- Email: [email protected]
Interactions of anionic surfactants with methemoglobin
Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Interactions of anionic surfactants with methemoglobin Lidia Gebicka ∗ , Ewa Banasiak Institute of Applied Radiation Chemistry, Faculty of Chemistry, Technical University of Lodz, Wroblewskiego 15, 93-590 Lodz, Poland
a r t i c l e
i n f o
Article history: Received 10 September 2010 Received in revised form 29 October 2010 Accepted 8 November 2010 Available online 17 November 2010 Keywords: Anionic surfactants Hemichrome Methemoglobin Stopped-flow spectrophotometry
a b s t r a c t Interactions of two anionic surfactants, sodium dodecyl sulphate (SDS) and sodium bis(2-ethylhexyl) sulfosuccinate (AOT) at concentrations below and above critical micelle concentration with methemoglobin (metHb) have been investigated by conventional as well as by stopped-flow absorption and fluorescence spectroscopy. The absorption spectra of metHb in AOT reverse micelles have been also analyzed. Both surfactants in their monomeric form convert metHb to reversible hemichrome. This is connected with a diminution of peroxidase-like activity of metHb and with an increase of the susceptibility of heme for a damage by H2 O2 . In micellar solutions of AOT and SDS as well as in AOT reverse micelles pentacoordinated ferric species seems to be the predominant form of this protein. It has been concluded, basing on a kinetic analysis, that conformational changes in the heme environment of metHb as induced by both surfactants occur independently of the alterations in the tertiary structure of this protein. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Hemoglobin (Hb) is the major heme protein of the red blood cells responsible for the transport of oxygen to the tissues. This function of Hb is possible only when iron is in the ferrous state. The most common ferric form of hemoglobin is a methemoglobin (metHb) with water or hydroxide as a sixth coordination ligand to the iron. The amount of metHb in normal blood cells does not exceed 3%, but it can increase during the interaction of Hb with some molecules ([1], and references therein). Under some conditions exogeneous ligands of Hb and metHb are replaced by the endogeneous side chain histidine and a low-spin hemichrome is formed [2]. In vivo hemichromes are involved in Heinz bodies formation and in the elimination of the older, less functional erythrocytes [2]. Mammalian hemoglobins show a low level of hemichrome in physiological conditions at room temperature [3]. The formation of hemichromes induced by several factors may proceed without or with disruption of the native conformation. Increasing the pressure, lowering the temperature or oxygenation of NO adducts facilitate the formation of reversible hemichrome [2,3]. Irreversible hemichrome formation has been postulated to take place under denaturing conditions, e.g. addition of surfactants or dehydration [2], although partially reversible conversion
Abbreviations: AOT, sodium bis(2-ethylhexyl) sulfosuccinate; Brij35, 2(dodecyloxy)ethanol; cmc, critical micelle concentration; CTAB, hexadecyl trimethyl ammonium bromide; DTAB, dodecyl trimethylammonium bromide; GDA, n-dodecylammonium ␣-glutamate; Hb, hemoglobin; LPS, bacterial lipopolisaccharide; Mb, myoglobin; metHb, methemoglobin; SDS, sodium dodecyl sulphate. ∗ Corresponding author. Tel.: +48 42 6313160; fax: +48 426840043. E-mail address: [email protected] (L. Gebicka). 0927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2010.11.017
of metHb to hemichrome has been observed in the presence of bacterial lipopolisaccharide (LPS) containing long hydrophobic chain of fatty acids [4], deoxycholate or ursodeoxycholate [5]. The widespread use of surfactants in analytical biochemistry has stimulated interest in the surfactant–protein interactions. The interactions of ionic surfactants, sodium dodecyl sulphate (SDS), hexadecyl trimethyl ammonium bromide (CTAB), dodecyl trimethylammonium bromide (DTAB), dodecylammonium ␣-glutamate (GDA) and glutamic acid-based gemini surfactants with hemoglobin (including giant extracellular hemoglobin) and methemoglobin have been studied [6–12]. The interaction between natural lipopeptide surfactant, C15, with Hb has been also reported [13]. It has been shown that ionic surfactants below critical micelle concentration (cmc) enhance the autooxidation of oxyhemoglobin to metHb followed by the formation of hemichrome [6,7,9]. On the other hand some authors have reported that autooxidation process is inhibited by SDS for [SDS]/[Hb] ratio below 50 [14] and metHb is reduced to deoxyhemoglobin when [SDS]/[Hb] ratio is below 10 [15]. The kinetics of hemichrome formation in the reaction of metHb with SDS at concentrations below cmc has been also studied [7]. The results reported in this work extend the earlier investigations concerning interactions of metHb with anionic surfactants. Sodium bis(2-ethylhexyl) sulfosuccinate (AOT) and SDS are used by us. Some results concerning the interaction of metHb with a nonionic surfactant 2-(dodecyloxy)ethanol (Brij35) are also shown for comparison.
2. Materials and methods Hemoglobin from bovine blood was from Sigma. Methemoglobin was prepared by oxidation of Hb with a 2-fold excess of
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
potassium ferricyanide and passage through a column containing Sephadex G-25 using 10 mM phosphate buffer, pH 7.0, as an eluent. Concentration of metHb was determined spectrophotometrically using ε405 = 1.79 × 105 M−1 cm−1 , expressed per heme [16]. AOT, SDS and Brij35 were from Sigma. AOT and SDS were dried under vacuum over P2 O5 . AOT reverse micelles were formed by injection of appropriate amounts of aqueous stock solutions in phosphate buffer, into 0.1 M AOT in n-heptane (Fluka) to obtain the desired wo , i.e. the ratio of [H2 O] to [AOT], where water indicates the buffer alone or buffer containing metHb. A mixture was shaken until a completely transparent solution was obtained. All measurements were done in 10 mM phosphate buffer. Water from MilliQ Plus was used. Measurements were done at ambient temperature 24 ± 1 ◦ C. Cmc in 10 mM phosphate buffer determined by a conductometric method was 2.5 and 4.2 mM for AOT and SDS, respectively. Rapid absorption changes of metHb were studied on an Applied Photophysics SX 18MV stopped-flow spectrofluorimeter with 1 cm cell. A Hewlett–Packard 8452A diode-array spectrophotometer was used for absorption measurements in a long-time regime. Fluorescence measurements were also carried out on an Applied Photophysics stopped-flow apparatus working in a fluorescence mode (0.2 cm cell). The excitation wavelength was 280 nm and a wideband interference filter (320–450 nm) was used.
3. Results It is known that absorption spectra of heme proteins are influenced by heme microenvironment. Methemoglobin, at neutral pH, possesses a characteristic absorption spectrum peaking at 406 nm (Soret), around 500 nm (Q) and at 630 nm (ligand-to-metal charge transfer, LMCT). We found that the absorption spectra in the Soret region of 2.5 M metHb in the presence of AOT at concentrations below 4 × 10−5 M, at pH 7.0, slightly decreased without any shift of an absorption maximum. The shift of the Soret band of metHb to 408 nm took place at 4 × 10−5 M of AOT. In the presence of 0.2–1 mM of AOT the absorption spectrum in the Soret region was characteristic for hemichrome (414 nm). Similar observations have been made after adding metHb and metmyoglobin (metMb) to SDS, CTAB, DTAB or GDA solutions at concentrations below cmc [7,8,10,12,17–19]. The absorption spectra of hemoglobin species in the presence of AOT at concentration ≥2 mM shifted towards shorter wavelengths with the concomitant decrease of an absorbance to about one half of the initial metHb absorbance (Fig. 1). A blue shift of the Soret band from 414 to 406 nm is related to the formation of pentacoordinated species [20–22]. The broad and asymmetric Soret band may indicate the coexistence of two high-spin pentacoordinated species, i.e. a species with proximal histidine as fifth ligand and a pentacoordinated free heme with water molecule as fifth ligand, without direct contact with polypeptide chains (in other words, heme is dissociated from the protein) [20–22]. In Table 1, we compared the wavelengths of the absorption maxima of hemoglobin species in the Soret region in the presence of AOT and SDS. The absorption spectra of hemoglobin species in the presence of AOT taken in the visible region are shown in Fig. 2. The band at 534 nm and a shoulder around 565 nm, characteristic for hemichrome [23], appeared in the presence of AOT at concentration range from 0.2 to 3 mM, i.e. for the ratio [AOT]/[metHb] from 10 to 125 (initial metHb concentration was 20 M). Unfortunately, we were not able to mix 20 M metHb with AOT at a concentration higher than 3 mM. In the case of SDS we observed hemichrome formation in the visible absorption spectra of hemoglobin species when [SDS]/[metHb] ≥7 (data not shown). The band at 630 nm disappears when a low-spin hemichrome is formed (Fig. 2). An increase of the LMCT band observed in the
117
Fig. 1. Absorption spectra in the Soret region of 2.5 M metHb in a buffer solution (1) and in the presence of 0.1 mM (2), 1 mM (3), 2 mM (4) and 3 mM (5) AOT. Table 1 The wavelengths of the absorption maxima of hemoglobin species at the Soret region in the presence of AOT and SDS. [metHb] = 2.5 M. [Surfactant], mM
max in the presence of AOT
max in the presence of SDS
0.01 0.02 0.04 0.05 0.1 0.15 0.2 0.4 0.6 1 2 2.5 3 4 5 7
406 406 408 410 410 412 414 414 414 414 412 412 – 410 408 406
406 408 – 410 412 414 414 414 414 414 412 412 412 412 406 404
presence of 3 mM AOT (line 5) indicates the formation of pentacoordinated high-spin species. Similar observation has been made for metHb in the presence of SDS or DTAB at concentrations higher than cmc [10].
Fig. 2. Absorption spectra in the visible region of 20 M metHb in a buffer solution (1) and in the presence of 0.1 mM (2), 1 mM (3), 2 mM (4) and 3 mM (5) AOT.
118
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
Fig. 3. Absorption spectra in the Soret region of 2.5 M metHb in a buffer solution (1) and in AOT/n-heptane reverse micelle at wo = 20 (2) and wo = 40 (3).
We also studied the absorption spectra of metHb incorporated into reverse micelles formed by AOT in n-heptane at various wo . We found that they resembled those obtained in the presence of aqueous micellar solutions of AOT or SDS (Fig. 3). This indicates that under such conditions pentacoordinated metHb species exist. Nonionic surfactants do not electrostatically interact with proteins and thus they are considered as “soft” [24]. Although the absorption spectra of metHb in the presence of Brij35 at concentrations below and above cmc (∼0.1 mM) do not resemble those of hemichrome or pentacoordinated species, some conformational changes of metHb occur as a result of an interaction with nonionic Brij35 (Fig. 4). It has been found that the addition of albumin to hemichrome formed from metHb in the presence of LPS, deoxycholate or ursodeoxycholate causes the re-formation of metHb [4,5]. We performed similar experiments with hemichrome generated from metHb in the presence of AOT and SDS. MetHb was incubated for 5 min in a surfactant solution to produce hemichrome (Soret absorbance maximum at 414 nm) and then different amounts of bovine serum albumin (BSA) were added. Under conditions when [SDS]/[BSA] < 10 we observed a shift of absorption maximum of hemoglobin species from 414 to 406 nm, i.e. to the value characteristic for metHb (Fig. 5). It means that reversible hemichrome is formed in the presence of SDS (below cmc). When appropriate amount of BSA is present in the system, SDS binds to BSA rather than
Fig. 5. Reversibility of hemichrome formation from metHb in the presence of SDS and BSA. 2 M metHb (1) was incubated with 0.16 mM SDS for 5 minutes to form hemichrome (2) and then 5 M (3), 20 M (4), 50 M (5) and 0.1 mM (6) BSA was added.
to ferric hemoglobin, and hemichrome transforms back to metHb. Similar observations we made for AOT. The kinetics of the formation of hemichrome in the reaction of metHb with AOT and SDS below and above cmc was studied by a stopped-flow technique by following the changes in the absorbance at 406 nm (absorption maximum of metHb) as well as at 534 nm (absorption maximum of hemichrome). The kinetic traces of absorption decrease observed at 406 nm for surfactant concentrations in the range from 0.2 to 3 mM could be fitted well with a double exponential function (A = A1 exp(−k1 t) + A2 exp(−k2 t) + An ) and two pseudo-first-order rate constants k1 and k2 , dependent on surfactant concentration, were obtained (Table 2). Similarly, a double exponential fit could be obtained for kinetic traces of absorption increase observed at 534 nm. The kinetics of the reaction of metHb with AOT and SDS at micellar concentrations was also studied. We found that at 534 nm the initial absorbance increase was followed by a slower absorbance decrease. Fig. 6 shows exemplary kinetic traces recorded at 534 nm in the presence of AOT. The first order rate constant of the slower process is independent of surfactant concentration and equals 2.4 and 13.2 s−1 for AOT and SDS, respectively. We were not able to extract an additional kinetic step in the Soret region. The kinetics of the overall protein conformation changes caused by anionic surfactants at concentrations below cmc were observed by monitoring an increase of fluorescence using the excitation wavelength 280 nm and appeared to be single exponential. The first order rate constants, kfl , increase on increasing surfactant concentration, but their values are different than those of k1 and k2 measured for absorbance changes at 406 nm (Table 2). Table 2 The observed first-order rate constants of metHb absorbance decrease at 406 nm and of fluorescence increase induced by monomeric forms of AOT and SDS as a function of surfactants concentration [metHb] = 2.5 M. [Surfactant], mM
Fig. 4. Absorption spectra in the Soret region of 1.8 M metHb in a buffer solution (1) and in the presence of 0.05 mM (2), 0.1 mM (3) and 1 mM (4) Brij35.
0.2 0.4 0.6 0.8 1 2 3
Absorbance decrease k1 , s−1
Absorbance decrease k2 , s−1
Fluorescence increase kfl , s−1
AOT
SDS
AOT
SDS
AOT
SDS
5 22 40 70 99 95 –
6 20 57 100 250 490 550
0.3 0.7 2.7 3.5 3.7 6.8 –
0.4 2.5 5 6.3 24 63 60
0.1 1.7 3.2 3.8 9.0 18 –
0.25 2.4 3.7 7.4 13 35 40
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
119
Fig. 6. Time courses of the absorption changes at 534 nm after mixing 20 M metHb with 0.2 mM (1), 1 mM (2) and 3 mM (3) AOT.
The reaction of hydrogen peroxide with metHb results in the formation of ferryl hemoglobin (Fe(IV) = O) (ferrylHb) with a globin based radical [25,26]. A globin radical in ferrylHb is very unstable and rapidly decays to a longer lived radical-free ferryl form [27]. Ferryl hemoglobin is able to oxidize biologically relevant molecules. FerrylHb undergoes slow acid-catalyzed autoreduction to the ferric (met) form [28], but the regenerated metHb is unidentical to the initial met state: some hemichrome and dimers of Hb subunits are also formed [29]. When hydrogen peroxide is in a large excess towards methemoglobin, no regeneration of metHb takes place. Instead, a continuous decrease of the absorption spectrum with maximum at 416 nm is observed which is probably connected with the damage of heme (Fig. 7A). Methemoglobin exhibits low peroxidase-like activity, i.e. it is able to oxidize some substrates in the presence of hydrogen peroxide [30]. It has been found that peroxidase-like activity of cytochrome c is greatly enhanced by partial unfolding of the protein, e.g. in the presence of ionic surfactants [31–33]. We checked whether methemoglobin exhibits the same behaviour. Peroxidase-like activity measured in the presence of SDS and AOT at submicellar concentrations was about 3 times lower than that measured in a buffer and did not depend on a surfactant concentration. A lowering of peroxidase-like activity has been also observed in the case of bis-histidyl Fe(III)- and Fe(II)-␣Hb, which are formed in the presence of ␣-hemoglobin-stabilized protein (AHSP) [34]. We did not detect peroxidatic activity of methemoglobin in the presence of micellar amounts of surfactants. The sensitivity of hemichrome towards H2 O2 was also studied. MetHb was mixed with SDS at concentrations below cmc and incubated for 5 minutes to produce hemichrome. Then 0.1 mM H2 O2 was added and the time-resolved absorption spectra were measured. The decrease of the Soret band with time was detected (Fig. 7B). The pseudo-first order rate constant of this decay was significantly higher than the rate constant for the decrease of the Soret band of ferrylHb in the presence of the same amount of H2 O2 and increased with surfactant concentration (Table 3). 4. Discussion It is known that ionic surfactants, even at submillimolar concentrations, can alter protein conformation. In the case of hemoglobin such interactions lead to the formation of hemichrome. It has been postulated that the interactions of free fatty acids and LPS containing long hydrophobic chain of fatty acids with Hb lead to the formation of metHb and hemichrome [4,35]. Since the reaction of ionic surfactants with oxyhemoglobin is very complex
Fig. 7. Time-resolved absorption spectra of 2.2 M metHb (A) and hemichrome formed by a 5 min incubation of 2.2 M metHb with 0.2 mM SDS (B) taken after mixing with 0.1 mM H2 O2 . Spectra were recorded every 2 min during first 10 min of the process and then every 4 min.
and is not completed over a period of hours [7], we studied, via UV/vis spectroscopy, the formation of hemichrome induced by the interaction of anionic surfactants with metHb, the product of hemoglobin autooxidation. The absorption spectra characteristic for hemichrome were observed in the presence of SDS at a concentration range from 0.15 to 4 mM and in the presence of AOT at a concentration range from 0.2 to 2.5 mM (Table 1.), i.e. in the presence of monomeric forms of surfactants. The formation of hemichrome was also connected with a diminution of peroxidaselike activity of metHb. Moreover, conformational changes induced by anionic surfactants at concentrations below cmc facilitated also heme damage by H2 O2 (Fig. 7B). When surfactant concentration approaches cmc value a new high-spin pentacoordinated ferric heme species appears. From the visible absorption spectra (Fig. 2) one can conclude that under our reaction conditions hemichrome → pentacoordinated species transition is not completed. It has been shown by Liu et al. [10] that the amount of anionic SDS needed for hemichrome → pentacoordinated species transition increases with pH of the system, whereas the amount of cationic DTAB needed to cause the same effect decreases with pH. The same authors have noticed that when pH of the solution Table 3 The observed first-order rate constants of the decay of the Soret band of ferrylHb (in the absence of SDS) and hemichrome induced by 0.1 mM H2 O2 . [metHb] = 2.2 M. [SDS], mM
k × 104 , s−1
0 0.2 0.4 0.6 0.8 1 2 4
3.1 4.4 9.1 13.4 16.3 17.4 28.2 44.4
120
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
is lower than the isoelectric point (pI) of hemoglobin, the interactions of anionic surfactants with protein are both electrostatic and hydrophobic, while the interaction of cationic surfactants with hemoglobin is mainly hydrophobic. When pH of the solution is higher than pI, an opposite conclusion can be drawn. They have found, however, that both surfactants at concentrations above cmc induce the release of heme from metHb [10]. The release of heme and its solubilization in CTAB as well as in SDS micellar media has been reported for normal and metal substituted hemoglobin (MHb, where M = Ni and Cu) [36]. Since pI of bovine metHb is 6.86 [10] the overall charge of metHb, under our reaction conditions, is slightly negative. Thus an electrostatic repulsion between metHb and ionic surfactant takes place and the interaction between metHb and AOT or SDS is mostly hydrophobic. We are not able to determine which of the pentacoordinated species, one with proximal histidine or free heme with water as a fifth ligand, is predominant under our reaction conditions. We can only speculate that the lack of peroxidase-like activity of metHb in the presence of micellar amounts of AOT or SDS suggests drastic changes in protein structure which favour heme release. Reverse micelles, which are formed by certain amphiphilic molecules in apolar solvents can entrap water and hydrophilic molecules, including proteins, within their hydrophilic core. It is generally accepted that a layer of water separates and thus protects the protein surface from the inner surface of a surfactant layer. It has been suggested, however, that even in the case of hydrophilic proteins, which are located exclusively in a water pool, electrostatic interactions of charged residues of the protein with the charges of ionic surfactant molecules forming reverse micelles are of considerable importance to the protein structure [24]. In the interior of reverse micelles electrostatic attraction as well as repulsion between ionic surfactant and a protein seem to be important factors inducing conformational changes of entrapped protein. The absorption spectrum of metHb entrapped within reverse micelles formed by AOT in n-heptane is characteristic for pentacoordinated ferric species. The absorption spectra of metHb in CTAB/n-C5 H11 OH/H2 O or GDA/n-C5 H11 OH/H2 O water in oil microemulsions have been reported [8,37]. The authors have concluded that in such systems heme monomer releases from the hydrophobic cavity of hemoglobin. The kinetics of hemichrome formation in the reaction of metHb with SDS has been studied by Sau et al. [7]. The authors have found that three reaction steps could be separated: the breakage of the iron oxygen bond of water molecule coordinated to the heme, binding of the distal histidine to the heme and breakage of the hydrogen-bonding network of the distal histidine with the solvent. They have suggested that at least four molecules of SDS interact with one molecule of metHb. A kinetic analysis of the interactions between myoglobin (Mb) and ionic surfactants has been also published [18,19]. The authors have shown that interaction between Mb and ionic surfactants is a multipathway and multistep process and is dependent on SDS concentration. Two main events occurring in the process have been proposed: a moderate conformational changes in the tertiary structure leading to the formation of hemichrome and a real unfolding (in micellar solution) which brings about the detachment of heme from the protein [18]. Our studies made for AOT and SDS at concentrations below cmc showed that the kinetics of the changes of heme-absorbance could be well fitted with a double exponential function. This implies that conformational changes induced by monomeric forms of AOT and SDS proceed at least in two-steps. We propose, following Sau et al. [7], that the first step is the binding of surfactant to metHb with the formation of metHb–surfactant complex:
metHb + (surfactant)mono [metHb : (surfactant)mono ]
(1)
In the next step hemichrome [metHb:(surfactant)mono ] complex:
is
[metHb : (surfactant)mono ] → hemichrome
formed
from (2)
This reaction should not depend on a surfactant concentration. The observed concentration dependence could reflect incomplete saturation degree of metHb by surfactant. Comparing the kinetics of absorption and fluorescence changes of metHb in the presence of AOT and SDS we could notice that these changes do not follow each other: the kinetics of fluorescence changes is a single exponential and the values of the pseudo-first order rate constants are different (Table 2). This suggests that conformational changes in the heme environment of metHb induced by both surfactants occur independently of the changes in a tertiary structure. Our results are in contrast to those obtained for Mb–surfactant interactions [19]. The strong similarity between kinetic data obtained by fluorescence of tryptophan and heme absorption has prompted the authors to conclusion that conformational changes are global in nature and affect both the heme environment and the aromatic residues to the same extent. It should be noted that the rates of absorbance and fluorescence changes of metHb observed by us in the presence of monomeric form of AOT are smaller than those observed for the same SDS concentration. Similar observation we have earlier made studying interaction of cytochrome c with AOT and SDS [38]. This has been explained by the differences in the geometry of the hydrocarbon part of a surfactant molecule. The presence of two hydrophobic chains in an AOT molecule makes it less flexible than the SDS molecule. Additional kinetic step observed at 534 nm, i.e. an absorbance decrease after initial fast increase could be connected with hemichrome → pentacoordinated heme species (5cHS) transition: hemichrome + (surfactant)mic [hemichrome : (surfactant)mic ] → 5cHS
(3)
Independence of the first-order rate constant of metHb absorption decrease measured at 534 nm in micellar solution of AOT and SDS on a surfactant concentration suggests that the formation of pentacoordinated high-spin species is much slower than the formation of [hemichrome:(surfactant)mic ] complex. Feis et al. [18] who have investigated interaction between myoglobin and SDS, have proposed the additional reaction route which predominates in micellar solution, when SDS concentration exceeds 10 mM: Mb + (SDS)mic [Mb : (SDS)mic ] → free heme
(4)
We are not able to conclude if this reaction path plays a role also in our system. 5. Conclusions AOT and SDS at concentrations below cmc interact with metHb to form reversible hemichrome. Kinetic analysis shows that hemichrome formation proceeds at least in two steps. Contrary to cytochrome c, peroxidase-like activity of metHb in the presence of both surfactants in their monomeric forms is lowered. It has been concluded from the absorption spectra that in micellar solutions of AOT and SDS pentacoordinated high-spin species are formed. The same species seem to appear after incorporation of metHb into reverse micelles formed by AOT in n-heptane. References [1] J. Umbreit, Am. J. Hematol. 82 (2007) 134–144. [2] J.M. Rifkind, O. Abugo, A. Levy, J. Heim, Methods Enzymol. 231 (1994) 449–481.
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121 [3] A. Vergara, L. Vitagliano, C. Verde, G. di Prisco, L. Mazzarella, Methods Enzymol. 436 (2008) 425–444. [4] W. Kaca, R.I. Roth, K.D. Vandegriff, G.C. Chen, F.A. Kuypers, R.M. Winslow, J. Levin, Biochemistry 34 (1995) 11176–11185. [5] T. Hayashi, K. Uchida, G. Takebe, K. Takahashi, Free Radic. Biol. Med. 36 (2004) 1025–1033. [6] D. Ajloo, A.A. Moosavi-Movahedi, G.H. Hakimelahi, A.A. Saboury, H. Gharibi, Colloid Surf. B 26 (2002) 185–196. [7] A.K. Sau, D. Currell, S. Mazumdar, S. Mitra, Biophys. Chem. 98 (2002) 267–273. [8] W. Liu, X. Guo, R. Guo, Int. J. Biol. Macromol. 37 (2005) 232–238. [9] A.L. Poli, L.M. Moreira, M. Tabak, H. Imasato, Colloid Surf. B 52 (2006) 96–104. [10] W. Liu, X. Guo, R. Guo, Int. J. Biol. Macromol. 41 (2007) 548–557. [11] M. Mojtahedi, H. Parastar, M. Jalali-Heravi, J. Chamani, F.C. Chilaca, A.A. Moosavi-Movahedi, Colloid Surf. B 63 (2008) 183–191. [12] Y. Wang, R. Guo, J. Xi, J. Colloid Int. Sci. 331 (2009) 470–475. [13] A. Zou, J. Liu, V.M. Garamus, K. Zheng, R. Willumeit, B. Mu, Biomacromolecules 11 (2010) 593–599. [14] D.M. Reza, A.A. Moosavi-Movahedi, N. Parviz, S. Shahrokh, J. Biochem. Mol. Biol. 35 (2002) 364–370. [15] A.A. Moosavi-Movahedi, M.R. Dayer, P. Norouzi, M. Shamsipur, A. Yeganeh-faal, M.J. Chaichi, H.O. Ghourchian, Colloid Surf. B 30 (2003) 139–146. [16] E. Antonini, M. Brunori, Hemoglobin and Myoglobin in their Reactions with Ligands, North Holland, Amsterdam, 1971. [17] L. Tofani, A. Feis, R.E. Snoke, D. Berti, P. Baglioni, G. Smulevich, Biophys. J. 87 (2004) 1186–1195. [18] A. Feis, L. Tofani, G. De Sanctis, M. Coletta, G. Smulevich, Biophys. J. 92 (2007) 4078–4087. [19] K.K. Andersen, P. Westh, D.E. Otzen, Langmuir 24 (2008) 399–407.
121
[20] A. Boffi, T.K. Das, S.D. Longa, C. Spagnudo, D.L. Rousseau, Biophys. J. 77 (1999) 1143–1149. [21] L.M. Moreira, A.L. Poli, A.J. Costa-Filho, H. Imasato, Biophys. Chem. 124 (2006) 62–72. [22] L.M. Moreira, A.L. Poli, A.J. Costa-Filho, H. Imasato, Int. J. Biol. Macromol. 42 (2008) 103–110. [23] E.A. Rachmilewitz, J. Peisach, W.E. Blumberg, J. Biol. Chem. 246 (1971) 3356–3366. [24] G. Savelli, N. Spreti, P. Di Profio, Curr. Opt. Colloid Int. Sci. 5 (2000) 111–117. [25] C.C. Winterbourn, Methods Enzymol. 186 (1990) 265–272. [26] C. Giulivi, K.J.A. Davies, Methods Enzymol. 231 (1994) 490–496. [27] K.M. McArthur, M.J. Davies, Biochim. Biophys. Acta 1202 (1993) 173–181. [28] L. Tesoriere, M. Allegra, D. D’Arpa, D. Butera, M.A. Livrea, J. Pineal Res. 31 (2001) 114–119. [29] A. Kowalczyk, M. Puchala, K. Wesolowska, E. Serafin, Biochim. Biophys. Acta 1774 (2007) 86–92. [30] J. Everse, N. Hsia, Free Radic. Biol. Med. 22 (1997) 1075–1099. [31] J. Hamachi, A. Fujita, T. Kunitake, J. Am. Chem. Soc. 116 (1994) 8811–8812. [32] L. Gebicka, Res. Chem. Intermed. 27 (2001) 717–723. [33] R.E.M. Diederix, S. Busson, M. Ubbink, G.W. Canters, J. Mol. Catal. B 27 (2004) 75–82. [34] L. Feng, S. Zhou, L. Gu, D.A. Gell, J.P. Mackay, M.J. Weiss, A.J. Gow, Y. Shi, Nature 435 (2005) 697–701. [35] A.A. Akherm, G.M. Andreyuk, M.A. Kisel, P.A. Kiselev, Biochim. Biophys. Acta 992 (1989) 191–194. [36] B. Venkatesh, S. Venkateshrao, S. Jayadevan, J.M. Rifkind, P.T. Manoharan, Biopolymers 80 (2005) 18–25. [37] Y. Wang, X. Guo, R. Guo, J. Colloid Int. Sci. 317 (2008) 568–576. [38] L. Gebicka, J.L. Gebicki, J. Prot. Chem. 18 (1999) 165–172.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Interactions of anionic surfactants with methemoglobin Lidia Gebicka ∗ , Ewa Banasiak Institute of Applied Radiation Chemistry, Faculty of Chemistry, Technical University of Lodz, Wroblewskiego 15, 93-590 Lodz, Poland
a r t i c l e
i n f o
Article history: Received 10 September 2010 Received in revised form 29 October 2010 Accepted 8 November 2010 Available online 17 November 2010 Keywords: Anionic surfactants Hemichrome Methemoglobin Stopped-flow spectrophotometry
a b s t r a c t Interactions of two anionic surfactants, sodium dodecyl sulphate (SDS) and sodium bis(2-ethylhexyl) sulfosuccinate (AOT) at concentrations below and above critical micelle concentration with methemoglobin (metHb) have been investigated by conventional as well as by stopped-flow absorption and fluorescence spectroscopy. The absorption spectra of metHb in AOT reverse micelles have been also analyzed. Both surfactants in their monomeric form convert metHb to reversible hemichrome. This is connected with a diminution of peroxidase-like activity of metHb and with an increase of the susceptibility of heme for a damage by H2 O2 . In micellar solutions of AOT and SDS as well as in AOT reverse micelles pentacoordinated ferric species seems to be the predominant form of this protein. It has been concluded, basing on a kinetic analysis, that conformational changes in the heme environment of metHb as induced by both surfactants occur independently of the alterations in the tertiary structure of this protein. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Hemoglobin (Hb) is the major heme protein of the red blood cells responsible for the transport of oxygen to the tissues. This function of Hb is possible only when iron is in the ferrous state. The most common ferric form of hemoglobin is a methemoglobin (metHb) with water or hydroxide as a sixth coordination ligand to the iron. The amount of metHb in normal blood cells does not exceed 3%, but it can increase during the interaction of Hb with some molecules ([1], and references therein). Under some conditions exogeneous ligands of Hb and metHb are replaced by the endogeneous side chain histidine and a low-spin hemichrome is formed [2]. In vivo hemichromes are involved in Heinz bodies formation and in the elimination of the older, less functional erythrocytes [2]. Mammalian hemoglobins show a low level of hemichrome in physiological conditions at room temperature [3]. The formation of hemichromes induced by several factors may proceed without or with disruption of the native conformation. Increasing the pressure, lowering the temperature or oxygenation of NO adducts facilitate the formation of reversible hemichrome [2,3]. Irreversible hemichrome formation has been postulated to take place under denaturing conditions, e.g. addition of surfactants or dehydration [2], although partially reversible conversion
Abbreviations: AOT, sodium bis(2-ethylhexyl) sulfosuccinate; Brij35, 2(dodecyloxy)ethanol; cmc, critical micelle concentration; CTAB, hexadecyl trimethyl ammonium bromide; DTAB, dodecyl trimethylammonium bromide; GDA, n-dodecylammonium ␣-glutamate; Hb, hemoglobin; LPS, bacterial lipopolisaccharide; Mb, myoglobin; metHb, methemoglobin; SDS, sodium dodecyl sulphate. ∗ Corresponding author. Tel.: +48 42 6313160; fax: +48 426840043. E-mail address: [email protected] (L. Gebicka). 0927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2010.11.017
of metHb to hemichrome has been observed in the presence of bacterial lipopolisaccharide (LPS) containing long hydrophobic chain of fatty acids [4], deoxycholate or ursodeoxycholate [5]. The widespread use of surfactants in analytical biochemistry has stimulated interest in the surfactant–protein interactions. The interactions of ionic surfactants, sodium dodecyl sulphate (SDS), hexadecyl trimethyl ammonium bromide (CTAB), dodecyl trimethylammonium bromide (DTAB), dodecylammonium ␣-glutamate (GDA) and glutamic acid-based gemini surfactants with hemoglobin (including giant extracellular hemoglobin) and methemoglobin have been studied [6–12]. The interaction between natural lipopeptide surfactant, C15, with Hb has been also reported [13]. It has been shown that ionic surfactants below critical micelle concentration (cmc) enhance the autooxidation of oxyhemoglobin to metHb followed by the formation of hemichrome [6,7,9]. On the other hand some authors have reported that autooxidation process is inhibited by SDS for [SDS]/[Hb] ratio below 50 [14] and metHb is reduced to deoxyhemoglobin when [SDS]/[Hb] ratio is below 10 [15]. The kinetics of hemichrome formation in the reaction of metHb with SDS at concentrations below cmc has been also studied [7]. The results reported in this work extend the earlier investigations concerning interactions of metHb with anionic surfactants. Sodium bis(2-ethylhexyl) sulfosuccinate (AOT) and SDS are used by us. Some results concerning the interaction of metHb with a nonionic surfactant 2-(dodecyloxy)ethanol (Brij35) are also shown for comparison.
2. Materials and methods Hemoglobin from bovine blood was from Sigma. Methemoglobin was prepared by oxidation of Hb with a 2-fold excess of
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
potassium ferricyanide and passage through a column containing Sephadex G-25 using 10 mM phosphate buffer, pH 7.0, as an eluent. Concentration of metHb was determined spectrophotometrically using ε405 = 1.79 × 105 M−1 cm−1 , expressed per heme [16]. AOT, SDS and Brij35 were from Sigma. AOT and SDS were dried under vacuum over P2 O5 . AOT reverse micelles were formed by injection of appropriate amounts of aqueous stock solutions in phosphate buffer, into 0.1 M AOT in n-heptane (Fluka) to obtain the desired wo , i.e. the ratio of [H2 O] to [AOT], where water indicates the buffer alone or buffer containing metHb. A mixture was shaken until a completely transparent solution was obtained. All measurements were done in 10 mM phosphate buffer. Water from MilliQ Plus was used. Measurements were done at ambient temperature 24 ± 1 ◦ C. Cmc in 10 mM phosphate buffer determined by a conductometric method was 2.5 and 4.2 mM for AOT and SDS, respectively. Rapid absorption changes of metHb were studied on an Applied Photophysics SX 18MV stopped-flow spectrofluorimeter with 1 cm cell. A Hewlett–Packard 8452A diode-array spectrophotometer was used for absorption measurements in a long-time regime. Fluorescence measurements were also carried out on an Applied Photophysics stopped-flow apparatus working in a fluorescence mode (0.2 cm cell). The excitation wavelength was 280 nm and a wideband interference filter (320–450 nm) was used.
3. Results It is known that absorption spectra of heme proteins are influenced by heme microenvironment. Methemoglobin, at neutral pH, possesses a characteristic absorption spectrum peaking at 406 nm (Soret), around 500 nm (Q) and at 630 nm (ligand-to-metal charge transfer, LMCT). We found that the absorption spectra in the Soret region of 2.5 M metHb in the presence of AOT at concentrations below 4 × 10−5 M, at pH 7.0, slightly decreased without any shift of an absorption maximum. The shift of the Soret band of metHb to 408 nm took place at 4 × 10−5 M of AOT. In the presence of 0.2–1 mM of AOT the absorption spectrum in the Soret region was characteristic for hemichrome (414 nm). Similar observations have been made after adding metHb and metmyoglobin (metMb) to SDS, CTAB, DTAB or GDA solutions at concentrations below cmc [7,8,10,12,17–19]. The absorption spectra of hemoglobin species in the presence of AOT at concentration ≥2 mM shifted towards shorter wavelengths with the concomitant decrease of an absorbance to about one half of the initial metHb absorbance (Fig. 1). A blue shift of the Soret band from 414 to 406 nm is related to the formation of pentacoordinated species [20–22]. The broad and asymmetric Soret band may indicate the coexistence of two high-spin pentacoordinated species, i.e. a species with proximal histidine as fifth ligand and a pentacoordinated free heme with water molecule as fifth ligand, without direct contact with polypeptide chains (in other words, heme is dissociated from the protein) [20–22]. In Table 1, we compared the wavelengths of the absorption maxima of hemoglobin species in the Soret region in the presence of AOT and SDS. The absorption spectra of hemoglobin species in the presence of AOT taken in the visible region are shown in Fig. 2. The band at 534 nm and a shoulder around 565 nm, characteristic for hemichrome [23], appeared in the presence of AOT at concentration range from 0.2 to 3 mM, i.e. for the ratio [AOT]/[metHb] from 10 to 125 (initial metHb concentration was 20 M). Unfortunately, we were not able to mix 20 M metHb with AOT at a concentration higher than 3 mM. In the case of SDS we observed hemichrome formation in the visible absorption spectra of hemoglobin species when [SDS]/[metHb] ≥7 (data not shown). The band at 630 nm disappears when a low-spin hemichrome is formed (Fig. 2). An increase of the LMCT band observed in the
117
Fig. 1. Absorption spectra in the Soret region of 2.5 M metHb in a buffer solution (1) and in the presence of 0.1 mM (2), 1 mM (3), 2 mM (4) and 3 mM (5) AOT. Table 1 The wavelengths of the absorption maxima of hemoglobin species at the Soret region in the presence of AOT and SDS. [metHb] = 2.5 M. [Surfactant], mM
max in the presence of AOT
max in the presence of SDS
0.01 0.02 0.04 0.05 0.1 0.15 0.2 0.4 0.6 1 2 2.5 3 4 5 7
406 406 408 410 410 412 414 414 414 414 412 412 – 410 408 406
406 408 – 410 412 414 414 414 414 414 412 412 412 412 406 404
presence of 3 mM AOT (line 5) indicates the formation of pentacoordinated high-spin species. Similar observation has been made for metHb in the presence of SDS or DTAB at concentrations higher than cmc [10].
Fig. 2. Absorption spectra in the visible region of 20 M metHb in a buffer solution (1) and in the presence of 0.1 mM (2), 1 mM (3), 2 mM (4) and 3 mM (5) AOT.
118
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
Fig. 3. Absorption spectra in the Soret region of 2.5 M metHb in a buffer solution (1) and in AOT/n-heptane reverse micelle at wo = 20 (2) and wo = 40 (3).
We also studied the absorption spectra of metHb incorporated into reverse micelles formed by AOT in n-heptane at various wo . We found that they resembled those obtained in the presence of aqueous micellar solutions of AOT or SDS (Fig. 3). This indicates that under such conditions pentacoordinated metHb species exist. Nonionic surfactants do not electrostatically interact with proteins and thus they are considered as “soft” [24]. Although the absorption spectra of metHb in the presence of Brij35 at concentrations below and above cmc (∼0.1 mM) do not resemble those of hemichrome or pentacoordinated species, some conformational changes of metHb occur as a result of an interaction with nonionic Brij35 (Fig. 4). It has been found that the addition of albumin to hemichrome formed from metHb in the presence of LPS, deoxycholate or ursodeoxycholate causes the re-formation of metHb [4,5]. We performed similar experiments with hemichrome generated from metHb in the presence of AOT and SDS. MetHb was incubated for 5 min in a surfactant solution to produce hemichrome (Soret absorbance maximum at 414 nm) and then different amounts of bovine serum albumin (BSA) were added. Under conditions when [SDS]/[BSA] < 10 we observed a shift of absorption maximum of hemoglobin species from 414 to 406 nm, i.e. to the value characteristic for metHb (Fig. 5). It means that reversible hemichrome is formed in the presence of SDS (below cmc). When appropriate amount of BSA is present in the system, SDS binds to BSA rather than
Fig. 5. Reversibility of hemichrome formation from metHb in the presence of SDS and BSA. 2 M metHb (1) was incubated with 0.16 mM SDS for 5 minutes to form hemichrome (2) and then 5 M (3), 20 M (4), 50 M (5) and 0.1 mM (6) BSA was added.
to ferric hemoglobin, and hemichrome transforms back to metHb. Similar observations we made for AOT. The kinetics of the formation of hemichrome in the reaction of metHb with AOT and SDS below and above cmc was studied by a stopped-flow technique by following the changes in the absorbance at 406 nm (absorption maximum of metHb) as well as at 534 nm (absorption maximum of hemichrome). The kinetic traces of absorption decrease observed at 406 nm for surfactant concentrations in the range from 0.2 to 3 mM could be fitted well with a double exponential function (A = A1 exp(−k1 t) + A2 exp(−k2 t) + An ) and two pseudo-first-order rate constants k1 and k2 , dependent on surfactant concentration, were obtained (Table 2). Similarly, a double exponential fit could be obtained for kinetic traces of absorption increase observed at 534 nm. The kinetics of the reaction of metHb with AOT and SDS at micellar concentrations was also studied. We found that at 534 nm the initial absorbance increase was followed by a slower absorbance decrease. Fig. 6 shows exemplary kinetic traces recorded at 534 nm in the presence of AOT. The first order rate constant of the slower process is independent of surfactant concentration and equals 2.4 and 13.2 s−1 for AOT and SDS, respectively. We were not able to extract an additional kinetic step in the Soret region. The kinetics of the overall protein conformation changes caused by anionic surfactants at concentrations below cmc were observed by monitoring an increase of fluorescence using the excitation wavelength 280 nm and appeared to be single exponential. The first order rate constants, kfl , increase on increasing surfactant concentration, but their values are different than those of k1 and k2 measured for absorbance changes at 406 nm (Table 2). Table 2 The observed first-order rate constants of metHb absorbance decrease at 406 nm and of fluorescence increase induced by monomeric forms of AOT and SDS as a function of surfactants concentration [metHb] = 2.5 M. [Surfactant], mM
Fig. 4. Absorption spectra in the Soret region of 1.8 M metHb in a buffer solution (1) and in the presence of 0.05 mM (2), 0.1 mM (3) and 1 mM (4) Brij35.
0.2 0.4 0.6 0.8 1 2 3
Absorbance decrease k1 , s−1
Absorbance decrease k2 , s−1
Fluorescence increase kfl , s−1
AOT
SDS
AOT
SDS
AOT
SDS
5 22 40 70 99 95 –
6 20 57 100 250 490 550
0.3 0.7 2.7 3.5 3.7 6.8 –
0.4 2.5 5 6.3 24 63 60
0.1 1.7 3.2 3.8 9.0 18 –
0.25 2.4 3.7 7.4 13 35 40
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
119
Fig. 6. Time courses of the absorption changes at 534 nm after mixing 20 M metHb with 0.2 mM (1), 1 mM (2) and 3 mM (3) AOT.
The reaction of hydrogen peroxide with metHb results in the formation of ferryl hemoglobin (Fe(IV) = O) (ferrylHb) with a globin based radical [25,26]. A globin radical in ferrylHb is very unstable and rapidly decays to a longer lived radical-free ferryl form [27]. Ferryl hemoglobin is able to oxidize biologically relevant molecules. FerrylHb undergoes slow acid-catalyzed autoreduction to the ferric (met) form [28], but the regenerated metHb is unidentical to the initial met state: some hemichrome and dimers of Hb subunits are also formed [29]. When hydrogen peroxide is in a large excess towards methemoglobin, no regeneration of metHb takes place. Instead, a continuous decrease of the absorption spectrum with maximum at 416 nm is observed which is probably connected with the damage of heme (Fig. 7A). Methemoglobin exhibits low peroxidase-like activity, i.e. it is able to oxidize some substrates in the presence of hydrogen peroxide [30]. It has been found that peroxidase-like activity of cytochrome c is greatly enhanced by partial unfolding of the protein, e.g. in the presence of ionic surfactants [31–33]. We checked whether methemoglobin exhibits the same behaviour. Peroxidase-like activity measured in the presence of SDS and AOT at submicellar concentrations was about 3 times lower than that measured in a buffer and did not depend on a surfactant concentration. A lowering of peroxidase-like activity has been also observed in the case of bis-histidyl Fe(III)- and Fe(II)-␣Hb, which are formed in the presence of ␣-hemoglobin-stabilized protein (AHSP) [34]. We did not detect peroxidatic activity of methemoglobin in the presence of micellar amounts of surfactants. The sensitivity of hemichrome towards H2 O2 was also studied. MetHb was mixed with SDS at concentrations below cmc and incubated for 5 minutes to produce hemichrome. Then 0.1 mM H2 O2 was added and the time-resolved absorption spectra were measured. The decrease of the Soret band with time was detected (Fig. 7B). The pseudo-first order rate constant of this decay was significantly higher than the rate constant for the decrease of the Soret band of ferrylHb in the presence of the same amount of H2 O2 and increased with surfactant concentration (Table 3). 4. Discussion It is known that ionic surfactants, even at submillimolar concentrations, can alter protein conformation. In the case of hemoglobin such interactions lead to the formation of hemichrome. It has been postulated that the interactions of free fatty acids and LPS containing long hydrophobic chain of fatty acids with Hb lead to the formation of metHb and hemichrome [4,35]. Since the reaction of ionic surfactants with oxyhemoglobin is very complex
Fig. 7. Time-resolved absorption spectra of 2.2 M metHb (A) and hemichrome formed by a 5 min incubation of 2.2 M metHb with 0.2 mM SDS (B) taken after mixing with 0.1 mM H2 O2 . Spectra were recorded every 2 min during first 10 min of the process and then every 4 min.
and is not completed over a period of hours [7], we studied, via UV/vis spectroscopy, the formation of hemichrome induced by the interaction of anionic surfactants with metHb, the product of hemoglobin autooxidation. The absorption spectra characteristic for hemichrome were observed in the presence of SDS at a concentration range from 0.15 to 4 mM and in the presence of AOT at a concentration range from 0.2 to 2.5 mM (Table 1.), i.e. in the presence of monomeric forms of surfactants. The formation of hemichrome was also connected with a diminution of peroxidaselike activity of metHb. Moreover, conformational changes induced by anionic surfactants at concentrations below cmc facilitated also heme damage by H2 O2 (Fig. 7B). When surfactant concentration approaches cmc value a new high-spin pentacoordinated ferric heme species appears. From the visible absorption spectra (Fig. 2) one can conclude that under our reaction conditions hemichrome → pentacoordinated species transition is not completed. It has been shown by Liu et al. [10] that the amount of anionic SDS needed for hemichrome → pentacoordinated species transition increases with pH of the system, whereas the amount of cationic DTAB needed to cause the same effect decreases with pH. The same authors have noticed that when pH of the solution Table 3 The observed first-order rate constants of the decay of the Soret band of ferrylHb (in the absence of SDS) and hemichrome induced by 0.1 mM H2 O2 . [metHb] = 2.2 M. [SDS], mM
k × 104 , s−1
0 0.2 0.4 0.6 0.8 1 2 4
3.1 4.4 9.1 13.4 16.3 17.4 28.2 44.4
120
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121
is lower than the isoelectric point (pI) of hemoglobin, the interactions of anionic surfactants with protein are both electrostatic and hydrophobic, while the interaction of cationic surfactants with hemoglobin is mainly hydrophobic. When pH of the solution is higher than pI, an opposite conclusion can be drawn. They have found, however, that both surfactants at concentrations above cmc induce the release of heme from metHb [10]. The release of heme and its solubilization in CTAB as well as in SDS micellar media has been reported for normal and metal substituted hemoglobin (MHb, where M = Ni and Cu) [36]. Since pI of bovine metHb is 6.86 [10] the overall charge of metHb, under our reaction conditions, is slightly negative. Thus an electrostatic repulsion between metHb and ionic surfactant takes place and the interaction between metHb and AOT or SDS is mostly hydrophobic. We are not able to determine which of the pentacoordinated species, one with proximal histidine or free heme with water as a fifth ligand, is predominant under our reaction conditions. We can only speculate that the lack of peroxidase-like activity of metHb in the presence of micellar amounts of AOT or SDS suggests drastic changes in protein structure which favour heme release. Reverse micelles, which are formed by certain amphiphilic molecules in apolar solvents can entrap water and hydrophilic molecules, including proteins, within their hydrophilic core. It is generally accepted that a layer of water separates and thus protects the protein surface from the inner surface of a surfactant layer. It has been suggested, however, that even in the case of hydrophilic proteins, which are located exclusively in a water pool, electrostatic interactions of charged residues of the protein with the charges of ionic surfactant molecules forming reverse micelles are of considerable importance to the protein structure [24]. In the interior of reverse micelles electrostatic attraction as well as repulsion between ionic surfactant and a protein seem to be important factors inducing conformational changes of entrapped protein. The absorption spectrum of metHb entrapped within reverse micelles formed by AOT in n-heptane is characteristic for pentacoordinated ferric species. The absorption spectra of metHb in CTAB/n-C5 H11 OH/H2 O or GDA/n-C5 H11 OH/H2 O water in oil microemulsions have been reported [8,37]. The authors have concluded that in such systems heme monomer releases from the hydrophobic cavity of hemoglobin. The kinetics of hemichrome formation in the reaction of metHb with SDS has been studied by Sau et al. [7]. The authors have found that three reaction steps could be separated: the breakage of the iron oxygen bond of water molecule coordinated to the heme, binding of the distal histidine to the heme and breakage of the hydrogen-bonding network of the distal histidine with the solvent. They have suggested that at least four molecules of SDS interact with one molecule of metHb. A kinetic analysis of the interactions between myoglobin (Mb) and ionic surfactants has been also published [18,19]. The authors have shown that interaction between Mb and ionic surfactants is a multipathway and multistep process and is dependent on SDS concentration. Two main events occurring in the process have been proposed: a moderate conformational changes in the tertiary structure leading to the formation of hemichrome and a real unfolding (in micellar solution) which brings about the detachment of heme from the protein [18]. Our studies made for AOT and SDS at concentrations below cmc showed that the kinetics of the changes of heme-absorbance could be well fitted with a double exponential function. This implies that conformational changes induced by monomeric forms of AOT and SDS proceed at least in two-steps. We propose, following Sau et al. [7], that the first step is the binding of surfactant to metHb with the formation of metHb–surfactant complex:
metHb + (surfactant)mono [metHb : (surfactant)mono ]
(1)
In the next step hemichrome [metHb:(surfactant)mono ] complex:
is
[metHb : (surfactant)mono ] → hemichrome
formed
from (2)
This reaction should not depend on a surfactant concentration. The observed concentration dependence could reflect incomplete saturation degree of metHb by surfactant. Comparing the kinetics of absorption and fluorescence changes of metHb in the presence of AOT and SDS we could notice that these changes do not follow each other: the kinetics of fluorescence changes is a single exponential and the values of the pseudo-first order rate constants are different (Table 2). This suggests that conformational changes in the heme environment of metHb induced by both surfactants occur independently of the changes in a tertiary structure. Our results are in contrast to those obtained for Mb–surfactant interactions [19]. The strong similarity between kinetic data obtained by fluorescence of tryptophan and heme absorption has prompted the authors to conclusion that conformational changes are global in nature and affect both the heme environment and the aromatic residues to the same extent. It should be noted that the rates of absorbance and fluorescence changes of metHb observed by us in the presence of monomeric form of AOT are smaller than those observed for the same SDS concentration. Similar observation we have earlier made studying interaction of cytochrome c with AOT and SDS [38]. This has been explained by the differences in the geometry of the hydrocarbon part of a surfactant molecule. The presence of two hydrophobic chains in an AOT molecule makes it less flexible than the SDS molecule. Additional kinetic step observed at 534 nm, i.e. an absorbance decrease after initial fast increase could be connected with hemichrome → pentacoordinated heme species (5cHS) transition: hemichrome + (surfactant)mic [hemichrome : (surfactant)mic ] → 5cHS
(3)
Independence of the first-order rate constant of metHb absorption decrease measured at 534 nm in micellar solution of AOT and SDS on a surfactant concentration suggests that the formation of pentacoordinated high-spin species is much slower than the formation of [hemichrome:(surfactant)mic ] complex. Feis et al. [18] who have investigated interaction between myoglobin and SDS, have proposed the additional reaction route which predominates in micellar solution, when SDS concentration exceeds 10 mM: Mb + (SDS)mic [Mb : (SDS)mic ] → free heme
(4)
We are not able to conclude if this reaction path plays a role also in our system. 5. Conclusions AOT and SDS at concentrations below cmc interact with metHb to form reversible hemichrome. Kinetic analysis shows that hemichrome formation proceeds at least in two steps. Contrary to cytochrome c, peroxidase-like activity of metHb in the presence of both surfactants in their monomeric forms is lowered. It has been concluded from the absorption spectra that in micellar solutions of AOT and SDS pentacoordinated high-spin species are formed. The same species seem to appear after incorporation of metHb into reverse micelles formed by AOT in n-heptane. References [1] J. Umbreit, Am. J. Hematol. 82 (2007) 134–144. [2] J.M. Rifkind, O. Abugo, A. Levy, J. Heim, Methods Enzymol. 231 (1994) 449–481.
L. Gebicka, E. Banasiak / Colloids and Surfaces B: Biointerfaces 83 (2011) 116–121 [3] A. Vergara, L. Vitagliano, C. Verde, G. di Prisco, L. Mazzarella, Methods Enzymol. 436 (2008) 425–444. [4] W. Kaca, R.I. Roth, K.D. Vandegriff, G.C. Chen, F.A. Kuypers, R.M. Winslow, J. Levin, Biochemistry 34 (1995) 11176–11185. [5] T. Hayashi, K. Uchida, G. Takebe, K. Takahashi, Free Radic. Biol. Med. 36 (2004) 1025–1033. [6] D. Ajloo, A.A. Moosavi-Movahedi, G.H. Hakimelahi, A.A. Saboury, H. Gharibi, Colloid Surf. B 26 (2002) 185–196. [7] A.K. Sau, D. Currell, S. Mazumdar, S. Mitra, Biophys. Chem. 98 (2002) 267–273. [8] W. Liu, X. Guo, R. Guo, Int. J. Biol. Macromol. 37 (2005) 232–238. [9] A.L. Poli, L.M. Moreira, M. Tabak, H. Imasato, Colloid Surf. B 52 (2006) 96–104. [10] W. Liu, X. Guo, R. Guo, Int. J. Biol. Macromol. 41 (2007) 548–557. [11] M. Mojtahedi, H. Parastar, M. Jalali-Heravi, J. Chamani, F.C. Chilaca, A.A. Moosavi-Movahedi, Colloid Surf. B 63 (2008) 183–191. [12] Y. Wang, R. Guo, J. Xi, J. Colloid Int. Sci. 331 (2009) 470–475. [13] A. Zou, J. Liu, V.M. Garamus, K. Zheng, R. Willumeit, B. Mu, Biomacromolecules 11 (2010) 593–599. [14] D.M. Reza, A.A. Moosavi-Movahedi, N. Parviz, S. Shahrokh, J. Biochem. Mol. Biol. 35 (2002) 364–370. [15] A.A. Moosavi-Movahedi, M.R. Dayer, P. Norouzi, M. Shamsipur, A. Yeganeh-faal, M.J. Chaichi, H.O. Ghourchian, Colloid Surf. B 30 (2003) 139–146. [16] E. Antonini, M. Brunori, Hemoglobin and Myoglobin in their Reactions with Ligands, North Holland, Amsterdam, 1971. [17] L. Tofani, A. Feis, R.E. Snoke, D. Berti, P. Baglioni, G. Smulevich, Biophys. J. 87 (2004) 1186–1195. [18] A. Feis, L. Tofani, G. De Sanctis, M. Coletta, G. Smulevich, Biophys. J. 92 (2007) 4078–4087. [19] K.K. Andersen, P. Westh, D.E. Otzen, Langmuir 24 (2008) 399–407.
121
[20] A. Boffi, T.K. Das, S.D. Longa, C. Spagnudo, D.L. Rousseau, Biophys. J. 77 (1999) 1143–1149. [21] L.M. Moreira, A.L. Poli, A.J. Costa-Filho, H. Imasato, Biophys. Chem. 124 (2006) 62–72. [22] L.M. Moreira, A.L. Poli, A.J. Costa-Filho, H. Imasato, Int. J. Biol. Macromol. 42 (2008) 103–110. [23] E.A. Rachmilewitz, J. Peisach, W.E. Blumberg, J. Biol. Chem. 246 (1971) 3356–3366. [24] G. Savelli, N. Spreti, P. Di Profio, Curr. Opt. Colloid Int. Sci. 5 (2000) 111–117. [25] C.C. Winterbourn, Methods Enzymol. 186 (1990) 265–272. [26] C. Giulivi, K.J.A. Davies, Methods Enzymol. 231 (1994) 490–496. [27] K.M. McArthur, M.J. Davies, Biochim. Biophys. Acta 1202 (1993) 173–181. [28] L. Tesoriere, M. Allegra, D. D’Arpa, D. Butera, M.A. Livrea, J. Pineal Res. 31 (2001) 114–119. [29] A. Kowalczyk, M. Puchala, K. Wesolowska, E. Serafin, Biochim. Biophys. Acta 1774 (2007) 86–92. [30] J. Everse, N. Hsia, Free Radic. Biol. Med. 22 (1997) 1075–1099. [31] J. Hamachi, A. Fujita, T. Kunitake, J. Am. Chem. Soc. 116 (1994) 8811–8812. [32] L. Gebicka, Res. Chem. Intermed. 27 (2001) 717–723. [33] R.E.M. Diederix, S. Busson, M. Ubbink, G.W. Canters, J. Mol. Catal. B 27 (2004) 75–82. [34] L. Feng, S. Zhou, L. Gu, D.A. Gell, J.P. Mackay, M.J. Weiss, A.J. Gow, Y. Shi, Nature 435 (2005) 697–701. [35] A.A. Akherm, G.M. Andreyuk, M.A. Kisel, P.A. Kiselev, Biochim. Biophys. Acta 992 (1989) 191–194. [36] B. Venkatesh, S. Venkateshrao, S. Jayadevan, J.M. Rifkind, P.T. Manoharan, Biopolymers 80 (2005) 18–25. [37] Y. Wang, X. Guo, R. Guo, J. Colloid Int. Sci. 317 (2008) 568–576. [38] L. Gebicka, J.L. Gebicki, J. Prot. Chem. 18 (1999) 165–172.