J. Phys. Chem. B 2008, 112, 4441-4446
4441
Pulse Radiolysis Studies on Reactions of Hydroxyl Radicals with Selenocystine Derivatives B. Mishra,† L. B. Kumbhare,‡ V. K. Jain,‡ and K. I. Priyadarsini*,† Radiation and Photochemistry DiVision, Chemistry DiVision, Bhabha Atomic Research Centre, Trombay, Mumbai-400085, India ReceiVed: October 10, 2007; In Final Form: January 22, 2008
Reactions of hydroxyl radicals (•OH) with selenocystine (SeCys) and two of its analogues, diselenodipropionic acid (SeP) and selenocystamine (SeA), have been studied in aqueous solutions at pHs of 1, 7, and 10 using the pulse radiolysis technique coupled with absorption detection. All of these diselenides react with •OH radicals with rate constants of ∼1010 M-1 s-1, producing diselenide radical cations (∼1-5 µs after the pulse), with an absorption maximum at 560 nm, by elimination of H2O or OH- from hydroxyl radical adducts. Assignment of the 560 nm band to the diselenide radical cation was made by comparing the transient spectra with those produced upon reaction of diselenides with specific one-electron oxidants, Cl2•- (pH 1) and Br2•radicals (pHs of 7 and 10). SeP having a carboxylic acid functionality showed quantitative conversion of hydroxyl radical adducts to radical cations. The compounds SeCys and SeA, having an amino functional group, in addition to the radical cations, produced a new transient with λmax at 460 nm, at later time scales (∼20-40 µs after the pulse). The rate and yield of formation of the 460 nm band increased with increasing concentrations of either SeCys or SeA. In analogy with similar studies reported for analogous disulfides, the 460 nm transient absorption band has been assigned to a triselenide radical adduct. The one-electron reduction potentials of the compounds were estimated to be 0.96, 1.3, and 1.6 V versus NHE, respectively, for SeP, SeCys, and SeA at pH 7. From these studies, it has been concluded that the electron-donating carboxylic acid group decreases the reduction potential and facilitates quantitative conversion of hydroxyl radical adducts to radical cations, while the electron-withdrawing NH3+ group not only increases the reduction potential but also leads to fragmentation of the hydroxyl radical adduct to selenyl radicals, which are converted to triselenide radical adducts.
Introduction Sulfur and selenium belong to the same group of the periodic table and are among the essential elements in biological systems.1-3 While sulfur compounds are more abundant in nature and their chemical reactions are well studied, the role of selenium was recognized only recently and is believed to play a very important role in human biology.2-5 Until now, at least 32 selenium-containing proteins have been identified in mammals,6 for example, glutathione peroxidase (GPx) and thioredoxin reductase (TrxR). The activity of such selenoenzymes depends mainly on the redox state of selenium.3,7 Both sulfur and selenium occur in variable oxidation states, and depending on the nature of the substituents, the structure, and the oxidation state, they play a crucial role in redox reactions.2-7 The redox processes in selenium-containing amino acids and proteins can take place either as one-electron-transfer or two-electron-transfer reactions. Understanding of such reactions is necessary to exploit the biological activity of such compounds.1-7 At present, several research groups are involved in developing new organoselenium compounds as enzyme mimics. Compounds which show promising activity are aliphatic and aromatic selenides, selenoethers, selenoaminoacids, and diselenides.4,5,8-10 Among these, diselenides were initially considered nonexistent in natural proteins. However, recent reports suggest that the formation of diselenide bonds is an important step in the selenoprotein redox reactions under physiological conditions.11 * To whom correspondence should be addressed. † Radiation and Photochemistry Division. ‡ Chemistry Division.
Recently, we have initiated studies on redox reactions of some water-soluble organoselenium compounds that show promising GPx activity.12 The redox reactions of some of these compounds were carried out by the nanosecond pulse radiolysis technique coupled with absorption detection.13-15 From these studies, it has been observed that compared to the sulfur compounds, selenium compounds undergo easy oxidation, and the transient species formed during oxidation show stronger absorption bands. In the literature, one- and two-electron-transfer processes in many natural and synthetic sulfur compounds have been reported extensively using techniques like pulse radiolysis, cyclic voltammetry, conductivity, EPR spectroscopy, DFT studies, and so forth.16-22 However, such detailed studies with selenium compounds are not many, except for a few reports on aryl and aliphatic mono- and diselenides using pulse radiolysis.12-15, 23-28 Thus, to understand the nature and interaction of selenium-centered radical cations and radical adducts in diselenides, we have examined reactions of hydroxyl radicals with selenocystine (SeCys), diselenodipropionic acid (SeP), and selenocystamine (SeA) (Scheme 1, where R-Se-Se-R represents a diselenide). The transients formed during the reactions were detected by absorption spectroscopy. Experimental Section Materials. All of the chemicals and reagents were of “Analar” grade and used as such. Solutions were prepared in “nanopure” water with a conductivity of 0.1 µS cm-1, obtained from a Millipore Elix3/A10 water purification system. Freshly prepared
10.1021/jp709880b CCC: $40.75 © 2008 American Chemical Society Published on Web 03/15/2008
4442 J. Phys. Chem. B, Vol. 112, No. 14, 2008
Mishra et al. is 0.6 µmol/J. At pH 1, •OH radical reactions were studied under aerated conditions with a G value of 0.32 µmol/J.33 The specific one-electron oxidants, Cl2•- at pH 1 and Br2•at pHs of 7 and 10 with their respective one-electron reduction potentials of 2.3 and 1.7 V versus NHE, were produced by pulse radiolysing an aerated aqueous solution of 0.1 M KCl (G value of 0.35 µmol/J) and a N2O-saturated aqueous solution of 0.1 M KBr (G value of 0.69 µmol/J) respectively.34
SCHEME 1
solutions were used for each experiment. The absorption spectra of the parent compounds were recorded on a Hitachi spectrometer, model 330. The pH of the solutions was adjusted using HClO4, phosphate salts. SeP was synthesized according to the literature method,29 while SeCys and SeA were purchased commercially from Sigma/Aldrich as hydrochloride salts. The purity of the compounds was verified by 1H, 13C{1H}, and 77Se{1H} NMR spectra recorded on a Bruker DPX-300 NMR spectrometer operating at 300, 75.47, and 57.24 MHz, respectively. Chemical shifts are relative to internal solvent peaks for 1H and 13C and external Me Se (secondary reference Ph Se δ 2 2 2 463 ppm) for 77Se NMR measurements. Characterization of SeP. 1H NMR (CD3OD) δ: 2.81 (t, SeCH2); 3.10 (t, CH2CO) (COOH proton exchanged with CD3OD). 13C{1H} NMR (CD3OD) δ: 23.4 (s, SeCH2); 35.4 (s, CH2CO); 174.3 (s, CO). 77Se{1H} NMR (CD3OD) δ: 322 (s). IR spectra: ν(CdO) ) 1694 cm-1. Mass spectral data: m/z ) 306 (molecular ion); 288 (M - H2O). The m/z value given here and elsewhere is based on the 80Se isotope with a natural abundance of 49.82%. Characterization of SeA‚2HCl. 1H NMR (D2O) δ: 3.13 (t, 6.4 Hz, SeCH2); 3.35 (t, 6.4 Hz, CH2N). 13C{1H} NMR (D2O) δ: 24.7 (s, SeCH2); 40.7 (s, CH2N). 77Se{1H} NMR (D2O) δ: 276 (s) ppm. Mass spectral data: m/z ) 248.8 (M+ - 2HCl, molecular ion); 203.8 (M+ - 2HCl-EtNH2). Characterization of SeCys‚2HCl. 1H NMR (D2O as DCl) δ: 3.32-3.52 (m, SeCH2); 4.35 (d, t unresolved CH) (NH2 and COOH proton exchanged with D). 13C{1H} NMR (D2O) δ: 27.1 (s, SeCH2, 1J(77Se-13C): 86 Hz); 53.3 (s, CHN); 170.0 (s, CO). 77Se{1H} NMR (D O) δ: 288 (s) ppm. 2 Pulse Radiolysis Studies. Pulse radiolysis experiments were carried out with high-energy electron pulses (7 MeV, 500 ns) obtained from a linear electron accelerator, and the details are given elsewhere.30 An aerated aqueous solution of KSCN (1 × 10-2 M) was used for determining the dose delivered per pulse using G475 ) 2.59 × 10-4 m2/J for the transient (SCN)2•species.31,32 G denotes the radiation chemical yield in mol/J, and is the molar absorption coefficient in m2 mol-1. The transient species formed upon pulse radiolysis were detected by the optical absorption method.32,33 The dose per pulse was close to 12 Gy (1 Gy ) 1 J kg-1). Radiolysis of a N2-saturated neutral aqueous solution leads to the formation of three highly reactive species (•H, •OH, eaq-) in addition to the less reactive or inert molecular products (H2, H2O2, H3O•+). The reaction with •OH radicals at pH 7 was carried out in N2O-saturated solutions, where eaq- is converted to •OH radicals and the G value for •OH under such conditions
Results Table 1 lists the ground-state pKa of the diselenides employed in the present study along with their absorption maxima. SeP is completely protonated at pH < 2 and exists in the dianionic form at pH > 3. The NH2 groups of SeA are protonated (NH3+) up to pH 9.7. In the case of SeCys, both the carboxylic acid groups and the NH2 groups are protonated at pH < 3, but in the pH range of 3-9.7, the carboxylic acid groups are ionized while the NH2 groups are protonated, making the molecule zwitterionic. At pH > 9.7, both the carboxylic acid groups and the amino groups are deprotonated, and the molecule is in the dianionic form.35 Since these diselenides do not have significant absorption, where the transients absorb, no correction for the ground-state absorption was made. The details of the transient spectra, decay kinetics, and one-electron reduction potential for the transformation of the diselenides to their respective radical cations are given below. Reactions with SeCys. The absorption spectrum of the transients produced by the reaction of •OH with SeCys at pH 1 at ∼5 µs after the pulse is given in Figure 1a. Comparing this spectrum with that produced by the reaction of SeCys with Cl2•radicals (Supporting Information Figure Sfig. 1) suggested that the transient absorbing at 560 nm may correspond to the SeCys radical cation (SeCys•+) formed by one-electron oxidation. At the 5 mM concentration of SeCys, assuming that all of the Cl2•radicals quantitatively convert SeCys to SeCys•+, the percentage of •OH radicals reacting with SeCys to form SeCys•+ was found to be ∼62%. The spectra of the transients under the same reaction conditions at ∼40 µs after the pulse showed decay of the 560 nm band and appearance of a new absorption band with a maximum at 460 nm (Figure 1b). These experiments, when carried out at low SeCys (100 µM) concentration, showed only the formation of the 560 nm band, and the 460 nm band was absent (Figure 1a). Careful examination of the decay kinetics at 560 nm and the formation kinetics at 460 nm indicated that while the transient decay at 560 nm did not vary significantly when the SeCys concentration was changed from 100 µM to 5 mM, the formation rate and the absorbance at 460 nm increased significantly with increasing SeCys concentrations (Supporting Information Figure Sfig.2). Insets c and d of Figure 1 show the absorption-time traces for the decay at 560 nm and formation at 460 nm, respectively, at a SeCys concentration of 5 mM. These concentration dependence experiments indicate that the 460 nm band may be due to the reaction of a species that is not directly formed by the reaction of SeCys•+ with SeCys but due
TABLE 1: Ground-State Absorption Maxima and pKa of the Diselenides, Rate Constant of RSeSeR with the •OH Radical, and Yield of the 560 nm Transient
RSeSeR SeCys SeP SeA
λmax (nm) 300 300 300
pKa 1.7, 2.3, 8.0, 8.7 ∼2.0 8.8, 9.2
k(RSeSeR + •OH) (M-1 s-1)/ 109
G value (µmol/J) and % of (RSeSeR)·+ formed by •OH reaction
pH 1
pH 7
pH 1
pH 7
11 ( 2.0 19 ( 0.3 11 ( 0.6
9.6 ( 0.6 18 ( 3.0 13 ( 3.0
0.2 ( 0.01 (∼62%) 0.27 ( 0.02 (83%) 0.19 ( 0.01 (∼60%)
0.55 ( 0.02 (92%) 0.32 ( 0.02 (53%)
Reactions of •OH with SeCys Derivatives
Figure 1. Difference absorption spectra of the transients obtained at pH 1, on reactions of hydroxyl radicals with (a) 100 µM SeCys at 5 µs after the pulse and (b) 5 mM SeCys at 40 µs after the pulse. Insets (c) and (d): Absorption-time plots at 560 and 460 nm, respectively, under the same experimental conditions as those in (b).
J. Phys. Chem. B, Vol. 112, No. 14, 2008 4443
Figure 3. Difference absorption spectra of the transient obtained upon pulse radiolysis of aerated aqueous solutions of 2 mM SeP at pH 1 (a) in the absence of 0.1 M KCl and 10 µs after the pulse and (b) in the presence of 0.1 M KCl, at 5 µs after the pulse. The inset shows transient absorption spectra obtained upon pulse radiolysis of a N2O-saturated aqueous solution of 100 µM SeP at pH 7 (c) at 10 µs after the pulse and (d) in the presence of 0.1 M KBr at 20 µs after the pulse.
Figure 2. Difference absorption spectra of the transient obtained upon pulse radiolysis of N2O-saturated aqueous solutions of 100 µM SeCys at 20 µs after the pulse at pH 7. Insets (a) and (b) show absorptiontime plots at 560 nm obtained upon pulse radiolysis of 200 µM SeCys in presence of 0.1 M KBr at pHs of 2.0 and 4.5, respectively.
to secondary reactions involving different intermediate species and excess SeCys. At pH 7, due to solubility limitations of SeCys, the reaction with •OH radicals could be studied only at a 100 µM concentration of SeCys. The transient spectrum at this concentration showed absorption maximum at 460 nm, and the 560 nm absorbing transient was absent in the spectrum even at the fastest detectable time scale (Figure 2). At pH 10, also with the maximum solubility of ∼100 µM, a similar transient absorption spectrum (λmax 460 nm) was observed (results not shown). The decay of the transient at 560 nm became faster with increasing the pH from 1 to 5 (insets a and b of Figure 2 at pHs of 2 and 4.5, respectively), indicating that the decay kinetics of SeCys•+ is influenced by the presence of OH- ions, and this may be one of the reasons for not observing the 560 nm absorbing species at pH > 5. To understand the role of the amino and carboxylic acid functional groups on the reactions of •OH with SeCys, studies were carried out with its structural analogues SeP and SeA. Reactions with SeP. The transient spectra obtained by the reaction of SeP with the •OH and Cl2•- radicals at pH 1 and with the •OH and Br2•- radicals at pH 7 are given in Figure 3. The spectra in all of these systems showed absorption in the wavelength region of 300-700 nm, with a maximum at 560 nm and another weak absorption maximum at 350-360 nm. Similarly at pH 10, the transient spectrum and the absorption maximum matched with those obtained at pHs of 1 and 7 (results not shown). As discussed earlier, the 560 nm absorption band
Figure 4. (a) and (b): Difference absorption spectra of the transient obtained upon pulse radiolysis of an aerated aqueous solution at pH 1, containing 2 mM SeA (a) at 1.2 µs and (b) at 40 µs after the pulse. Insets (c) and (d): Absorption-time plots at 560 and 460 nm, respectively, under the same experimental conditions.
has been assigned to the radical cation of SeP (SeP•+), and its decay kinetics did not show significant change with the pH of the solution from 1-7 (Table 2). At later time scales, the transient spectrum did not show any features attributable to the formation of new transients even at higher concentrations of SeP (∼5 mM). By comparing the yields of the 560 nm absorbing species obtained from the reactions of •OH radicals and oneelectron oxidants at a SeP concentration of 5 mM, the percentage of •OH radicals leading to the formation of SeP•+ was found to be ∼86 and ∼92% at pHs of 1 and 7, respectively, confirming quantitative conversion of SeP to SeP•+ in this pH range by •OH radicals. Reactions with SeA. The reaction of •OH radicals with SeA at pH 1 produced a transient with a broad absorption spectrum and an apparent maximum at ∼560 nm (Figure 4a). The spectrum recorded at ∼40 µs after the pulse showed a significant decrease in the 560 nm absorbing transient and the appearance of a sharp band at 460 nm (Figure 4b). The absorption-time plots recorded at 460 nm, as given in the insets c and d of Figure 4, distinctly show a slow-forming transient whose rate and yield increased with increasing concentration of SeA from 200 µM to 4 mM. Reaction of Cl2•- radicals at pH 1 also produced a 560 nm band (Supporting Information Figure Sfig. 3), and on
4444 J. Phys. Chem. B, Vol. 112, No. 14, 2008
Mishra et al.
Figure 5. Time-resolved absorption spectra of the transient obtained upon pulse radiolysis of a N2O-saturated aqueous solution containing 2 mM SeA at pH 7 at (a) 1.4 µs and (b) 40 µs after the pulse. Insets (c) and (d) show absorption-time plots at 560 nm obtained upon pulse radiolysis of a N2O-saturated aqueous solution containing 0.1 M KBr and 0.5 mM SeA at pHs of 3.6 and 7.4, respectively.
Figure 6. (a): Variation of (kobs)/[SeP] with the ratio of the concentrations of PMZ and SeP at pH 7. Here, kobs corresponds to the observed rate constant for the decay of PMZ•+ at 500 nm. Inset (b): Variation in the equilibrium absorbance at 500 nm with the ratio of the concentrations of PMZ and SeP at pH 7. Line fits show least-square fitting of the data according to the procedure given in reference 37. Inset (b) shows a representative absorption-time plot at 500 nm due to PMZ•+ at redox equilibrium conditions ([PMZ] ) 5 mM and [SeP] ) 5 mM).
560 nm absorption band increased significantly with the pH of the solution, as shown in the insets c and d of Figure 5 at pHs of 3.6 and 7.4, respectively. By comparing the yields of the 560 nm absorbing species obtained from the reactions of •OH radicals and one-electron oxidants at a SeA concentration of 5 mM, the percentage of •OH radicals leading to the formation of SeA•+ was found to be ∼ 62 and 53% at pHs of 1 and 7, respectively. Estimation of Rate Constants. Rate constants for the reaction of •OH radicals with SeCys, SeP, and SeA were determined at pHs of 1 and 7 by the competition kinetics method using KSCN as a reference solute,36 and the resulting data has been summarized in Table. 1. The rate constants for the decay of the transients at 460 and 560 nm, obtained from the reactions of diselenides with •OH radicals at pHs of 1, 7, and 10, are given in Table 2. In almost all of the cases, the 560 nm transient decay obeyed first-order kinetics and remained unaffected by the diselenide concentration. The decay of the transient absorbing at 460 nm was slightly less than that at 560 nm and also showed first-order kinetics. Estimation of Reduction Potential for the Formation of Radical Cations at pH 7. The energetics for the conversion of the diselenides to their corresponding radical cations has been estimated in terms of their reduction potentials. The studies have been carried out only at pH 7. The one-electron reduction potentials were estimated by determining the equilibrium constants (K) for the reversible electron transfer between the diselenide couple and a reference couple, as reported earlier.37 Various standards were employed as reference couples, such as promethazine radical cation PMZ•+/PMZ; (E ) 0.98 V vs NHE), dibromide radical anion Br2•-/2Br- (E ) 1.7 V vs NHE), and azide radical N3•/N3- (E ) 1.33 V vs NHE).33 The individual electron-transfer equilibria are represented by eqs 1-3 kf
SeCys•+ + N3- {\ } SeCys + N3• k
(1)
b
kf
} PMZ + SeP•+ PMZ•+ + SeP {\ k
(2)
b
kf
} SeA•+ + 2BrSeA + Br2•- {\ k
(3)
b
the basis of this, the 560 nm band has been assigned to the SeA radical cation (SeA•+), and its decay rate was found to remain unaffected by increasing SeA concentration. A similar transient spectral behavior was observed during the reaction of SeA with •OH radicals at pH 7. At 1.4 µs after the pulse, a broad spectrum with an absorption maximum in the 560 nm region (Figure 5a) was observed, and at ∼40 µs after the pulse, the absorption in the 560 nm region decreased, and the 460 nm band appeared (Figure 5b). The rate and yield of the 460 nm absorbing transient increased with increasing SeA concentration from 250 µM to 5 mM. The decay rate of the
Under these conditions, reversibility is confirmed by ensuring the increase in the decay or formation kinetics of either of the transients, monitored at a suitable wavelength, with the increase in concentration of the reactants. The K values were determined either by following the equilibrium kinetics of the transient radicals or by estimation of the radical yield at the equilibrium according to known methods.37 The difference in the potential between the two couples was determined from the K values, according to the Nernst’s equation [∆E ) (0.059)log K]. Figure 6 gives representative plots for estimation of redox equilibrium
TABLE 2: Rate Constants for the Decay of 460 and 560 nm Absorbing Transients and the Equilibrium Constants for the Formation of Radical Adducts
compounds
pH
kdecay (s-1) at 460 nm
kdecay (s-1) at 560 nm
SeCys
1 7 1 7 1 7
5.6 ( 1.0 × 103 4.3 ( 2.0 × 103 7.3 ( 0.9 × 103 7.0 ( 1.1 × 103
7.5 ( 1.6 × 103 1.0 ( 0.7 × 104 1.3 ( 0.5 × 104 1.4 ( 1.0 × 104 1.7 ( 1.6 × 105
SeP SeA
equilibrium constant (K′) for selenuranyl radical formation (M-1) 4.2 ( 0.2 × 103
1.1 ( 0.3 × 103 1.8 ( 0.1 × 103
Reactions of •OH with SeCys Derivatives
J. Phys. Chem. B, Vol. 112, No. 14, 2008 4445
TABLE 3: Redox Equilibrium Constants and One-Electron Reduction Potentials of the Diselenides at pH 7 couple
Kaverage
∆E (V)
E (V) vs NHE
reference couple vs NHE
(SeCys•+/SeCys) (SeP•+/SeP) (SeA•+/SeA)
7.3 ( 7.0 0.53 ( 0.25 55.0 ( 31.0
0.05 0.02 0.10
1.3 ( 0.1 0.96 ( 0.01 1.6 ( 0.1
N3•/N3-(E ) 1.33 V) PMZ•+/PMZ (E ) 0.98 V) Br2•-/2Br-(E ) 1.7 V)
constants between the couples, SeP•+/SeP and PMZ•+/PMZ, by following the absorbance and observed decay rate constants (kobs) at 500 nm in the presence of varying concentrations of PMZ and SeP. The inset of Figure 6 gives an absorption-time plot under such equilibrium conditions. The details of the K and ∆E values and the estimated one-electron reduction potentials of these compounds are listed in Table 3. The values range from 1.0 to 1.6 V and are in the order SeP < SeCys < SeA. Discussion The reaction of •OH radicals with selenocystine (SeCys), an amino acid, and a diselenide has been studied at pHs of 1, 7 and 10, and the transients were monitored by absorption spectroscopy. In order to understand the influence of the two functional groups on the absorption spectrum of the transient, studies have been carried out with its structural analogues having either a carboxylic acid group (SeP) or an amino group (SeA). From the rate constants given in Table 1 and as observed with many sulfur analogues, this fast reaction of •OH radicals is attributed to the formation of hydroxyl radical adducts,16,38-40 represented by the general reaction I in Scheme 2. Here, •OH radical adds to the diselenide bond. In the above reactions, depending on the concentration, pH, substitution, and time point, the transient spectra showed two distinctly different transients with absorption maxima at 560 and 460 nm. The 560 nm band has been assigned to diselenide radical cation [(RSeSeR)•+], the formation of which has been verified by studying the reactions of the diselenides with specific one-electron oxidants, Cl2•- and Br2•- radicals. The reaction paths II and III of Scheme 2 show the steps leading to formation of (RSeSeR)•+ from the •OH radical reaction and the specific one-electron oxidant. Formation of (RSeSeR)•+ from hydroxyl radical adducts can be envisaged by the loss of either water molecule (H2O) at pH 1 or OH- at pH 7.38,39 In organoselenium compounds, since the electron density is the highest on selenium, the radical cations are expected to be selenium centric. Such selenium-centered radical cations, like their sulfur analogues, are generally unstable and gain stability by coordination with the lone pair of either another selenium atom or a heteroatom in the molecule.13-15,25 Since the SCHEME 2
compounds employed in the present study are diselenides, the radical cation can attain stability through the p orbital of the two selenium atoms, forming a two-center-three-electron bond, in which two electrons are in the bonding π orbital and one electron is in the antibonding π* orbital.16 The above studies also indicate that (RSeSeR)•+ in all of the cases, irrespective of substitution, absorb at 560 nm. The yields of radical cations produced by •OH reaction have been found to be different for different diselenides. While >80% of •OH radicals cause oneelectron oxidation in SeP, only ∼60% of them cause oxidation of SeCys and SeA. One of the reasons for the high yield of radical cations in SeP could be due to the substitution of an electron-donating carboxylate group, which increases the electron density on the Se-Se bond and makes it easier to undergo oxidation. This is also supported by the lowest value for the one-electron reduction potential for the couple SeP•+/SeP (Table 3). Compounds SeCys and SeA, having amino functional groups, showed different results. In both cases, in addition to the initially formed (RSeSeR)•+, a new transient having a sharp absorption band at 460 nm is observed on later time scales. The 460 nm absorption became more prominent when the concentration of the parent diselenide was increased to the millimolar range (except for SeCys at pH 7, which showed formation of the 460 nm band at 100 µM). As the two compounds are in the hydrochloride form, the Cl- ion concentration increases with increasing diselenide concentration, which is likely to contribute to the formation of Cl2•- radicals. This may leave a doubt that the 460 nm absorption band observed at high diselenide concentration may be due to the reactions of Cl2•- radicals. However, our independent experiments on Cl2•- radical reactions with a low concentration of diselenide (Supporting Information Figure Sfig.1) ruled out any such possibility. The above observations indicate that the 460 nm band is not formed directly from the decay of radical cations. As is evident from Table 2, in these compounds, only ∼60% of hydroxyl radical adducts are converted to radical cations. Therefore, one possible pathway for the formation of the 460 nm absorbing transient may be from the hydroxyl radical adducts decaying by a different pathway to an intermediate selenyl radical (RSe•)38-40 which reacts with excess diselenide to give a
4446 J. Phys. Chem. B, Vol. 112, No. 14, 2008 triselenide radical adduct (reaction IV and V of Scheme 2). It is also likely that some of the (RSeSeR)•+ may be converted to RSe•, as given in reaction VI. Similar reactions were proposed by pulse radiolysis during the reactions of •OH radicals and one-electron oxidants with substituted disulfides like dimethyldisulfide, cystine, cystamine, and so forth.38-41 Both reactions V and VI become increasingly favorable with an increase in the pH.38-40 At low solute concentrations, the RSe• radicals may be converted back to the diselenide by radical-radical recombination (reaction VII of Scheme 2). The nature of the triselenide radical adduct formed in reaction V may correspond to a selenuranyl type of radical, in which a triselenide bridge is formed and the unpaired electron is shared between the three Se atoms.41,42 Due to the hemibond nature, it would undergo selenium-selenium bond rupture, and the process is reversible. From the concentration-dependent change in the absorbance at 460 nm, the equilibrium constants (K′) for selenuranyl radical formation (reaction V) have been estimated for SeCys at pH 1 and SeA at pHs of 1 and 7 according to a procedure described earlier,13 and the values are listed in Table 2. The supporting figure (Supporting Information Figure Sfig.4) gives linear plots and a brief description of the method employed for these measurements. The increase in the one-electron reduction potential of the diselenide due to a decrease in electron density caused by the electron-withdrawing NH3+ group may be a possible reason for the formation of selenyl and triselenide radicals from aminosubstituted diselenides, although other reaction sequences cannot be ruled out. Keeping in mind the complexity of the reactions and the limitations of absorption spectroscopic detection, complementary experimental techniques, like time-resolved conductivity measurements, would assist in confirming this reaction sequence, which are being planned by us in the future. Conclusions Reactions of •OH radicals with SeCys, a diselenide, amino acid, and its two structural analogues, SeP and SeA, have been studied at pHs of 1, 7, and 10. The hydroxyl radical adducts formed in these reactions follow two reaction pathways, (1) a radical cation formation by the loss of H2O or OH- absorbing at 560 nm and (2) fragmentation to form a selenyl radical. SeP, having only carboxylic acid substitution, is the easiest to undergo oxidation, and therefore, the hydroxyl radical adducts are quantitatively converted to radical cations. Compounds SeCys and SeA, having an amino substitution, are difficult to undergo oxidation. Therefore, the hydroxyl radical adducts, in addition to radical cation formation, undergo fragmentation to form selenyl radicals, which, in the presence of excess diselenide, are converted to triselenide radical adducts, as observed with analogous disulfides. Acknowledgment. The authors are thankful to Drs. T. Mukherjee, S. K. Sarkar, and D. Das for the support and encouragement. The authors also like to express their gratefulness to one of the reviewers for many useful suggestions in the revision of the manuscript. Supporting Information Available: Absorption spectra, absorption-time plots, and double reciprocal plots. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Bock, A. In Encyclopedia of Inorganic Chemistry; Bruce-King, R., Ed.; Wiley: New York, 1994; Vol. 7, pp 3700-3708.
Mishra et al. (2) Johanssona, L.; Gafvelinb, G.; Arne´r, E. S. J. Biochim. Biophys. Acta 2005, 1726, 1-13. (3) Jacob, C.; Giles, G. I.; Giles, N. M.; Sies, H. Angew. Chem., Intl. Ed. 2003, 42, 4742-4758. (4) Mugesh, G.; du Mont, W.; Sies, H. Chem. ReV. 2001, 101, 21252180. (5) Nogueira, C. W.; Zeni, G.; Rocha, J. B. T. Chem. ReV. 2004, 104, 6255-6286. (6) Rayman, M. P. Lancet 2000, 356, 233-241. (7) Nauser, T.; Dockheer, S.; Kissner, R.; Koppenol, W. H. Biochemistry 2006, 45, 6038-6043. (8) Back, T. G.; Moussa, Z. J. Am. Chem. Soc. 2003, 125, 1345513460. (9) Mishra, B.; Priyadarsini, K. I.; Mohan, H.; Mugesh, G. Bioorg. Med. Chem. Lett. 2006, 16, 5334-5338. (10) Kim, H. Y.; Gladyshev, V. N. PLoS Biol. 2005, 3, 2080-2089. (11) Metanis, N.; Keinan, E.; Dawson, P. E. J. Am. Chem. Soc. 2006, 128, 16684-16691. (12) Kunwar, A.; Mishra, B.; Barik, A.; Kumbhare, L. B.; Pandey, R.; Jain, V. K.; Priyadarsini, K. I. Chem. Res. Toxicol. 2007, 20, 1482-1487. (13) Mishra, B.; Maity, D. K.; Priyadarsini, K. I.; Mohan, H.; Mittal, J. P. J. Phys. Chem. A 2004, 108, 1552-1559. (14) Mishra, B.; Priyadarsini, K. I.; Mohan, H. J. Phys. Chem. A 2006, 110, 1894-1900. (15) Priyadarsini, K. I.; Mishra, B.; Maity, D. K.; Mohan, H. Phosphorus, Sulfur Silicon Relat. Elem. 2005, 180, 985-988. (16) Asmus, K.-D. In Sulfur-Centered ReactiVe Intermediates as Studied by Radiation Chemical and Complementary Techniques, S-Centered Radicals; Alfassi, Z. B., Ed.; John Wiley: New York, 1999, pp 142-145. (17) Feroci, G.; Fini, A. Inorg. Chim. Acta 2007, 360, 1023-1031. (18) Everett, S. A.; Scho¨neich, C.; Stewart, J. H.; Asmus, K. J. Phys. Chem. 1992, 96, 306-314. (19) Glass, R. S. Sulfur Radical Cations: Topics in Current Chemistry; Springer-Verlag: Berlin, Germany, 1999; Vol. 205, pp 1-88. (20) Mohan, H.; Mittal, J. P. J. Phys. Chem. A 2002, 106, 6574-6580. (21) Scho¨neich, C.; Zhao, F.; Madden, K. P.; Bobrowski, K. J. Am. Chem. Soc. 1994, 116, 4641-4644. (22) Korzeniowska-Sobczuk, A.; Hug, G. L.; Carmichael, I.; Bobrowski, K. J. Phys. Chem. A 2002, 106, 9251-9260. (23) Engman, E.; Lind, J.; Merenyi, G. J. Phys. Chem 1994, 98, 31743182. (24) Badiello, R. In Chemistry of Organic Selenium and Tellurium Compounds; Patai, S., Rappoport, Z., Eds.; Wiley: New York, 1986; Vol. 1. pp 287-297. (25) Assmann, A.; Bonifacˇic´, M.; Briviba, K.; Sies, H.; Asmus, K.-D. Free Radical Res. 2000, 32, 371-376. (26) Mu¨ller, R.; Heinze, J. U. Phosphorus, Sulfur Silicon Relat. Elem. 1998, 141, 111-116. (27) Scho¨neich, C.; Narayanaswami, V.; Asmus, K.-D.; Sies, H. Arch. Biochem. Biophys. 1990, 282, 18-25. (28) Badiello, R.; Tamba, M. Int. J. Radiat. Biol. 1973, 23, 435-442. (29) Arnold, A. P.; Tan, K. S.; Rabenstein, D. L. Inorg. Chem. 1986, 25, 2433-2437. (30) Guha, S. N.; Moorthy, P. N.; Kishore, K.; Rao, K. N. Proc. Indian Acad. Sci., Chem. Sci. 1987, 99, 261-271. (31) Buxton, G. V.; Stuart, C. C. J. Chem. Soc., Faraday Trans. 1995, 91, 279-281. (32) Fielden, E. M. In The Study of Fast Processes and Transient Species by Electron Pulse Radiolysis; Baxendale, J. H., Busi, F., Eds.; D. Reidel: Boston, MA, 1984; pp 58-59. (33) Buxton, G. V.; Mulazzani, Q. G. In Radiation-Chemical Techniques, Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH: Weinheim, Germany, 2001; Vol. 1, pp 503-557. (34) Candeias, L. P.; Wardman, P.; Mason, R. P. Biophys. Chem. 1997, 67, 229-237. (35) Dawson, R. M. C.; Elliott, D. C.; Elliott, W. H.; Jones, K. M. Data for Biochemical Research; Clarendon Press: Oxford, U.K., 1986; pp 213231. (36) Spinks, J. W. T.; Woods, R. J. Introduction to Radiation Chemistry; John Wiley: New York, 1991, pp 243-244. (37) Barik, A.; Priyadarsini, K. I.; Mohan, H. Radiat. Phys. Chem. 2004, 70, 687-696. (38) Bonifacˇic´, M.; Asmus, K.-D. J. Phys. Chem. 1976, 80, 2426-2430. (39) Mockel, H.; Bonifacˇic´, M.; Asmus, K.-D. J. Phys. Chem. 1974, 78, 282-284. (40) Elliot, A. J.; McEachern, R. J.; Armstrong, D. A. J. Phys. Chem. 1981, 85, 68-75. (41) Bonifacˇic´, M.; Asmus, K.-D. J. Phys. Chem. 1984, 88, 6286-6090. (42) Bonifacˇic´, M.; Asmus, K.-D. Int. J. Radiat. Biol. 1984, 46, 3545.
