Calcium-activated butyrylcholinesterase in human skin protects acetylcholinesterase against suicide inhibition by neurotoxic organophosphates

Calcium-activated butyrylcholinesterase in human skin protects acetylcholinesterase against suicide inhibition by neurotoxic organophosphates

Biochemical and Biophysical Research Communications 355 (2007) 1069–1074 www.elsevier.com/locate/ybbrc Calcium-activated butyrylcholinesterase in hum...

215KB Sizes 0 Downloads 10 Views

Biochemical and Biophysical Research Communications 355 (2007) 1069–1074 www.elsevier.com/locate/ybbrc

Calcium-activated butyrylcholinesterase in human skin protects acetylcholinesterase against suicide inhibition by neurotoxic organophosphates Karin U. Schallreuter a,c,*, Nicholas C.J. Gibbons a,c, Souna M. Elwary a, Susan M. Parkin b, John M. Wood a a

Clinical and Experimental Dermatology, Department of Biomedical Sciences, University of Bradford, Bradford BD7 1DP, West Yorkshire, UK b Department of Biomedical Sciences, University of Bradford, Bradford, UK c Institute for Pigmentary Disorders in Association with EM Arndt University of Greifswald, Germany and University of Bradford, UK Received 17 January 2007 Available online 23 February 2007

Abstract The human epidermis holds an autocrine acetylcholine production and degradation including functioning membrane integrated and cytosolic butyrylcholinesterase (BuchE). Here we show that BuchE activities increase 9-fold in the presence of calcium (0.5 · 10-3 M) via a specific EF-hand calcium binding site, whereas acetylcholinesterase (AchE) is not affected. 45Calcium labelling and computer simulation confirmed the presence of one EF-hand binding site per subunit which is disrupted by H2O2-mediated oxidation. Moreover, we confirmed the faster hydrolysis by calcium-activated BuchE using the neurotoxic organophosphate O-ethyl-O-(4-nitrophenyl)-phenylphosphonothioate (EPN). Considering the large size of the human skin with 1.8 m2 surface area with its calcium gradient in the 103 M range, our results implicate calcium-activated BuchE as a major protective mechanism against suicide inhibition of AchE by organophosphates in this non-neuronal tissue.  2007 Elsevier Inc. All rights reserved. Keywords: Butyrylcholinesterase; Acetylcholinesterase; Calcium; EF-hand; Human epidermis; Organophosphates; Hydrogen peroxide

The role of butyrylcholinesterase (BuchE, EC 3.1.1.18) in human cells and tissues is still obscure because no specific endogenous substrate has been identified for this enzyme. Moreover, although its expression in human serum is high, peripheral tissues have overall lower levels compared to acetylcholinesterase (AchE) [1]. Earlier it has been proposed that BuchE functions as a protective mechanism for AchE [2]. Interestingly BuchE has 53% sequence homology with AchE and both enzymes share the same active site catalytic triad of Glu, His and Ser. Moreover, both enzymes contain * Corresponding author. Address: Clinical and Experimental Dermatology, Department of Biomedical Sciences, University of Bradford, Bradford BD7 1DP, West Yorkshire, UK. Fax: +44 1274 236489. E-mail address: [email protected] (K.U. Schallreuter).

0006-291X/$ - see front matter  2007 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2007.02.078

aromatic gorges for substrate binding where AchE has 14 aromatic residues lining the gorge, while BuchE has only eight such residues [3]. Based on these structures, AchE owns a very specific substrate binding site for the neurotransmitter acetylcholine, whereas BuchE is non-specific, hydrolysing many ester substrates including potent neurotoxins e.g., organophosphates and cocaine [4–6]. In this context it is interesting that the human epidermis holds the full capacity for a non-neuronal cholinergic signal transduction [7]. Only very recently it was recognised that this tissue also holds the capacity of a functioning BuchE in the cytosol and on cell membranes [8]. Enzyme activities are significantly higher compared to AchE, implicating an important role for BuchE in this compartment. Moreover, it was also recognised that both

1070

K.U. Schallreuter et al. / Biochemical and Biophysical Research Communications 355 (2007) 1069–1074

BuchE and AchE are regulated by hydrogen peroxide (H2O2) in a concentration dependent manner. Both enzymes are inhibited/deactivated by H2O2 concentrations >0.5 · 103 M, whereas concentrations <0.5 · 103 M activate transcription and enzyme activities [8,9]. BuchE is more sensitive to this reactive oxygen species (ROS) than AchE. This was explained due to H2O2-mediated oxidation of methionine (Met) and tryptophan (Trp) residues in the active site as well as in the tetramerisation domain yielding a more severe structural alteration in the case of BuchE compared to AchE, where only the enzyme active site is affected [8,9]. Considering the strong expression of both enzymes in the outermost layer of human skin which in turn presents a first line target for many exogenous insults including UV-generated H2O2, it was tempting to invoke a major function for BuchE as an important protection/defence mechanism in this tissue. Preliminary in situ results on protein expression of membrane integrated and cytosolic BuchE suggested that this enzyme followed the well-established calcium gradient in the human epidermis [8,10]. This result is in agreement with previously reported data on membrane integrated BuchE in other tissues [1]. To study the direct effect of calcium on enzyme activities, we used pure human serum BuchE and rh AchE following the kinetics in the presence and absence of calcium in a concentration dependent manner. The results reveal that calcium increases BuchE activities 9-fold over the lanthanum inhibited calcium free enzyme, while AchE activities were not affected. In order to substantiate the nature of the calcium effect on BuchE, we utilised radiolabelled 45calcium identifying stoichiometric binding of one calcium atom per enzyme subunit. Since H2O2-mediated oxidation severely altered the active site and the tetramerisation domain of the enzyme, we asked the question whether calcium binding would be also affected by this ROS [8]. These results show that calcium binding is severely altered after oxidation by H2O2. Evaluation of the primary sequence of BuchE suggested the presence of a single EF-hand binding domain per subunit which was confirmed by computer simulation. After H2O2-mediated oxidation of Met and Trp residues of the protein a significant shift in the backbone structure results in disruption of the calcium binding site. Calcium-activated BuchE revealed a 9-fold faster turnover of the organophosphate EPN. In summary, our results show for the first time that BuchE is regulated by calcium via a single EF-hand binding domain per subunit while AchE is not influenced by this ion. Moreover, we show that this calcium binding is regulated by H2O2. Since the human epidermis with its size of 1.8 m2 presents a major source for BuchE, we conclude that the function of this enzyme in this outermost tissue of the human body provides a major protection mechanism for AchE to prevent suicide inhibition of this enzyme by neurotoxins such as organophosphates.

Materials and methods Materials. Human serum BuchE, human recombinant AchE, butyrylthiocholine iodide, acetylthiocholine iodide, O-ethyl-O-(4-nitrophenyl) phenylphosphonothioate (EPN), 5 0 ,5 0 -dithiobis-2-nitrobenzoic acid (DTNB), lanthanum nitrate and all other chemicals were obtained from Sigma (Poole Dorset, UK). Radiolabelled 45calcium chloride (12.2 mCi/ mg) was obtained from ICN (Basingstoke, Hants, UK). BuchE enzyme assay. For enzyme kinetics pure human serum BuchE was used. In order to ensure a calcium free enzyme, lanthanum (103 M) was utilised to exchange any bound calcium from the native enzyme prior to the kinetic analysis. One unit of enzyme hydrolysed 1.0 lmol of butyrylthiocholine iodide per minute at room temperature (pH 7.4). Enzyme activities were determined spectrophotometrically at 405 nm by following the reduction of DTNB according to Ellman et al. [11]. BuchE and AchE activities were measured in the presence and absence of CaCl2 (0– 2 · 103 M) under saturating substrate conditions (103 M). In order to test the effect of calcium on the hydrolysis of the organophosphate EPN (104 M), reaction rates for its turnover were determined in the presence and absence of 103 M CaCl2. AchE enzyme assay. AchE enzyme activities were measured at 405 nm using acetylthiocholine as substrate under saturating conditions (103 M) coupled to the reduction of DTNB [11]. Enzyme activities were determined in the presence of CaCl2 (0– 1 · 103 M). Quantitation of calcium binding to BuchE using the isotope 45calcium. To follow the specific binding of calcium to BuchE, 2.2 mg of pure enzyme in 0.5 ml distilled water was incubated with 2 · 103 M 45CaCl2 for 30 min at room temperature. The labelled enzyme was applied to a G25 Sephadex column (1.5 · 6 cm) and eluted with distilled water in 0.5 ml fractions. BuchE (340 kDa) eluted in the void volume. In order to detect the enzyme, the protein content was determined in each fraction at 280 nm [12]. 45 calcium binding was followed in 100 ll aliquots from each fraction and counted in 3 ml scintillation fluid (Ready Safe, Beckman Coulter, Fullerton, CA, USA) at the 14C channel of a Tricarb 2001TR scintillation counter (Packard Instruments, Meriden, CT, USA). To prove the specificity of 45calcium binding, we utilised lanthanum (103 M) which has a stronger irreversible affinity for calcium binding sites. For this purpose the radiolabelled enzyme was preincubated with lanthanum for 15 min at room temperature and separated by Sephadex chromatography. The exchange of the isotope was followed by counting the 45calcium in each fraction. Hyperchem molecular modelling of the single EF-hand calcium binding site per subunit of BuchE. The molecular 3D structure of BuchE based on the X-ray crystallographic analysis was obtained from the Protein Data Bank [3]. This structure was compared with the primary sequence of the enzyme and it was realised that the crystal structure from the Data Bank missed two Asp residues in its sequence (Asp378 and Asp379). Since analysis of the sequence implicated these two acidic residues in the putative EF-hand site, both residues were added to generate the complete structure of the proposed EF-hand site prior to the Deep View analysis (Swiss Institute of Bioinformatics, Lausanne, Switzerland) by sealing the loop and minimising the structure in water using Hyperchem software (Hypercube Gainsville, FL, USA). In order to follow the effect of H2O2mediated oxidation of Met and Trp residues on the protein structure in the vicinity of the EF-hand binding domain, the enzyme was oxidised and compared to the native enzyme structure.

Results BuchE activities are increased 9-fold by calcium but AchE activities are not affected In order to follow the effect of calcium on serum BuchE, we pre-incubated the enzyme with lanthanum (103 M) for 15 min to ensure that any residual calcium is displaced

K.U. Schallreuter et al. / Biochemical and Biophysical Research Communications 355 (2007) 1069–1074

from the native enzyme. BuchE activities were determined in the presence and absence of 103 M calcium following the reaction rates/minute over 10 min. The results show that the native enzyme contains indeed residual calcium. Fig. 1A presents the data of this study yielding a 9-fold increase in the presence of saturating calcium concentration. In order to determine the Vmax of the enzyme in the presence and absence of calcium, reaction rates were followed over a calcium concentration range (0–2 · 103 M). These results yield a Vmax at 0.5 · 103 M calcium (Fig. 1B). BuchE obeys normal Michaelis–Menten kinetics at pH 7.0 with increasing substrate concentrations resulting in an apparent Km value of 2 · 105 M when butyrylthiocholine is the substrate. AchE activities were determined under the same experimental conditions. Calcium has no effect on enzyme activities (data not shown).

3.94 calcium atoms per BuchE tetramer resulting in binding of one calcium atom per subunit. The specificity of calcium binding was proven by the exchange of the radiolabel in the presence of lanthanum (103 M) showing a 93% loss of 45calcium after 15 minutes incubation (Fig. 2B). To test whether H2O2-mediated oxidation was possibly affecting calcium binding, we used 45calcium labelled BuchE and incubated the enzyme with 103 M H2O2 for 60 min. The results show a release of calcium indicating that oxidation alters severely calcium binding on the enzyme (Fig. 2C). Calcium-activated BuchE increases the hydrolysis of toxic organophosphates BuchE has been shown to hydrolyse a large number of toxic organophosphates while these compounds are suicide inhibitors for AchE [2,5,6]. Based on the above results we postulated that hydrolysis of those organophosphates would be faster in the presence of calcium compared to native BuchE. For this experiment we used the toxic organophosphorus compound EPN as one representative. Calcium increases the hydrolysis of EPN 9-fold compared to calcium free enzyme (Fig. 3A), while AchE is inhibited with an apparent Ki of 45 lM (Fig. 3B). This result supports increased efficacy of calcium-activated BuchE in the turnover of organophosphates and fosters this enzyme as an efficient protective mechanism for AchE.

Radiolabelled 45calcium binding to human BuchE identifies one high affinity single binding site for calcium per subunit which is affected by H2O2-mediated oxidation Purified human serum BuchE was pre-incubated for 30 min with 103 M 45calcium chloride as outlined in methods. The enzyme eluted in fractions 2–5. Hundred microliter aliquots from each fraction were counted in a liquid scintillation counter on the 14C channel. Fig. 2A shows the high affinity binding of 45calcium to BuchE. Based on protein content and 45calcium labelling the stoichiometry of calcium binding was calculated as

B

0.8

0.16

Reduction of DTNB rate / 2 min

BuChe Activity DTNB OD 405 nm / 10 min

A

0.6

0.4

0.2

0

1071

0.12

0.08

0.04

0 0

2

4

6

Time (min)

8

10

0

0.5

1

1.5

2

Ca2+ (10-3 M)

Fig. 1. The effect of calcium on BuchE and AchE activities. (A) Calcium activates BuchE 9-fold. Reactions were measured at 405 nm in Hepes buffer (0.5 M, pH 7.4) containing saturating butyrylthiocholine iodide (1 mM), BuchE (5 U) and DTNB (0.5 mM) in a total volume of 1.0 ml in the presence and absence of calcium chloride (103 M) over 10 min. The enzyme was incubated for 15 min with lanthanum (103 M) to ensure calcium free enzyme (D–D). Calcium (103 M) failed to displace lanthanum from the enzyme (x–x). Native BuchE from Sigma contained calcium (e–e). Activities of saturating calcium show a 9-fold increase over calcium free enzyme activities (h–h). (B) V vs. S analysis confirms a 9-fold increase of BuchE activity by calcium. Reactions rates/2 min were determined in the presence and absence of calcium chloride (0–2 mM) yielding a 9-fold increase in the Vmax of BuchE in the presence of saturating calcium (0.5 · 103 M). The apparent Km value was 2 · 105 M. Calcium at the same concentrations had no effect on AchE activities (data not shown).

C 0.3

2

1.5

4 0.2 0

0 0 1 2 3 4 5 6

45

0.2 1 0.1

0

0

0.1 0.5

0

6

8

10

12

14

0

Fraction Number

Fraction Number

Fraction Number

-3

1

1 3 5 7 9 11 13 15 17

7

x10

0.2

45

cpm -3

Ca

2

x10

0.3

Bound

0.4

Absorbance 280 nm

-3 45

Ca

8

0.6

2+

x10

0.8

Bound

Absorbance 280 nm

12

cpm

0.4 1

Ca

3

B

Bound

16

A 1.2

cpm/ μl

K.U. Schallreuter et al. / Biochemical and Biophysical Research Communications 355 (2007) 1069–1074

Absorbance 280 nm

1072

Fig. 2. Stoichiometric binding of 45calcium to BuchE is affected by lanthanum and H2O2. (A) 45Calcium binds to BuchE. Sephadex G25 was used for separation of 45calcium binding on BuchE. The enzyme eluted between fractions 2–5 as measured by OD280 nm. Specific 45calcium binding to the enzyme followed the protein profile indicating strong binding of this fast exchange ion to the enzyme. Based on this result the stoichiometry yields one calcium atom bound to each subunit of BuchE. (protein (h–h), 45calcium labelled enzyme (–)). (B) Calcium binding to BuchE is specific. Radiolabelled enzyme (14,000 cpm) was incubated with lanthanum (103 M) for 15 min and subjected to Sephadex G25 separation (fraction volume 0.25 ml). The peak fraction of the protein was counted for residual calcium (995 cpm). This result confirms the specific exchange of calcium by lanthanum. (protein (h–h), 45calcium labelled enzyme (D–D)). (C) H2O2 affects calcium binding on BuchE. 45calcium labelled enzyme (8583 cpm) was incubated with H2O2 (103 M) for 60 min and subjected to Sephadex G25 separation (fraction volume 0.5 ml). The decrease in counts (390 cpm) indicated alteration of calcium binding after H2O2mediated oxidation of BuchE (H2O2 oxidised protein (h–h), 45calcium labelled enzyme (D–D)).

A

B

1.2

0.8

DTNB Reduction OD 405 nm

DTNB Reduction OD 405 nm

EPN

0.8

0.4

0

0.4

0 0

5 Time (min)

10

0

4 Time (min)

8

Fig. 3. Calcium increases the turnover of neurotoxic organophosphates. (A) Calcium-activated BuchE increases the hydrolysis of the organophosphate EPN 9-fold. Reaction rates of BuchE were determined with saturating substrate in the presence of calcium (103 M) (h–h) and in the absence of calcium (e–e). (B) AchE activities are decreased by EPN. AchE activities were followed for 4 min in the presence of saturating acetylthiocholine (103 M). At that point 0.1 mM EPN was added to the reaction mixture (indicated by the arrow). The result shows a decrease in the reaction rate indicating that EPN is a potent inhibitor of AchE with an apparent Ki for this organophosphate of 45 lM (data not shown).

Hyperchem molecular modelling confirms a single calcium binding site per subunit of BuchE and shows alteration of this domain by H2O2-mediated oxidation of Met and Trp residues The corrected crystal structure of BuchE (see Materials and methods) was examined for potential helix loop helices for a putative EF-hand binding site for calcium. Only one site was found between residues Glu363 and Tyr396, where residues Glu363 and Tyr373 as well as Asn384 and Tyr396

are present on the E and F helices, respectively. The binding sequence of O-donor residues such as two Asp, Glu, Pro, Asn and Gln are present in this putative EF-hand domain. Comparing this sequence with Kretsinger’s established canonical consensus sequence, the putative EF-hand loop in BuchE yields 86% homology [13] (Fig. 4A). In the sequence of BuchE Gly381 is substituted with Arg381. Potential ligands for a traditional canonical EF-hand site in the enzyme are Asp375, Asp378, Gln380, Pro382 and Glu387 which is severely altered after H2O2-mediated oxi-

K.U. Schallreuter et al. / Biochemical and Biophysical Research Communications 355 (2007) 1069–1074

A

Canonical BchE

1073

α-helix E ~ X * Y * Z G # I -X * * -Z ~ α-helix F α-helix E ~ D * D * Q R # E N * * E ~ α -helix F

X,Y, Z,-X and -Z refer to calcium coordinating amino acids (Asp, Asn, Ser, Thr, Glu, Gln). In BuchE: D=Asp, Q=Gln, R=Arg, E=Glu, N=Asn, #=Pro

B

C

Fig. 4. Computer simulation of the corrected BuchE 3D-structure identifies a single EF-hand binding site for calcium per subunit which is severely disrupted after H2O2-mediated oxidation of Met and Trp residues. (A) EF-hand binding consensus sequences of a classical canonical calcium binding loop compared to the proposed BuchE-binding loop yields 86% homology. (B) Close-up view of the EF-hand binding site in BuchE. Asp375, Asp378 and Asn380 contribute single O-donors while the carbonyl group of Pro382 contributes a fourth O-donor atom. Glu387 serves as a bidentate calcium ligand. (C) Comparison of the calcium binding domain before and after H2O2-mediated oxidation of Met and Trp residues (Deep view analysis) shows severe disruption of the EF-hand calcium binding site. The region of the EF-hand binding domain reaching from Tyr363 to Glu396 of the sequence contains several Met and Trp residues as potential targets for H2O2-mediated oxidation which in turn lead to major shifts of the backbone structure (a) native enzyme and (b) oxidised enzyme.

dation of the protein. This result confirms the experimental data supporting disruption of the calcium binding site after H2O2-mediated oxidation (Fig. 4B and C). Discussion The human epidermis holds the capacity for auto- and paracrine cholinergic signal transduction [7–9,14]. Briefly, choline is transported and acetylcholine is synthesised via choline acetyltransferase (chAT, EC 2.3.1.6), transported and stored by the vesicular acetylcholine transporter (VachT) and degraded by both AchE and BuchE [7,8,14,15]. Moreover, the presence of all five muscarinic receptors and the nicotinic ion channel receptors has been demonstrated on epidermal keratinocytes and melanocytes (for review see Grando et al. [7]). Only recently it was shown that BuchE activities in the human epidermis are double those of non-neuronal AchE implicating this avascular tissue in a unique position vis a vis acetylcholine degradation compared to other tissues and organs [1,8]. Moreover, it was realised that H2O2 regulates both enzymes in a concentration dependent manner [8,9]. BuchE has been shown to hydrolyse a large number of toxic organophosphates which in turn are functioning as suicide inhibitors of AchE due to irreversible esterification at the Ser200 active site leading to complete disruption of the cholinergic neurotransmission signal. Hence, it has been postulated that BuchE functions as a protective system for AchE [4,6,16–19].

The outer layer of the skin is target to many insults including H2O2 in the millimolar range originating from UV-light [20,21]. It has been documented that calcium plays a pivotal role in the redox balance as well as controlling many mechanisms including differentiation of epidermal keratinocytes [7,22] and a linear calcium gradient in the epidermis has been documented [10]. Taking into consideration that epidermal acetylcholine has been implicated in this differentiation process, it was tempting to follow the influence of calcium on both AchE and BuchE activities in this tissue. Our results show that BuchE is activated 9-fold by this ion while AchE is not affected. Moreover, we identified a specific EF-hand calcium binding site per subunit of BuchE resulting in four bound calcium atoms per active tetramer. The specificity for calcium activation of BuchE was confirmed by lanthanum substitution. Considering that epidermal BuchE activities are double those of nonneuronal AchE in the same compartment, our results support that AchE in this tissue is protected against organophosphate-mediated suicide inhibition by calcium-activated BuchE as suggested earlier for other tissues by Soreq et al. [2]. In this context it is noteworthy that BuchE is subject to regulation by H2O2 where concentrations <0.5 · 103 M activate, while >0.5 · 103 M H2O2 deactivate this enzyme by disrupting its active site, the tetramerisation domain and the calcium binding site [8]. Given that the human epidermis holds calcium concentrations in the 103 M range in the suprabasal layers including the stratum corneum together with cellular and high membrane

1074

K.U. Schallreuter et al. / Biochemical and Biophysical Research Communications 355 (2007) 1069–1074

integrated BuchE levels, we propose that epidermal calcium-activated BuchE in this avascular tissue provides a major first line protection mechanism for AchE in the same compartment. Most importantly, this novel mechanism offers an efficient strategy against the deleterious effects of neurotoxins entering the skin. References [1] A. Chatonnet, O. Lockridge, Comparison of butyrylcholinesterase and acetylcholinesterase, Biochem. J. 260 (1989) 625–634. [2] H. Soreq, H. Zakut, Amplification of butyrylcholinesterase and acetylcholinesterase genes in normal and tumor tissues: putative relationship to organophosphorous poisoning, Pharm. Res. 7 (1990) 1–7. [3] Y. Nicolet, O. Lockridge, P. Masson, J.C. Fontecilla-Camps, F. Nachon, Crystal structure of human butyrylcholinesterase and of its complexes with substrate and products, J. Biol. Chem. 278 (2003) 41141–41147. [4] H. Sun, J. El Yazal, O. Lockridge, L.M. Schopfer, S. Brimijoin, Y.P. Pang, Predicted Michaelis–Menten complexes of cocaine-butyrylcholinesterase. Engineering effective butyrylcholinesterase mutants for cocaine detoxication, J. Biol. Chem. 276 (2001) 9330–9336. [5] B.P. Doctor, L. Raveh, A.D. Wolfe, D.M. Maxwell, Y. Ashani, Enzymes as pretreatment drugs for organophosphate toxicity, Neurosci. Biobehav. Rev. 15 (1991) 123–128. [6] L. Raveh, J. Grunwald, D. Marcus, Y. Papier, E. Cohen, Y. Ashani, Human butyrylcholinesterase as a general prophylactic antidote for nerve agent toxicity. In vitro and in vivo quantitative characterization, Biochem. Pharmacol. 45 (1993) 2465–2474. [7] S.A. Grando, M.R. Pittelkow, K.U. Schallreuter, Adrenergic and cholinergic control in the biology of epidermis: physiological and clinical significance, J. Invest. Dermatol. 126 (2006) 1948–1965. [8] K.U. Schallreuter, N.C.J. Gibbons, C. Zothner, S.M. Elwary, H. Rokos, J.M. Wood, Butyrylcholinesterase is present in the human epidermis and is regulated by H2O2: more evidence for oxidative stress in vitiligo, Biochem. Biophys. Res. Commun. 349 (2006) 931– 938. [9] K.U. Schallreuter, S.M. Elwary, N.C. Gibbons, H. Rokos, J.M. Wood, Activation/deactivation of acetylcholinesterase by H2O2: more evidence for oxidative stress in vitiligo, Biochem. Biophys. Res. Commun. 315 (2004) 502–508.

[10] G.K. Menon, S. Grayson, P.M. Elias, Ionic calcium reservoirs in mammalian epidermis: ultrastructural localization by ion-capture cytochemistry, J. Invest. Dermatol. 84 (1985) 508–512. [11] G.L. Ellman, K.D. Courtney, V. Andres Jr., R.M. Feather-Stone, A new and rapid colorimetric determination of acetylcholinesterase activity, Biochem. Pharmacol. 7 (1961) 88–95. [12] V.F. Kalb Jr., R.W. Bernlohr, A new spectrophotometric assay for protein in cell extracts, Anal. Biochem. 82 (1977) 362–371. [13] R.H. Kretsinger, Calcium-binding proteins, Annu. Rev. Biochem. 45 (1976) 239–266. [14] S.M. Elwary, B. Chavan, K.U. Schallreuter, The vesicular acetylcholine transporter is present in melanocytes and keratinocytes in the human epidermis, J. Invest. Dermatol. 126 (2006) 1879–1884. [15] R.V. Haberberger, U. Pfeil, K.S. Lips, W. Kummer, Expression of the high affinity-choline transporter, CHT1, in the neuronal and nonneuronal cholinergic system of human and rat skin, J. Invest. Dermatol. 119 (2002) 943–948. [16] H. Soreq, H. Zakut, Expression and in vivo amplification of the human acetylcholinesterase and butyrylcholinesterase genes, Prog. Brain Res. 84 (1990) 51–61. [17] C.A. Prody, P. Dreyfus, R. Zamir, H. Zakut, H. Soreq, De novo amplification within a ‘‘silent’’ human cholinesterase gene in a family subjected to prolonged exposure to organophosphorous insecticides, Proc. Natl. Acad. Sci. USA 86 (1989) 690–694. [18] B. Veronesi, K. Jones, C. Pope, The neurotoxicity of subchronic acetylcholinesterase (AChE) inhibition in rat hippocampus, Toxicol. Appl. Pharmacol. 104 (1990) 440–456. [19] A. Blair, S.H. Zahn, K.P. Cantor, P.A. Stewart, Estimating exposure to pesticides in epidemiological studies of cancer, ACS Symp. Ser. 382 (1989) 38–46. [20] V. Maresca, E. Flori, S. Briganti, E. Camera, M. Cario-Andre, A. Taieb, M. Picardo, UVA-induced modification of catalase charge properties in the epidermis is correlated with the skin phototype, J. Invest. Dermatol. 126 (2006) 182–190. [21] K.U. Schallreuter, J. Moore, J.M. Wood, W.D. Beazley, D.C. Gaze, D.J. Tobin, H.S. Marshall, A. Panske, E. Panzig, N.A. Hibberts, In vivo and in vitro evidence for hydrogen peroxide (H2O2) accumulation in the epidermis of patients with vitiligo and its successful removal by a UVB-activated pseudocatalase, J. Invest. Dermatol. Symp. Proc. 4 (1999) 91–96. [22] J.J. Wille, M.R. Pittelkow, G.D. Shipley, R.E. Scott, Integrated control of growth and differentiation of normal human prokeratinocytes cultered in serum-free medium: clonal analyses, growth kinetics and cell cycle studies, J. Cell Physiol. 121 (1984) 31–44.