- Top of page
- Materials and methods
To establish the possible roles of oxidative stress, inflammatory processes and other unknown mechanisms in neurodegeneration, we investigated brain gene alterations in N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) mice model of Parkinson's disease using Atlas mouse cDNA expression array membrane. The expression of 51 different genes involved in oxidative stress, inflammation, glutamate and neurotrophic factors pathways as well as in still undefined processes, such as cell cycle regulators and signal transduction molecules, was differentially affected by the treatment. The present study indicates the involvement of an additional cascade of events that might act in parallel to oxidative stress and inflammation to converge eventually into a common pathway leading to neurodegeneration. The attenuation of these gene changes by R-apomorphine, an iron chelator-radical scavenger drug, supports our previous findings in vivo where R-apomorphine was neuroprotective.
Parkinson's disease (PD) is a progressive neurodegenerative disorder that results in degeneration of nigro-striatal dopamine neurons with the deficiency of dopamine in the striatum (Bernheimer et al. 1973). The causes and mechanism for the degeneration of dopaminergic neurons is still elusive. There have been numerous hypotheses concerning the etiology of PD, including genetic aberrations, involvement of endogenous and exogenous derived neurotoxins and initiation of oxidative stress (OS) as a consequence of accumulation of reactive oxygen species (ROS). However, in majority of idiopathic PD the role of OS has gained support mainly because the neurochemical changes that have been observed occur specifically in substantia nigra pars compacta (SNPC) and not in substantia nigra (SN) pars reticulata, where melanin containing dopamine (DA) neurons degenerate. The neurochemical lesions in SNPC are accompanied by a progressive accumulation of iron and ferritin within reactive microglia and melanin containing dopamine neurons (Jellinger et al. 1990). It is thought that the chelatable iron has a pivotal role in the process of neurodegeneration and participates in the Fenton reaction with hydrogen peroxide to generate the most reactive of all ROS, namely the hydroxyl radical (Youdim et al. 1999). Support for OS in dopaminergic neurodegeneration has come from animal studies with the use of neurotoxins 6-hydroxydopamine (6-OHDA) and N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) (Gerlach and Riederer 1996). Both neurotoxins are considered relevant models of the disease and are thought to induce neurodegeneration via OS, since iron chelators (e.g. desferrioxamine and apomorphine) and radical scavengers (vitamin E and α-lipoic acid) pretreatment induces neuroprotection against the two neurotoxins (Olanow 1996; Grünblatt et al. 1999). Post mortem human PD brain and animal studies revealed that dopaminergic neurodegeneration might constitute a cascade of events, including OS, leading to demise of the neurons.
The major problem concerning a better therapeutic approach to the treatment, neuroprotection and prevention of the disease, is the enigma of its underlying cause. Knowledge of highly selective gene expression, as well as sequence homology to a known gene family, can provide a convenient shortcut for implicating a target in a given pathway of disease. The advent of cDNA microarrays provided a potential tool for gene expression profiling analysis. The most attractive application of cDNAs microarrays is in the study of differential gene expression in disease and animal models (Debouck and Goodfellow 1999). Detailed profiling of gene expression in the MPTP Parkinson model may yield additional insight into cellular, animal and human physiology, which is critical to the discovery and validation of therapeutic targets. Since OS has been implicated in a number of neurodegenerative diseases (Alzheimer's disease, Huntington chorea, amyotrophic lateral sclerosis and PD), the knowledge of the specific cascades of events leading and causing the neurodegeneration will be the key factor in developing and using neuroprotective drugs.
In the present study, we applied a cDNA array including 1200 gene fragments for comparing gene expression in brains of control and MPTP treated mice. The results were then compared with those of quantitative reverse-transcription polymerase chain reaction (RT-PCR) and in situ hybridization, and the effects of R-apomorphine (R-APO), a neuroprotective drug, was explored. The data show that alterations of gene expression detected by means of cDNA array, display a relatively high reliability and that they all implicate to a highly structured cascade of events involving not only OS, inflammation and glutamate toxicity but also cell cycle and signal transduction pathways. Monitoring differential gene expression profile provides a better insight for understanding the molecular mechanism of neurodegeneration and protection by neuroprotective drugs, such as R-APO.
- Top of page
- Materials and methods
The technique of cDNA expression array is being extensively used to study global changes in gene expression in disease, model systems and in response to drug treatment (DeRisi et al. 1996; Heller et al. 1997; Gray et al. 1998; Marton et al. 1998; Backert et al. 1999; Lee et al. 1999). This may lead to a better understanding of disease pathology and development of more specific and effective drugs. In the present work, we have used this technique for the first time to examine differential gene expression in a model of PD using the Parkinson-inducing neurotoxin, MPTP and neuroprotective drugs, R- and S-APO. Our objective was to gain an insight to the processes involved in dopaminergic neurodegeneration, which is not fully established by biochemical means and to determine gene expression in response to dopaminergic neuroprotective drugs (e.g. R-APO, Gassen et al. 1996; Grünblatt et al. 1999). In addition to understanding the pathways causing cell death it may help to develop other neuroprotective drugs and strategies.
There have been a number of reports indicating that in idiopathic PD, as well as in 6-OHDA and MPTP models, OS may have a role in the mechanism of DA neuron degeneration. Although increased inflammatory and glutaminergic excitotoxicity have also been implicated, by no means they have been established in vivo. We have identified 51 major gene changes in the brains of MPTP-mice model of PD. In addition pretreatment with R-APO reversed most of the alterations. The present gene expression analysis has clearly indicated that the process of dopaminergic neurodegeneration is a complex cascade of events that simple OS cannot explain. In addition to OS, glutaminergic-excitatory, nitric oxide mediated and inflammatory processes reported to be involved in the process of neurodegeneration, we also identified other novel genes as growth factors, cell cycle regulators as cyclin B2 and cytochrome P450 who may also be involved in the process of neuronal cell death (Fig. 4).
Figure 4. Current hypothesis for neurodegeneration cascade of events in the MPTP model of PD. The dashed lines represent potential targets for neuroprotection.
Download figure to PowerPoint
The general increase in interleukin (IL)-1β, IL-6, and IL-7, as well as in IL-1R, IL-2R, IL-3R and IL-4R induced by MPTP, confirms the concept of inflammation in neurodegeneration (Mogi et al. 1996, 1998; Bessler et al. 1999). Indeed, in PD and MPTP models there is a proliferation of reactive microglia around and on top of dying dopamine neurons (Jellinger et al. 1990), suggesting an on-going microglia-induced inflammatory process. In line with these findings, pretreatment with R-APO attenuated the elevation in most of those genes in the mouse model. The increase of the anti-inflammatory cytokine, IL-10 mRNA by MPTP might reflect an attempt to protect the neurons from degeneration. The evidence for OS includes the noted increase in chelatable-iron levels in microglia and nigrostriatal DA neurons (Jellinger et al. 1990) that may lead to activation of the redox and iron-sensitive NF-κB (Schreck et al. 1991; Youdim et al. 1999). Increased iron in macrophages and microglia, as seen in PD (Jellinger et al. 1990), may lead to iron dependent activation of NF-κB and gene regulation of IL-1β, IL-6 and TNFα (Lin et al. 1997; Bowie and O'Neill 2000). Indeed, a 70-fold increase in immunoreactive NF-κB in the nucleus of melanized dopaminergic neurons of PD patients was recently reported (Hunot et al. 1997). In the present study, the precursor of the NF-κB p50 subunit, NF-κB p105, mRNA expression was increased as a consequence of chronic MPTP treatment, whereas pretreatment of animals with R-APO prevented this effect. Inversely, inhibitor-κB (I-κB) was decreased by MPTP and increased upon pretreatment with R-APO, indicating a tight regulation of both proteins in neurodegeneration. Antioxidant and specifically iron chelators were found to be potent inactivators of NF-κB (Schreck et al. 1991; Lin et al. 1997; Youdim et al. 1999), suggesting a pivotal role for iron in NF-κB activation. Thus, the reversal in MPTP-induced NF-κB and I-κB mRNA expression pattern by R-APO could be assigned to its antioxidant and iron chelating properties (Gassen et al. 1996).
Stimulation of NMDA receptor results in calcium influx into the neuron, resulting in activation of NOS to form NO−, which is released in bursts (Garthwaite et al. 1989). It has been established that MPP+ releases glutamate (Carboni et al. 1990) and that this may be associated with nigral cell death. In the present work we have shown for the first time decreased gene expression of NMDA2B receptors by R-APO pretreatment, while no changes occurred in the expression of AMPA-1 receptor and iNOS mRNA by MPTP or R-APO treatments. The fact that R-APO pretreatment decreased only NMDA receptor mRNA expression, without affecting expression of the AMPA1 receptor, suggests a mechanism of action through NMDA, and of no intervention by an AMPA receptor.
The increased expression of neurotrophic factors may reflect a compensatory mechanism by stimulating the sprouting of the surviving neurons. GDNF and EGF have been shown to exert growth-promoting and survival effects on dopaminergic neurons (Hadjiconstantinou et al. 1991; Lin et al. 1993). In the present study, an extensive increase of GDNF mRNA was observed in SN and the hippocampus in response to acute and chronic MPTP administration. Nonetheless, in post mortem analysis of PD patients, GDNF mRNA expression was undetectable (Hunot et al. 1996). This result may be explained by the regenerative properties of nigro-striatal dopamine system in rodents and non-human primates models of PD (Gash et al. 1998). Both enantiomers of APO prevented the increase of GDNF. TNFα-induced protein, a novel gene recently found to be expressed in human endothelial cells in response to the inflammatory cytotoxic cytokine TNFα (Liberatore et al. 1999), was detected in SN, hippocampal and cortical neurons as a result of MPTP treatment. However, R- and S-APO pretreatment completely abolished its expression, further supporting the notion for the involvement of inflammatory process and OS in neurodegeneration. It is possible that the increased expression of TNFα-induced protein activates NF-κB, which in turn induces iNOS and cytotoxic cytokines gene expression leading to neurodegeneration (Swift et al. 1998). R-APO may act as a chain breaker by chelating the iron and scavenging ROS, thus preventing NF-κB activation.
A decrease in several mRNAs coding for different cyclins (Table 3) was also observed as a consequence of MPTP treatment. At present we cannot draw any conclusions regarding the role the group of cyclins have in neurodegeneration, since their expressions have not been examined in idiopathic PD. Nevertheless, cell cycle arrest may be related to the inflammatory processes occurring in PD and in MPTP model of PD, which results in increased neurotrophic factors (Mogi et al. 1996, 1998; Bessler et al. 1999), as also observed in our work. Exposure of human monocytes to transforming growth factor-β (TGF-β) causes cell cycle arrest in G1/S phase and inactivation of cyclin B2 (Liu et al. 1999). On the other hand, in our studies the expression of cyclin B2 was up-regulated by MPTP in the SN, as observed by in situ hybridization. In addition, PC12 cell culture treatment with nerve growth factor (NGF) induces a decrease in the expression of cyclin F (Movsesyan et al. 1996). Iron, which is known to initiate OS and inflammation via liberation of the hydroxyl radical and membrane lipid peroxidation, has been shown (Philpott et al. 1998) to result in cell cycle arrest and decrease in the expression of G1 cyclins Cln1 and Cln2. In PD as well as in the MPTP-model of PD, iron is selectively increased in the SNPC (Jellinger et al. 1990). It is more than possible that the alterations in some of the expression of cyclins may be related to iron accumulation. The reversal of these gene expression by exposure to R-APO, an iron chelator and radical scavenger, further support this notion.
Three novel genes corresponding to a stress-response protein functional group were also found to be affected by MPTP and to be reversed by treatment with R-APO. Osp94 is a member of a recently described HSP110/SSE subfamily of heat shock and osmotic stress proteins which was shown to be down-regulated in response to hydrogen peroxide (Santos et al. 1998). MPTP decreased the expression of this gene supporting the role of hydrogen peroxide-induced OS in mechanism of MPTP neurotoxicity. Up-regulation of this gene by R-APO in control and MPTP-treated mice, confirms the protection provided by this drug against hydrogen peroxide and 6-OHDA-induced OS in pheochromocytoma cells (Gassen et al. 1998). Furthermore, our results point to possible gene targets for R-APO action, since R-APO itself alters the expression of many genes (Table 3). Cytochrome P450 1A1 is one of dozens of individual members of the cytochrome gene superfamily, which forms a major part of the body's defense against toxin exposure. Three P450 families (CYP1, CYP2 and CYP3) appear to be responsible for most drug metabolism (Wrighton and Stevens 1992), but they may also be involved in endogenous signals, which have not been identified (Dey et al. 1999). In several families with PD, genetic polymorphisms of P450 were reported (Takakubo et al. 1996; Riedl et al. 1998). In our opinion, this gene was down-regulated by MPTP treatment while R-APO prevented the effect. These findings point to the similarity of MPTP model to PD etiology, since the expression of this gene may also be crucial to the process of neurodegeneration by drug metabolizing mechanisms. The third novel gene, whose expression was unaltered by MPTP treatment but was down-regulated by R-APO treatment, is A170. This protein is not an antioxidant, but was found to be induced by exposure of cells to OS (Ishii et al. 1997). Its close structural similarity to a signal transduction protein, 60–62 kDa human lymphocyte protein, suggests that the A170 protein could play a role in signal transduction to induce cellular responses under OS which occurs with MPTP.
Most interesting is the effect of R-APO on Huntingtin-associated protein 1 (HAP1), as a suppressor. HAP1 is specifically expressed in neurons, and is associated with Huntington Chorea (Li et al. 1998). One of the possible functions of HAP1 is to activate the mutated huntingtin protein, and therefore causing the disease. The fact that R-APO inhibits the expression of HAP1 gene suggest that iron chelator and/or radical scavengers may be beneficial in treatment of Huntington Chorea neurodegenerative disease, where OS has been implicated (Polidori et al. 1999) and should be further investigated.
Although the 1200 genes analyzed in this study represent only about 1–2 % of the mouse genome, this method has given us a more complex and wider view of molecular events in dopaminergic neurodegeneration previously not known. Although we did not isolate SN specifically, the gene analysis of the dissected brain shows significant and specific gene expression related to the mechanism of MPTP-induced neurodegeneration. This is further supported by the prevention of these gene changes by R-APO pretreatment, which also results in neuroprotection. The gene expression array data presented here provide the first global assessment of the processes involved in neurodegeneration of dopamine neurons and the neuroprotection afforded by drug treatment at molecular levels. We are now examining gene expression in 6-OHDA model and in SNPC from idiopathic PD to evaluate the homology between animal models and clinical manifestation of the disease. We are confident that these findings could lead to development of novel and effective neuroprotective drugs.