Address correspondence and reprint requests to David S. Goldstein, Clinical Neurocardiology Section, CNP/DIR/NINDS/NIH, 9000 Rockville Pike, Bldg. 10 Rm. 5N220, Bethesda, MD 20892-1620 USA. E-mail: firstname.lastname@example.org
Parkinson's disease entails profound loss of nigrostriatal dopaminergic terminals, decreased vesicular uptake of intraneuronal catecholamines, and relatively increased putamen tissue concentrations of the toxic dopamine metabolite, 3,4-dihydroxyphenylacetaldehyde (DOPAL). The objective of this study was to test whether vesicular uptake blockade augments endogenous DOPAL production. We also examined whether intracellular DOPAL contributes to apoptosis and, as α-synuclein oligomers may be pathogenetic in Parkinson's disease, oligomerizes α-synuclein. Catechols were assayed in PC12 cells after reserpine to block vesicular uptake, with or without inhibition of enzymes metabolizing DOPAL—daidzein for aldehyde dehydrogenase and AL1576 for aldehyde reductase. Vesicular uptake was quantified by a method based on 6F- or 13C-dopamine incubation; DOPAL toxicity by apoptosis responses to exogenous dopamine, with or without daidzein+AL1576; and DOPAL-induced synuclein oligomerization by synuclein dimer production during DOPA incubation, with or without inhibition of L-aromatic-amino-acid decarboxylase or monoamine oxidase. Reserpine inhibited vesicular uptake by 95–97% and rapidly increased cell DOPAL content (p = 0.0008). Daidzein+AL1576 augmented DOPAL responses to reserpine (p = 0.004). Intracellular DOPAL contributed to dopamine-evoked apoptosis and DOPA-evoked synuclein dimerization. The findings fit with the ‘catecholaldehyde hypothesis,’ according to which decreased vesicular sequestration of cytosolic catecholamines and impaired catecholaldehyde detoxification contribute to the catecholaminergic denervation that characterizes Parkinson's disease.
Parkinson's disease is associated with drastic depletion of dopamine in the corpus striatum (Ehringer and Hornykiewicz 1960) and dopaminergic neuron loss in the substantia nigra (Halliday et al. 1990), as well as decreased norepinephrine contents in several brain areas (Ehringer and Hornykiewicz 1960; Kish et al. 1984; Goldstein et al. 2011). Lewy bodies, a neuropathologic hallmark of Parkinson's disease, contain abundant α-synuclein, and synuclein oligomers are thought to play a key pathogenetic role in Parkinson's disease (Kazantsev and Kolchinsky 2008; Winner et al. 2011).
Mechanisms of loss of catecholaminergic neurons in Parkinson's disease and related disorders and relationships between synucleinopathy and endogenous catecholamines remain incompletely understood. The relatively selective involvement of monoaminergic neurons in Parkinson's disease suggests that the transmitters themselves might be sources of ‘autotoxins.’ Catecholamines leak continuously from storage vesicles into the cytosol and are recaptured by vesicular uptake (Eisenhofer et al. 2004a,b). Buildup of cytosolic dopamine is toxic in neurons lacking vesicles (Chen et al. 2008); elevation of cytosolic dopamine levels reduces survival of cultured midbrain neurons (Mosharov et al. 2009); pharmacologic depletion of dopamine in PC12 cells reduces toxicity exerted by subsequent treatment with the complex-1 inhibitor rotenone combined with the sympathomimetic amine methamphetamine (Dukes et al. 2005); and reduced vesicular storage of dopamine augments methamphetamine-induced toxicity (Fumagalli et al. 1999; Guillot et al. 2008).
A well-studied mechanism of catecholamine toxicity is spontaneous oxidation to form quinones (Graham 1978; Jana et al. 2011). Dopamine also undergoes enzymatic oxidation to form potentially harmful aldehydes (Blaschko 1952). Monoamine oxidase (EC 184.108.40.206) in the outer mitochondrial membrane catalyzes conversion of cytosolic dopamine to 3,4-dihydroxyphenylacetaldehyde (DOPAL). DOPAL is toxic (Panneton et al. 2010) by at least four mechanisms—protein cross-linking (Rees et al. 2009), oxidation to quinones (Anderson et al. 2011), production of hydroxyl radicals (Li et al. 2001), and oligomerization and precipitation of α-synuclein (Burke et al. 2008). Moreover, post-mortem putamen tissue from Parkinson's disease patients contains increased DOPAL relative to dopamine (Goldstein et al. 2011). In the setting of reduced vesicular uptake, impaired aldehyde detoxification might increase cytosolic DOPAL levels further, exacerbating the neurodegeneration.
Few studies have measured endogenous DOPAL, and none has expressed whether blockade of vesicular uptake augments DOPAL generation. Addressing this issue was the main objective of the present experiments. We used PC12 cells as a model of dopaminergic neurons, because PC12 cells synthesize, store, and release dopamine; they contain monoamine oxidase; they express vesicular and cell membrane catecholamine transporters; and they contain DOPAL (Lamensdorf et al. 2000) and two DOPAL metabolites, dihydroxyphenylacetic acid (DOPAC) and dihydroxyphenylethanol (DOPET). As PC12 cells express catechol-O-methyltransferase (EC 220.127.116.11), whereas dopaminergic neurons do not, the cells were pre-incubated for 24 h in medium containing the catechol-O-methyltransferase inhibitor, tolcapone. This cell line normally expresses little α-synuclein. Therefore, to determine whether DOPAL produced intracellularly oligomerizes synuclein, we used PC12 cells that conditionally over-express α-synuclein (Ito et al. 2010).
This study was designed to test whether (i) reserpine-induced blockade of vesicular uptake and the attendant increase in cytosolic dopamine enhance production of deaminated metabolites of endogenous dopamine, including DOPAL; (ii) inhibition of aldehyde dehydrogenase (EC 18.104.22.168) and aldehyde reductase (EC. 22.214.171.124) augments the DOPAL response to vesicular uptake blockade; (iii) reserpine at a dose that blocks vesicular uptake does not affect cell transmembrane uptake of the dopamine analogs, 6-fluorodopamine (Eisenhofer et al. 1989) or 13C-dopamine; (iv) inhibition of aldehyde dehydrogenase and aldehyde reductase augments toxicity of elevated intracellular dopamine; and (v) DOPAL mediates DOPA-induced oligomerization of α-synuclein.
Materials and methods
PC12 cells and the cell culture media, F12K and RPMI1640, were from the American Type Culture Collection; CellBIND culture dishes from Corning (Lowell, MA, USA); tolcapone (to block catechol-O-methyltransferase) from Orion Pharma (Espoo, Finland); dopamine, reserpine (to block vesicular uptake), daidzein (to inhibit aldehyde dehydrogenase), and hydrogen peroxide (30%) from Sigma, St Louis, MO, USA; AL1576 (to inhibit aldehyde reductase) from Alcon Laboratories (Fort Worth, TX, USA); epidermal growth factor from R&D Systems (Minneapolis, MS, USA); and Guava Nexin® reagent (for apoptosis assays) from Millipore (Haywood, CA, USA). 6F-Dopamine was obtained in solution (about 1.35 mM) from the NIH PET Department after having passed quality control for i.v. administration to humans. DOPAL standard was synthesized in the laboratory of and provided by Dr. Kenneth L. Kirk (NIDDK). d4-Dopamine internal standard was from CDN Isotopes (Quebec, Canada). 13C-Dopamine (13C at each of the six positions on the benzene ring) was obtained from Cambridge Isotope Laboratories (Andover, MA, USA). For detection of protein through western blotting, we used mouse anti-α-synuclein antibody (Invitrogen, Camarillo, CA, USA) and goat anti-mouse IRDye 800CW secondary antibody (LI-COR, Lincoln, NE, USA). PC12 cells conditionally over-expressing α-synuclein in a tetracycline-inducible manner (Ito et al. 2010) were kindly provided by Drs. Ito and Nakaso (Department of Neurology, Institute of Neurological Sciences, Faculty of Medicine, Tottori University, Japan). Human recombinant α-synuclein 1–140 (MW 14 460 kDa) was purchased from Calbiochem (La Jolla, CA, USA).
PC12 cell culture
PC12 cells were kept frozen in liquid nitrogen until passaged for experiments. Most experiments involved adherent cells; however, these were unavailable for a time, and some experiments involved non-adherent cells. There were no obvious differences in cell contents of catechols between adherent and non-adherent cell cultures. The PC12 cells were not differentiated.
Adherent cells were grown in F12K medium supplemented with 15% heat-inactivated horse serum, 2.5% fetal bovine serum, and 40 ng/mL epidermal growth factor on CellBIND culture dishes. Non-adherent cells were grown in RPMI1640 with 10% heat-inactivated horse serum and 5% fetal bovine serum. The cells were incubated at 37°C in an atmosphere of 5% carbon dioxide. Media were changed several times per week and passaged once per week.
Prior to plating for the experiments, cells were trypsinized, suspended in media, counted, and the suspensions diluted with medium to provide 200 000 cell/mL. One milliliter of these suspensions was transferred to each well. After 48 h, the medium was removed and replaced with 1 mL of the same medium, but containing 10 μM tolcapone. Experiments began after 24 h of incubation with tolcapone-containing medium (‘baseline’).
Inhibition of vesicular uptake
Effects of reserpine on levels of endogenous catechols in cells and medium were examined after 24 h of exposure to tolcapone, by the addition of 6.0 μL (final concentration 0.6% vol : vol) dimethylsulfoxide vehicle alone or vehicle containing reserpine, to attain a final concentration of 10 μM reserpine in the medium. At 10, 20, 30, 60, 120, 180, or 240 min after addition of reserpine or vehicle, the medium was removed and centrifuged. Adherent cells were scraped into 400 μL of a 20 : 80 mixture of 40 mM phosphoric acid and 200 mM acetic acid. For non-adherent cell cultures, the content of each well was centrifuged, the supernatant removed, and the sedimented cells taken up in 400 μL of the same solution used for adherent cells. The separated medium and the disrupted cell solutions were frozen until assayed for catechol contents, as described below.
For experiments about effects of aldehyde dehydrogenase inhibition by daidzein or aldehyde reductase inhibition by AL1576 on catechol levels and catechol responses to reserpine, combinations of daidzein (final concentration 10 μM), AL1576 (final concentration 1 μM) and dimethylsulfoxide vehicle were added to the medium at the same time as tolcapone. The cells were incubated for 24 h before acute experiments with reserpine. For the acute experiments, daidzein, AL1576, or daidzein+AL1576 were continued in the medium along with tolcapone.
6F-dopamine or 13C-dopamine incubation
For experiments with 6F-dopamine, the medium of cells that had been pre-incubated with tolcapone for 24 h was replaced with medium containing 1–2 μM of 6F-dopamine and 6.0 μL dimethylsulfoxide vehicle, with or without 10 μM reserpine. Cells and medium were separated at 20 or 180 min and frozen until assayed for endogenous and 6F-catechols, as described below. Analogous experiments were performed with 13C-dopamine, except that cells were incubated with reserpine for 10 min before 13C-dopamine was added to the medium.
To estimate the contribution of DOPAL to cytotoxicity induced by increasing cytosolic dopamine, we measured dopamine-induced apoptosis, with or without pre-incubation with daidzein+AL1576. We reasoned that if DOPAL contributed to dopamine-induced apoptosis, then pre-treatment with daidzein+AL1576 should exacerbate the apoptosis, and the extent of increase in apoptosis with versus without daidzein+AL1576 would provide a measure of the contribution of DOPAL. As a positive control, we used serum withdrawal for 24 h, as this is well known to evoke apoptosis.
For measuring apoptosis induced by dopamine, about 2 × 105 cells were plated (12 wells/plate) and incubated for 24 h in F12 medium containing 15% horse serum and 2.5% bovine serum at 37 oC in 5% CO2. The cells were then treated with dopamine (100 μM) for another 24 h in the same medium with 10 μM tolcapone added to block catechol-O-methyltransferase. After incubation, cells were collected and treated with Guava Nexin® Reagent (Millipore, Hayward, CA, USA) for 20 min in the dark. The Guava Nexin® assay detects phosphatidylserine on the external membrane of apoptotic cells. Apoptotic cells were counted by flow cytometry using the Guava device. Results were expressed in terms of apoptotic cells as a function of total cells.
Conditional over-expression of α-synuclein
To over-express α-synuclein, PC12 cells conditionally over-expressing α-synuclein in a tetracycline-inducible manner were cultured for 10 days using medium that did not contain doxycycline. For confirmation of α-synuclein expression, western blot analysis was performed.
Cells over-expressing α-synuclein were exposed to 100 μM L-DOPA and 10 μM copper (II) for 6 h, with or without 10 μM pargyline to inhibit monoamine oxidase or 10 μM carbidopa to inhibit l-aromatic-amino-acid decarboxylase. The cells were then lysed and cytosolic proteins isolated using a Qproteome cell compartment kit (Qiagen, Valencia, CA, USA). Cytosolic protein (500–600 μg) was further purified by immunoprecipitation (Pierce Classic kit, Thermo Scientific, Rockford, IL, USA). α-synuclein aggregation was analyzed by western blotting using an Odyssey Infrared Imaging System (LI-COR, Lincoln, NE, USA). Detection of protein was performed using mouse anti-α-synuclein antibody (1 : 200) and goat anti-mouse IRDye 800CW (1 : 10 000) secondary antibody. The intensity of bands was quantified using LI-COR Odyssey software. Cell lysates were also assayed for DOPAL and other catechols.
Effects of hydrogen peroxide versus DOPAL on α-synuclein dimerization
α-synuclein (1.6 μM) dissolved in 100 mM Tris-HCl buffer (pH 7.4) was incubated with DOPAL (1–30 μM) or hydrogen peroxide (10–300 μM) at 37°C for 60 min. Dimers of α-synuclein were quantified by western blotting and the LI-COR system.
Endogenous and 6F-labeled catechols in cells and medium were assayed in our laboratory by HPLC with electrochemical detection after batch alumina extraction (Holmes et al. 1994). Identity of the DOPAL standard was confirmed by mass spectrometry, nuclear magnetic resonance, and liquid chromatography with time-of-flight mass spectrometry. Injection of DOPAL into the HPLC-electrochemical apparatus resulted in an unusual, characteristically wide chromatographic peak (Fig. 1). When dopamine was incubated with monoamine oxidase-A, the combination generated the DOPAL peak. Incubation of dopamine with monoamine oxidase-A and pargyline to inhibit monoamine oxidase abolished production of the DOPAL peak, demonstrating its derivation from oxidative deamination of dopamine. We synthesized 6F-DOPAL from 6F-dopamine by the same approach, to provide a standard by which to identify 6F-DOPAL in chromatographs.
Cell catechol contents were assayed from 100–200-μL aliquots of the 400 μL of disrupted cells. For samples of medium, 500 μL was assayed. Catechol concentrations in cell lysates were expressed in units of picomoles per well. Cell counts were measured using a Cellometer device (Nexcelom Bioscience, Lawrence, MA, USA).
For assaying cell concentrations of endogenous and 13C-dopamine, we used ultra-HPLC with tandem quadrupole mass spectrometry. d4-Dopamine (2 ng) was added to each sample as an internal standard. For elution of catechols from alumina, 0.1% formic acid was used. The samples were injected into a Waters Quattro Premier XE system (Waters Corporation, Milford, MA, USA) under isocratic conditions.
As our methodology for simultaneous measurement of dopamine, 13C-dopamine, and d4-dopamine has not yet been published, we provide more details as follows.
The system consisted of a Waters Acquity UPLC Core System containing a Binary Solvent Manager and Sample Manager with Column Heater Module and a Quattro Premier XE tandem quadrupole mass spectrometer. Carrier gas was generated by a nitrogen generator (Peak Scientific, Billerica, MA, USA). The collision gas was high-purity (99.9%) argon (Roberts Oxygen). Vacuum was generated by an Edwards XDS dry vacuum scroll pump (Edwards, Sanborn, NY, USA). The sample manager was fitted with a 50 μL loop and a 250 μL syringe. The injection mode was partial loop with needle overfill. Column temperature was 30°C, sample compartment temperature 10°C. source temperature 140°C, and desolvation temperature 450°C. Gas flow was 1000 L/h and cone gas flow 50 L/h. Cone voltage was 34 V, capillary voltage 1.0 kV, collision voltage 17 V, and collision energy 17.
Results were acquired in multiple-reaction monitoring mode with positive ionization. Parent-to-daughter transitions for dopamine were 136.95–90.80 m/z, d4-dopamine 140.95–94.80 m/z, and 13C-dopamine 142.95–96.80 m/z. Interchannel delay was 0.005 s, interscan delay 0.005 s, and dwell time 0.05 s.
Peak integration was performed using Waters QuanLynx Method Editor V4.1 software, using the following integration parameters: Smoothing enabled, Method Mean, Iterations 2, Width 2; Apex track parameters: Peak to peak baseline noise 10, Peak width at 5% height 30, baseline start at threshold 0.0, baseline end at threshold 0.5, detect shoulder peaks: No.; Standard peak detect ion parameters: Peak to peak noise amplitude 5440, balance 30, splitting 90, detect shoulder peak threshold 30, reduce tail 50, reduce height 10; View threshold parameters: Threshold relative height 1.5, threshold absolute height 0, threshold relative area 2.0, threshold absolute area 0.
Dopamine concentrations (in units of pg/mL) were calculated from peak areas (PAs) using the following formula.
‘Extracted’ refers to compounds after batch alumina extraction.
Data analysis and statistics
Neurochemical data were displayed using KaleidaGraph 4.01 (Synergy Software, Reading, PA, USA). Fractional changes at 20 min from baseline in the same experiments were compared by two-tailed, dependent-means t-tests. Differences between drug treatments across experiments were compared by two-tailed, independent-means t-tests. For testing hypotheses involving three or more treatments, factorial analyses of variance were used with Fisher's PLSD post hoc test. A p value less than 0.05 defined statistical significance.
Dopamine was the main catechol detected in PC12 cells. In descending order of cell lysate concentrations, the other detected catechols were DOPAC > DOPET > norepinephrine > dihydroxyphenylalanine > DOPAL > dihydroxyphenylglycol (Fig. 2). No epinephrine was detected. DOPAL corresponded to 0.35% of dopamine. The pattern of catechol levels in the medium at baseline (after 24 h pre-incubation in medium containing tolcapone) differed substantially from that in cell lysates (Fig. 2), with concentrations of the deaminated metabolites DOPAC and DOPET exceeding and with the DOPAL concentration about the same as that of dopamine.
Reserpine depletes endogenous dopamine
Reserpine rapidly and markedly depleted cellular dopamine, by about 1/2 at 20 min (p < 0.0001 vs. baseline) and at least 90% at 180 min (p < 0.0001, n = 7; Fig. 3a). Dopamine in the medium increased slightly but not significantly (Fig. 3b).
Reserpine increases formation of endogenous deaminated metabolites of dopamine, including DOPAL
Reserpine rapidly increased cellular contents of the deaminated metabolites DOPAL, DOPET, and DOPAC; Fig. 3a and c). Peak increases in DOPET (p < 0.0001), DOPAC (p < 0.0001), and DOPAL (p = 0.0008) were attained at 20 min, when DOPAL averaged 2.7 times baseline. Thereafter, concentrations of the deaminated metabolites declined. Relative increases from baseline were similar for DOPAL, DOPET, and DOPAC. Reserpine also rapidly increased DOPAL concentrations in the medium (p = 0.0003 at 20 min; Fig. 3d). At both 20 min and 3 h of incubation with 6F-dopamine, a compound cochromatographing with 6FDOPAL was noted, with or without reserpine treatment (data not shown).
Inhibition of aldehyde dehydrogenase and aldehyde reductase increases endogenous DOPAL production
Both aldehyde dehydrogenase inhibition by daidzein and aldehyde reductase inhibition by AL1576 increased DOPAL levels in cells (Fig. 4) and medium (Fig. 5). Daidzein increased and AL1576 markedly decreased cell lysate and medium contents of DOPET (Figs 4b and 5b). Neither drug affected lysate or medium DOPAC (Figs 4c and 5c). Combined pre-incubation with daidzein+AL1576 produced larger increases in DOPAL than did either drug alone, in both cells and medium. Pre-incubation with daidzein or AL1576 augmented DOPAL responses to reserpine. The largest endogenous DOPAL responses observed in both cells and medium were with daidzein+AL1576+reserpine.
Reserpine blocks vesicular uptake without affecting cellular uptake
Reserpine treatment decreased 6F-dopamine content by 85.4% at 20 min (p < 0.0001) and 97.3% at 3 h (p < 0.0001; Fig. 6a and Table 1). In contrast, 6F-dihydroxyphenylacetic acid (6FDOPAC) was increased slightly by 3.2% at 20 min and 11.0% at 3 h. 6F-norepinephrine was detected at 3 h, indicating vesicular uptake and slow intravesicular conversion of 6F-dopamine to 6F-norepinephrine. In cells coincubated with 6F-dopamine and reserpine, no 6F-norepinephrine was detected. Between 20 min and 3 h of incubation with 6F-dopamine, 6F-dopamine in the medium often decreased to undetectable levels, while 6FDOPAC increased substantially. Reserpine treatment produced similar drastic effects on cell contents of endogenous and 13C-dopamine by 3 h (compare with Figs 2a and 6b). Vehicle treatment did not affect concentrations of 6F-dopamine, 6FDOPAC, or 13C-dopamine at either time point.
Table 1. Medium mean±SEM concentrations (pmol/mL) of endogenous and 6F-catechols
Data averaged for three to five experiments.
VEH + 6FDA
40 ± 16
1423 ± 90
352 ± 108
34 ± 9
RES + 6FDA
36 ± 15
1449 ± 86
346 ± 108
33 ± 8
VEH + 6FDA
64 ± 9
1470 ± 152
4 ± 3
547 ± 63
RES + 6FDA
44 ± 2
1650 ± 112
2 ± 2
586 ± 70
DOPAL contributes to DA-induced apoptosis
Incubation of PC12 cells with exogenous dopamine increased cellular DOPAL content (Fig. 7a) and evoked apoptosis that was about 80% that produced by serum withdrawal (Fig. 7b). When cells were incubated with dopamine and daidzein+AL1576, the amount of apoptosis increased to 90% of that produced by serum withdrawal. Incubation with daidzein+AL1576 augmented the cellular DOPAL response to exogenous dopamine (Fig. 7a). As indicated by the lines and brackets in Fig. 7b, coincubation with dopamine and daidzein+AL1576 augmented dopamine-induced apoptosis by 22% (p < 0.01).
DOPAL contributes to DA-induced synuclein oligomerization
As expected, PC12 cells conditionally over-expressing synuclein had increased synuclein content, based on western blotting (Fig. 8a). DOPA incubation time-dependently increased cell contents of DOPAL, DOPET, and DOPAC by about 100-fold each (Fig. 8b shows averaged results for 3 experiments). The peak cell DOPAL concentration attained during incubation with DOPA was about the same as that attained during incubation with reserpine+daidzein+AL1576.
Incubation of PC12 cells over-expressing α-synuclein with DOPA resulted in augmented formation of α-synuclein dimers (Fig. 8c shows a representative western blot and Fig. 8d, averaged results for four experiments). Pargyline (n = 3, Fig. 8d) or carbidopa (n = 2) treatment prevented DOPA-induced dimerization of α-synuclein. Neither pargyline nor carbidopa alone affected α-synuclein dimerization (data not shown).
As enzymatic deamination of dopamine produces DOPAL and hydrogen peroxide in a 1 : 1 molar ratio, DOPAL was compared with hydrogen peroxide across a range of concentrations in the ability to oligomerize α-synuclein. As shown in Fig. 8e, DOPAL potently evoked α-synuclein dimerization, whereas hydrogen peroxide did not.
Here, we report that in PC12 cells, which normally contain dopamine, the catecholaldehyde DOPAL, and the DOPAL metabolites DOPAC and DOPET, pharmacologic blockade of vesicular uptake increases cell and medium contents of endogenous DOPAL; and inhibition of enzymes that metabolize DOPAL to DOPAC and DOPET augments DOPAL responses to vesicular uptake blockade. We also obtained evidence that intracellular buildup of DOPAL contributes to both apoptosis and to α-synuclein oligomerization.
Catecholaminergic cells and neurons are characterized by ongoing passive leakage of vesicular catecholamines into the cytosol (Eisenhofer et al. 2004a), with active reuptake mediated by the vesicular monoamine transporter. Reserpine-induced blockade of vesicular uptake therefore would be expected to result in net translocation of stored dopamine into the cytosol, where dopamine is susceptible to oxidative deamination catalyzed by mitochondrial monoamine oxidase. Consistent with such translocation, reserpine treatment rapidly depleted cellular dopamine content, by about 1/2 at 20 min and at least 90% at 3 h, associated with simultaneous rapid increases in cell and medium contents of DOPAL. These results agree with previously reported findings in PC12 cells exposed to the complex I inhibitor rotenone (Lamensdorf et al. 2000), which can produce an animal model of Parkinson's disease (Sherer et al. 2003), redistributes dopamine from vesicles into the cytosol (Watabe and Nakaki 2008), and down-regulates the type 2 vesicular monoamine transporter (Sai et al. 2008); however, whether rotenone inhibits vesicular uptake of intraneuronal catecholamines, such as by following the fate of tracer-labeled dopamine, has not yet been demonstrated.
At 3 h of coincubation with 6F-dopamine or 13C-dopamine, reserpine markedly decreased cell 6F-dopamine content by 97% and 13C-dopamine content by 96%, confirming extremely efficient blockade of vesicular uptake by reserpine in PC12 cells. As reserpine did not affect concentrations of either 6FDA or 6FDOPAC in the medium, reserpine did not impede cellular catecholamine uptake.
DOPAL is converted to DOPAC by aldehyde dehydrogenase and to DOPET by aldehyde reductase. Normally the former route is favored, and the main endogenous metabolites of dopamine are acids (Kopin 1985). Thus, in this study, cellular content of the acid DOPAC exceeded that of the alcohol DOPET. As inhibition of aldehyde dehydrogenase by daidzein increased and inhibition of aldehyde reductase by AL1576 decreased DOPET levels, the two enzymes are alternatively available to act on cytosolic DOPAL. Inhibition of either enzyme increased cell DOPAL, inhibition of both increased DOPAL further, and reserpine exposure in the setting of AL1576+daidzein was associated with the highest cell and medium concentrations of endogenous DOPAL seen in the study, several times those in untreated cells.
These findings may help explain inconsistent literature about toxicity of DOPAL in nigrostriatal dopaminergic neurons. In human neuroblastoma SH-SY5Y cells, incubation with exogenous dopamine in the setting of disulfiram to block aldehyde dehydrogenase was found to increase DOPAL levels and to evoke substantial toxicity (Legros et al. 2004a). In rats treated semi-chronically with levodopa and benserazide to increase dopamine production in the brain, however, disulfiram treatment did not alter striatal dopamine content (Legros et al. 2004b). Based on the present results, perhaps levodopa treatment with aldehyde dehydrogenase inhibition would have exerted cytotoxicity, had the study included inhibition of vesicular uptake or of aldehyde reductase.
Mice with genetically determined very low activity of the type 2 vesicular monoamine transporter or with double knockout of the genes encoding aldehyde dehydrogenase 1A1 and 2 have aging-related behavioral and neuropathologic findings mimicking those in Parkinson's disease (Caudle et al. 2007; Wey et al. 2012). The current findings lead straightforwardly to the hypotheses that pharmacologic inhibition of aldehyde dehydrogenase and aldehyde reductase should exacerbate neurodegeneration in the former animal model, whereas pharmacologic inhibition of vesicular uptake should exacerbate neurodegeneration in the latter model. If so, this would provide much needed gene–environment interaction models for the pathogenesis of Parkinson's disease (Bronstein et al. 2009).
Incubation of PC12 cells with dopamine evoked apoptosis the severity of which corresponded to about 80% of that seen with serum withdrawal, in line with previous reports (Shinkai et al. 1997; Jana et al. 2011). To examine the contribution of DOPAL produced within the cells to dopamine-induced apoptosis, we incubated PC12 cells with dopamine, with or without daidzein+AL1576 to inhibit enzymatic breakdown of DOPAL. From the increase in apoptosis in the setting of daidzein+AL1576, we estimated that at least 22% of dopamine-evoked apoptosis was attributable to DOPAL.
Lewy bodies, cytosolic inclusions in monoaminergic neurons, are a pathologic hallmark of sporadic Parkinson's disease. Lewy bodies contain abundant α-synuclein (Spillantini et al. 1997), rare patients with familial Parkinson's disease have mutations or replication of the gene encoding α-synuclein abnormalities of (Polymeropoulos et al. 1997; Singleton et al. 2003), and genome-wide association studies have consistently reported statistical associations between Parkinson's disease and genotypic variants of the α-synuclein gene (Satake et al. 2009; Edwards et al. 2010). According to the ‘catecholaldehyde hypothesis,’ reduced vesicular uptake and impaired aldehyde detoxification contribute to the catecholaminergic denervation and synucleinopathy that characterize Parkinson's disease.
For the catecholaldehyde hypothesis to hold credence, one would expect a link between DOPAL and α-synucleinopathy. This link has indeed been found: exogenous DOPAL oligomerizes and precipitates α-synuclein (Burke et al. 2008). In this study, the findings of DOPA-induced dimerization of α-synuclein and prevention of this effect by either carbidopa to block conversion of DOPA to dopamine or pargyline to block conversion of dopamine to DOPAL indicate that DOPAL formed intracellularly from cytosolic dopamine contributes to α-synuclein oligomerization. Considering that α-synucleinopathy may directly or indirectly interfere with vesicular sequestration of intraneuronal catecholamines (Adamczyk et al. 2006; Park et al. 2007; Guo et al. 2008; Mosharov et al. 2009) and with enzymatic detoxification of catecholaldehydes (Lee et al. 2001; Jinsmaa et al. 2009; Nasstrom et al. 2011), induction of multiple positive-feedback loops could lead to rapidly progressive neurodegeneration. The present results do not imply, however, that increased oligomerization of α-synuclein accounts for DOPAL-induced apoptosis. One may reasonably doubt this causal sequence, because PC12 cells normally contain relatively little α-synuclein.
DOPAL production from DOPA necessarily entails hydrogen peroxide production, in a 1 : 1 molar ratio, upon enzymatic deamination of dopamine. To evaluate the relative contributions of DOPAL and hydrogen peroxide to synuclein oligomerization after DOPA incubation, we compared DOPAL with hydrogen peroxide in the ability to dimerize α-synuclein. Whereas hydrogen peroxide up to a concentration of 300 μM did not dimerize α-synuclein, DOPAL did so at concentrations above about 30 μM. Prevention by pargyline of DOPA-induced α-synuclein oligomerization therefore is much more likely to be from decreased DOPAL generation than from decreased hydrogen peroxide generation.
Whether levodopa treatment for Parkinson's disease accelerates the neurodegenerative process has been a persistently controversial and provocative topic in clinical neurology. Based on the findings that DOPA oligomerizes α-synuclein, that the oligomerization occurs via increased DOPAL production, and that the attained intracellular DOPAL concentration upon exposure to exogenous DOPA is within a physiologic range, perhaps better understanding about determinants of production and metabolism of DOPAL and about DOPAL-induced synuclein oligomerization in individual patients will help resolve this important matter.
In summary, blockade of vesicular uptake increases endogenous DOPAL production, inhibition of aldehyde dehydrogenase and aldehyde reductase attenuates DOPAL metabolism, the combined manipulations augment DOPAL levels further, and DOPAL produced within cells contributes to apoptosis and α-synuclein oligomerization. The results therefore fit with the catecholaldehyde hypothesis. Further studies about potential effects of genetically determined α-synucleinopathy on endogenous DOPAL production and about factors influencing DOPAL-induced synuclein oligomerization seem indicated.
The research reported here was supported by the Intramural Research Program of the National Institute of Neurological Disorders and Stroke. The authors thank Gilberto Carmona, Nelson Cole, Richard Imrich, and Roshanak Mansouri for preliminary work that enabled us to carry out the current experiments. The authors have no conflicts of interest to disclose.
D.S.G. was involved in conception and design, data analysis, data interpretation, drafting the article, and final approval. P.S., R.S. and D.J.G. were involved in data acquisition and data analysis. A.C. was involved in data acquisition, drafting the article, revising the article, and data analysis. Y.J. was involved in data acquisition, data analysis, conception and design, drafting the article, and revising the article. CH. was involved in data acquisition, data analysis, drafting the article, and revising the article. I.K. and Y.S. were involved in conception and design, data analysis, drafting the article, and revising the article critically for important intellectual content.