Analytical applications of the optical properties of ferric hemoglobin: A theoretical and experimental study

Microchemical Journal 114 (2014) 175–181

Contents lists available at ScienceDirect

Microchemical Journal journal homepage: www.elsevier.com/locate/microc

Analytical applications of the optical properties of ferric hemoglobin: A theoretical and experimental study Vanesa Sanz, Susana de Marcos, Javier Galbán ⁎ Analytical Biosensors Group (GBA), Analytical Chemistry Department, Faculty of Sciences, Aragon Institute of Nanosciences (INA), Zaragoza-50009, Spain

a r t i c l e

i n f o

Article history: Received 16 October 2013 Received in revised form 23 December 2013 Accepted 23 December 2013 Available online 31 December 2013 Keywords: Ferric hemoglobin Autoindicating absorption properties Mathematical model Hydrogen peroxide Peracetic acid Glucose/glucose oxidase

a b s t r a c t Like other hemeproteins, ferric hemoglobin (HbIII) not only can act as a catalyzer in the oxidation of substrates by H2O2 but also can act alone as a H2O2 reducer. In this paper, the kinetic mechanism of the reaction between HbIII and H2O2 is studied, and a mathematical model is developed, which permits the determination of the kinetic constants from the absorbance measurement; the results indicate that each HbIII molecule is able to oxidize up to 5 H2O2 molecules without external regeneration. Second, it is demonstrated that the optical properties of HbIII can also be used as an indicator of this oxidation reaction, and analytical methods can be developed using this approach. A method for H2O2 peroxide determination is presented, which permits the analyte to be determined in a concentration range that depends on the HbIII concentration used and the measurement time, with an RSD better than 5%. In addition, methods for peracetic acid (PPA), the simultaneous determination of PAA + H2O2 and the use of HbIII/H2O2 as an indicator for oxidase enzymatic reactions (glucose/glucose oxidase) are outlined. © 2013 Elsevier B.V. All rights reserved.

1. Introduction The development of analytical methods based on the use of enzymatic reactions attracts considerable interest, owing to their specificity and reversibility. Within these analytical methods, several research groups have focused on the use of the spectroscopic properties of proteins [1–5], mainly fluorescence. However, as far as we know, the autoindicating fluorescence properties can only be used in those enzymes having flavin groups as coenzymes. Although many interesting analytes are involved in reactions catalyzed by these kind of enzymes (glucose, cholesterol, or choline), there are still many other compounds for which such reactions are not available. In these cases, researchers have been designed alternatives methodologies based on the use of the molecular absorption properties of proteins, mainly hemeproteins, because they are able of self-detecting hydrogen peroxide produced in a previous reaction. The hemoproteins, present the hemo group as a cofactor, consist of a porphyrin ring linked to Fe atoms through central nitrogen. This hemo group has an intense absorption band in the UV region (Soret band or γ) and weak absorption bands in the visible and in the near IR region [6–9]. These absorption bands are highly dependent on the kind of hemo group and its axial ligands. Moreover, these optical properties also depend on the oxidation state of the hemo group and can

⁎ Corresponding author. Tel.: +34 976761291. E-mail address: [email protected] (J. Galbán). 0026-265X/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.microc.2013.12.016

consequently be used for the monitoring of the enzymatic reaction in which the hemoprotein takes part. In previous work [10,11], we developed an analytical methodology that makes it possible to design and to optimize an autoindicating biosensor using hemoproteins both as bio-receptors and transducers. This methodology is based on the molecular absorption properties of HRP and its variations during the enzymatic reaction with H2O2. In this work, ferric hemoglobin (HbIII) was selected from among the different hemoproteins due to its function as a transport protein and because it is able to react with oxygen peroxide through a pseudoperoxidase catalytic cycle with reaction intermediates similar to that of HRP (peroxidase). However, there are differences between HbIII and HRP, which confirm that the autoindicating methodology could be valid for any hemoprotein. Like all heme proteins, HbIII is made up of an active redox heme group (in this case, ferrous) and its amino acidic surround. During the enzymatic reaction of Hb with H2O2, the reaction intermediates (compound I and compound II) and the native HbIII show different molecular absorption spectra. It is therefore possible to follow the enzymatic reaction through the changes in its molecular absorption spectrum owing to the different reaction intermediates that are proportional to H2O2 concentration. A mathematical model relating the HbIII absorbance variation during the reaction with the H2O2 concentration has been developed to provide theoretical support for the method and to predict its application to other compounds. The kinetic of the HbIII with H2O2 has also been clarified. This paper also provides new information about the HbIII

176

V. Sanz et al. / Microchemical Journal 114 (2014) 175–181

cycle, the redox behavior of heme proteins and spectral information about reaction intermediates.

the absence of an external reducer (substrate). The whole process includes several steps and combines several alternative pathways but can be summarized as follows (Fig. 1) [12–15]:

2. Experimental

1. Just after mixing HbIII and H2O2, the heme group is oxidized to an unstable compound (compound I, having a formal + 5 oxidation state and represented as HbV). 2. Compound I is not shown in Fig. 1 because it immediately decays to the intermediate compound II, with the heme group in a +4 oxidation state (HbIV). The mechanism driving this reduction is the socalled intramolecular reduction (IntraR) in which the heme group is reduced by its amino acid environment (tryptophan, tyrosine, or cysteine). As a result, this proteinic part of the hemoglobin becomes oxidized (this is represented in Fig. 1 by a superscript in the corresponding species indicating the formal oxidation state of the proteinic part). 3. Compound II with the proteinic part oxidized (HbIV+) decays to the original ferric state following one of two alternatives pathways: a) by another IntraR to give HbIII2+ b) by the so-called intermolecular reduction process (InterR) in which HbIV+ is reduced by the proteinic part of another hemoglobin molecule (HbIII), so both molecules give the same species: HbIII+. 4. The ferric hemoglobin (HbIII+, HbIII2 +) thus obtained is ready to react with another molecule of H2O2, and the process represented by steps 1 to 3 is repeated. This cycle can take place several times until the full oxidization of the amino acidic residues is achieved. However, two problems arise: ⁎ A fraction of the hemoglobin does not return to the original ferric state and remains inactive mainly as compound II (and partially as compound I). ⁎ The kinetic constants in each cycle become slower than in the previous cycle. 5. Finally, it is important to emphasize that at very high H2O2 concentrations the degradation of the hemoglobin takes place [16–18]. In this process, HbIV+ reacts with the excess of H2O2 to give a superoxide intermediate (HbIII-O•2 ‐). The superoxide anion in the heme site

2.1. Material and methods 2.1.1. Apparatus The molecular absorption measurements were carried out with a Hewlett-Packard 8452A diode-array spectrometer and a Perkin-Elmer Lambda 5 (spectral bandwidth 2 nm) spectrometer. A quartz cuvette with a 1-cm optical pathway was used. 2.1.2. Reagents The buffer solution was a 0.1-M phosphate solution of pH 6 (from solid KH2PO4 and solid Na2HPO4). Ferric human hemoglobin and ferrous human hemoglobin were supplied by Sigma (Sigma H-7379 and Sigma H-0267 respectively). Glucose oxidase was taken from Aspergillus niger, and EC 1.1.3.4 (Sigma G-7141) was of 179000 IU g−1 of lyophilized solid. Hydrogen peroxide solutions were prepared from dilution of the commercial solution (Panreac, 30% (w/v)). The hydrogen peroxide concentration was determined by redox titration with permanganate. Peracetic acid solutions were prepared from dilution of the commercial solution (Fluka 77240 39% d = 1.15 g mL−1). Glucose solutions were prepared from solid β-D-glucose (Sigma G-5250) in buffer solution (the solution was left for two hours to achieve the equilibrium between β-D-glucose and α-D-glucose). Triton X-100 from Sigma was also used to stabilize the hemoglobin solutions. 3. Theoretical study of the process: mechanism of the HbIII/H2O2 reaction 3.1. Kinetic pathways As occurs with other hemeproteins, the reaction mechanism between ferric hemoglobin (HbIII) and hydrogen peroxide is complex in

H2O HbIII-O•-2

2n+ HbV HbV2n+

k2,n+1

HO H 2O2 22 k1,n+1

Biliverdin

HbIV(2n+1)+ k3,n+1

HbIV+

Dipyrrolic compounds

kHb d Fe3+

Hb(2n+2)+

kHb α

Hb+ HbIII HbIII

H2O2 Fig. 1. Reaction mechanism of HbIII with H2O2.

2 O2 22 k1HHO

V. Sanz et al. / Microchemical Journal 114 (2014) 175–181

opens the phorphyrin ring giving the biliverdin intermediate which reacts with additional H2O2 to give dipirrolic fragments and free Fe3+.

3.2. Optical properties of hemoglobin In terms of the UV–vis absorption properties, hemoglobin also behaves like other hemeproteins [19]. The molecular absorption spectrum of hemoglobin depends on the heme oxidation state and does not depend on the oxidation state of the proteinic part; this means that only three different spectra can be observed corresponding to HbIII, compound II (+4, heme group oxidation state) and compound I (+5, heme group oxidation state). However, because of the kinetic constants (see below), it is difficult to obtain some of these spectra for hemoglobin. Unlike the case of other hemeproteins, the HbIII + H2O2 reaction (see kinetic constants later) is slow compared to the IntraR or InterR process. This means that a stoichiometric mixture of both compounds leads to an incomplete conversion to any of the compounds; this makes necessary to work in H2O2 excess conditions. The evolution of the molecular absorption spectra while increasing the concentrations of H2O2 added to HbIII is shown in Fig. S1. As can be seen, an isosbestic point is observed between two spectra, one corresponding to HbIII and the other to compound II. With a H2O2:HbIII ratio of 18:1, a complete conversion to compound II is obtained. The spectrum of this species is shown in Fig. 2. The molecular absorption spectra of compound I deserves special attention given that it is very difficult to obtain owing to its instability [20]. For other hemoproteins, the addition of high H2O2 concentrations triggers several oxidation-IntraR cycles and permits, in some cases, compound I to be stabilized. In the case of HbIII, the addition of high H2O2 concentrations leads to heme degradation (see below). To avoid heme degradation, HbIII was subjected to reaction with a continuous but low H2O2 concentration. This was obtained by using the redox interaction between hemeproteins and flavoenzymes. In this interaction, the ferric hemeprotein reacts with the FAD group to give compound I (which decays by intramolecular reduction to compound II) and reduced flavoenzyme (E-FADH2), which is oxidized by O2 to give an oxidized enzyme and H2O2. This reaction was carried out with HbIII and the flavoenzyme glucose oxidase (GOx). HbIV+ produced decays by the inter- and intramolecular reduction mechanism to give HbIII+ and HbIII2+, respectively, which reacted with GOx and the H2O2 generated by GOx reoxidization. After a long reaction time, the HbV2n+ species were stabilized. The absorption spectra variation of the HbIII and GOx mixture is shown in the supplementary material section (Fig. S2). During the first minutes of the reaction, a HbIII/HbIV transition was obtained, which turned into a complex HbIII/HbIV/HbV transition until HbV stabilization was reached. Solutions obtained from this reaction had the typical green color of hemeprotein compound I. Its spectrum is shown in Fig. 2.

177

Heme degradation takes place through a HbIII-O•2 ‐ intermediate, which decays to biliverdin and dipyrrolic fragments. The experimental results indicate that the extent of the degradation process depends on the excess H2O2 added. (A) When the H2O2 excess is less than 50:1, in the first seconds of the reaction a transition between two intermediates with an isosbestic point at 588 nm is observed (see insert in Fig. S3). The presence of biliverdin in the heme degradation mechanism has been described in the literature. Since biliverdin has an absorption maximum at 670 nm and weak absorption in the 300to 400-nm region, these spectral changes are ascribed to a HbIV/biliverdin transition [21,22]. (B) When the H2O2 excess is higher, heme degradation takes place. The absorption spectra variation when high H2O2 concentrations are added to HbIII is shown in Fig. S3 (H2O2:HbIII ratio higher than 50:1). As can be seen, HbIII gives HbIV+, which decays after subsequent reaction with the H2O2 excess to give dipyrrolic fragments and free Fe3+, which leads to a gradual absorbance decrease in the absorption spectra [16]. The molecular absorption spectra of the heme degradation products are shown in Fig. 2. The molecular absorption spectra of the different oxidation states of the hemoglobin heme group are shown in Fig. 2. These spectra enable the reaction to be followed from the absorbance variations observed. HbIII has absorption maxima at 496 and 630 nm with shoulders at 536 and 572 nm, compound II has an absorption maximum at 540 with a shoulder at 578 nm and compound I has an absorption maximum at 590 nm. Apart from the absorption maxima, isosbestic points are important from the kinetic point of view given that they permit an individual species to be followed. Isosbestic points for HbIII/HbIV are at 470, 516, 614 and 658 nm, for HbV/HbIV at 568 nm and for HbIII/ HbV at 540 nm. This last wavelength (540 nm) also corresponds with the maximum of compound II and this is the wavelength used for analytical purposes. 3.3. Kinetic model and kinetic constants determination 3.3.1. Determination of the k1,n constants As has been indicated above, when a HbIII aliquot is submitted to a reaction with a stoichiometric aliquot of H2O2, the whole compound becomes oxidized to HbIV+ and is later regenerated by the IntraM mechanism, but a part of the HbIII remains inactivated (partially as compound I and partially as compound II). If the same HbIII is sequentially treated with several stoichiometric aliquots of H2O2, in each new cycle, the amount of inactivated hemoglobin increases until the protein completely loses its activity. Fig. 3 shows this process. In the figure, the absorbance variation at 540 nm is shown during 5 consecutive additions of H2O2. As can be seen, after H2O2 addition, the absorbance (due to HbIV(2n + 1)+) increases until a maximum is reached and later

10000

b

9000 8000

a

c

ε (M-1cm-1)

7000 6000 5000

d

4000 3000 2000 1000 0 450

500

550

600

650

700

750

λ (nm) Fig. 2. Molecular absorption spectra of reaction intermediates of HbIII/H2O2 reaction: (a) HbIII, (b) compound II, (c) compound I and (d) dipyrrolic compounds.

178

V. Sanz et al. / Microchemical Journal 114 (2014) 175–181

0.2

150

a

100

%

0.18 0.16

50

Δ Abs540,t

0.14

0

0.12

0

b c d e

0.1 0.08 0.06 0.04 0.02

2

4

cycle

0 0

500

1000

1500

2000

2500

3000

time (s) Fig. 3. Absorbance variation at 540 nm when HbIII reacts with stoichiometric additions of H2O2. Each H2O2 addition corresponds to an oxidation-reduction cycle: (a) cycle 1, (b) cycle 2, (c) cycle 3, (d) cycle 4 and (e) cycle 5. HbIII initial concentration 1.27 × 10−4 M.

decreases (intraM reduction). The final absorbance value is higher than the initial value, which indicates that a percentage of HbIII becomes inactivated at the end of the cycle. From the spectra in Fig. 2, the maximum molar absorptivity of each species can be seen for the three compounds (495, 540 and 585 for HbIII, compound II and compound I, respectively), and from these values, the percentage of each compound after each cycle is obtained (insert in Fig. 3). As can be seen, hemoglobin undergoes 5 oxidation-reduction cycles. Moreover, from the absorbance signals in Fig. 3, it is possible to H2 O2 obtain the kinetic constant values of k1;nþ1 for each cycle (n = 0 is cycle 1). To do this, a mathematical model was developed that relates absorbance variations with the kinetic of the process. As compound I remains at a very low concentration during the reaction, it is possible to discard its variations during the process. Therefore, after the H2O2 addition, the variation of the HbIV(2n + 1)+ concentration as a function of time is given by the following expression h i d HbIVð2nþ1Þþ dt

H O

h

2nþ

2 2 ¼ k1;nþ1 HbIII

i

h

ð2nþ1Þþ

½H2 O2 ‐k3;nþ1 HbIV

i

ð1Þ

During the first seconds of the reaction (small t values), the following polynomial approximation can be applied to (5): HbIII2nþ o ‐½H2 O2 o Þk1 2 eð½

H O2 t

¼¼ N

≈1 þ

h

2nþ

HbIII

i o

 H2 O2 ‐½H2 O2 o k1;nþ1 t

ð6Þ

h i H2 O2 h i HbIII2nþ ½H2 O2 0 k1;nþ1 t ð2nþ1Þþ 0 ¼ HbIV   H2 O2 2nþ k t 1 þ HbIII 0 1;nþ1

Now, the absorbance at 540 can be easily related to [HbIV(2n + 1) +]. The absorbance measured at the beginning of each cycle (Abs0,n + 1) and during the cycle time (Abst,n + 1) (before reaching the maximum) at any wavelength is given by h i 2nþ Abs0;nþ1 ¼ εIII HbIII þ εIV ½HbIV0;nþ1

ð7Þ

h i  h i 2nþ ð2nþ1Þþ þ εIV ½HbIV0;nþ1 þ HbIV Abst;nþ1 ¼ εIII HbIII

ð8Þ

0

and the corresponding mass balance for total hemoglobin (Hb) (i.e., the added amount) is given by

In the first seconds of the reaction, HbIV(2n + 1)+ consumption is negligible compared to HbIV(2n + 1)+ formation (the k3,n + 1 value is low) so Eq. (1) can be simplified to:

h i 2nþ ½Hb0 ¼ HbIII þ ½HbIV0; nþ1

h i d HbIVð2nþ1Þþ

The combination of Eqs. (7)–(9) gives the following expression for the absorbance at 540 nm:

dt

H O

h

2nþ

2 2 ¼ k1;nþ1 HbIII

i

½H2 O2 

ð2Þ ΔAbs ¼ Abst;nþ1 ‐Abs0;nþ1

The following mass balance can be applied: h

2nþ

HbIII

i 0

h i h i 2nþ ð2nþ1Þþ þ HbIV ¼ HbIII

h i ð2nþ1Þþ ½H2 O2 0 ¼ ½H2 O2  þ HbIV

ð3Þ

ð4Þ

where [HbIII2n+]o is the initial HbIII2n+ concentration for each reaction cycle which remains active from the previous cycle. By substituting Eqs. (3) and (4) in Eq. (2) and solving the differential equation with the initial condition (t = 0, [HbIV(2n + 1)+] = 0), the following expression is obtained: h

HbIV

ð2nþ1Þþ

i

¼

  h i H2 O 2 2nþ HbIII2nþ ½H2 O2 0 1‐eð½HbIII 0 ‐½H2 O2 0 Þk1;nþ1 t 0

H2 O 2   2nþ ½H2 O2 0 ‐ HbIII2nþ 0 eð½HbIII 0 ‐½H2 O2 0 Þk1;nþ1 t

h i H2 O2 ðεIV ‐εIII Þ HbIII2nþ ½H2 O2 0 k1;nþ1 t 0 ¼   H2 O2 2nþ k t 1 þ HbIII 0 1;nþ1

ð10Þ

The reverse of this equation linearly relates the absorbance change H2 O2 for the five cycles can (1/ΔAbs) with the reaction time and the k1;nþ1 be obtained from the slope of the line. Table 1 shows these values. As can be seen, the values hardly change with the cycles. This is an interesting difference compared to other hemeproteins (such as peroxidase) [11], which permit at least 17 cycles and thus the corresponding k1 values become lower with each cycle. Table 1 H2 O2 Kinetic constant values for k1;nþ1 and k3,n + 1 as a function of the cycle number. It was not H2 O2 possible to calculate k5;nþ1 given that its low precision. Cycle

ð5Þ

ð9Þ

0

1 2 3 4

2 2 k1;nþ1 (M−1 s−1)

H O

5.7 (±1.5) 4.6 (±0.8) 6.3 (±0.7) 9.7 (±0.9)

× × × ×

2

10 102 102 102

k3,n + 1 (s−1) 0.00101 0.00097 0.00075 0.00079

V. Sanz et al. / Microchemical Journal 114 (2014) 175–181

179

3.3.2. Determination of the k3,n and other constants In the previous section, the k1,n values have been determined using the first part (before the maximum) of the Abs = f(t) recorders. Using the second part of this representation, k3,n values can also be estimated. In this case, the HbIV(2n + 1)+ formation is negligible and only the consumption is relevant, so Eq. (1) can be simplified to

in the Supplementary Material section in which a kd value of 0.00146 ± 0.00001 M−1 s−1 fits well with the experimental results.

h i d HbIVð2nþ1Þþ

4.1. Determination of H2O2

dt

h i ð2nþ1Þþ ¼ ‐k3;nþ1 HbIV

ð11Þ

The solution of the differential equation with the initial condition [t = tm, [HbIV(2n + 1)+] = [HbIV(2n + 1)+]m] where the subscript “m” refers to the maximum is as follows: h

i HbIVð2nþ1Þþ  ¼ ‐k3;nþ1 ðt‐t m Þ ln  HbIVð2nþ1Þþ m

ð12Þ

Using a similar procedure, the absorbance at 540 can easily be related to [HbIV(2n + 1) +]. To do this, equations indicating the absorbance at the maximum (Absm) and during the cycle time (Abst) (after reaching the maximum) at any wavelength have to be defined as follows: h i  h i  2nþ ð2nþ1Þþ Absm ¼ εIII HbIII þ εIV ½HbIV0;nþ1 þ HbIV

ð13Þ

h i  h i ð2nþ2Þþ ð2nþ1Þþ þ εIV ½HbIV0;nþ1 þ HbIV Abst ¼ εIII HbIII

ð14Þ

m

m

4. Analytical applications of the intrinsic absorption properties of HbIII

The changes in the molecular absorption spectra of hemoglobin during its reaction with H2O2 can be used to design analytical methods. The candidate analytes are not only H2O2 but also other peroxide (bio) chemical species, which participate in reactions which generate H2O2. In this section, the analytical possibilities for H2O2 will be described. In Fig. S5, the absorbance variations for different H2O2 concentrations using the same hemoglobin concentration are shown. The absorbance variations are shown at 540 nm, the isobestic point for HbIII/ HbV where HbIV can be spectrophotometrically followed. Another wavelength could be selected (except the isosbestic points for HbIII/ HbIV) given that HbV is present at very low concentrations during the reaction. Several parameters can be chosen for analytical purposes, such as the absorbance variation at the maxima (ΔAbsm) or the Abs = f(t) area. We have tried to develop theoretical expressions relating these parameters with the [H2O2]0 and [Hb]0 concentrations, but the mathematical equations obtained were complex and difficult to manage. Therefore, the absorbance variation during the first moment of the reaction was finally proposed as an analytical parameter, so the general equation deduced from Eq. (5)

Also, the corresponding mass balance for HbIII is given by: h i 2nþ ½Hb0 ¼ HbIII

0;nþ1

h

2nþ

HbIII

i 0

h i h i 2nþ ð2nþ1Þþ þ HbIII þ HbIV m

ð15Þ

m

h i h i ð2nþ2Þþ ð2nþ1Þþ þ HbIV ¼ HbIII

ð16Þ

ΔAbs540;t ¼ ðεIV ‐εIII Þ

  H O ½Hb ‐½H O  k 2 2 t ½Hb0 ½H2 O2 0 1‐eð 0 2 2 0 Þ 1 2 ½H2 O2 0 ‐½Hb0 eð½Hb0 ‐½H2 O2 0 Þk1

H O2

t

ð18Þ

or the simplified equation similar to Eq. (10) H O

Combining Eqs. (12)–(16) and Eqs. (7)–(10) gives ln

ΔAbs540;t ¼

Abst ‐Abs0;nþ1 ¼ ‐k3;nþ1 ðt‐t m Þ Absm ‐Abs0;nþ1

ð17Þ Abs ‐Abs

From the data given in Fig. 3, the representations ln Absmt ‐Abs0;nþ1 ¼ 0;nþ1 f ðt‐t m Þ are obtained, and the slopes of the corresponding lines give the k3,n + 1 values (Table 1). As can be seen, the k3 values become smaller as the number of cycles increases. It is also possible to obtain the kinetic constant for the degradation of heme group (kd) from the absorbance changes that take place during the process. The methodology (and its validity) is explained in detail

2 2 ðεIV ‐εIII Þ½Hb0 ½H2 O2 0 k1;nþ1 t

1þ

H2 O2 t ½Hb0 k1;nþ1

ð19Þ

can be applied. Note that in both cases, the k1 for the first cycle is used and the hemoglobin concentration appears to correspond to the total added ([Hb]0). Fig. 4 shows the absorbance variation ΔAbs540,t as a function of the H2O2 concentration and the fitting of the data to Eq. (18). From this fitting, a value of 3.3 (±0.2) × 102 M−1 s−1 for the kinetic H O constant k1 2 2 is obtained, which is in accordance with that previously obtained (see Table 1). The figure caption shows the standard deviation of the measurements; in all cases, the RSD obtained were ≤7%. Finally, the figure shows that the linear response range (Eq. (19)) between

0.12

Δ Abs540,20

0.1 0.08 0.06 0.04 0.02 0 0

0.0001

0.0002

0.0003

0.0004

0.0005

[H2O2]0 M Fig. 4. Absorbance variation ΔAbsλ,t as a function of H2O2 concentration. HbIII concentration 8.18 × 10−5 M, t = 20 s. The theoretical curve according to Eq. (18) is shown in grey. Linear response range is show by a straight line. The RSD obtained for the different concentrations were (n = 5): 7% (2.5 × 10 − 5 M), 6.1% (5.0 × 10 − 5 M), 5.4% (1.0 × 10−4 M), 5.0% (2.5 × 10−4 M) and 4.0% (5.0 × 10−4 M).

180

V. Sanz et al. / Microchemical Journal 114 (2014) 175–181

Table 2 Upper limits for the linear range as a function of the measuring time and HbIII concentration predicted by Eqs. (18) and (19). [Hb]0 M

20 s

2 × 10−5 4 × 10−5 6 × 10−5 1 × 10−4 1.2 × 10−4

6 7 8 1.1 1.2

× × × × ×

[GOx]0 M

30 s 10−5 10−5 10−5 10−4 10−4

5 6 7 1 1.1

× × × × ×

Table 3 Effect of glucose oxidase concentration on the absorbance variation at 540 nm (HbIII concentration, 1.1.4 × 10−4 M; glucose concentration, 2.78 × 10−5 M).

10−5 10−5 10−5 10−4 10−4

ΔAbs540,t and the analyte concentration is fulfilled up to about 10−4 M. Obviously, the linear response range will depend on the [Hb]0 concentration used and the measurement time (see Eq. (7)). Table 2 shows the upper limits for the linear response range predicted by Eq. (19) for different conditions. To verify the proposed model, the effect of the total hemoglobin concentration ([Hb]0) on the analytical parameter was studied. Fig. 5 shows the experimental results obtained. Again, the fitting of the H O experimental data to the proposed model gives a value for k1 2 2 of 3.0 2 −1 −1 (±0.2) × 10 M s .

4.2. Determination of peracetic acid (PAA) One of the most relevant applications of H2O2 is as a water or wastewater disinfectant. For this application, H2O2 is combined with peracetic acid (PAA), an organic peroxide with high antimicrobial action, so the simultaneous determination of H2O2 and PAA is of interest. To study the kinetic of the reaction of HbIII with PAA, the absorbance variations for different PAA concentrations were measured. Fig. S6 shows the absorbance variation of these assays at two wavelengths: 516 nm, the isosbestic point for HbIII/HbIV (this wavelength gives the variations of [HbV]), and 540 nm, the proposed wavelength for obtaining the [HbIV] variations (isosbestic point for HbIII/HbIV). This figure also shows the absorbance variation using H2O2. The following conclusions can be derived: 1) PAA reacts with HbIII following a similar mechanism as H2O2. This means that the mathematical model given in Eqs. (18) and (19) can also be applied for PAA. 2) The absorbance variation at 540 nm is lower than that obtained with H2O2, which indicates that the k1 value for PAA is lower than that obtained for H2O2. From the results shown in Fig. S6 and following a similar procedure to that used for H2O2, a kPAA of 1.0 1 (±0.2) × 102 M−1 s−1 was obtained. 3) The absorbance variation at 516 nm is higher than that obtained with H2O2, which means that a higher HbV is stabilized in the presence of PAA.

6.57 8.71 9.79 1.08

× × × ×

10−7 10−7 10−7 10−6

ΔAbs540,max

tmax (s)

0.0090 0.0165 0.0135 0.0150

112 122 112 119

Considering the differences in sensitivity of H2O2 and PAA at 540 and 516 nm, it would be possible to develop a method for the simultaneous determination of both analytes by measuring the absorbance at both wavelengths. 4.3. Coupling of oxidase reactions to HbIII/H2O2 reaction Preliminary assays were performed in order to combine HbIII/H2O2 with a previous enzymatic reaction producing H2O2. The glucose oxidase (GOx) reaction was used as a model of the H2O2 generating enzyme. When the absorbance variation at 540 nm is measured in a solution containing GOx, glucose ([G]) and Hb, the analytical signal obtained is similar to that described above for the Hb/H2O2 system. If the GOx concentration is high enough to achieve a complete conversion of glucose into H2O2, it is possible to apply the model described in Eq. (18) by substituting [H2O2]0 by the initial glucose concentration H O [G]0 and replacing k1 2 2 with a new constant considering the GOx/ HbIII interaction mentioned above (kG 1 ). This model relates the absorbance variations at a selected wavelength with the glucose concentration in the sample:

ΔAbs540;t ¼ ðεIV ‐εIII Þ

  G ½Hb0 ½G0 1‐eð½Hb0 ‐½G0 Þk1 t ½G0 ‐½Hb0 eð½Hb0 ‐½G0 Þk1 t G

ð20Þ

If the GOx concentration is low, a wider peak is obtained owing to the slower contribution of H2O2 through the GOx/Glucose reaction. Therefore, the effect of the GOx concentration on the analytical signal was studied in order to optimize this concentration. The optimal GOx concentration is the minimum concentration required to achieve a non-GOx concentration depending on the absorbance variations. Table 3 shows the effect of the GOx concentration on the absorbance variations at the maximum and the time at which this value is reached. As can be seen, for GOx concentrations higher than 8 × 10−7 M, the analytical signal does not depend on the GOx concentration. It is important to emphasize that the GOx concentration used for this determination is about 150 times lower than those used for obtaining the spectra of compound I, so its contribution to the total absorbance at 540 nm is negligible.

0.1 0.09

Δ Abs540,20

0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 0

0.00002 0.00004 0.00006 0.00008 0.0001 0.00012 0.00014 0.00016 0.00018

[Hb]0 M Fig. 5. Absorbance variations ΔAbsλ,t as a function of hemoglobin concentration. H2O2 concentration 1.25 × 10−4 M, t = 20 s. The theoretical curve according to Eq. (18) is shown in grey.

V. Sanz et al. / Microchemical Journal 114 (2014) 175–181

Appendix A. Supplementary data

Table 4 Effect of Triton concentration on the kinetic of HbIII/H2O2 reaction. HbIII concentration 1.46 × 10−4 M, H2O2 concentration 4.40.10−4 M, in phosphate-buffered solution 0.1 M pH 6. Concentración de Tritón (%)

ΔAbs540,max

0.0025 0.025 0.05 0.25 1

0.1271 0.0925 0.0898 0.0495 0.0310

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.microc.2013.12.016. References

4.4. Stability of the hemoglobin solutions A problem that could arise from the use of hemoglobin as a reagent is its instability in solutions of high ionic strength; it is necessary to stabilize hemoglobin in these particular cases only. This can be achieved with a surfactant such as Triton; for example, 0.25% m/V Triton X-100 stabilizes HbIII during at least 150 min (see Table S4). On the other hand, Triton affects hemoglobin activity. In Table 4, the effect of Triton on the kinetic of HbIII/H2O2 reaction is shown. As can be seen, as Triton X-100 concentration increases, hemoglobin activity decreases. In Fig. S7, it is shown that this decreases linearly depends on the Ln[Triton X-100], which can be understood as if Triton X-100 would modify the k1 constant according to:

k1 ðTritonÞ ≈

k1 ½Triton X‐100

181

ð21Þ

So its effect can be easily implemented into the mathematical model.

5. Conclusions This paper demonstrates that the intrinsic molecular absorption properties of ferric hemoglobin can be used for analytical purposes. The compound can be used for H2O2 determination. The basis for the simultaneous determination of H2O2/PAA and compounds giving H2O2 in previous enzymatic reactions has been given. Comparing the kinetic results obtained for hemoglobin with those previously reported for peroxidase (HRP) indicates that their properties and therefore their analytical possibilities are different. From this point, it would be of interest to study the analytical properties of other hemeproteins (e.g., myoglobin). The solubility problems of hemoglobin (or other hemeproteins) can be solved by using appropriate concentrations of surfactant (see Supplementary Material). They can thus be considered as serious candidates for use as indicators of combined enzymatic reactions.

Acknowledgements This work was supported by the Ministry of Economy and Competitiveness (MINECO) of Spain within the project CTQ20012- 34774, by the University of Zaragoza (UZ2011-CIE-03) and by the Government of Aragón within the founding for Research groups (DGA-FEDER), which is gratefully acknowledged.

[1] P. Chan, P. So, D. Ma, Y. Zhao, T.S. Lai, W.H. Chung, K.C. Chan, K.F. Yiu, H.W. Chan, F.M. Siu, C.W. Tsang, Y.C. Leung, K.Y. Wong, Fluorophore-labeled beta-lactamase as a bionsensor for beta-lactam antibiotics: a study of a biosensing process, J. Am. Chem. Soc. 130 (2008) 6351. [2] S. D'Auria, J. Lakowicz, Enzyme fluorescence as a sensing tool: new perspectives in biotechnology, Curr. Opin. Biotechnol. 12 (2001) 99. [3] J. Galban, I. Sanz-Vicente, E. Ortega, M. del Barrio, S. de Marcos, Reagentless fluorescent biosensors based on proteins for continuous monitoring systems, Anal. Bioanal. Chem. 402 (2012) 3039. [4] J. Stocker, D. Balluch, M. Gsell, H. Harms, J. Feliciano, S. Daunert, K. Malik, J. Van der Meer, Development of a set of simple bacterial biosensors for quantitative and rapid measurements of arsenite and arsenate in potable water, Environ. Sci. Technol. 37 (2003) 4743. [5] W. Trettnak, O. Wolfbeis, Fully reversible fibre-optic glucose biosensor based on the intrinsic fluorescence of glucose-oxidase rid D-2855-2009, Anal. Chim. Acta. 221 (1989) 195. [6] W. Blumberg, J. Peisach, B.A. Wittenbe, J.B. Wittenbe, electronic Structure of protoheme proteins.I. An electron paramagnetic resonance and optical study of horseradish peroxidase and its derivatives, J. Biol. Chem. 243 (1968)1854. [7] B.A. Wittenbe, L. Kampa, J.B. Wittenbe, W. Blumberg, J. Peisach, Electronic structure of protoheme proteins.2. An electron paramagnetic resonance and optical study of cytochrome c peroxidase and its derivatives, J. Biol. Chem. 243 (1968)1863. [8] J. Peisach, W. Blumberg, B.A. Wittenbe, J.B. Wittenbe, Electronic structure of protoheme proteins. 3. Configuration of heme and its Ligands, J. Biol. Chem. 243 (1968)1871. [9] G. Loew, Structure, spectra, and function of heme sites, Int. J. Quantum Chem. 77 (2000) 54. [10] V. Sanz, S. de Marcos, J. Galban, Hydrogen peroxide and peracetic acid determination in waste water using a reversible reagentless biosensor, Anal. Chim. Acta. 583 (2007) 332. [11] V. Sanz, S. de Marcos, J.R. Castillo, J. Galban, Application of molecular absorption properties of horseradish peroxidase for self-indicating enzymatic reactions and analytical methods, J. Am. Chem. Soc. 127 (2005) 2038. [12] W.G. Zijlstra, A. Buursma, Spectrophotometry of hemoglobin: absorption spectra of bovine oxyhemoglobin, deoxyhemoglobin, carboxyhemoglobin and methemoglobin, Comp. Biochem. Physiol. 118B (1997) 743. [13] D.A. Sviturenko, Reaction of haem containing proteins and enzymes with hydroperoxides: the radical view, Biochim. Biophys. Acta 1707 (2005) 127. [14] M.I. González-Sánchez, F. García-Carmona, H. Macià, E. Valero, Catalase-like activity of human methemoglobin: a kinetic and mechanistic study, Arch. Biochem. Biophys. 516 (2011) 10. [15] W. de Jesús-Bonilla, J.E. Cortés-Figueroa, F.A. Souto-Bachiller, L. Rodriguez, J. López-Garriga, Formation of compound I and compound II ferryl species in the reaction of hemoglobin I from Lucina pectinata with hydrogen peroxide, Arch. Biochem. Biophys. 390 (2001) 304. [16] S.B. Brown, H. Hatzikonstantinou, D.G. Herries, The role of peroxide in haem degradation. A study of the oxidation of ferrihaems by hydrogen peroxide, Biochem. J. 174 (1978) 901. [17] E. Nagababu, J.M. Rifkind, Reaction of hydrogen peroxide with ferrylhemoglobin: superoxide production and hene degradation, Biochemistry 39 (2000) 12503. [18] E. Nagababu, J.M. Rifkind, Herm degradation by reactive oxygen species, Antioxid. Redox Signal. 6 (2004) 967. [19] M.I. González-Sánchez, M.C. Manjabacas, F. García-Carmona, E. Valero, Mechanism of acetaminophen oxidation by the peroxidase-like activity of methemoglobin, Chem. Res. Toxicol. 22 (2009) 1841. [20] T. Egawa, H. Shimada, Y. Ishimura, Formation of compound I in the reaction of native myoglobins with hydrogen peroxide, J. Biol. Chem. 275 (2000) 34858. [21] W. Zhu, A. Wilks, I. Stojiljkovic, Degradation of heme in Gram-negative bacteria: the product of the hemo gene of Neisseriae as a heme oxygenase, J. Bacteriol. 182 (2000) 6783. [22] Y. Liu, P.R. Ortiz de Montellano, Reaction intermediates and single turnover rate constants for the oxidation of heme by human heme oxygenase-1, J. Biol. Chem. 275 (2000) 5297.