
(2015) 479491.
A. Krel, W. Leniak, M. Jeowska-Bojczuk, P. Mynarz, J. Brasu, H. Kozowski, W.
Bal, J. Inorg. Biochem. 84 (2001) 7788.
K.L. Chan, I. Bakman, A.R. Marts, Y. Batir, T.L. Dowd, D.L. Tierney, B.R. Gibney, Inorg.
Chem. 53 (2014) 63096320.
I.V. Sukhno, V.Y. Buzko, L.D. Pettit, Chem. Int. (2005) 27(3).
J. Nagaj, K. Stokowa-Sotys, E. Kurowska, T. Frczyk, M. Jeowska-Bojczuk, W. Bal,
Inorg. Chem. 52 (2013) 1392713933.
E. Kimura, T. Shiota, T. Koike, M. Shiro, M. Kodoma, J. Am. Chem. Soc. 112 (1990)
58055811.
A. Krel, W. Maret, J. Biol. Inorg. Chem. 11 (2006) 10491062.
A.E. Martell, R.M. Smith, Critical Stability Constants, Plenum Press, New York, 1974.