Oxidative metabolism of dopamine: A colour reaction from human midbrain analysed by mass spectrometry

Biochimica et Biophysica Acta 1784 (2008) 1687–1693

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / b b a p a p

Oxidative metabolism of dopamine: A colour reaction from human midbrain analysed by mass spectrometry Angela De Iuliis a,⁎, Giorgio Arrigoni b, Liselotte Andersson b, Pamela Zambenedetti c, Alessandro Burlina d, Peter James b, Paola Arslan a, Fabio Vianello e a

Department of Medical Sciences, University of Padua, Padova, Italy Department of Protein Technology, University of Lund, Lund, Sweden Pathology Division and Brain Bank, General Hospital of Dolo-Venice, Italy d Neurological Unit, San Bassiano Hospital, Bassano del Grappa, Italy e Department of Biological Chemistry, University of Padua, Padova, Italy b c

a r t i c l e

i n f o

Article history: Received 14 March 2008 Received in revised form 2 July 2008 Accepted 3 July 2008 Available online 15 July 2008 Keywords: Dopamine Dopaminochrome Dopamine peroxidizing activity Human brain Parkinson's disease Mass spectrometry

a b s t r a c t In order to identify the protein responsible for a dopamine peroxidizing activity, previously described in human normal and parkinsonian substantia nigra by our group, we developed non-denaturing polyacrylamide gel electrophoresis conditions, mimicking the characteristic colour in vitro reaction, resulting from cyclic oxidation of dopamine (DA). After separating protein mixtures from human normal midbrain homogenates on two sets of identical native gels, one gel set was subjected to specific activity staining by using DA and hydrogen peroxide. An activity red/orange band appeared in midbrain tissue lanes, similarly to the lane where commercial horseradish peroxidase (HRP) was present as control of peroxidative activity. The second set of gels, stained with Coomassie Blue, showed other, not enzymatically active protein bands. Mass spectrometry analysis of the bands containing the activity and the corresponding Coomassie Blue bands revealed the presence of proteins that may play a role in neurodegenerative disease, highlighting a possible functional link among dopamine/dopaminochrome redox cycle and protein metabolism. © 2008 Elsevier B.V. All rights reserved.

1. Introduction Over the past years, the role of dopamine metabolites has been focus of neurodegeneration studies in Parkinson's disease (PD). A characteristic autoptic finding of this disease is a depigmented substantia nigra with loss of dopaminergic neuromelanin-containing neurons and post-mortem studies have consistently indicated oxidative damage in PD pathogenesis [1,2]. The primary cause of oxidative stress has not been clarified but the leading candidate is the metabolism of DA itself, since it gives rise to various toxic species. Reactive products of partially reduced oxygen, such as superoxide, hydrogen peroxide, and hydroxyl radical are generated from monoamine oxidase (MAO)-deamination, during normal catabolism of DA, and also from catechol oxidation, during oxidative pathways in which DA can serve as substrate for neuromelanin synthesis [3]. The pathway of the reactions converting dopamine to neuromelanin is not yet well understood, but it is thought to proceed through DA oxidation to dopamine o-quinone, cyclization of dopamine o-quinone to dopaminochrome, leading to the formation of leukoaminochrome, oxidation of leukoaminochrome to dopaminochrome and polymerization of dopaminochrome to neuromelanin ⁎ Corresponding author. Tel.: +39 49 827 6143; fax: +39 49 807 3310. E-mail address: [email protected] (A. De Iuliis). 1570-9639/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2008.07.002

(Fig. 1) [4,5]. The DA oxidation can occur spontaneously [3], is accelerated by transition metal ions (Mn2+ or Fe2+) [6], or can be catalyzed by a number of different enzymes [7–11]. The quinone consequently produced has an electron-deficient ring, which readily forms covalent bonds with available nucleophiles [12], especially the cellular important sulphydryl groups, such as L-cysteine, glutathione, and proteins containing cysteinyl residues [13]. Several in vitro studies have demonstrated the binding of quinone formed by DA to protein sulphydryl groups, leading to inactivation of enzymes of vital importance for the cell function, and such reactions have been implicated in the neurodegeneration process of catecholaminergic neurons [14–17]. The in vivo occurrence of dopamine oxidation has been demonstrated by the recovery of dopamine adducts in human and in other mammalian brain extracts [18–22]. However, in the absence of competing nucleophiles, the amine side chain of DA is readily available for 1,4-intramolecular ring closure and oxidation, forming relatively more stable dopaminochrome (2,3-1H-indole-5,6quinone), the potential precursor of neuromelanin [12]. Dopaminochrome is a member of the family of red to violet coloured indoline5,6-quinones, known as aminochromes, which are readily obtained on oxidation of the corresponding catecholamines [23]. Occurrence in the brain of dopaminochrome comes from the fact that this metabolite was identified as part of neuromelanin, along with 5,6dihydroxyindole, a rearrangement product of dopaminochrome itself

1688

A. De Iuliis et al. / Biochimica et Biophysica Acta 1784 (2008) 1687–1693

Fig. 1. Neuromelanin formation. The reaction pathway converting dopamine to neuromelanin is a normal process in the substantia nigra, which it is thought to involve several steps: dopamine oxidation to dopamine o-quinone catalyzed by metals, oxygen, peroxynitrite or peroxidative activity of several enzymes (PX), such as prostaglandin H synthase, cytochrome P450 1A2, xantine oxidase, cyclization of dopamine o-quinone to dopaminochrome via an addition, at physiological pH values, leading to the formation of unstable leukoaminochrome and oxidation of leukoaminochrome to dopaminochrome and polymerization of dopaminochrome to neuromelanin, via dihydroxyindole formation (see the text).

and noradrenochrome [24–26]. Hastings [8] described dopaminochrome in vitro formation from Prostaglandin H synthase (PHS)mediated dopamine oxidation, while Mattamal showed the in vivo reaction, by a mass spectrometry study [9]. Among possible other candidates able to oxidize DA to dopaminochrome, cytochrome P450 1A2 [10], xanthine oxidase [11] and peroxynitrite [27,28] were reported, but to date, none of them has been identified as specifically responsible for such reaction in nigral dopaminergic neurons. We have previously provided spectrophotometric evidence of the presence of an enzymatic activity in human substantia nigra [29]. This activity, catalyzing the formation of dopaminochrome from dopamine and hydrogen peroxide, was first demonstrated in rat brain fraction [30]. We subsequently observed increased dopamine peroxidation in the midbrain and basal ganglia of Parkinsonian brain, obtained at autopsy [31]. The colour reactions resulting from the cyclic oxidation of catecholamines formed the basis of qualitative and quantitative assay procedures [23]. In an attempt to identify the protein responsible for dopamine peroxidizing activity in substantia nigra, we have analysed the characteristic, coloured reaction forming the dopaminochrome by using a proteomic approach. We report here the specific staining procedure, developed for in gel-detection of human DA peroxidizing activity, after electrophoresis of midbrain tissue homogenates, and the mass spectrometry analysis, which revealed the presence of proteins in the activity bands. 2. Materials and methods 2.1. Chemicals and instruments Dopamine, hydrogen peroxide, reagents used for homogenate buffer, and horseradish peroxidase (TYPE VI-A, P6783) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Acrylamide, TEMED, ammonium persulfate, Coomassie Brilliant Blue G-250 as well as all the reagents and apparatus for gel electrophoresis were obtained from Bio-Rad Laboratories, Inc. (Hercules, CA, USA). Spectrophometric measurements were performed at 25 °C, with a Beckman DU-640 recording instruments equipped an Epson FX-850

recorder. Brain homogenates were obtained by a Beckman, J2-21 model fixed angle rotor, centrifuge. Mass spectra were acquired with a Q-TOF Ultima mass spectrometer (Waters Manchester, UK). 2.2. Sample preparation Human normal midbrain tissues from four different specimens were obtained from the Brain Bank of the General Hospital of Dolo (Venice, Italy). Each specimen was accompanied by a histopathological report, certifying they were not affected by neurodegenerative diseases, neither brain vasculopathy involving mesencephalic regions (sex: 3 males, 1 female, aged 75 ± 5 years). The autopsies were performed within 24 h of death. Midbrain has been excised from thawed autoptic specimens, using ceramic tools to avoid metal contamination of the samples. When present, blood clots were removed before sample processing. Brain specimens (1–3 g) were homogenized and the supernatant was spectrophotometrically tested for the enzymatic activity, as described [31]. Protein content was estimated by Bio-Rad assay. 2.3. Gel electrophoresis Two identical sets of three non-denaturing polyacrylamide gel (3,5% acrylamide stacking gel and 7% separating gel) were performed in Bio-Rad Mini Protean II cell (1 mm thickness, 10 cm × 10 cm gel size). Ten microliters of homogenates from midbrain tissues, corresponding to 10 μg of total proteins, was loaded onto each gel, along with 2 ml (2 μg) of horseradish peroxidase (HRP), used, without further purification, as control of peroxidative activity. Electrophoresis was carried out with a current of 10 mA/gel at room temperature, one extra hour after the dye front had migrated to the end of the gel. 2.4. Peroxidative activity and protein staining After electrophoresis, one gel set was soaked in substrate solution (0. 2 M sodium phosphate buffer, pH 7.4, containing 2 mM dopamine and 30 mM hydrogen peroxide), with gentle shaking at room temperature, until an orange activity band appeared.

A. De Iuliis et al. / Biochimica et Biophysica Acta 1784 (2008) 1687–1693

1689

The corresponding second control gel set was stained with Coomassie Brilliant Blue, according to Chrambach et al. [32]. 2.5. In gel tryptic digestion The band of interest was excised from the gel, shrunk in 100% acetonitrile and then reswollen in 50 μl of 10 mM DTT and incubated at 56 °C for 1 h. The gel piece was then rinsed and treated with 50 μl of 55 mM iodoacetamide for 45 min at room temperature in a dark place. The band was then dried under vacuum and reswollen in 10 μl of trypsin (12.5 ng/μl) at 4 °C for 45 min. The gel piece was covered with 25 mM NH4HCO3 and the digestion was carried on overnight at 37 °C. The tryptic peptides were extracted once with 25 mM NH4HCO3 and then with three changes of 50% acetonitrile/5% formic acid. The solution was then concentrated under vacuum and used for MS/MS analysis. 2.6. MS/MS analysis of tryptic peptides A 5 μl portion of the protein digest was separated by nanoscale liquid chromatography quadrupole time-of-flight tandem mass spectrometry (nanoHPLC-Q-TOF MS/MS) using a C18 PepMap100 capillary column (3 μm, 100 Å, 15 cm, 75 μm I.D, LC Packings, CA, USA) and a CapLC HPLC system (Waters) interfaced to an Ultima Q-TOF (Waters). The peptides were separated with a linear gradient of acetonitrile/formic acid 0.1% from 7% to 85% in 60 min. The analysis was done in a Data Dependent Acquisition mode: during the run the three most intense ions at each run time were automatically selected for the MS/MS analysis if their threshold was above 30 counts/s. The MS/MS acquisition time was limited to 10 s or interrupted if the signal intensity dropped under 10 counts/s. A scan time of 1 s was used for both MS and MS/MS mode using a 400–1500 Da mass range for MS mode and 50–1500 Da for MS/MS mode. MS/MS spectra were analysed using Mascot Search Engine (Matrix Sciences, London, UK) to identify tryptic peptide sequences matched to the National Center for Biotechnology Information (NCBI) non-redundant protein database. Only proteins identified by at least 2 different peptides with a significant score (p b 0.05) were considered correctly identified. All the MS/MS spectra were manually inspected for further confirmation. 3. Results 3.1. Detection of peroxidative activity from human midbrain on native gel Protein mixtures of midbrain tissues homogenates from four different specimens were separated in two corresponding gel sets. Each set was comprised three native gels. Electrophoresis running was simultaneously carried out for both gel sets. One gel set was soaked in the specific substrate solution to visualize the expected dopamine peroxidizing activity and the other identical gel set was stained with Coomassie Blue. Fig. 2 shows the specific staining of dopamine peroxidizing activity in native PAGE (panel A). The gel depicted in the image is representative of three gels that were performed. Within 10 min from the dopamine/hydrogen peroxide staining, a red/orange band appeared in all the midbrain tissue lanes, similarly to the lane where the commercial HRP was present as control of peroxidative activity. This corresponds to dimeric HRP with an apparent molecular weight of 80 kDa at the top of the separating gel and to monomeric HRP with a apparent molecular weight of 40 kDa. In the Coomassie Blue-stained gel, other proteins present in the midbrain tissue homogenates, but not reacting with the substrate, appeared (panel A′ of Fig. 2). 3.2. Protein identification by mass spectrometry To identify the colour forming fraction of midbrain tissue, we cut the orange bands and the corresponding Coomassie Blue-stained

Fig. 2. Enzymatic detection in native PAGE. Native electrophoresis was applied to human midbrain tissue homogenates from four different specimens (lanes 1–4, 1′–4′) on 7% polyacrylamide gel, along with horseradish peroxidase (HRP) used as positive control (lane 6, 6′). Lanes 5, 5′ were empty. Within 10 min from the dopamine/ hydrogen peroxide staining, a red/orange band appeared, in all the midbrain tissue lanes, as indicated by the arrow, similarly to the lane where the commercial HRP was present as control of peroxidative activity. The gel depicted is representative of three gels performed. Panel A: Specific staining for the dopamine peroxidizing activity was obtained by soaking the gel in substrate solution as described in Materials and methods. Panel A′: Protein staining with Coomassie Brilliant Blue R-250 dye allowed to visualize the complete protein pattern of brain tissue homogenate. The arrow indicates the Coomassie Blue band corresponding to the red–orange activity band. Other proteins, present in the midbrain tissue homogenates but not reacting with the substrate, appear.

bands, (these bands are indicated with arrows in the figure) and analysed them by matrix assisted laser desorption ionization-time-offlight (MALDI-TOF) and subsequently by nanoHPLC-Q-TOF mass spectrometry. As blank sample we analysed a colourless band, cut immediately below the reactive band. In the enzymatically active bands we identified a series of different proteins which corresponded to those present in the Coomassie Bluestained bands. MALDI-TOF mass analysis revealed the presence of the glial fibrillary acidic protein (GFAP) (data not shown). Q-TOF analysis confirmed the presence of GFAP and identified eleven more proteins. Table 1 lists all the peptides that have been identified by nanoHPLC-Q-TOF mass spectrometry analysis. The molecular weights of the corresponding proteins range about between 12 and 63 kDa.

1690

A. De Iuliis et al. / Biochimica et Biophysica Acta 1784 (2008) 1687–1693

Table 1 Proteins identified in activity band and in the corresponding Coomassie Blue-stained band by means of Q-TOF mass spectrometry Protein ID

1

1P00293276

Protein name

Macrophage migration inhibitory factor (MIF) 2 IPI00329351 Peroxiredoxin-1 3 IPI004119501 FTH1, protein ferritin heavy chain 4 IPI00375676 Ferritin light chain (ferritin L subunit) 5 IPI00418262 ALDOC protein (fructose-bisphosphate aldolase, brain-type aldolase) 6 IPI00395757 Fructose-bisphosphate aldolase A 7 IPI00220644 Pyruvate kinase isozymes M1 8 IPI00465248 Isoform alpha-enolase of Alpha-enolase 9 IPI00169383 Phosphoglycerate kinase 1 10 IPI00219018 Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) 11 IPI00257508 Dihydropyriminidase-related protein 2 12 IPI0025363 Glial fibrillary acidic protein, astrocyte (GFAP)

Molecular % weight Sequence coverage 12,900

9

22,110 21,094 19,888 39,325

5 14 17 7

39,289 57,931 47,169 44,483 35,922

15 6 16 10 11

62,297 49,880

6 17

Macrophage migration inhibitory factor (MIF) and peroxiredoxin-1 (PRX-1) were present in the reactive band, as well as ferritin heavy and light chains. Ferritin iron was utilized to synthesize neuromelanin in vitro, as Fe3+ is a known effective catalyst of dopamine [33] and oxidation of adrenaline in adrenochrome by ferritin iron and hydrogen peroxide was also described [34]. Thus, we have investigated the single possible action of ferritin on the dopamine peroxidating activity. We carried out spectrophotometric assay of DA peroxidation by ferritin, as described [30] by means of commercial horse spleen protein. No catalytic activity was observed adding ferritin in the concentration range 5–20 μg/l. Native electrophoresis was then carried out in duplicate, with different amount of ferritin (10 ng, 100 ng, 1 μg) using one midbrain tissue sample and HRP as controls of enzymatic activity. One gel was stained in buffer solution containing only hydrogen peroxide (30 mM) without DA, to observe if ferritin iron reacting with H2O2 was able to catalyze Fenton's reaction [34]. The second gel was instead stained in substrate solution as above described. In the first case we did not observe any colour reaction, in none of the lanes. In the second case, no colour reaction was visible in 1 ng and 100 ng ferritin lanes, while the 1 μg ferritin lane appeared red even without any staining (data not shown). The following six proteins, detected in the reactive band as well, are involved on glucose metabolism: glyceraldehyde-3-phosphate dehydrogenase (GAPDH), phosphoglycerate kinase 1 (PGK1), fructosebisphosphate aldolase C (FBC), fructose-bisphosphate aldolase A (FBA), pyruvate kinase (PK), and alpha-enolase. Finally, dihydropyriminidinase-related protein 2 (DRP-2), a protein which promotes axonal growth, was also present, along with the above cited GAFP. The proteins identified in the colourless band, and therefore not involved in the enzymatic reaction, were: malate dehydrogenase (MDH), isoform 1 of alpha-adducin, retinal dehydrogenase 1 (Raldh 1), and IGKV1-5 protein (Table 2). The lack of the reaction, in the gel band containing these proteins, seems to be consistent with their known function. Brain adducin was reported to promote binding of brain spectrin to actin and as substrate of Proteinase C [35], while Raldh1, catalyzing the oxidation of retinaldehyde into retinoic acid, is a specific marker of DA progenitors in the midbrain, although its role in the brain remains to be fully elucidated [36]. IGKV1-5 is Ig-kappa chain II region of Ig-like family [37]. Conversely, MDH, protein of energy metabolism, behaves differently from other enzymes of energy metabolism identified in the reactive band.

4. Discussion Gel staining method evidencing catalytic activity has been validated for several enzymes [38–40] and the detection of dopachrome isomerase from Manduca sexta larvae and dopaminochrome isomerase from Rhinoceros oryctes larvae was also reported [41–43]. Here we report, for the first time, peroxidative activity staining, after electrophoresis, from human midbrain tissue, by using dopamine and hydrogen peroxide as substrates for the enzymatic reaction. This reaction is likely catalyzed by proteins that could be responsible for a DA peroxidizing activity [31]. DA in vitro peroxidation catalyzed by several peroxidases, such as myeloperoxidase, lactoperoxidase, HRP develops a characteristic red/ orange colour [7,30]. We have previously utilized HRP/hydrogen peroxide system [7] as enzymatic model of dopamine peroxidizing catalysis, in the in vitro assays [29,30]. Procedures for detection of peroxidative activity of HRP, following SDS-polyacrylamide gel (SDSPAGE) were validated by Schmidt and Trojanowski [44]. We have optimised a similar procedure to investigate non-denaturing-polyacrylamide gel electrophoresis conditions in order to isolate the human midbrain protein fraction containing DA peroxidizing activity. To determine the conditions for the staining of brain tissues we have modified the protocol reported by Smith and Trojanowski [44]. We could not visualize any enzymatically active bands in SDSPAGE containing 10% acrylamide, after SDS extraction, as described in [44], neither in PAGE containing acrylamide concentrations ranging from 8 to 10%. The optimal conditions for PAGE were instead reached utilizing 7% acrylamide and carrying out the electrophoretic process for an extra hour more, after the dye front had migrated at the end of the gel. In these conditions all the enzymatically active bands appeared, after the substrate staining. Mass spectrometry identifications of enzymatic activity were shown to be performed by non-denaturing two-dimensional (2D) gel, after detergent extraction or microscale non-denaturing 2D gel [45–47]. Here, we have tried to use mass spectrometry analysis of dopamine peroxidizing activity, after in situ detection on native mono-dimensional gel. Embedded DA and hydrogen peroxide did not interfere with the mass spectrometric identification of the peptides, as demonstrated by comparing with the parallel analysis of the corresponding non-denaturing gel band, stained in Coomassie Blue. The fractionation of the proteins in their native conformation by gel electrophoresis cannot distinguish between the effects of size, shape, and charge on electrophoretic mobility. As a consequence, proteins with different molecular weight can show the same mobility of those really involved in the enzymatic activity, overlapping the DA peroxidation reaction. As shown in Table 1, the reactive band appears to comprise a series of proteins, apparently involved in the oxidative metabolism of DA. Our attention is focusing on macrophage migration inhibitory factor (MIF) and peroxiredoxin-1 (Prx1). Among the others, they may be likely major candidates to DA peroxidation, since they are the only proteins known possessing putative peroxidase activity. MIF is an evolutionary conserved protein, with pleiotropic effects, ranging from immunity to redox regulation. MIF has been reported to possess a thiol-protein oxidoreductase activity and demonstrated to play a role in cellular redox signaling, since it was shown to modulate cellular glutathione levels [48]. This protein is highly expressed in

Table 2 Proteins identified in colourless, non-reactive band

1 2 3 4

Protein ID

Protein name

Molecular weight

% Sequence coverage

IPI00291005 IPI00019901 IPI00218914 IPI00419424

Malate dehydrogenase, cytoplasmic Isoform 1 of Alpha-adducin Retinal dehydrogenase 1 IGKV1-5 protein

36,426 80,955 54,862 26,234

7 8 11 18

A. De Iuliis et al. / Biochimica et Biophysica Acta 1784 (2008) 1687–1693

many tissues including brain, where its function has not been well established. However, it was postulated a MIF involvement in the metabolism of oxidized catecholamines and related compounds. MIF was characterized for a D-dopachrome tautomerase activity, which converts D-dopachrome methylester to 5,6-dihydroxyindole-2-carboxymethylester (DHICA) [49,50]. Successive studies have shown that such a protein is able to catalyze the conversion of dopaminochrome and noradrenochrome to indoledihydroxy derivatives [51], potential precursor of neuromelanin, thus suggesting a possible role of the protein in neuromelanin synthesis [52]. Recently, MIF was demonstrated to form amyloid fibrils similar to those surrounding dead neurons in parkinsonian and Alzheimer's brain [53]. Interestingly, an opposite action was instead described for the dopaminochrome, which seems to be involved in the inhibition of alphasynuclein filament formation [54]. The Peroxiredoxins (Prxs) have received considerable attention in recent years as members of a new expanding family of multifunctional, antioxidant thioredoxin-dependent peroxidases, whose major functions comprise regulation of cell proliferation, cellular protection against oxidative stress and modulation of intracellular signaling cascades that apply hydrogen peroxide as a second messenger molecule [55]. Very significantly, specific interaction between MIF and PAG, a peroxiredoxin, was shown to involve a mixed disulphide, which could also mutually influence the enzymatic activity of these proteins [56]. Light (L) and heavy (H) chain ferritin, the iron storage protein, are described in all the brain regions [57] and in substantia nigra as well, where iron appears also to be bound to neuromelanin [58]. In addition to a range of neurologic activities, such as cognition and myelogenesis, iron is required for the synthesis of dopamine [59,60] and perturbed iron metabolism can lead to molecular and cellular damage via oxidative stress [61]. Moreover, previous studies have shown that iron, ferritin and neuromelanin undergo dramatic changes in PD [62]. In our case, ferritin seems not directly involved in the enzymatic reaction, and, actually, ferritin was not classified as peroxidase [34]. With regard to GDNF, it has been shown to protect and restore dopamine neurons, and was recently demonstrated to reduce oxidative stress in animal model of PD [63]. We suppose that the presence in the activity band of enzymes of glycolytic pathway should be ascribed to the their electrophoretic mobility, depending on both the protein's charge and its hydrodynamic size that colocalize them at the level of the enzymatic activity. However, glycolytic enzymes were frequently described in several different proteomic studies, as involved in many different pathologies. Concerning the brain, GAPDH, PGK1, FBC, PK, along with GFAP and peroxiredoxin-1 were found to be present in human substantia nigra, from control and parkinsonian patients by Basso et al. [64], apparently without significative differential expression between normal and pathological brain tissues. In other studies, proteins of energy metabolism are shown to be implicated in oxidative stress: alphaenolase levels appeared increased in 6-hydroxydopamine-treated substantia nigra of PD animal model [65]. In Alzheimer's disease (AD) brain, alpha-enolase was also identified as oxidatively modified along with GAPDH, PGK1, FBC, FBA, PK, and dihydropyriminidinerelated protein 2 (DRP-2), a protein involved in the formation of neuronal connection [66]. Association between dopaminergic impairment and glucose hypometabolism was further demonstrated in PD, and elsewhere studies of post-mortem brain tissues from AD and PD patients have provided evidence for increased levels of oxidative stress, mitochondrial dysfunction and impaired glucose uptake in vulnerable neuronal populations [67,68]. On the other hand, glycolytic enzymes, which were believed to have only metabolic functions, were shown to play other unsuspected, so called “moonlighting” roles, [69] particularly in the brain, where they were identified as multiprotein complex components. Neuronal Aldolase A and Aldolase C were described as specifically interacting

1691

with light neurofilament (NF-L) mRNA, exhibiting ribonuclease activity in regulatory pathway of the neuron form and function [70]. GAPDH is increasingly recognized as multifunctional protein involved in processes distinct from glycolysis, including the induction of neuronal apoptotic death. Evidence also links GAPDH to the pathogenesis of classical PD, as GAPDH colocalizes with alphasynuclein in Lewy bodies, well known hallmarks of PD and AD diseases [1,2], and nuclear translocation of GAPDH has been detected in nigral neurons in post-mortem PD brains [71,72]. In addition, GAPDH is a part of RNA binding cAMP responsive-multiprotein complex that was described to be inhibited by DJ-1 [73], a protein acting as redox-regulated chaperone [74], oxidative stress sensor [75] and regulator of tyrosine hydroxylase transcription [76]. Very significant, several mutations of the DJ-1 gene have been associated with familiar cases of PD [77,78]. 5. Conclusions In summary, dopamine peroxidizing activity was isolated from human midbrain by means of non-denaturing gel electrophoresis. Mass spectrometry analysis of the enzymatically active bands of the gel led to the identification of a series of proteins, apparently involved in the oxidative metabolism of dopamine. In the past, studies on the dopaminergic cell degeneration, focusing on the oxidative metabolism of DA, described cytotoxic and genotoxic potential of this neurotransmitter through DA adducts formation [79]. Today, a new scenario comes from familial PD-linked genes discoveries, associated with toxicological studies and epidemiological investigations that show several, intersecting pathways underlying PD pathogenesis. Oxidative metabolism of DA seems to be involved in the impairment of mitochondrial function, protein misfolding and abnormal aggregation, as well as in the signaling pathway in dopaminergic neuronal survival, such as possible red thread of multiple etiological events triggering the disease mechanism [1,2,80–82]. On the light of these latest studies, our results suggest a possible functional link between oxidative species and protein metabolism. Future research will allow us to further dissect the details of the dopamine peroxidizing activity. The presence of MIF and Peroxiredoxin-1 needs additional investigations: the properties of both proteins are consistent with their role in the maintenance of redox potential within cells. By defining the interplay of dopamine– dopaminochrome/protein system, as well as understanding how this interaction is regulated, invaluable insights into protein function and DA metabolism could be forthcoming. Acknowledgements We are very grateful to prof. A. Bindoli and prof. M.P. Rigobello for their helpful discussions. References [1] J.T. Greenamyre, T.G. Hastings, Parkinson's-divergent causes, convergent mechanisms, Science 304 (2004) 1120–1122. [2] P. Jenner, Oxidative stress in Parkinson's disease, Ann. Neurol. 53 (2003) S26–S38. [3] D.G. Graham, Oxidative pathways for catecholamines in the genesis of neuromelanin and citotoxic quinone, Mol. Pharmacol. 14 (1978) 633–643. [4] D.G. Graham, S.M. Tiffany, W.R. Bell, W.F. Gutknecht, Autoxidation versus covalent binding of quinones as the mechanism of toxicity of dopamine, 6-hydroxydopamine, and related compound toward C1300 neuroblastoma cells in vitro, Mol. Pharmacol. 14 (1978) 644–653. [5] L. Zecca, D. Tampellini, M. Gerlach, P. Riederer, G.R. Fariello, D. Sulzer, Substantia nigra neuromelanin: structure, synthesis, and molecular behaviour, J. Clin. Pathol. Mol. Pathol. 54 (2001) 414–418. [6] B. Halliwell, J.M.C. Gutteridge, Oxygen toxicity, oxygen radicals, transition metals and disease, J. Biochem. 219 (1984) 1–14. [7] A. Napolitano, O. Crescenzi, A. Pezzella, G. Prota, Generation of the neurotoxin 6hydroxydopamine by peroxidase/H2O2 oxidation of dopamine, J. Med. Chem. 38 (1995) 917–922.

1692

A. De Iuliis et al. / Biochimica et Biophysica Acta 1784 (2008) 1687–1693

[8] T.G. Hastings, Enzymatic oxidation of dopamine: the role of prostaglandin H synthase, J. Neurochem. 64 (1995) 912–924. [9] M. Mattamal, R. Strong, V.M. Lakshmi, H.D. Chung, A.H. Stephenson, Prostaglandin H synthetase-mediated metabolism of dopamine: implication for Parkinson's disease, J. Neurochem. 64 (1995) 1645–1653. [10] J. Segura-Aguilar, Peroxidase activity of liver microsomal vitamin D 25hydroxylase and cytochrome P450 1A2 catalyzes 25-hydroxylation of vitamin D3 and oxidation of dopamine to aminochrome, Biochem. Mol. Med. 58 (1996) 122–129. [11] C. Foppoli, R. Coccia, C. Cini, A.M. Rosei, Catecholamine oxidation by xanthine oxidase, Biochim. Biophys. Acta. 1334 (1997) 200–206. [12] A. Bindoli, M.P. Rigobello, D. Deeble, Biochemical and toxicological properties of the oxidation products of catecholamines, Free Radic. Biol. Med. 13 (1992) 391–405. [13] S. Ito, T. Kato, K. Fujita, Covalent binding of catechols to proteins through the sulphydryl group, Biochem. Pharmacol. 37 (1988) 1707–1710. [14] C.S. Tse, R.L. Mc Creery, R.N. Adams, Potential oxidative pathways of brain catecholamines, J. Med. Chem. 19 (1976) 37–40. [15] F. Zhang, G. Dryhurst, Effects of L-cysteine on the oxidation chemistry of dopamine: new reaction pathways of potential relevance to idiopathic Parkinson's disease, J. Med. Chem. 37 (1994) 1084–1098. [16] T.G. Hastings, D.A. Lewis, M.J. Zigmond, Role of oxidation in the neurotoxic effects of intrastriatal dopamine injections, Proc. Natl. Acad. Sci. 93 (1996) 1956–1961. [17] M.J. La Voje, T.G. Hastings, Dopamine quinone formation and protein modification associated with the striatal neurotoxicity of metamphetamine: evidence against a role for extracellular dopamine, J. Neurosci. 19 (1999) 1484–1491. [18] B. Fornstedt, E. Rosengren, A. Carlsson, Occurrence and distribution of 5-S-cysteinyl derivatives of dopamine, DOPA and DOPAC in the brain of eight mammalian species, Neuropharmacology 25 (1986) 451–454. [19] B. Fornstedt, A. Carlsson, A marked rise in 5-S-cysteinyl-dopamine levels in guinea-pig striatum following reserpine treatment, J. Neural. Transm. 76 (1989) 155–161. [20] B. Fornstedt, I. Bergh, E. Rosengren, A. Carlsson, An improved HPLC-electrochemical detection method for measuring brain levels of 5-cysteinyldopamine, 5cysteinyl-3,4-dihydroxyphenylalanine, and 5-S-cysteinyl-3,4-dihydroxyphenylacetic acid, J. Neurochem. 54 (1990) 578–586. [21] B. Fornstedt, E. Pileblad, A. Carlsson, In vivo autoxidation of dopamine in guinea pig striatum increases with age, J. Neurochem. 55 (1990) 655–659. [22] B. Fornstedt, A. Carlsson, Vitamin C deficiency facilitates 5-S-cysteinyldopamine formation in guinea pig striatum, J. Neurochem. 56 (1991) 407–414. [23] R.A. Heacock, W.S. Powell, Adrenochrome and related compounds, in: G.P. Ellis, G.B. West (Eds.), Progress in Medicinal Chemistry, vol. 9, 1973, pp. 275–339, Amsterdam. [24] R. Carstam, C. Brinck, A. Hindemith-Augustsson, H. Rorsman, E. Rosengren, The neuromelanin of the human substantia nigra, Biochim. Biophys. Acta. 1097 (1991) 152–160. [25] G. Odh, R. Carstam, J. Paulson, A. Wittbejer, E. Rosengren, H. Rorsman, Neuromelanin of the human substantia nigra: a mixed-type melanin, J. Neurochem. 62 (1994) 2030–2036. [26] J. Smythies, On the function of neuromelanin, Proc. R. Soc. Lond. B 263 (1996) 487–489. [27] C. Daveau, C. Servy, M. Dendane, P. Marin, C. Ducrocq, Oxidation and nitration of catecholamines by nitrogen oxides derived from nitric oxide, Nitric Oxide 3 (2002) 234–243. [28] D. Rettori, Y. Tang, L.C. Dias Jr., E. Cadenas, Pathways of dopamine oxidation mediated by nitric oxide, Free Rad. Biol. Med. 33 (2002) 685–890. [29] L. Galzigna, A. De Iuliis, L. Zanetta, Enzymatic dopamine peroxidation in substantia nigra of human brain, Clin. Chim. Acta. 300 (2000) 131–138. [30] L. Galzigna, M.P. Schiappelli, A. Rigo, M. Scarpa, A rat brain fraction and different purified peroxidases catalyzing the formation of dopaminochrome from dopamine, Biochim. Biophys. Acta 1427 (1999) 329–336. [31] A. De Iuliis, A.P. Burlina, R. Boschetto, P. Zambenedetti, P. Arslan, L. Galzigna, Increased dopamine peroxidation in post-mortem Parkinsonian brain, Biochim. Biophys. Acta. 1573 (2002) 63–67. [32] A. Chrambach, R.A. Reisfeld, M. Wyckoff, J. Zaccari, A procedure for rapid and sensitive staining of protein fractionated by polyacrylamide gel electrophoresis, Anal. Biochem. 20 (1967) 154–164. [33] J.E. Gorton, R.F. Jameson, Complexes of doubly chelating ligands. I. Proton and copper-(II) complexes of L-β-(3,4-dihydroxyphenyl)alanine (DOPA), J. Chem. Soc. A (1968) 2615–2618. [34] A. Mazur, S. Green, E. Shoor, The oxidation of adrenaline by ferritin iron and hydrogen peroxide, J. Biol. Chem. 213 (1956) 227–235. [35] V. Bennett, K. Gardner, J.P. Steiner, Brain adducin: a protein kinase C substrate that may mediate site-directed assembly at the spectrin–actin junction, J. Biol. Chem. 263 (1988) 5860–5869. [36] S.-L. Ang, Transcriptional control of midbrain dopaminergic neuron development, Development 133 (2006) 3499–3506. [37] D.K. Smith, H. Xue, Sequence profiles of immunoglobulin and immunoglobulinlike domains, Mol. Biol. 274 (1997) 530–545. [38] C.-L. Lin, H.-J. Chen, W.-C. Hou, Activity staining of glutathione peroxidase after electrophoresis on native and sodium dodecyl sulphate polyacrylamide gels, Electrophoresis 23 (2002) 513–516. [39] M.-H. Lee, M.-T. Chuang, W.-C. Hou, Activity staining of plasma amine oxidase after polyacrylamide gel electrophoresis and its application to natural inhibitor screening, Electrophoresis 23 (2002) 2369–2372. [40] K. Nellaiappan, K. Valivittan, Differentiation of tyrosine hydroxylase and phenol oxidase after electrophoresis, Biotech. Histochem. 68 (1993) 281–283.

[41] K. Nellaiappan, G. Nicklas, M. Sugumaran, Detection of dopachrome isomerase activity on gels, Anal. Biochem. 220 (1994) 122–128. [42] G. Nicklas, M. Sugumaran, Detection of dopachrome isomerase on gels and membranes, Anal. Biochem. 230 (1995) 248–253. [43] K. Valivittan, A method for detecting dopaminechrome isomerase activity on gels, Biotech. Histochem. 74 (1999) 213–218. [44] M.L. Schmidt, J.Q. Trojanowski, Enzymatic detection of native and derivatized horseradish peroxidase in sodium dodecyl sulphate polyacrylamide gels, Anal. Biochem. 155 (1986) 371–375. [45] C.-W. Kho, et al., Activity staining of glutathione peroxidase after two-dimensional gel electrophoresis, Mol. Cells 18 (2004) 369–373. [46] Y. Shimazaki, Y. Sugawara, Y. Ohtsuka, T. Manabe, Analysis of the activity and identification of enzymes after separation of cytosol proteins in mouse liver by microscale nondenaturing two-dimensional electrophoresis, Proteomics 3 (2003) 2002–2007. [47] Y. Shimazaki, Y. Hiraka, Y. Uesugi, T. Manabe, Simultaneous analysis of esterase and transferase activities in cytosol proteins from the bovine retina by using microscale non-denaturing two-dimensional electrophoresis, Biochim. Biophys. Acta. 1696 (2004) 51–57. [48] M. Thiele, J. Bernhagen, Link between macrophage migration inhibitory factor and cellular redox regulation, Antioxid. Redox. Signal 7 (2005) 1234–1248. [49] G. Odh, A. Hindemith, A.M. Rosengren, E. Rosengren, H. Rorsman, Isolation of a new tautomerase monitored by the conversion of D-dopachrome to 5,6dihydroxyindole, Biochem. Biophys. Res. Commun. 197 (1993) 619–624. [50] E. Rosengren, E. Bucala, P. Aman, L. Jacobsson, G. Odh, C.N. Metz, H. Rorsman, The immunoregulatory mediator macrophage inhibitory factor (MIF) catalyzes a tautomerization reaction, Mol. Med. 1 (1996) 143–149. [51] J. Matsunaga, D. Sinha, F. Solano, C. Santis, G. Wistow, V. Hearing, Enzyme activity of macrophage migration inhibitory factor toward oxidized catecholamines, J. Biochem. 274 (1999) 3268–3271. [52] J. Matsunaga, D. Sinha, F. Solano, C. Santis, G. Wistow, V. Hearing, Macrophage migration inhibitory factor (MIF)—its role in catecholamine metabolism, Cell. Mol. Biol. 45 (1999) 1035–1040. [53] H.A. Lashuel, B. Aljabari, E.M. Sigurdsson, C.N. Metz, L. Leng, D.J.E. Callaway, R. Bucala, Amyloid fibril formation by macrophage inhibitory factor, Biochem. Biophys. Res. Commun. 338 (2005) 973–980. [54] E.H. Norris, L.B.I. Giasson, R. Hodara, S. Xu, J.Q. Trojanowski, H. Ischiropoulos, V.M.-Y. Lee, Reversible inhibition of α-synuclein fibrillization by dopaminochrome-mediated conformational alterations, J. Biol. Chem. 280 (2005) 21212–21219. [55] S.G. Rhee, H.Z. Chae, K. Kim, Peroxiredoxins: a historical overview and speculative preview of novel mechanism and emerging concepts in cell signalling, Free Rad. Biol. Med. 12 (2005) 1543–1552. [56] H. Jung, T. Kim, H.Z. Chae, K.-T. Kim, H. Ha, Regulation of macrophage migration inhibitory factor and thiol-specific antioxidant protein PAG by direct interaction, J. Biol. Chem. 276 (2001) 15504–15510. [57] J.R. Connor, S.L. Menzies, J.R. Burdo, P.J. Boyer, Iron and iron management proteins in neurobiology, Pediatr. Neurol. 23 (2001) 118–129. [58] L.K. Double, et al., Iron-binding characteristics of neuromelanin of the human substantia nigra, Biochem. Pharmacol. 66 (2003) 489–494. [59] Q. Chen, J.L. Beard, B.C. Jones, Abnormal brain monoamine metabolism in iron deficiency anemia, J. Nutr. Biochem. 6 (1995) 486–493. [60] L. Zecca, M.B.H. Youdim, P. Riederer, J.R. Connor, R.R. Crichton, Iron, brain aging and neurodegenerative disorders, Nat. Rev. 5 (2004) 863–873. [61] K.L. Doulble, M. Gerlack, M.H. Youdim, P. Riederer, Impaired iron homeostasis in Parkinson's disease, J. Neural Transm. 60 (2000) 37–58. [62] J. Galazka-Friedman, E.R. Bauminger, D. Koziorowski, A. Friedman, Mossbauer spectroscopy and ELISA studies reveal differences between Parkinson's disease and control substantia nigra, Biochim. Biophys. Acta. 1668 (2004) 130–136. [63] M.P. Smith, W.A. Cass, GDNF reduces oxidative stress in 6-hydroxydopamine model of Parkinson's disease, Neurosci. Lett. 412 (2007) 259–263. [64] M. Basso, S. Giraudo, D. Corpillo, B. Bergamasco, L. Lopiano, M. Fasano, Proteome analysis of human substantia nigra in Parkinson's disease, Proteomics 4 (2004) 3943–3952. [65] A. De Iuliis, J. Grigoletto, A. Recchia, P. Giusti, P. Arslan, A proteomic approach in the study of an animal model of Parkinson's disease, Clin. Chim. Acta. 357 (2005) 202–209. [66] A. Castegna, M. Aksenov, V. Thongbookerd, J.B. Klein, W.M. Pierce, R. Booze, et al., Proteomic identification of oxidatively modified proteins in Alzheimer's disease brain. Part II: dihydropyrimidinase-related protein 2, alpha-enolase and heat shock cognate 71, J. Neurochem. 82 (2002) 1524–1532. [67] M.P. Mattson, W.A. Pedersen, D. Wenzhen, C. Culmsee, S. Camandola, Cellular and molecular mechanism underlying perturbed energy metabolism and neuronal degeneration in Alzheimer's and Parkinson's diseases, Ann. N.Y. Sci. 893 (1999) 154–175. [68] G. Berding, P. Odin, D.J. Brooks, G. Nikkhah, C. Matthies, T. Peschel, et al., Resting regional cerebral glucose metabolism in advanced Parkinson's disease studied in the off and on conditions with [(18)F]FDG-PET, Mov. Disord. 16 (2000) 1014–1022. [69] C.J. Jeffery, Moonlighting proteins: old proteins learning new tricks, Trends Genet. Sci. 19 (2003) 415–417. [70] R. Canete-Soler, K.S. Reddy, D.R. Tolan, J. Zhai, Aldolases A and C are ribonucleolytic components of a neuronal complex that regulates the stability of the lightneurofilament mRNA, J. Neurosci. 25 (2005) 4353–4364. [71] M.A. Sirover, New insights into an old protein: the functional diversity of mammalian glyceraldehyde-3-phosphate dehydrogenase, Biochim. Biophys. Acta. 1432 (1999) 159–184.

A. De Iuliis et al. / Biochimica et Biophysica Acta 1784 (2008) 1687–1693 [72] J. Mazzola, M.A. Sirover, Alteration of intracellular structure and function of glyceraldehyde-3-phosphate dehydrogenase: a common phenotype of neurodegenerative disorders? Neurotoxicology 23 (2002) 603–609. [73] Y. Hod, S.N. Pentyala, T.C. Whyard, M.R. El-Maghrabi, Identification and characterization of a novel protein that regulates RNA–protein interaction, J. Cell. Biol. 72 (1999) 435–444. [74] S.J. Lee, S.J. Kim, J. Ko, C.S. Jeong, et al., Crystal structures of human DJ-1 and Escherichia coli Hsp31, which share an evolutionarily conserved domain, Proc. Nat. Acad. Sci. 100 (2003) 9256–9261. [75] T. Taira, Y. Saito, E. Niki, S.M. Iguchi-Ariga, K. Takahashi, H. Ariga, DJ-1 has a role in antioxidative stress to prevent cell death, EMBO Rep. 5 (2004) 213–218. [76] N. Zhong, C.Y. Kim, P. Rizzu, C. Geula, D.R. Porter, E.N.F. Pothos, F. Squitieri, P. Heuntik, J. Xu, DJ-1 transcriptionally up-regulates the human tyrosine hydro-

[77]

[78] [79] [80] [81] [82]

1693

xylase by inhibiting the sumoylation of pyrimidine tract-binding proteinassociated splicing factor, J. Biol. Chem. 281 (2006) 20940–20948. J. Xu, et al., The Parkinson's disease-associated DJ-1 protein is a transcriptional coactivator that protects against neuronal apoptosis, Hum. Mol. Genet. 14 (2005) 1231–1241. V. Bonifati, P. Rizzo, M.J. van Baren, et al., Mutations in the DJ-1 gene associated with autosomal recessive early-onset parkinsonism, Science 299 (2003) 256–259. J. Smythies, L. Galzigna, The oxidative metabolism of catecholamines in the brain: a review, Biochim. Biophys. Acta 1380 (1998) 159–162. A. Wood-Kacmar, S. Gandhi, N.W. Wood, Understanding the molecular causes of Parkinson's disease, Trends in Mol. Med. 12 (2006) 521–528. M.R. Cookson, W. Dauer, T. Dawson, E.A. Fon, M. Guo, J. Shen, The roles of kinases in familial Parkinson's disease, J. Neurosci. 27 (2007) 11865–11868. B. Thomas, F.M. Beal, Parkinson's disease, Hum. Mol. Gen. 16 (2007) R183–R194.