Proteasome inhibitor MG-132 induces dopaminergic degeneration in cell culture and animal models

Impairment in ubiquitin-proteasome system (UPS) has recently been implicated in Parkinson's disease, as demonstrated by reduced proteasomal activities, protein aggregation and mutation of several genes associated with UPS. However, experimental studies with proteasome inhibitors failed to yield consensus regarding the effect of proteasome inhibition on dopaminergic degeneration. In this study, we systematically examined the effect of the proteasome inhibitor MG-132 on dopaminergic degeneration in cell culture and animal models of Parkinson's disease. Exposure of immortalized dopaminergic neuronal cells (N27) to low doses of MG-132 (2-10 microM) resulted in dose- and time-dependent cytotoxicity. Further, exposure to MG-132 (5 microM) for 10 min led to dramatic reduction of proteasomal activity (>70%) accompanied by a rapid accumulation of ubiquitinated proteins in these cells. MG-132 treatment also induced increases in caspase-3 activity in a time-dependent manner, with significant activation occurring between 90 and 150 min. We also noted a 12-fold increase in DNA fragmentation in MG-132 treated N27 cells. Similarly, primary mesencephalic neurons exposed to 5 microM MG-132 also induced >60% loss of TH positive neurons but only a minimal loss of non-dopaminergic cells. Stereotaxic injection of MG-132 (0.4 microg in 4 microl) into the substantia nigra compacta (SNc) in C57 black mice resulted in significant depletion of ipisilateral striatal dopamine and DOPAC content as compared to the vehicle-injected contralateral control sides. Also, we observed a significant decrease in the number of TH positive neurons in the substantia nigra of MG-132-injected compared to the vehicle-injected sites. Collectively, these results demonstrate that the proteasomal inhibitor MG-132 induces dopamine depletion and nigral dopaminergic degeneration in both cell culture and animal models, and suggest that proteasomal dysfunction may promote nigral dopaminergic degeneration in Parkinson's disease.     

The pattern of neuronal loss and survival may reflect differential expression of proteasome activators in Parkinson's disease

The basis of neuronal vulnerability, degeneration, and sparing in PD are unknown, but there is increasing evidence to suggest that the ubiquitin‐proteasome system (UPS) plays an important role in the pathogenesis of the disorder. In this study, we employed an immunocytochemical approach to determine if the differential expression of key UPS components in various brain regions and cells might underlie the pattern of neuronal degeneration and survival seen in PD. We showed that the ubiquitin activating enzyme (E1), ubiquitin conjugating enzyme (E2), and 26/20S proteasome α‐ and β‐subunits, are abundantly expressed in the substantia nigra pars compacta (SNc) and in cultured dopaminergic neurons. Although the proteasome activator PA700 is expressed in the medial SNc, levels are low in the lateral region, and expression of the other proteasome activator, PA28, is near absent in the entire SNc. PA28 (but not PA700) was found to be poorly expressed in noradrenergic neurons in the locus coeruleus (LC) compared with adjacent cells in the mesencephalic nucleus. PA700 and PA28 are also poorly expressed in dopaminergic neurons compared with other cell types in culture. Inhibition of proteasomal function, generation of misfolded proteins, induction of oxidative stress or impairment of mitochondrial complex I activity, caused a compensatory upregulation in PA700 and PA28 in a variety of cells but not in dopaminergic neurons in culture. These findings are consistent with the demonstration that, in sporadic PD, proteasomal activity and levels of PA700/PA28 are reduced in the SNc butare markedly upregulated in regions/cells that are spared from the neurodegenerative process. Thus, the differential distribution and activity of proteasome activations could play a significant role in the pathogenesis of PD. Synapse 64:241–250, 2010. © 2009 Wiley‐Liss, Inc.     

蛋白酶体抑制剂对神经细胞的双重作用

目的探索蛋白酶体抑制剂在不同浓度时对多巴胺能神经元的作用及原因。方法用6-羟多巴(6-OHDA)50uM处理的人SH-SY5Y细胞作为细胞受损伤的模型,加用不同浓度的蛋白酶体抑制剂lactacystin,镜下细胞计数和SRB法测定细胞活力,平行对照组测定蛋白酶体活性。用选择性MEKl/2抑制剂PD98059验证蛋白酶体抑制剂是否通过MAPK途径起作用。结果蛋白酶体抑制剂lactacystin在0.1、0.25、0.5uM浓度时提高细胞存活率,在2、5uM时降低细胞存活率,相应的蛋白酶体活性分别是对照组的83.43%、73.84%、66.14%、24.11%、12.36%,加用PD98059后,0.25、0.5uMlactacystin的保护作用被阻断。结论蛋白酶体抑制剂在低浓度时对多巴胺能神经元有保护作用,高浓度时对多巴胺能神经元有毒性作用,这种不同作用的原因可能与蛋白酶体抑制的程度有关。蛋白酶体抑制剂的保护作用可能通过MAPK途径起作用。     

Lack of nigrostriatal pathology in a rat model of proteasome inhibition

Systemic administration of ubiquitin-proteasome system inhibitors to rodents has been reported to induce certain behavioral and neuropathological features of Parkinson's disease. The goal of this study was to replicate these observations by administering a proteasome inhibitor (PSI) to rats using McNaught and colleagues' protocol. No alterations in locomotor activity or striatal dopamine and its metabolites were observed. Differences in nigral dopaminergic cell number between proteasome inhibitor- and vehicle-treated rats and inclusion bodies were not found. Extending the time of survival after administration and using different solvents failed to alter results, indicating this proteasome inhibitor does not consistently produce the selective toxicity and pathological hallmarks characterizing Parkinson's disease.     

Inhibition of proteasome activity sensitizes dopamine neurons to protein alterations and oxidative stress

Impairment in the capacity of the ubiquitin-proteasome pathway to clear unwanted proteins has been implicated in the cell death that occurs in Parkinson's disease (PD). In support of this concept, defects in proteasomal structure and function, as well as protein aggregates and increased levels of oxidized proteins are found in the substantia nigra of PD patients. We have previously demonstrated that inhibition of proteasome activity in mesencephalic cultures induces degeneration of dopaminergic neurons coupled with the formation of proteinaceous intracellular inclusions. In this study we examined the effect of proteasome inhibition on cultured dopamine neurons when combined with oxidative stress and protein misfolding, in order to better simulate the condition in PD. We demonstrate that two structurally unrelated inhibitors of proteasome activity, lactacystin and carbobenzoxy-L-leucul-L-leucyl-L-leucinal (MG132), cause dose-dependent cell loss that preferentially affects dopaminergic neurons. Conditions that promote protein damage and misfolding such as oxidative stress, heat shock, and canavanine also induce neuronal degeneration with preferential loss of dopamine neurons and cell death is markedly increased when any of these is combined with a proteasome inhibitor. These studies demonstrate a synergistic effect between conditions that promote the formation of damaged proteins and those in which proteasomal function is impaired, and provide further support for the notion that cell loss in PD could be related to a defect in protein handling.     

Failure of proteasome inhibitor administration to provide a model of Parkinson's disease in rats and monkeys

McNaught and colleagues reported recently that systemic administration of proteasome inhibitors PSI (Z-Ileu-Glu(OtBu)-Ala-Leu-CHO) or epoxomicin recapitulated many of the degenerative changes seen in Parkinson's disease including loss of striatal dopamine and cell loss in the substantia nigra, locus ceruleus, dorsal motor nucleus of the X cranial nerve, and nucleus basalis of Meynert. Intracytoplasmic inclusions resembling Lewy bodies were also described. All experiments administering PSI to rats using identical procedures and multiple attempts failed to induce any of the previously described changes. Furthermore, administration of PSI or epoxomicin to monkeys in an attempt to extend the model to a primate species failed. Currently, systemic proteasome inhibition is not a reliable model for Parkinson's disease.     

Elevation of heme oxygenase‐1 by proteasome inhibition affords dopaminergic neuroprotection

Postmortem studies have shown that heme oxygenase‐1 (HO‐1) immunoreactivity is increased in patients with Parkinson disease. HO‐1 expression is highly upregulated by a variety of stress. Since the proteasome activity is decreased in patients with Parkinson disease, we investigated whether proteasome activity regulates HO‐1 content. MG‐132, a proteasome inhibitor, increased the amount of HO‐1 protein mainly in astrocytes of primary mesencephalic cultures. Quantitative RT‐PCR analysis revealed that lactacystin upregulated HO‐1 mRNA expression. Proteasome inhibition with MG132 also increased the cytomegalovirus promoter‐driven expression of Flag‐HO‐1 protein and resulted in an accumulation of ubiquitinated Flag‐HO‐1 in Flag‐HO‐1‐overexpressing PC12 cells. In addition, a cycloheximide chase assay demonstrated that the degradation of Flag‐HO‐1 protein was slowed by MG‐132. Next, the function of HO‐1 which was upregulated by proteasome inhibitors was examined. Proteasome inhibitors protected dopaminergic neurons from 6‐hydroxydopamine (6‐OHDA)‐induced toxicity and this neuroprotection was abrogated by co‐treatment with zinc protoporphyrin IX, a HO‐1 inhibitor. Furthermore, 6‐OHDA‐induced toxicity was blocked by bilirubin and carbon monoxide, products of the HO‐1‐catalyzed degradation of heme. These results suggest that mesencephalic HO‐1 protein level is regulated by proteasome activity and the elevation by proteasome inhibition affords neuroprotection. © 2010 Wiley‐Liss, Inc.     

Estrogen effects on the tuberoinfundibular dopaminergic system in the female rat brain

Estrogen effects on tyrosine hydroxylase (TH), monoamine oxidase types A and B (MAO), and dopamine (DA) in microdissected regions of the hypothalamus, preoptic area and substantia nigra (SNR) of the female rat brain were investigated. Ovariectomized (OVX) young adult female rats were implanted with single silastic capsules containing 100% estradiol valerate (EV). Control rats received empty silastic capsules. Two weeks following capsule insertion, EV decreased TH activity and DA concentration in the arcuate nucleus (AN) while no significant changes in TH activity or DA concentration were observed in the SNR, ventromedial nucleus (VMN), suprachiasmatic nucleus, paraventricular nucleus, medial preoptic nucleus, or the periventricular preoptic nucleus. Although estrogen suppressed TH and DA in the AN, 2 weeks following removal of the estrogen containing capsules, TH activity and DA concentration were restored to control (OVX) levels. Suppression of MAO activity occurred in both the AN and the VMN of rats implanted with EV capsules and returned to OVX levels following the removal of the estradiol load. These results revealed that estrogen effects on TH and MAO activities and DA concentration in the midbrain are region specific and reversible; and that among the dopaminergic systems studied, estrogen effects on TH and DA are confined to the tuberoinfundibular dopaminergic system (TIDAS). Furthermore, these results support our hypothesis that estrogen is a key regulator of DA function in the TIDAS via effects on TH. The importance of these findings to the control of gonadotropin secretion and reproductive cyclicity is discussed.     

Increase in mesolimbic electrical self-stimulation after chronic haloperidol: reversal by L-DOPA or lithium

After chronic neuroleptic drug treatment, an increase in electrical intracranial self-stimulation (ICSS) rate is seen from electrodes in the A10 dopaminergic nucleus. This increase, which persists for approximately 3 weeks following drug withdrawal, is believed to represent a behavioral manifestation of drug-induced dopaminergic synaptic supersensitivity. Chronic L-DOPA caused a partial reversal of haloperidol-induced ICSS increase. Lithium carbonate, given concurrently with the haloperidol, partially prevented the development of ICSS supersensitivity. It is concluded that dopaminergic synaptic sensitivity has a two-way modulatory capability.     

Controversies on new animal models of Parkinson's disease pro and con: the rotenone model of Parkinson's disease (PD)

A general complex I deficit has been hypothesized to contribute to neurodegeneration in Parkinson's disease (PD) and all toxins used to destroy dopaminergic neurons are complex I inhibitors. With MPTP or 6-OHdopamine, this hypothesis can not be tested since these toxins selectively accumulate in the dopaminergic neurons. However with rotenone, which penetrates all cells, the hypothesis can be tested. Thus, the proof of the hypothesis is whether or not rotenone-induced neurodegeneration mimics the degenerative processes underlying PD. Low doses of rotenone (1.5 or 2.5 mg/kg in oil i.p.) were administered to Sprague Dawley rats on a daily basis. After about 20 days of treatment, signs of parkinsonism occurred and the concentrations of NO and peroxidase products rose in the brain, especially in the striatum. After 60 days of treatment, rotenone had destroyed dopaminergic neurons. Behaviourally, catalepsy was evident, a hunchback posture and reduced locomotion. Other transmitter systems were not, or much less affected. L-DOPA-methylester (10 mg/kg plus decarboxylase inhibition) potently reversed the parkinsonism in rats. Also when infused directly into the dopaminergic neurons, rotenone produced parkinsonism which was antagonized by L-DOPA. Some peripheral symptoms of PD are mimiced by rotenone too, for example a low testosterone concentration in the serum and a loss of dopaminergic amacrine cells in the retina. These results support the hypothesis of an involvement of complex I in PD and render the rotenone model as a suitable experimental model. The slow onset of degeneration make it suitable also to study neuroprotective strategies. Evidence that rotenone-induced neurodegeneration spreads beyond the dopaminergic system is not contradictory given that, according to the new staging studies, also degeneration in PD is not confined to dopamine neurons.     

Fas/FasL-mediated apoptosis in the arcuate nucleus and medial preoptic area of male ArKO mice is ameliorated by selective estrogen receptor alpha and estrogen receptor beta agonist treatment, respectively

The aromatase (ArKO) knockout mouse is estrogen deficient. Our previous analysis revealed apoptosis of dopaminergic neurons in the arcuate nucleus (Arc) and medial preoptic area (MPO) of 1-year-old male ArKO mice. We sought to determine which estrogen receptor (ER) is involved in the anti-apoptotic action of estrogen. Male ArKO (9.5-month-old) mice were treated with 16alpha-LE(2) (ERalpha-specific agonist) or 8beta-VE(2) (ERbeta-specific agonist). Daily injections (6 weeks) with 16alpha-LE(2) prevented dopaminergic cell death in the Arc of male ArKO mice, with no significant effect of 8beta-VE(2) treatment. In contrast, 8beta-VE(2) prevented dopaminergic cell death in the MPO, while 16alpha-LE(2) had no significant effect. Concomitant decreases in Fas and FasL protein levels were found upon 16alpha-LE(2) and 8beta-VE(2) treatment in the Arc and MPO, respectively. Our results indicate that anti-apoptotic effects of estrogen are ER mediated, and the specific ER subtype involved in regulating apoptosis depends on the particular brain nucleus in question.     

Recovery of function and basal ganglia [14C]2-deoxyglucose uptake after nigrostriatal injury

Rats with unilateral 6-hydroxydopamine injections along the mesotelencephalic dopaminergic projection showed a profound impairment in localizing somatosensory stimuli on the contralateral body surface at 3 days postoperatively. Approximately one-half of the affected animals recovered the ability to localize tactile stimuli during 6 weeks postoperatively, whereas the remainder did not. When it occurred, the recovery of sensorimotor function began between the third and fifth day postoperatively and plateaued between days 14 and 21. Unilateral damage to these dopaminergic neurons resulted in hemispheric asymmetries of [14C]2-deoxyglucose incorporation at 3 days postoperatively. For structures that normally receive a dopaminergic innervation (e.g., neostriatum, nucleus accumbens septi, olfactory tubercle) the autoradiographic density of the injured side was decreased relative to the intact hemisphere. For structures that receive striatal inputs (globus pallidus, entopeduncular nucleus, substantia nigra pars reticulata), the autoradiographic density was increased on the side of the injury. This pattern of altered [14C]2-deoxyglucose incorporation was still present at 6 weeks postoperatively in animals that showed no recovery of somatosensory localization during that time. In contrast, rats that did recover showed no hemispheric asymmetries within the anterior neostriatum, globus pallidus, or substantia nigra pars reticulata at 6 weeks postoperatively, and the time course of normalization of metabolic activity in these structures was similar to that for behavioral restoration. These results directly demonstrate the importance of the neostriatum and particular structures efferent to it in the recovery of sensorimotor functions after striatal dopamine depletion. The types of neuronal plasticity within this basal ganglia circuitry responsible for the normalization of [14C]2-deoxyglucose incorporation and behavior are discussed.     

Dopaminergic neurotoxins require excitotoxic stimulation in organotypic cultures

We have investigated the properties of the dopaminergic neurotoxins 6-hydroxydopamine, 1-methyl-4-phenylpyridinium and rotenone using an organotypic culture that included slices of substantia nigra, striatum and cortex maintained for about 20 days in vitro. At this age, the organotypic culture contains dopaminergic neurons, visualized using tyrosine hydroxylase (TH) immunohistochemistry, that project into the striatal slice and extend up to 1 mm into the cortical slice. Using TH immunohistochemistry to assess survival of dopaminergic neurons, we found that the three dopaminergic toxins alone were not selectively neurotoxic. However, the addition of a low concentration of N-methyl-d-aspartate together with each individual toxin resulted in profound injury to the dopaminergic neurons, reflected by the loss of cell bodies and the fragmentation of processes. The combined toxicity was completely blocked by MK801. To assess the specificity of the injury, we measured the diameter of cell nuclei in the organotypic culture stained with Hoechst 33342 because the nucleus shrinks when neurons are injured. These measurements showed that the combined toxin treatment selectively injured only the TH immunoreactive cells. Thus, in a model culture system where dopaminergic neurons innervate appropriate targets, excitotoxicity appears to be essential for the manifestation of the toxic actions of 6-hydroxydopamine, 1-methyl-4-phenylpyridinium and rotenone.     

Neurobehavioral evidence for mesolimbic specificity of action by clozapine: studies using electrical intracranial self-stimulation

The ability of chronic treatment with the atypical neuroleptic clozapine to induce functional dopaminergic hypersensitivity in laboratory rats was assessed. The intracranial electrical self-stimulation paradigm, known to be sensitive to changes in functional dopaminergic sensitivity, was used. Animals with electrodes in the ventral tegmental nucleus (mesolimbic dopamine cell body area) showed a marked increase in self-stimulation rate following 3 weeks of chronic clozapine. This increase was similar in magnitude and duration to that shown by animals given 3 weeks of chronic haloperidol. In contrast, animals with electrodes in the substantia nigra (nigrostriatal dopamine cell body area) showed no change in self-stimulation rate following 3 weeks of chronic clozapine. These data are interpreted in the light of previous suggestions that clozapine and other atypical neuroleptics may possess functional selectivity for the mesolimbic dopamine system.     

Vasoactive intestinal peptide fibers innervate neuroendocrine dopaminergic neurons

Hypothalamic neuroendocrine dopaminergic neurons exhibit a diurnal rhythm. Higher level input to these neurons has not been described. In the present study, we identified fibers known to originate in the suprachiasmatic nucleus (SCN), which were associated with neuroendocrine dopaminergic neurons. Hypothalamic sections were obtained from either ovariectomized (OVX) female rats or OVX female rats implanted with estrogen and progesterone (E+P). Confocal microscopic images were acquired from the periventricular nucleus, as well as the rostral, dorsomedial, ventrolateral, and caudal regions of the arcuate nucleus. Using antibodies directed against vasoactive intestinal peptide (VIP) and tyrosine hydroxylase (TH) the rate-limiting enzyme in dopamine synthesis, fine VIP fibers in close apposition to TH-immunoreactive (IR) soma and proximal dendrites were revealed. Of the antibodies for the two VIP receptor subtypes (VIP1R and VIP2R), only VIP2R was found on TH-IR neurons. E+P significantly increased the incidence and density of neuroendocrine dopaminergic neurons expressing VIP2R, when compared to OVX animals. E+P did not affect the percent of neuroendocrine dopaminergic neurons associated with VIP fibers. No VIP fibers or VIP2R were found on dopaminergic neurons in the zona incerta. Brain sections triple labeled for Synapsin (a protein localized in synaptic vesicles) VIP, and TH demonstrated that Synapsin was colocalized with VIP fibers that were associated with TH-IR neurons in the arcuate nucleus. Double-label immuno-electron microscopy of hypothalamic sections labeled with antibodies for VIP and TH revealed VIP boutons associated with TH-IR soma and proximal dendrites. These results suggest VIPergic neurons may directly regulate neuroendocrine dopaminergic neuron activity, and ovarian steroids may play a modulatory role.     

Proteasome inhibition results in increased poly-ADP-ribosylation: implications for neuron death

This study demonstrates the ability of proteasome inhibitors (lactacystin, MG 115, MG 132) adenosine diphosphate to induce a time- and dose-dependent increase in poly-ADP-ribosylation (PAR) in the neural PC6 cell line, a subclone of PC12 cells. Elevated levels of PAR contribute to the toxicity associated with impaired proteasome activity, based on the ability of PAR inhibitors to ameliorate the toxicity associated with the application of lactacystin, MG 115, and MG 132. Proteasome inhibitors induced the accumulation of PAR and neuron death in primary hippocampal neuron cultures, which were both ameliorated by treatment with PAR inhibitors. Together, these data indicate a role for increased PAR in the toxicity associated with proteasome inhibition, and suggest that inhibitors of PAR may provide neuroprotection in conditions where proteasome inhibition occurs.     

c-Jun N-terminal kinase mediates lactacystin-induced dopamine neuron degeneration

Parkinson disease is characterized by the progressive loss of dopaminergic neurons in the substantia nigra pars compacta. It has been proposed that dysfunction of the ubiquitin proteasome system plays an important role in the pathogenesis of Parkinson disease, but the mechanisms underlying ubiquitin proteasome system-related neuron degeneration are unknown. Here, we demonstrate that the proteasome inhibitor lactacystin induces phosphorylation of c-Jun N-terminal kinase (JNK) and c-Jun, the release of cytochrome c, activation of both caspase-9 and caspase-3, and sequential apoptosis of dopaminergic neurons in vitro. Most of these effects can be attenuated by the JNK inhibitor SP600125. Furthermore, infusion of lactacystin in rats in vivo also leads to phosphorylation of JNK before nigral neuron loss; chronic administration of SP600125 also blocks this loss. These results indicate that JNK is involved in proteasome inhibition-induced dopaminergic neuron degeneration through caspase-3-mediated apoptotic pathways, suggesting that this kinase may be a therapeutic target for the prevention of substantia nigra pars compacta degeneration in Parkinson disease patients.     

Recovery of function and basal ganglia [C]2-deoxyglucose uptake after nigrostriatal injury

Rats with unilateral 6-hydroxydopamine injections along the mesotelencephalic dopaminergic projection showed a profound impairment in localizing somatosensory stimuli on the contralateral body surface at 3 days postoperatively. Approximately one-half of the affected animals recovered the ability to localize tactile stimuli during 6 weeks postoperatively, whereas the remainder did not. When it occurred, the recovery of sensorimotor function began between the third and fifth day postoperatively and plateaued between days 14 and 21.Unilateral damage to these dopaminergic neurons resulted in hemispheric asymmetries of [¹⁴C]2-deoxyglucose incorporation at 3 days postoperatively. For structures that normally receive a dopaminergic innervation (e.g., neostriatum, nucleus accumbens septi, olfactory tubercle) the autoradiographic density of the injured side was decreased relative to the intact hemisphere. For structures that receive striatal inputs (globus pallidus, entopeduncular nucleus, substantia nigra pars reticulata), the autoradiographic density was increased on the side of the injury.This pattern of altered [¹⁴C]2-deoxyglucose incorporation was still present at 6 weeks postoperatively in animals that showed no recovery of somatosensory localization during that time. In contrast, rats that did recover showed no hemispheric asymmetries within the anterior neostriatum, globus pallidus, or substantia nigra pars reticulata at 6 weeks postoperatively, and the time course of normalization of metabolic activity in these structures was similar to that for behavioral restoration.These results directly demonstrate the importance of the neostriatum and particular structures efferent to it in the recovery of sensorimotor functions after striatal dopamine depletion. The types of neuronal plasticity within this basal ganglia circuitry responsible for the normalization of [¹⁴C]2-deoxyglucose incorporation and behavior are discussed.     

Proteasomal inhibition causes loss of nigral tyrosine hydroxylase neurons

Dysfunction of the ubiquitin-proteasomal system (UPS) has been implicated in the pathogenesis of Parkinson's disease. The systemic administration of UPS inhibitors has been reported to induce nigrostriatal cell death and model Parkinson's disease pathology in rodents. We administered a synthetic, specific UPS inhibitor (PSI) subcutaneously to rats and quantified substantia nigral tyrosine hydroxylase-positive dopaminergic neurons by stereology. PSI caused a 15% decrease in UPS activity at 2 weeks and a 42% reduction in substantia nigra pars compacta tyrosine hydroxylase-positive neurons at 8 weeks. Systemic inhibition of the UPS warrants further evaluation as a means to model Parkinson's disease.