Calabrese 20et 20al 20 2007 20J 20Neurochem 20101 20709 717

Document Sample
Calabrese 20et 20al 20 2007 20J 20Neurochem 20101 20709 717 Powered By Docstoc
					Journal of Neurochemistry, 2007, 101, 709–717                                                            doi:10.1111/j.1471-4159.2006.04367.x

In vivo induction of heat shock proteins in the substantia nigra
following L-DOPA administration is associated with increased
activity of mitochondrial complex I and nitrosative stress in rats:
regulation by glutathione redox state

Vittorio Calabrese,* Cesare Mancuso,  Agrippino Ravagna,* Marzia Perluigi,à Chiara Cini,à
Carlo De Marco,à D. Allan Butterfield§ and Anna Maria Giuffrida Stella*
*Department of Chemistry, Biochemistry and Molecular Biology Section, Faculty of Medicine, University of Catania, Catania, Italy
 Institute of Pharmacology, Catholic University School of Medicine, Rome, Italy
àDepartment of Biochemical Sciences, University of Rome ‘La Sapienza,’ Rome, Italy
§Department of Chemistry, Center of Membrane Sciences, and Sander-Brown Center on Aging, University of Kentucky, Lexington,
Kentucky, USA

Abstract                                                                 chondrial complex I whereas no significant changes were
Increasing evidence suggests a critical role for oxidative and           found in the activity of complex IV. In the same experimental
nitrosative stress in the pathogenesis of most important neu-            conditions, a significant decrease in reduced glutathione was
rodegenerative disorders. Parkinson’s disease (PD) is a                  observed, which was associated with an increased content of
neurodegenerative disease characterized by a severe deple-               oxidized glutathione content as well as nitric oxide (NO) syn-
tion in number of dopaminergic cells of the substantia nigra             thase activity, NO metabolites and nitrotyrosine immunore-
(SN). Administration of L-DOPA (LD) is the more effective                activity. Interestingly, Hsp70 induction, iNOS up-regulation
treatment for patients with PD. However, the vast majority of            and nitrotyrosine formation have been confirmed also in SN
patients suffer LD-related complications, which represent the            and striatum of rats treated with LD and carbidopa, this latter
major problem in the clinical management of PD. In the pre-              being an inhibitor of the peripheral DOPA decarboxylase. Our
sent study, LD administration to rats resulted in a significant           data are in favor of the importance of the heat shock signal
dose-dependent increase in Hsp70 synthesis which was                     pathway as a basic mechanism of defense against neuro-
specific for the SN. The amount of 70 kDa protein increased               toxicity elicited by free radical oxygen and nitrogen species
after 6 h treatment reaching the maximal induction after 24–             produced in aging and neurodegenerative disorders.
48 h. Induction of Hsp70 in the SN was associated with a                 Keywords: glutathione, heat shock protein, levodopa, mito-
significant increase in constitutive Hsc70 and mitochondrial              chondria, nitrosative stress, substantia nigra.
Hsp60 stress proteins, and with increased expression of mito-            J. Neurochem. (2007) 101, 709–717.

Parkinson’s disease (PD) is a neurodegenerative disorder                    Address correspondence and reprint requests to Vittorio Calabrese,
characterized by a severe depletion in number of dopamin-                Department of Chemistry, Biochemistry Molecular Biology Section,
                                                                         Faculty of Medicine, University of Catania, Catania, Italy.
ergic cells of the substantia nigra (SN). As an effect of this
reduction in dopaminergic neurons, a significant fall in brain               Abbreviations used: BBB, blood–brain barrier; DA, dopamine;
dopamine (DA) levels occurs (Hald and Lotharius 2005).                   DOPAC, 3,4-dihydroxyphenylacetic acid; GSH, reduced glutathione;
Although several hypotheses have been raised, including (i)              GSSG, oxidized glutathione; HPLC/ED, high-performance liquid chro-
                                                                         matography/electrochemical detection; Hsc, heat shock cognate (con-
                                                                         stitutive form); Hsp, heat shock protein; i.p., intraperitoneal; LD,
                                                                         L-DOPA; MAO-A, mono-amino-oxidase type A; NOS, nitric oxide
Received October 2, 2006; revised manuscript received October 2, 2006;   synthase; PD, Parkinson’s disease; RNS, reactive nitrogen species; ROS,
accepted November 3, 2006.                                               reactive oxygen species; SN, substantia nigra; St, striatum.

Ó 2007 The Authors
Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717                                      709
710 V. Calabrese et al.

defective DNA repair mechanisms, (ii) specific genetic                   A common pathway through which tissues counteract
defects, (iii) mitochondrial dysfunction, or (iv) toxic com-         either oxidative or nitrosative stress is the induction of the
pounds in the environment, none of these, alone, completely          heat shock protein (Hsp) system, including Hsp32 (known as
explains the cascade of events responsible for the onset and         heme oxygenase-1), Hsp60, and Hsp70 (Calabrese et al.
progression of PD (Calabrese et al. 2000a, 2006; Hald and            2006). In mammalian cells, Hsps synthesis is induced not
Lotharius 2005; Poon et al. 2005). A large body of evidence          only by hyperthermia, but also following stimuli which alter
demonstrates that free radicals play a key role in the               the intracellular redox environment, exposure to heavy
pathogenesis of PD. In fact, there is a 10-fold increase in          metals, amino acid analogs or cytotoxic drugs and requires
hydroperoxide levels in SN in PD, and dopaminergic neurons           the activation and translocation to the nucleus of one or more
produce hydrogen peroxide either enzymatically, through the          heat shock transcription factors which control the expression
activity of mono-amine oxidase-A (MAO-A), or non-enzy-               of a specific set of genes (Calabrese et al. 2000a,b, 2002,
matically via the intracellular autoxidation of DA (Dexter           2006).
et al. 1994; Hald and Lotharius 2005). Once formed,                     In view of recent finding indicating a role for Hsps in
hydrogen peroxide by reacting with the reduced form of               protecting brain cells against free radical injury (Sultana
transition metals, such as Fe (II) and Cu(I), gives rise to the      et al. 2005; Abdul et al. 2006; Joshi et al. 2006; Perluigi
powerful oxidant hydroxyl radical and oxidative damage to            et al. 2006), the present study was aimed at investigating,
nigral membrane lipids, proteins, and DNA ensues (Cala-              in vivo, the effect of LD on: the expression of Hsps and
brese et al. 2000a). Reduced glutathione (GSH) significantly          mitochondrial complexes, as well as the GSH status, NOS
contributes to the detoxification of hydroxyl radical, in fact it     activity, NO metabolites, and protein nitration, in different rat
reacts with the free thiol group of GSH which is oxidized to         brain regions.
GSSG (Calabrese et al. 2000a). Unfortunately, SN has very
low levels of GSH compared with other brain areas and this
contributes to the free radicals triggered pathogenesis of PD        Materials and methods
(Perry et al. 1982).
   Because of its inability to cross blood–brain barrier (BBB),      Animals
DA cannot be used as a drug in patients affected by PD.              Male Wistar rats (200 ± 60 g weight) were used for experiments.
Therefore, the administration of L-DOPA (LD), the precursor          Levodopa (50–600 mg/kg body weight) was dissolved in saline and
                                                                     administered intraperitoneally (i.p.). In selected experiments, LD
of DA, is the more efficient line of treatment for patients with
                                                                     50 mg/kg plus carbidopa 12.5 mg/kg were dissolved in saline and
PD. In fact, LD can cross BBB and, once arrived in the brain,
                                                                     given by the same route. At the proper time points, animals were
is converted by DOPA decarboxylase in DA (Ahlskog 2001).             killed, brains quickly removed and dissected into the cerebral cortex,
However, the vast majority of patients treated with LD suffer        hippocampus, septal area, and striatum (St), according to a
drug-related complications, which represent the major prob-          standardized procedure, in a cold anatomical chamber and following
lem in the clinical management of PD. The severity of                a protocol that allows a maximum of 50 s time-variability for each
disease and duration of LD therapy are the major factors             sample across animals. SN was dissected from the deepest part of
determining the incidence of these complications, which              the interpeduncolar fossa. Samples of different brain areas were
include motor fluctuations, dyskinesia, and psychiatric               homogenized for 2 min in 0.05 mol/L Tris–HCl buffer, pH 7.4
problems (Stocchi 2005). Therefore, understanding the                (1 : 9). All the experimental procedures on animals have been
molecular mechanisms by which LD damages SN is                       approved by the Ethical Committee of the University of Catania.
important in designing therapeutic strategies to alleviate
                                                                     Enzyme assays
LD-induced damage (Calabrese et al. 2001).
                                                                     Brain homogenate was centrifuged at 1000 · g for 10 min. The
   Nitrosative stress represents a mechanism by which LD             nuclear pellet was washed once and the supernatants were
can exert its toxic effects. In fact, both LD and DA have been       centrifuged at 20 000 · g for 30 min to obtain the supernatant
shown to increase the expression of the three isoforms of            and the mitochondrial pellet, respectively. The pellet was washed
nitric oxide synthase (NOS) in brain cells and, consequently,        twice and the mitochondria were resuspended in 40 mmol/L Tris–
the synthesis of nitric oxide (NO) (Soliman et al. 2002; Chen        HCl (pH 6.8), 2.5% sodium dodecyl sulfate (SDS). For the
et al. 2003). This latter can react with the reactive oxygen         measurement of mitochondrial enzymatic activities, the pellet was
species (ROS) derived from LD autoxidation (Soliman et al.           homogenized in 0.8 mL of 0.9% NaCl and sonicated for 10 s at 0–
2002; Hald and Lotharius 2005), thus promoting the                   2°C in an ultrasonic disintegrator (power 150 W). Contamination of
formation of reactive nitrogen species (RNS) which are               the mitochondrial extract by cytosol, determined by measurement
                                                                     of lactate dehydrogenase activity was estimated to be less than 10%
highly toxic for brain cells (Boje 2004). In support of this
                                                                     of the measured activity. NADH-CoQ1 reductase (complex I)
theory, it has been reported that the selective inhibition of
                                                                     activity was determined according to Schapira et al.(1990) whereas
nNOS prevents 1-methyl-4-phenyl-1,2,3,6-tetrahydropyri-              cytochrome c oxidase (complex IV) was determined by the method
dine-induced parkinsonism in experimental animals (Beal              of Warton and Tzagoloff (1967).

                                   Ó 2007 The Authors
                                   Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717
                                                                                        L-DOPA and cellular stress response in rat brain 711

Western blot analysis                                                   with 1 mL of buffer 1. Aliquots (865 lL) of the combined eluates
Hsp70, Hsp60, Hsc70, inducible NOS (iNOS), mitochondrial                were added to a cuvette with 250 nmol of DTNB and 0.5 U of
complexes I and IV, and protein nitration were evaluated by western     GSSG reductase. The assay then proceeded as in the measurement
blot analysis. In brief, an equal amount of proteins (40 lg) for each   of total GSH. GSH and GSSG standards in the ranges between 0–
sample was separated by SDS-polyacrylamide gel electrophoresis          10 nmol and 0.010–10 nmol, respectively, added to control samples
(SDS-PAGE), and transferred overnight to nitrocellulose mem-            were used to obtain the relative standard curves, and the results were
branes, and the non-specific binding of antibodies was blocked with      expressed in nmol of GSH or GSSG, respectively, per mg protein.
3% non-fat dried milk in phosphate-buffered saline (PBS). Immu-
nodetection of iNOS and protein nitrotyrosine were performed using      Nitric oxide synthase assay
a polyclonal rabbit anti-iNOS antibody (sc-651, Santa Cruz              Nitric oxide synthase activity assay was performed spectrophoto-
Biotechnology, Santa Cruz, CA, USA; 1 : 500 dilution in PBS, pH         metrically by exploiting the reaction of NO with oxyhemoglobin
7.5) and polyclonal rabbit anti-nitrotyrosine antibody (06-284,         (HbO2) to form methemoglobin, according to Salter and Knowles
Upstate, Charlottesville, VA, USA; 1 : 1000 dilution in PBS, pH         (1998). The reaction mixture contained in a final volume of 1 mL:
7.5), respectively. Immunodetection of Hsp70, Hsc70, and Hsp60          1 mmol/L L-arginine, 1 mmol/L CaCl2, 0.1 mmol/L NADPH,
was performed by using a monoclonal mouse anti-Hsp70 antibody           12 lmol/L THB4, 5 lmol/L HBO2, 4 lmol/L FAD, 100 mmol/L
(SPA-810, Stressgen, Ann Arbor, MI, USA; 1 : 1000 dilution in PBS       HEPES (pH 7.5) and 0.3 mL brain cytosol. The enzyme activity was
pH, 7.5), monoclonal mouse anti-Hsc70 (sc-7298, Santa Cruz,             monitored by absorption spectrophotometry by following the
1 : 1000 dilution in PBS pH, 7.5) and a polyclonal goat anti-Hsp60      controlled oxidation of HBO2 to methemoglobin. The oxidation of
antibody (sc-1052, Santa Cruz, 1 : 1000 dilution in PBS pH, 7.5),       HBO2 to methemoglobin sensitive to L-NMMA (1 mmol/L)
respectively. As far as mitochondrial complexes I and IV are            inhibition and in the presence of 1 lmol/L superoxide dismutase
concerned, monoclonal mouse antibodies against the 39 kDa subunit       and catalase was followed at 411–401 nm in a double-beam
of complex I (A-21344, Molecular Probes, Invitrogen, Carlsbad, CA,      spectrophotometer Perkin-Elmer 559 with a multiple wavelength
USA; 2 lg/mL in PBS pH, 7.5) and 17 kDa subunit of complex IV           program at 22°C. NOS activity was measured in the absence and
(A-21363, Molecular Probes; 0.2 lg/mL in PBS pH, 7.5) have been         presence of: (i) 0.1 mmol/L aminoethyl-isothiourea (ITU), which is
used. A goat polyclonal antibody specific for b-actin was used as a      a specific iNOS inhibitor (Saetre et al. 1998), and (ii) 1 mmol/L
loading control (sc-1615, Santa Cruz; 1 : 1000 dilution in PBS pH,      methyl-L-arginine (L-NMMA), a non-specific NOS inhibitor which
7.5). All blots were then visualized using a horseradish peroxidase-    inhibits all three NOS isoforms (Jansson et al. 1999).
conjugated goat anti-rabbit or anti mouse IgG, followed by enhanced
chemiluminescence (Amersham, Piscataway, NJ, USA). Immunore-            Nitrite assay
active bands were scanned by a laser densitometer (LKB Ultroscan        Nitrite was measured in brain cytosol by the Griess reaction after the
XL, Pharmacia, GE Healthcare, Little Chalfont, UK).                     reduction of nitrate by nitrate reductase (Granger et al. 1996).

Reduced and oxidized glutathione assay                                  L-DOPA and metabolites assay
Reduced glutathione and oxidized glutathione (GSSG) were                Rat brain areas were homogenized in ice-cold trichloracetic acid
measured by the NADPH-dependent GSSG reductase method as                0.1 mol/L, containing 0.05 mmol/L of ascorbic acid for about 20 s.
previously reported (Calabrese et al. 2000b). Rat brain areas were      After centrifugation (5000 · g for 5 min), the supernatants were
homogenized on ice for 10 s in 100 mmol/L potassium phosphate,          filtered through 0.2-lm cellulose membranes and supernatants
pH 7.5, which contained 12 mmol/L disodium EDTA. For total              injected onto a high-performance liquid chromatography/electro-
glutathione, aliquots (0.1 mL) of homogenates were immediately          chemical detection (HPLC/ED) system. Levels of DA and its
added to 0.1 mL of a cold solution containing 10 mmol/L DTNB            metabolite 3,4-dihydroxyphenylacetic acid (DOPAC) were assayed
and 5 mmol/L EDTA in 100 mmol/L potassium phosphate, pH 7.5.            by HPLC/ED as described by Wagner et al. (1982).
The samples were then mixed by tilting and centrifuged at
12000 · g for 2 min at 4°C. An aliquot (50 lL) of the supernatant       Statistical analysis
was added to a cuvette containing 0.5 U of GSSG reductase in            Results were expressed as means ± SEM of 6 separate experiments.
100 mmol/L potassium phosphate and 5 mmol/L EDTA, pH 7.5                Data were analyzed by one-way ANOVA, followed by Duncan’s new
(buffer 1). After 1 min of equilibration, the reaction was initiated    multiple-range test. Differences were considered significant at
with 220 nmol of NADPH in buffer 1 for a final reaction volume of        p < 0.05.
1 mL. The formation of a GSH-DTNB conjugate was then
measured at 412 nm. The reference cuvette contained equal
concentrations of DTNB, NADPH, and enzyme, but not sample.              Results
For assay of GSSG, aliquots (0.5 mL) of homogenates were                Rats treated with LD 50–600 mg/kg i.p. exhibited a signi-
immediately added to 0.5 mL of a solution containing 10 mmol/L
                                                                        ficant dose-dependent increase of Hsp70 synthesis which was
N-ethylmaleimide (NEM) and 5 mmol/L EDTA in 100 mmol/L
                                                                        specific for the SN (Fig. 1a). This increase became statisti-
potassium phosphate, pH 7.5. The sample was mixed by tilting and
centrifuged at 12000 · g for 2 min at 4°C. An aliquot (500 lL) of
                                                                        cally significant at the dose of 400 mg/kg. Therefore, this last
the supernatant was passed at one drop/s through a SEP-PAK C18          dose has been used for further experiments. Interestingly, this
Column (Waters, Framingham, MA, USA) that had been washed               dose of LD is in the same order of magnitude than that used
with methanol followed by water. The column was then washed             in other articles in which the pathogenetic mechanisms

Ó 2007 The Authors
Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717
712 V. Calabrese et al.

Fig. 1 Hsp70, Hsc70, and Hsp60 protein expression in substatntia          kg i.p.). (d) Time-course studies of Hsp60 expression in both St (white
nigra (SN) and striatum (St) after L-DOPA administration to rat. Hsp70,   columns) and SN (black columns) of LD (400 mg/kg i.p.)-treated rats.
Hsc70, and Hsp60 protein levels have been measured by western blot        (a), (c) and (d) bar graphs and (b) line graph represent the densito-
using specific antibodies as described in ‘Materials and methods’. (a)     metric evaluation of specific bands after normalization with the densi-
Dose–response curve of Hsp70 expression in rat SN 24 h after the          tometric values of b-actin, used as a loading control. Values are
administration of LD (50–600 mg/kg i.p.). (b, c) Time-course studies of   expressed as mean ± SEM of six separate experiments. (a–d) repre-
both Hsp70 and Hsc70 expressions in rat SN treated with LD (400 mg/       sentative immunoblots are shown. *p < 0.01 versus control or time 0.

underlying acute LD toxicity have been addressed (Camp                    specifically in the SN, without significant changes either in
et al. 2000; Ishida et al. 2000). Finally, it is noteworthy that          the St or in other brain regions examined (Fig. 1d and data
the dose of 400 mg/kg, corrected for body weight, surface                 not shown). Notably, time course experiments showed that
area, and concomitant carbidopa administration is very close              LD 400 mg/kg i.p. enhanced nigral mitochondrial complex I
to the maximum dosage commonly used in PD patients                        expression and activity (Fig. 3). A significant increase at 6 h,
(1000–1200 mg/day) (Koller et al. 1994; Aminoff 2001). As                 with maximal values at 24 h, compared with control animals,
shown in Fig. 1b, time course-studies revealed that Hsp70                 in absence of concomitant changes in the expression or
induction was an early event appearing as early as 6 h of                 activity of complex IV has been found (Fig. 3). In the same
treatment with LD 400 mg/kg i.p. and remained still                       experimental conditions, a significant decrease in GSH was
significantly higher than control after 48 h. Induction of                 observed in the SN but not in other brain regions examined
Hsp70 in the SN was associated with a significant increase in              and this was associated with increased GSSG content
both constitutive Hsc70 (Fig. 1c) and mitochondrial Hsp60                 (Fig. 4).
stress proteins (Fig. 1d).                                                   To verify that the above-mentioned results were related to
   Induction of Hsps in the SN was paralleled by a significant             DA, converted by the brain starting from LD, we measured
increase in iNOS expression. As shown in Fig. 2a, LD                      both DA and DOPAC (one of its metabolite) in SN and St.
400 mg/kg i.p. caused a significant induction of iNOS; the                 As shown in Fig. 5, 6 h after the administration of LD
amount of 130 kDa protein increased significantly in the SN                400 mg/kg i.p., DA was the major product assayed in brain
after 6 h of treatment, with maximal expression after 24 h.               areas whereas DOPAC has been found to a lesser extent.
At this time point, we observed, in the same brain region, a              Interestingly, the conversion of LD in DA was significantly
significant increase in nitrotyrosine immunoreactivity                     higher in SN than in St, thus confirming the primary role of
(Fig. 2b) which was associated with maximal NOS activity                  DA in this brain area.
and total nitrite production (Figs 2c and d).                                Similar results on nitrosative stress induction and heat
   The interesting finding emerging from the present study is              shock response have been obtained by treating rats with LD
that the above-described LD-induced changes occurred                      50 mg/kg plus carbidopa 12.5 mg/kg i.p. This dose regimen

                                       Ó 2007 The Authors
                                       Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717
                                                                                           L-DOPA and cellular stress response in rat brain 713

Fig 2 Nitrosative stress induction in the substantia nigra (SN) after L-   24 h after LD treatment (as above) by western blot using a specific
DOPA (LD) administration to rats. Rats were treated with LD (400 mg/       antibody. (c) NOS activity and (d) total nitrite levels have been
kg i.p.). (a) iNOS protein levels have been measured 6 and 24 h after      measured 6 and 24 h after treatment. For further information on the
treatment by western blot using a specific antibody. The bar graph          analytical techniques, see ‘Materials and methods’. (a), (b), and (d)
represents the densitometric evaluation of the bands of the immuno-        data are expressed as mean ± SEM of six separate experiments. In
blot put below after normalization with the densitometric values of b-     (a) and (b), representative immunoblots are shown. *p < 0.05 and
actin, used as a loading control. (b) Protein nitration was measured       **p < 0.01 versus control. C, control; LD, L-DOPA.

Fig. 3 Complexes I and IV activities and protein levels by L-DOPA in
the substantia nigra (SN) and striatum (St) of rats. Rats were treated
with LD (400 mg/kg i.p.). (a) Specific activity of complexes I and IV
were assayed 6 and 24 h after LD treatment in both SN and St.
Complexes I [(b) and (d)] and IV [(c) and (e)] protein levels were         Fig. 4 L-DOPA-induced changes in the reduced glutathione (GSH)
measured in SN and St 6 and 24 h after LD treatment by western blot        and GSSG levels of rat brain areas. Rats were treated with LD (400 mg/
using a specific antibody as described in ‘Materials and methods’. In       kg i.p.). (a) GSH and (b) GSSG concentrations were measured in rat
(b)–(e), representative immunoblots are shown. Data in (a) are ex-         brain areas 6 and 24 h after LD treatment as described in ‘Materials and
pressed as the specific activity of each complex, mean ± SEM of six         methods’. Data are expressed as mean ± SEM of six separate
separate experiments. *p < 0.05 and **p < 0.01 versus control. C,          experiments. *p < 0.01 versus control. Cb, cerebellum; Hp, hippo-
control; LD, L-DOPA.                                                       campus; LD, L-DOPA; SN, substantia nigra; Sp, septum; St, striatum.

Ó 2007 The Authors
Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717
714 V. Calabrese et al.

                                                                        et al. 2005). Post-mortem analysis demonstrated a twofold
                                                                        increase in protein carbonyl levels, which are indicative of
                                                                        protein oxidation, in SN of PD patients compared with
                                                                        normal subjects (Floor and Wetzel 1998). Furthermore,
                                                                        levels of 4-hydroxy-2,3-nonenal, an aldehyde generated
                                                                        during lipid peroxidation and 8-hydroxyguanosine, a
                                                                        nucleoside oxidation product, are increased six- and
                                                                        16-fold, respectively, in SN of PD patients compared with
                                                                        control (Yoritaka et al. 1996; Zhang et al. 1999). Further-
                                                                        more, in vitro studies demonstrated a significant reduction
Fig. 5 L-DOPA metabolism in rat substantia nigra (SN) and striatum
after L-DOPA treatment. Rats were treated with LD (400 mg/kg i.p.).
                                                                        in GSH together with an increase in GSSG levels thus
Dopamine and DOPAC concentrations were measured in SN and St            reflecting formation of ROS originating from DA oxidation
6 h after LD treatment as described in ‘Materials and methods’. Data    (Hald and Lotharius 2005).
are expressed as mean ± SEM of six separate experiments. *p < 0.01         The more efficient therapy for PD is represented by LD.
versus control. C, control; DA, dopamine; DOPAC, 3,4-dihydrox-          Unfortunately, in many cases, LD treatment has to be
yphenylacetic acid; LD, L-DOPA.                                         suspended because of its toxic effects, mainly motor
                                                                        fluctuations, dyskinesia, and psychiatric problems (Stocchi
resembles the one currently used for humans, in particular at           2005). Currently, the molecular events which underlie these
the initial stage of PD (500 mg LD/day) (Aminoff 2001).                toxic effects are poorly understood. It is interesting to note
As shown in Fig. 6, LD in combination with carbidopa                    as LD-derived DA further increases the level of oxidative
significantly increased Hsp70 expression not only in rat SN              stress in brain by at least two mechanisms. The first one is
but also in St, and this effect was significant after 6 h and still      related to the enzymatic transformation of DA into DOPAC
maintained after 24 h from treatment. With regard to the                by MAO-A with formation of hydrogen peroxide whereas
nitrosative stress induction, LD 50 mg/kg plus carbidopa                the second one is due to the autoxidation of DA in
12.5 mg/kg i.p. increased iNOS expression and nitrotyrosine             dopamine-quinone (Hald and Lotharius 2005). As a conse-
immunoreactivity in both SN and St following the same                   quence, ROS increase and significant changes in GSH/
time-course of the Hsp70 expression.                                    GSSG ratio derive. Our in vivo results corroborated this
                                                                        hypothesis on the depletion of GSH following LD admin-
                                                                        istration. Consistent with this notion, we demonstrated that
                                                                        reduction of GSH levels after LD is specific for SN,
Oxidative stress has been implicated in the pathogenesis of             whereas in other brain area this depletion did not occur. In
PD, a neurodegenerative disorder whose primary neurolog-                addition, our data demonstrated that the transformation of
ical feature is the massive degeneration of SN dopaminergic             LD in DOPAC occur through MAO-A, thus implying the
neurons (Jenner 2003; Hald and Lotharius 2005; Moore                    generation of superoxide anion/hydrogen peroxide and the

                                                                                                  Fig. 6 Hsp70 and nitrosative stress
                                                                                                  markers in substantia nigra (SN) and stria-
                                                                                                  tum (St) of rats treated with L-DOPA plus
                                                                                                  carbidopa. Rats were treated with LD
                                                                                                  (50 mg/kg i.p.) plus carbidopa (12.5 mg/kg
                                                                                                  i.p.). Hsp70, iNOS and nitrotyrosine levels
                                                                                                  were detected in both SN [(b) and (d) upper
                                                                                                  panels and (e)] and St [(b) and (d) lower
                                                                                                  panels and (f)] 6 and 24 h after treatment.
                                                                                                  (a) and (c) The bar graphs represent the
                                                                                                  densitometric evaluation of the bands of the
                                                                                                  immunoblots put in (b) and (d), respectively,
                                                                                                  after normalization with the densitometric
                                                                                                  values of b-actin, used as a loading control.
                                                                                                  Black columns, SN; white columns, St. Data
                                                                                                  are expressed as mean ± SEM of six
                                                                                                  separate experiments. *p < 0.01 versus
                                                                                                  control. C, control; LD, L-DOPA; 3-NT,

                                      Ó 2007 The Authors
                                      Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717
                                                                                        L-DOPA and cellular stress response in rat brain 715

consequent increase in the pro-oxidant status of the cell.              derived from DA autoxidation (Soliman et al. 2002), and
The relationship between increased ROS levels and GSH                   forms the toxic species peroxynitrite, whose toxicity is
depletion is of special importance considering that nigral              further increased by the reduction in GSH levels (Drake et al.
cells also contain neuromelanin, a pigment substance                    2003; Muyderman et al. 2004). Our results which show a
related to lipofuscin and derived from DA. Neuromelanin                 significative increase in iNOS induction and NO formation
has been demonstrated to have high affinity for Fe(III), and             along with a sustained condition of nitrosative stress as a
this iron–melanin interaction might have pathogenetic                   result of LD treatment, as demonstrated by the formation of
implications. In fact, the synthesis of neuromelanin from               nitrotyrosine, are in good agreement with this theory and
DA is known to produce more oxidative damage than its                   provide a strong in vivo evidence about the ability of LD per
synthesis from other catecolamines (Li and Dryhurst 1997)               se to trigger nitrosative conditions. The mechanism by which
and, in addition, neuromelanins polymerize from pheomel-                LD-derived DA can induce the three isoforms of NOS in
anin in a process that requires cysteine for synthesis, thus            brain cells are still elusive (Chen et al. 2003). A possible
competing with c-glutamyl cysteine synthetase which                     mechanism regards the interaction of DA with transcription
utilizes cysteine for GSH synthesis (Calabrese et al.                   factors such as nuclear factor-kappaB (NF-jB). It has been
2000a). Under these circumstances, the GSH system in                    demonstrated that DA increased both the phosphorylation of
the SN could be in a position of increased demand and                   the IkappaB protein, the inhibitory subunit of NF-jB in the
decreased synthetic capability, and hence contribute to the             cytoplasm and the nuclear translocation of NF-jB in PC12
high vulnerability of this region to peroxidative injury                cells (Panet et al. 2001; Shin et al. 2004; Wang et al. 2005),
(Spencer et al. 1994). Taken together, these data corroborate           probably by interacting with the D1 and D2 receptors (Chen
the hypothesis of an increased oxidative stress condition in            et al. 2003; Takeuchi and Fukunaga 2004). Furthermore, the
the SN of LD treated rats.                                              activation of NF-jB has been demonstrated in an experi-
   Consistent with the above-mentioned findings are our                  mental model of PD by using the neurotoxins 6-OH-DOPA
results showing the increased expression of Hsp70, Hsc70,               or tetrahydropapaveroline (Park et al. 2004; Shin et al.
and Hsp60 in SN and St after LD (Figs 1 and 6). Although                2004). As a direct consequence of NF-jB activation, an
the induction of Hsps is not always related to neuropro-                increased expression of iNOS takes place with the triggering
tection (Hansson et al. 2003; Olsson et al. 2004), our                  of nitrosative stress (Dehmer et al. 2004).
results could be interpreted as an early attempt of SN to                  Our data are in favor of the possible importance of the heat
react against LD-induced oxidative stress, in agreement                 shock signal pathway as a basic mechanism of defense
with some studies indicating that overproduction of Hsp70               against neurotoxicity elicited by free radical oxygen and
protects the nervous system against injury (Yenari et al.               nitrogen species produced in aging and neurodegenerative
1999; Calabrese et al. 2005a; Sultana et al. 2005; Joshi                disorders (Calabrese et al. 2005a,b,c; Lodi et al. 2006; Poon
et al. 2006). This increase in oxidative conditions following           et al. 2006). The present study, however, does not exclude
LD administration can explain our finding of an early and                the possibility that, over the long term, overwhelming
long-lasting specific increase both at the protein level and             conditions of increased oxidative and nitrosative stress may
activity of mitochondrial complex I in SN. This result is               insidiously develop and induce progressive degenerative cell
not surprising and is corroborated by previous work in                  death.
which the activation of the heat shock response is
paralleled by an increase in mitochondrial complex I
(Sammut et al. 2001). A possible explanation of this
phenomenon is that the heat shock response in SN may                    This work was supported by grants of Italian Cofin 2000, FIRB
increase mitochondrial complex I thereby protecting integ-              RBNE01ZK8F, and by NIH grants to D.A.B. [AG-05119; AG-
rated mitochondrial and hence protecting SN. Indeed, we                 10836].
cannot exclude that prolonged oxidative stress following
LD treatment can exhaust compensatory mechanisms, and a                 References
shift of complex I from an up-regulation to a down-
                                                                        Abdul H. M., Calabrese V., Calvani M. and Butterfield D. A. (2006)
regulation may occur, thus confirming the experimental
                                                                             Acetyl-L-carnitine-induced up-regulation of heat shock proteins
evidence of a complex I deficiency in PD (Li and Dryhurst                     protects cortical neurons against amyloid-beta peptide 1-42-medi-
1997; Perier et al. 2005).                                                   ated oxidative stress and neurotoxicity: implications for Alzhei-
   As far as nitrosative stress, its importance in the onset and             mer’s disease. J. Neurosci. Res. 84, 398–408.
progression of PD has been hypothesized (Jenner 1998;                   Ahlskog J. E. (2001) Parkinson’s disease: medical and surgical treat-
                                                                             ment. Neurol. Clin. 19, 579–605.
Moore et al. 2005). Recently, it has been shown that LD
                                                                        Aminoff M. J. (2001) Pharmacologic management of parkinsonism and
treatment causes a strong induction of iNOS with a following                 other movement disorders, in Basic and Clinical Pharmacology,
increase in NO production in C6 rat glioma cells (Soliman                    8th Edn (Foltin J., Ransom J., Nogueira I., Davis K. and Katzung
et al. 2002). NO in turn reacts with superoxide anion,

Ó 2007 The Authors
Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717
716 V. Calabrese et al.

     B.G., eds.), pp. 463–477. Lange Medical Books/McGraw-Hill,                 Granger D. L., Taintor R. R., Boockvar K. S. and Hibbs J. B. Jr (1996)
     Columbus, OH, USA.                                                               Measurement of nitrate and nitrite in biological samples using
Beal M. F. (1998) Excitotoxicity and nitric oxide in Parkinson’s disease              nitrate reductase and Griess reaction. Meth. Enzymol. 268, 142–
     pathogenesis. Ann. Neurol. 44, S110–S114.                                        151.
Boje K. M. (2004) Nitric oxide neurotoxicity in neurodegenerative               Hald H. and Lotharius J. (2005) Oxidative stress and inflammation in
     diseases. Front. Biosci. 9, 763–776.                                             Parkinson’s disease: is there a causal link? Exp. Neurol. 193, 279–
Calabrese V., Bates T. E. and Giuffrida Stella A. M. (2000a) NO syn-                  290.
     thase and NO dependent signal pathways in brain aging and neu-             Hansson O., Nylandsted J., Castilho R. F., Leist M., Jaattela M. and
     rodegenerative disorders: the role of oxidant/antioxidant balance.               Brundin P. (2003) Overexpression of heat shock protein 70 in R6/2
     Neurochem. Res. 25, 1315–1341.                                                   Huntington’s disease mice has only modest effects on disease
Calabrese V., Testa G., Ravagna A., Bates T. E. and Stella A. M. (2000b)              progression. Brain Res. 970, 47–57.
     HSP70 induction in the brain following ethanol administration in           Ishida Y., Hashiguchi H., Todaka K., Ishizuka Y. and Mitsuyama Y.
     the rat: regulation by glutathione redox state. Biochem. Biophys.                (2000) Repeated administration of high dose levodopa enhances
     Res. Commun. 269, 397–400.                                                       hydroxyl radical production in the rat striatum denervated with 6-
Calabrese V., Scapagnini G., Giuffrida Stella A. M., Bates T. E. and                  hydroxydopamine. Neurosci. Lett. 290, 33–36.
     Clark J. B. (2001) Mitochondrial involvement in brain function             Jansson A., Mazel T., Andbjer B., Rosen L., Guidolin D., Zoli M.,
     and dysfunction: relevance to aging, neurodegenerative disorders                 Sykova E., Agnati L. F. and Fuxe K. (1999) Effects of nitric oxide
     and longevity. Neurochem. Res. 26, 739–764.                                      inhibition on the spread of biotinylated dextran and on extracellular
Calabrese V., Scapagnini G., Ravagna A., Fariello R. G., Giuffrida Stella             space parameters in the neostriatum of the male rat. Neuroscience
     A. M. and Abraham N. G. (2002) Regional distribution of heme                     91, 69–80.
     oxygenase, HSP70, and glutathione in brain: relevance for                  Jenner P. (1998) Oxidative mechanisms in nigral cell death in Parkin-
     endogenous oxidant/antioxidant balance and stress tolerance.                     son’s disease. Mov. Disord. 13, 24–34.
     J. Neurosci. Res. 68, 65–75.                                               Jenner P. (2003) Oxidative stress in Parkinson’s disease. Ann. Neurol. 53,
Calabrese V., Colombrita C., Guagliano E. et al. (2005a) Protective ef-               S26–S36.
     fect of carnosine during nitrosative stress in astroglial cell cultures.   Joshi G., Perluigi M., Sultana R., Agrippino R., Calabrese V. and But-
     Neurochem. Res. 30, 797–807.                                                     terfield D. A. (2006) In vivo protection of synaptosomes by ferulic
Calabrese V., Ravagna A., Colombrita C., Guagliano E., Scapagnini G.,                 acid ethyl ester (FAEE) from oxidative stress mediated by 2,2-
     Calvani M., Butterfield D. A. and Giuffrida Stella A. M. (2005b)                  azobis(2-amidino-propane)dihydrochloride (AAPH) or Fe(2+)/
     Acetylcarnitine induces heme oxygenase in rat astrocytes and                     H(2)O(2): insight into mechanisms of neuroprotection and rele-
     protects against oxidative stress: involvement of the transcription              vance to oxidative stress-related neurodegenerative disorders.
     factor Nrf2. J. Neurosci. Res. 79, 509–521.                                      Neurochem. Int. 48, 318–327.
Calabrese V., Ravagna A., Scapagnini G., Catalano C., Pennisi G.,               Koller W. C., Silver D. E. and Lieberman A. (1994) An algorithm for the
     Butterfield D. A. and Giuffrida Stella A. M. (2005c) Oxidative                    management of Parkinson’s disease. Neurology 44, S1–S52.
     stress, mitochondrial dysfunction and cellular stress response in          Li H. and Dryhurst G. (1997) Irreversible inhibition of mitochondrial
     Friedreich’s ataxia. J. Neurol. Sci. 233, 145–162.                               complex I by 7-(2-aminoethyl)-3,4-dihydro-5-hydroxy-2H-1,4-
Calabrese V., Butterfield D. A., Scapagnini G., Stella A. M. and Maines                benzothiazine-3-carboxylic acid (DHBT-1): a putative nigral
     M. D. (2006) Redox regulation of heat shock protein expression by                endotoxin of relevance to Parkinson’s disease. J. Neurochem. 69,
     signaling involving nitric oxide and carbon monoxide: relevance to               1530–1541.
     brain aging, neurodegenerative disorders, and longevity. Antioxid.         Lodi R., Tonon C., Calabrese V. and Schapira A. H. (2006) Friedreich’s
     Redox Signal. 8, 444–477.                                                        ataxia: from disease mechanisms to therapeutic interventions.
Camp D. M., Loeffler D. A. and LeWitt P. A. (2000) L-DOPA does not                     Antioxid. Redox Signal. 8, 438–443.
     enhance hydroxyl radical formation in the nigrostriatal dopamine           Moore D. J., West A. B., Dawson V. L. and Dawson T. M. (2005)
     system of rats with a unilateral 6-hydroxydopamine lesion.                       Molecular pathophysiology of Parkinson’s disease. Annu. Rev.
     J. Neurochem. 74, 1229–1240.                                                     Neurosci. 28, 57–87.
Chen J., Wersinger C. and Sidhu A. (2003) Chronic stimulation of D1             Muyderman H., Nilsson M. and Sims N. R. (2004) Highly selective and
     dopamine receptors in human SK-N-MC neuroblastoma cells                          prolonged depletion of mitochondrial glutathione in astrocytes
     induces nitric-oxide synthase activation and cytotoxicity. J. Biol.              markedly increases sensitivity to peroxynitrite. J. Neurosci. 24,
     Chem. 278, 28 089–28 100.                                                        8019–8028.
Dehmer T., Heneka M. T., Sastre M., Dichgans J. and Schulz J. B.                Olsson T., Hansson O., Nylandsted J., Jaattela M., Smith M. L. and
     (2004) Protection by pioglitazone in the MPTP model of Parkin-                   Wieloch T. (2004) Lack of neuroprotection by heat shock protein
     son’s disease correlates with I kappa B alpha induction and block                70 overexpression in a mouse model of global cerebral ischemia.
     of NF kappa B and iNOS activation. J. Neurochem. 88, 494–501.                    Exp. Brain Res. 154, 442–449.
Dexter D. T., Holley A. E., Flitter W. D., Slater T. F., Wells F. R., Daniel    Panet H., Barzilai A., Daily D., Melamed E. and Offen D. (2001)
     S. E., Lees A. J., Jenner P. and Marsden C. D. (1994) Increased                  Activation of nuclear transcription factor kappa B (NF-kappaB) is
     levels of lipid hydroperoxides in the parkinsonian substantia nigra:             essential for dopamine-induced apoptosis in PC12 cells. J. Neuro-
     an HPLC and ESR study. Mov. Disord. 9, 92–97.                                    chem. 77, 391–398.
Drake J., Sultana R., Aksenova M., Calabrese V. and Butterfield D. A.            Park S. H., Choi W. S., Yoon S. Y., Ahn Y. S. and Oh Y. J. (2004)
     (2003) Elevation of mitochondrial glutathione by gamma-glut-                     Activation of NF-kappaB is involved in 6-hydroxydopamine-but
     amylcysteine ethyl ester protects mitochondria against peroxyni-                 not MPP+ -induced dopaminergic neuronal cell death: its potential
     trite-induced oxidative stress. J. Neurosci. Res 74, 917–927.                    role as a survival determinant. Biochem. Biophys. Res. Commun.
Floor E. and Wetzel M. G. (1998) Increased protein oxidation in human                 322, 727–733.
     substantia nigra pars compacta in comparison with basal ganglia            Perier C., Tieu K., Guegan C. et al. (2005) Complex I deficiency primes
     and prefrontal cortex measured with an improved din-                             Bax-dependent neuronal apoptosis through mitochondrial oxida-
     itrophenylhydrazine assay. J. Neurochem. 70, 268–275.                            tive damage. Proc. Natl Acad. Sci. USA 102, 19 126–19 131.

                                           Ó 2007 The Authors
                                           Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717
                                                                                               L-DOPA and cellular stress response in rat brain 717

Perluigi M., Joshi G., Sultana R., Calabrese V., De Marco C., Coccia R.,      Spencer J. P., Jenner A., Aruoma O. I., Evans P. J., Kaur H., Dexter
      Cini C. and Butterfield D. A. (2006) In vivo protective effects of             D. T., Jenner P., Lees A. J., Marsden D. C. and Halliwell B. (1994)
      ferulic acid ethyl ester against amyloid-beta peptide 1-42-induced            Intense oxidative DNA damage promoted by L-dopa and its me-
      oxidative stress. J. Neurosci. Res. 84, 418–426.                              tabolites. Implications for neurodegenerative disease. FEBS Lett.
Perry T. L., Godin D. V. and Hansen S. (1982) Parkinson’s disease: a                353, 246–250.
      disorder due to nigral glutathione deficiency? Neurosci. Lett. 33,       Stocchi F. (2005) Optimising levodopa therapy for the management of
      305–310.                                                                      Parkinson’s disease. J. Neurol. 252, IV43–IV48.
Poon H. F., Frasier M., Shreve N., Calabrese V., Wolozin B. and Butter-       Sultana R., Ravagna A., Mohmmad-Abdul H., Calabrese V. and But-
      field D. A. (2005) Mitochondrial associated metabolic proteins are             terfield D. A. (2005) Ferulic acid ethyl ester protects neurons
      selectively oxidized in A30P alpha-synuclein transgenic mice – a              against amyloid beta- peptide(1-42)-induced oxidative stress and
      model of familial Parkinson’s disease. Neurobiol. Dis. 18, 492–498.           neurotoxicity: relationship to antioxidant activity. J. Neurochem.
Poon H. F., Calabrese V., Calvani M. and Butterfield D. A. (2006)                    92, 749–758.
      Proteomics analyses of specific protein oxidation and protein            Takeuchi Y. and Fukunaga K. (2004) Different activation of NF-kappaB
      expression in aged rat brain and its modulation by L-acetylcarni-             by stimulation of dopamine D2L and D2S receptors through cal-
      tine: insights into the mechanisms of action of this proposed                 cineurin activation. J. Neurochem. 90, 155–163.
      therapeutic agent for CNS disorders associated with oxidative           Wagner J., Vitali P., Palfreyman M. G., Zraika M. and Huot S. (1982)
      stress. Antioxid. Redox Signal. 8, 381–394.                                   Simultaneous determination of 3,4-dihydroxyphenylalanine, 5-hy-
Saetre T., Gundersen Y., Thiemermann C., Lilleaasen P. and Aasen A. O.              droxytryptophan, dopamine, 4-hydroxy-3-methoxyphenylalanine,
      (1998) Aminoethyl-isothiourea, a selective inhibitor of inducible             norepinephrine, 3,4-dihydroxyphenylacetic acid, homovanillic
      nitric oxide synthase activity, improves liver circulation and oxygen         acid, serotonin, and 5-hydroxyindoleacetic acid in rat cerebrospinal
      metabolism in a porcine model of endotoxemia. Shock 9, 109–115.               fluid and brain by high-performance liquid chromatography with
Salter M. and Knowles R. G. (1998) Assay of NOS activity by the                     electrochemical detection. J. Neurochem. 38, 1241–1254.
      measurement of conversion of oxyhemoglobin to methemoglobin             Wang L. Z., Sun W. C. and Zhu X. Z. (2005) Ethyl pyruvate protects
      by NO. Methods Mol. Biol. 100, 61–65.                                         PC12 cells from dopamine-induced apoptosis. Eur. J. Pharmacol.
Sammut I. A., Jayakumar J., Latif N., Rothery S., Severs N. J., Smo-                508, 57–68.
      lenski R. T., Bates T. E. and Yacoub M. H. (2001) Heat stress           Warton D. C. and Tzagoloff A. (1967) Cytochrome oxidase from bovine
      contributes to the enhancement of cardiac mitochondrial complex               heart mitochondria. Meth. Enzymol. 10, 245–257.
      activity. Am. J. Pathol. 158, 1821–1831.                                Yenari M. A., Giffard R. G., Sapolsky R. M. and Steinberg G. K. (1999)
Schapira A. H., Mann V. M., Cooper J. M., Dexter D., Daniel S. E.,                  The neuroprotective potential of heat shock protein 70 (HSP70).
      Jenner P., Clark J. B. and Marsden C. D. (1990) Anatomic and                  Mol. Med. Today 5, 525–531.
      disease specificity of NADH CoQ1 reductase (complex I) defici-            Yoritaka A., Hattori N., Uchida K., Tanaka M., Stadtman E. R. and
      ency in Parkinson’s disease. J. Neurochem. 55, 2142–2145.                     Mizuno Y. (1996) Immunohistochemical detection of 4-hydroxy-
Shin M. H., Jang J. H. and Surh Y. J. (2004) Potential roles of NF-kappaB           nonenal protein adducts in Parkinson disease. Proc. Natl Acad. Sci.
      and ERK1/2 in cytoprotection against oxidative cell death induced             USA 93, 2696–2701.
      by tetrahydropapaveroline. Free Radic. Biol. Med. 36, 1185–1194.        Zhang J., Perry G., Smith M. A., Robertson D., Olson S. J., Graham
Soliman M. K., Mazzio E. and Soliman K. F. (2002) Levodopa modu-                    D. G. and Montine T. J. (1999) Parkinson’s disease is associated
      lating effects of inducible nitric oxide synthase and reactive oxygen         with oxidative damage to cytoplasmic DNA and RNA in substantia
      species in glioma cells. Life Sci. 72, 185–198.                               nigra neurons. Am. J. Pathol. 154, 1423–1429.

Ó 2007 The Authors
Journal Compilation Ó 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 709–717

G4j0t9rI G4j0t9rI