Valproic acid neuroprotection in 6-OHDA lesioned rat, a model for parkinson’s disease

Background: Valproic acid (VPA) a long standing anti-epileptic and anti-manic drug has been recently investigated as a neuroprotective molecule, in relation to its action as an inhibitor of histone deacetylases (HDACs), favoring relaxed configuration of chromatin and thus promoting gene transcription. Methods: In the present study, chronic administration of VPA added to the diet, was tested for neuroprotection in a rat model of Parkinson’s disease. The model consists of multiple injections of the dopaminergic toxin, 6-hydroxydopamine (6-OHDA), unilaterally in the striatum with consequent degeneration of the dopaminergic neurons originating the nigro-striatal pathway. This model of neurodegeneration is widely used as a reliable animal model for Parkinson’s disease (PD). Results: Chronic VPA administration significantly reduced degeneration of dopaminergic neurons in the substantia nigra, and of dopaminergic terminals in the striatum, in rats subjected to the unilateral lesion of the nigrostriatal pathway. VPA treatment was also able to increase α-synuclein expression in the substantia nigra and to counteract the lesion-dependent decrease of the protein in the substantia nigra itself and in the striatum. Conclusions: Present data, which follow previous results obtained in the rotenone rat model of nigrostriatal degeneration, allow to propose VPA as a treatment to be tested for its effectiveness in other animal models of parkinsonism, in view of possible translation to


Background
Animal models of Parkinson's disease (PD) have been studied for many years to understand the etiology of the disease and to test putative therapeutic interventions [1][2][3][4]. In particular, the dopaminergic toxin, 6-hydroxydopamine (6-OHDA) has been widely used to lesion the nigro-striatal pathway and to reproduce in rodents main neuropathologic features of PD [5,6]. Multiple intrastriatal injections of this neurotoxin result in rapid loss of dopaminergic terminals in the striatum itself followed by slower (3-4 weeks) and partial retrograde degeneration of dopaminergic neurons in the substantia nigra [7]. This lesion reproduces in the rat model Parkinson-like conditions and is particularly useful when chronic pharmacologic treatments are devised to counteract the relatively slow and partial degeneration of dopaminergic nigral neurons [7].
Valproic acid (VPA), a long standing anti-epileptic and anti-manic drug, has recently revealed neuroprotective properties possibly related, among the several other targets, to its action as histone deacetylase (HDAC) inhibitor [8][9][10]. In particular, VPA neuroprotection has been demonstrated in several culture models: spontaneous age-related death and excitotoxic damage of cortical neurons [11,12], beta-amyloid toxicity in hippocampal neurons [13], non-depolarizing conditions and exposure to 6-OHDA in cerebellar granule neurons [14,15], ischemic conditions in cultured hippocampal slices [16]. In vivo, VPA was proven neuroprotective in several animal models of acute brain damage and neurodegenerative diseases: transient cerebral ischemia in rats [17], animal models of Huntington's disease [18,19] and Alzheimer disease [20,21], as well as in clinical trials on patients affected by spinal muscular atrophy [22]. Following reports on culture-based studies showing that HDAC inhibitors protect against dopaminergic neurotoxicity [15], we have recently demonstrated that VPA added to the diet protects nigral dopaminergic neurons in the rotenone rat model of PD [23]. We provide here evidence that the same treatment significantly protects dopaminergic neurons of the substantia nigra in the 6-OHDA model of multiple striatal toxin injections and that neuroprotection is related to the maintenance of increased a-synuclein expression in the substantia nigra. The positive results of VPA treatment in two different types of Parkinson-like degeneration suggest the opportunity to test such treatment in other animal models of the disease in view of possible translation to patients.

Methods
Male Wistar rats from Harlan Nossan, weighing 200-220 grams were used. All animals were maintained on a 12:12 h light/dark cycle and given food and water ad libitum. The experiments were carried out in agreement with the Italian and European Community laws on the use of animals and protocols were approved by a local bioethical committee. After acclimation (age of rats around 11-12 weeks) , rats of the VPA groups were fed for 4 weeks with standard chow added with 2% sodium valproate (Sigma, St Louis, MO). As previously reported [23], VPA feeding resulted in approximate daily intake of 1.4g/kg body weight, which has been shown to result in mean plasma concentration of 42.6 mg/L, close to the human therapeutic window [24]. Animals anesthetized with Zoletil (20 mg/Kg) were placed in a David Kopf stereotaxic instrument and four injections of 6-OHDA (7 μg free base/2μl each, dissolved in saline containing 0.02% ascorbic acid) were made at four locations into the left striatum according to the procedures previously described [7]. Injection were made through a 5 μl Hamilton syringe to the needle of which a thin (50-70 µm in diameter) glass capillary had been fixed with wax, in order to minimize penetration damage to the tissue. Coordinates of the injections (in mm) were the followings: AP: +1.3, +0.4, -0.4 -1.3 with respect to bregma; ML: 2.6, 3.0, 4.2, 4.5 with respect to midline; DV: 5.2 below the dura. After each injection, needle was left in place for 2 min and then slowly retracted. Equivalent injections of saline were made in the contralateral striatum. After surgery, rats survived for further 4 weeks during which the same previous feeding regimens were maintained. For western blot analysis, rats were killed by decapitation, the brains were sliced with a tissue chopper, samples of the substantia nigra and the striatum were dissected under the stereomicroscope on a cold plate and frozen in dry ice. Samples were homogenized in 50 mM Tris buffer containing 5 mM EDTA, 0.1% SDS, a protease inhibitor cocktail and standard western blot analysis was performed, using antibodies against tyrosine hydroxylase (TH, Sigma) and a-synuclein (clone P4D1, Santa Cruz) [15]. In the same homogenates, b-actin content was determined in parallel as reference protein with a specific antibody (Sigma). Densitometric results were statistically evaluated by one way ANOVA followed by Bonferroni's tests. For immunohistochemistry, rats were deeply anesthetized and perfusion-fixed through heart with 4% paraformaldehyde in 100 mM phosphate buffer pH 7.4. Brains removed from the skull were postfixed overnight in the same fixative and left in 18% sucrose until sunk. Frozen free-floating sections, 40 μm thick, were processed for immunohistochemistry using a TH monoclonal antibody (TH-16, Sigma, 1:5,000) and a HRP-linked secondary antibody. Reaction was visualized using diaminobenzidine as chromogen (DAB).

Results & Discussion
After 4 weeks from 6-OHDA injections into the left striatum, this region appeared shrunken and largely devoid of TH immunoreactivity in normally-fed rats while damage was less evident in VPA-fed rats (Figure 1). In the substantia nigra of normally-fed (control) rats immunoreactivity of TH-positive dopaminergic neurons (Figure 2A) was almost completely lost in the 6-OHDA injected side ( Figure 2B) while these neurons appeared significantly spared in the injected side of VPA-treated rats ( Figure 2C). Partial sparing of 6-OHDA lesioned dopaminergic neurons in VPA-fed rats was confirmed by western blot analysis for TH immunoreactivity in homogenates of substantia nigra and quantitative analysis based on band densitometry ( Figure 3A). While TH content was significantly decreased by more than 60% in the substantia nigra of control rats, this decrease was substantially recovered in VPA-fed rats ( Figure 3A). A similar trend of reduced TH loss in VPA-fed animals was also observed in homogenates of striata ( Figure 3B). Western blot analysis of substantia nigra homogenates revealed that VPA treatment increased the level of expression of a-synuclein with respect to controls and  that the decrease of this protein in the substantia nigra of control rats injected with 6-OHDA was completely restored in the same region of VPA-fed rats ( Figure 4A). Also regarding a-synuclein, doi: 10.7243/2050-0874-1-4 striatal homogenates showed trends of variation similar, albeit not statistically significant, to those described in the substantia nigra ( Figure 4A). Following previous results obtained with the rotenone model for PD [23], our novel demonstration of neuroprotection in the 6-OHDA model of the disease suggests that VPA may have a general neuroprotective role in PD. Our data supporting the beneficial effect of VPA on 6-OHDA lesion are presently based on qualitative immunohistochemistry for TH in both substantia nigra and striatum as well as on quantitative densitometry of western blots from tissue extracts. To better quantify the neuroprotective effect, future studies should include cell counts in the substantia nigra and high resolution analysis of TH-positive terminals in the striatum. From our previous results [23], we know that the regimen of VPA administration adopted here is able to effectively inhibit HDAC and to increase histone acetylation state in the substantia nigra and the striatum, at the time when 6-OHDA injection is made. While VPA is a relatively unspefic HDAC inhibitor, its prevalent effect is against class I HDACs which are highly expressed in brain [25]. Due to the numerous molecular targets of VPA [9], the link between its neuroprotective action and the increase of hystone acetylation is still correlative. It will be important for future research to test VPA in other models of PD, in particular those obtained through genetic manipulation of mice or targeted delivery of mutated genes to the rat brain using the viral vector technology [2]. This research could hopefully open novel therapeutic perspectives for patients. Previous results of limited clinical trials did not demonstrate any beneficial effect of VPA in PD patients [26] and suggested that reversible parkinsonism and cognitive impairment could be developed by patients treated with high doses of VPA for epilepsy [27]. However, a recent report based on three cases of PD patients demonstrated that daily doses of 500-1000 mg of VPA ameliorated compulsive behaviors associated with dopaminergic medications prescribed to patients and allowed to reduce drug dosages without worsening PD symptoms [28]. Thus VPA could act as a neuroprotectant against degeneration of dopaminergic neurons in parkinsonism (present results and reference 23) and fight at the same time compulsive behaviors due to treatments with levo-DOPA and dopaminergic agonists [28]. Taken together, these results may bring to devise novel trials for PD patients based on VPA treatment, particularly in the case that additional positive data derive from experiments based on VPA administration in genetic animal models.
The molecular targets of VPA are multifarious [9] and it is likely that multiple cellular pathways are involved in its neuroprotective effects. A specific molecular determinant of this neuroprotection emerging from the present study, is the VPA-related increase of a-synuclein levels in the substantia nigra, and the consequent protection from the decrease caused by the neurotoxic lesion. Previous results demonstrated a similar effect of VPA on a-synuclein expression in mice brain and linked the increased expression to neuroprotection from excitotoxicity [10]. This novel concept of VPA-mediated neuroprotective role of a-synuclein, fits with other recent studies on a-synuclein neuroprotection in culture and in vivo, including PD-like models of neurodegeneration [9,10,29].  . Western Blot analysis and relative densitometries of α-synuclein and β-actin, as reference protein, expression in substantia nigra (A) and striatum (B) from control, 6-OHDA-treated, VPA-treated and VPA+6-OHDA-treated animals. The level of α-synuclein was normalized for the β-actin content in each sample and the data were expressed as arbitrary units. Each bar represents the mean ± S.E. of 3-6 samples from different animals. *p<0.05 with respect to control, #p<0.05 with respect to 6-OHDA, § § § p<0.01 with respect to VPA; Bonferroni's test after ANOVA. doi: 10.7243/2050-0874-1-4

Conclusion
In conclusion, the demonstration that a chronic treatment with VPA added to the diet is neuroprotective in the rat against different types of nigrostriatal degeneration, allows to propose this treatment as a model to be tested for its effectiveness in other Parkinsonrelated animal studies in view of possible translation to patients. Present, as well as previous [9,10,15,23] data suggest the interest of focusing this research on the effects of VPA on histone-regulated transcriptional activity and on a-synuclein expression in brain regions relevant to PD.