Peripheral administration of the Class-IIa HDAC inhibitor MC1568 partially protects against nigrostriatal neurodegeneration in the striatal 6-OHDA rat model of Parkinson’s disease

Parkinson's disease (PD) is a neurodegenerative disorder characterised by nigrostriatal dopaminergic (DA) neurodegeneration. There is a critical need for neuroprotective therapies, particularly those that do not require direct intracranial administration. Small molecule inhibitors of histone deacetylases (HDAC) (HDIs) are neuroprotective in in vitro and in vivo models of PD, however it is unknown whether ClassIIa-specific HDIs are neuroprotective when administered peripherally. Here we show that 6-hydroxydopamine (6-OHDA) treatment induces protein kinase C (PKC)-dependent nuclear accumulation of the Class IIa HDAC5 in SH-SY5Y cells and cultured DA neurons in vitro. Treatment of these cultures with the Class-IIa specific HDI, MC1568, partially protected against 6-OHDA-induced cell death. In the intrastriatal 6-OHDA lesion in vivo rat model of PD, MC1568 treatment (0.5 mg/kg i.p.) for 7 days reduced forelimb akinesia and partially protected DA neurons in the substantia nigra and their striatal terminals from 6-OHDA-induced neurodegeneration. MC1568 treatment prevented 6-OHDA-induced increases in microglial activation in the striatum and substantia nigra. Furthermore, MC1568 treatment decreased 6-OHDA-induced increases in nuclear HDAC5 in nigral DA neurons. These data suggest that peripheral administration of Class-IIa specific HDIs may be a potential therapy for neuroprotective in PD.

Class IIa HDACs undergo nucleocytoplasmic shuttling to exert effects on neuronal survival and axonal growth (Cho and Cavalli, 2012). Class IIa HDACs, including HDAC4 and 5, are expressed in DA neurons in mouse SN, located mainly in the cytoplasm (Mazzocchi et al., 2019;Wu et al., 2017). Gene co-expression analysis has shown that HDAC5 and HDAC9 are co-expressed with DA markers in the human SN (Mazzocchi et al., 2021;Mazzocchi et al., 2019). Treatment with the class IIa HDIs, MC1568 and LMK235, protected against MPP + -and α-synucleininduced degeneration of DA neurons in vitro (Collins et al., 2015;Mazzocchi et al., 2021). These effects appear to be mediated through HDAC5 inhibition, as siRNAs against HDAC5, but not HDAC4 and HDAC7, protected against α-synuclein-induced neurite degeneration in SH-SY5Y cells (Mazzocchi et al., 2019). However, it is unclear whether Class IIa HDIs, and in particular HDAC5 inhibition, is neuroprotective against DA neurodegeneration in vivo.
There is no known selective inhibitor of HDAC5, but the small molecule MC1568 is the most well-characterised Class IIa-specific HDI (Mai et al., 2005). Intraperitoneal (i.p.) administration of MC1568 in rats reduces HDAC5 levels in nucleus accumbens (Taniguchi et al., 2017), highlighting the feasibility of a peripheral administration route. MC1568 (i.p.) has also been shown to reduce thimerosal-induced apoptosis in rat prefrontal cortex, by preventing upregulation of Class IIa HDACs (Guida et al., 2016). In the current study, we examined the neuroprotective potential of MC1568 in in vitro models of PD, then investigated the potential of i.p. administration of MC1568 to exert neuroprotective effects in the intrastriatal 6-OHDA lesion rat model of PD.

MTT assay
Culture media was removed and 300 μL of 0.5 mg/ml 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma) in HBSS was added to each well for 4 h at 37 • C in 5% CO2. MTT-HBSS solution was carefully aspirated and 100 μL of DMSO was added to each well to permeabilise any formazan produced by the cells. 75 μL of DMSO/formazan solution from each well was pipetted into a 96-well plate and absorbance of each sample measured at 620 nm using an A600 plate reader (Thermo Fisher).

Immunocytochemistry
For neurite length analysis, SH-SY5Y cells were labelled with the fluorescent vital dye calcein AM (1:1000 Invitrogen) for 1 h at the end of culture period. In primary cultures, DA neurons were identified by immunocytochemical staining for tyrosine hydroxylase (TH). Cell culture plates were fixed for 15 min using 4% paraformaldehyde. Following 3 × 5 min washes in PBS-T, cultures were incubated in 5% bovine serum albumin (BSA) in 10 mM PBS-T for 1 h at room temperature. They were incubated in primary antibodies: TH (Millipore MAB318; 1:200), HDAC5 (Santa-Cruz; sc-133106 1:200) or beta-actin (Santa-Cruz sc-47778; 1:200), diluted in 1% BSA in 10 mM PBS at 4 C for 16 h. Following 3 × 5 min washes in 10 mM PBS-T, cells were incubated in 594-conjugated Alexa-Fluor® secondary antibodies (Invitrogen; 1:500 A11005 or A11012) in 1% BSA in 10 mM PBS, prior to 3 × 5 min washes. Immunostained SH-SY5Y cells and E14 VM neurons were imaged using Olympus IX71 inverted microscope. For analysis of neurite growth, five non-overlapping images were captured from each well in each experimental group and 135-350 cells were analysed per group, as indicated in the figure legends. Fluorescence intensity of individual cells was measured by densitometry using Image J analysis software.

In vivo study design
32 adult female Sprague-Dawley rats were purchased from Envigo, UK, and maintained on a 12 h:12 h light:dark cycle with access to food and water ad libitum. Animals were randomly assigned to one of four experimental groups: Vehicle/Saline (n = 8), Vehicle/MC1568 (n = 8), 6-OHDA/Saline (n = 8) and 6-OHDA/MC1568 (n = 8). Vehicle/Saline group received stereotactic injection of saline with 0.01% ascorbate, followed by 7 daily i.p. injections of saline with 0.5% DMSO. Vehicle/ MC1568 group received stereotactic injection of saline with 0.01% ascorbate, followed by 7 daily i.p. injections of 0.5 mg/kg MC1568 in saline with 0.5% DMSO. 6-OHDA/Saline group received stereotactic injection of 7 μg of 6-OHDA in saline with 0.01% ascorbate, followed by 7 daily i.p. injections of saline with 0.5% DMSO. 6-OHDA/MC1568 group received stereotactic injection of 7 μg of 6-OHDA dissolved saline with 0.01% ascorbate followed by 7 daily i.p. injections of 0.5 mg/ kg MC1568 in saline with 0.5% DMSO. The dose of 0.5 mg/kg MC1568 i. p. was based on a study (Griffin et al., 2017) showing that this dose was safe and well tolerated; this was supported by the lack of an effect of this treatment on the body weight of the animals throughout our study ( Supplementary Fig. 1). Rats were housed in groups of four in standard housing cages with environmental enrichment. Behavioural testing was performed on the week before stereotactic surgery, and at day 8 and 12 post-surgery. All experiments were conducted in accordance with the European Directive 2010/63/EU and under authorisation granted by the Health Products Regulatory Authority Ireland (AE19130/P057).

6-OHDA lesion surgery
Stereotaxic surgery was performed under general anaesthesia induced by isoflurane inhalation. The mortality rate in this study was 3.1%, as one animal did not recover from stereotaxic surgery. Each animal was then placed into the stereotactic frame, an incision was made on the skull and four small holes were made into the skull. Four-site unilateral intrastriatal lesion was made by infusion of 6-OHDA hydrobromide (7 μg as free base (Sigma) in 3 μL saline with 0.01% ascorbate, into each of four sites) at a rate of 1 μL/min with 2 min for diffusion, at the following co-ordinates: AP + 1.3, ML ± 2.7; AP + 0.4, ML ± 3.1; AP − 0.4, ML ± 4.3; AP − 1.3, ML ± 4.7 from bregma and DV − 5.0 below dura, with incision bar at − 2.3 mm). After diffusion, the needle was withdrawn, and sutures were performed. As post-surgery care, animals received Carprofen (5 mg/kg s.c.) as analgesic and 5% glucose solution (i.p), then were allowed to recover fully on a heated mat before being returned to their home cages.

Behavioural analysis
The Stepping test for forelimb akinesia was performed as described previously (Olsson et al., 1995). In brief, animals were restrained to retain one forelimb, then hindlimbs and the free forelimb were placed on the countertop. The animal was then moved sideways at steady pace on the countertop across 90 cm in approximately 15 s. Numbers of adjusting steps made by the unrestrained forelimb on ipsilateral and contralateral sides were counted.

Tissue collection and processing
At 14 days after surgery, rats were sacrificed under terminal anaesthesia with sodium pentobarbital (50 mg/kg) and transcardially perfused with 4% paraformaldehyde. Brains were extracted and placed in 4% paraformaldehyde in PBS overnight, then placed in 30% sucrose for cryoprotection and subsequently snap-frozen in isopentane with liquid nitrogen. 30 μm cryosections were cut, then placed in cryoprotectant (10 mM PBS with 30% ethylene glycol, 25% glycerol, 20% water) until used for immunohistochemistry.

Immunohistochemistry
Immunohistochemistry was carried out as previously described . Sections were washed three times in 1.2% trisbuffered saline (TBS) solution for 5 min. Where appropriate, sections were quenched using 3% hydrogen peroxide/10% methanol in distilled water for 5 min to block endogenous peroxidase activity, then washed three times in TBS solution. For fluorescent immunostaining, nonspecific binding was blocked using 10% goat, horse or rabbit serum as appropriate, diluted in tris-buffered saline containing 0.02% Triton-X100 (TxTBS). Sections were incubated overnight shaking at 4 • C in primary antibody diluted in 1% TxTBS serum. The following primary antibodies were used: TH (Merck Millipore; MAB318 1:1000), IBA1 (Fujifilm; 019-19741 1:1000) or HDAC5 (Abcam; ab1439 1:500). Following three washes in TBS, sections were incubated for 2 h in secondary antibody diluted in TxTBS containing 1% serum. For fluorescence staining, Alexa Fluor 488-or 594-conjugated secondary antibodies were used (1:500; Invitrogen). Sections were washed three times in TBS and, where indicated, nuclei were stained by incubation in DAPI (Sigma; 1:3000) solution for 5 h in the dark, followed by three TBS washes, then cover-slipped using fluorescent mounting media (Dako Diagnostics). For chromogen detection, secondary antibodies were biotinylated goat-anti-rabbit IgG (1:200, Jackson Immunoresearch Lab), horse-anti-mouse IgG (1:200, Vector Labs) or rabbit-anti-goat IgG (1:200, Vector Labs). Following three TBS washes, sections were incubated in streptavidin-biotin-horseradish peroxidase solution (Vector Labs) for 2 h. Sections were washed three times in TBS before developing with DAB (Vector Labs). Sections were dehydrated using increasing concentrations of ethanol, cleared in Xylene and cover-slipped using DPX mounting media (BDH Chemicals). Images were taken using the Olympus BX53 Upright Microscope, Olympus FV1000 Confocal Laser Scanning Biological Microscope and Olympus FV10i Microscope.

Image analysis
For the analysis of in vitro experiments, three non-overlapping images of each well were acquired and n = 135 cells for each group were analysed using ImageJ software analysis. These values were averaged to give one n per group per experiment. Specifically, for neurite legth analysis, Image J was calibrated using the appropriate scale bar and the freehand tool was used to manually trace neurites from three cells in each image, which were then measured using the 'analyse' function. To quantify nuclear accumulation of HDAC5, cells were immunostained for HDAC5 (red) and DAPI (blue). The nuclear outline was traced using the DAPI image of three individual cells in each image, and the mean intensity value was measured. A similar strategy was used to measure HDAC5 nuclear accumulation in vivo, in TH-positive neurons in the SN. A similar traing approach was also used to measure the cell body area of IBA1 + microglia, by tracing the cell body, which was then quantified using the 'analyse' function. For immunohistochemial analysis of nigrostriatal pathway integrity, 3 individual brain slices of striatum or SN were imaged for each of the 8 animals per experimental group. To quantify TH immunoreactivity in striatum, images were converted to 8bit images using Image J, the segmented line tool was used to trace the outline of the striatum, and mean intensity values were measured to quantify TH-immunopositive staining on the ipsilateral and contralateral sides. To evaluate the numbers of TH neurons in the SN, and numbers of IBA1 + cells in both striatum and SN, images were converted to 8-bit images, 'adjust threshold function' was used to obtain a black and white image with a threshold to allow visualisation of each individual cell, this was maintained throughout the image analysis. Finally, the 'analyse particles' function was used to quantify the numbers of cells in each image.

Statistical analysis
Statistical analysis was performed using GraphPad Prism 8 (©2020 GraphPad Software, CA USA). Data are presented as mean (expressed as a percentage of control) ± SEM of the number of experimental replicates, rather than of the number of cells. In vivo data are presented as mean ± SEM as percentage of the contralateral side. Statistical differences were analysed using two-way ANOVA as appropriate, with posthoc test as indicated in the figure legends.

Delayed treatment with MC1568 partially protects against 6-OHDAinduced decreases in neurite length in cultured DA neurons.
Although concurrent treatment is useful in an experimental setting, in PD some degree of neurodegeneration would have already occurred before any therapeutic would be applied. To model this in vitro, we used a delayed treatment paradigm in which SH-SY5Y cells were firstly treated with 15 μM 6-OHDA for 24 h, prior to treatment with 0.1 μM MC1568 for 3DIV (Fig. 2E). Two-way ANOVA analysis of the MTT assay revealed significant effects of 6-OHDA (F (1− 8) = 1272, P < 0.0001) and in SH-SY5Y cells after treatment with or without 10 μM of the PKC activator, PMA, for 1 h prior to treatment with or without 15 μM 6-OHDA for 24 h. Graphs showing (C) nuclear HDAC5 expression and (D) neurite length, and (E) representative photomicrographs of TH-positive DA neurons, in primary cultures of E14 rat VM after treatment with or without 10 μM of the PKC activator, PMA, for 1 h prior to treatment with or without 5 μM 6-OHDA for 24 h. All data are presented as the mean ± SEM as a percentage of the control of n = 3 experiments. **** p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05 vs. control or as indicated; Two-way ANOVA with post-hoc Tukey's test.

Peripheral administration of MC1568 exerts partial protective effects on motor behaviour and nigrostriatal integrity in 6-OHDA-lesioned rats.
We next performed an in vivo experiment to determine whether i.p. administration of MC1568 could prevent against 6-OHDA-induced nigrostriatal neurodegeneration. For detailed description of the experimental design see the Methods section. In brief, adult female rats received unilateral intrastriatal stereotactic injection 6-OHDA or saline, followed by 7 daily i.p. injections of saline or 0.5 mg/kg MC1568. Motor behavior was assessed using the stepping test (Olsson et al., 1995) (H). All data are presented as the mean ± SEM as a percentage of the control of n = 3 experiments. **** p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05 vs. control or as indicated; Two-way ANOVA with post-hoc Tukey's test. significant decrease in the number of steps taken with the contralateral paw at day 8 and day 12 post surgery in the 6-OHDA + vehicle group when compared to controls, which was not seen in the 6-OHDA + MC1568 group (Fig. 3B, C).
To investigate whether the increased number of IBA1 + cells may also reflect an increase in microglia activation, we examined the somal area of IBA1 + cells, which has been used as a proxy measurement of microglial activation (Davis et al., 2017). We found that compared to the control group, 6-OHDA led to a significant increase in IBA1 + somal area (47.0 ± 2.3 vs 77.13 ± 3.6 mm2; p > 0.0001), which was partially prevented by MC1568 treatment (77.13 ± 3.6 mm2 vs 66.0 ± 3.0 mm2; p > 0.05).
Finally, we evaluated whether 6-OHDA resulted in an increase in nuclear accumulation of HDAC5 in DA neurons in the SN, and whether this could be prevented by MC1568. Both 6-OHDA and MC1568 treatments resulted in significant changes in HDAC5 nuclear levels (F (1− 28) = 18.27, P = 0.0002 and F (1− 28) = 31.70, P < 0.0001, respectively), and there was a significant 6-OHDA × MC1568 interaction (F (1− 28) = 12.94, P = 0.0012). Post-hoc testing revealed a significant increase in nuclear HDAC5 levels in TH-positive neurons in the SN in the 6-OHDA + vehicle group, which was prevented by MC1568 treatment (Fig. 4E, F). These data show that 6-OHDA lesion induced nuclear accumulation of HDAC5 in DA neurons, and that this could be prevented by peripheral administration of MC1568.

Discussion
Although the role of HDACs in PD is being increasingly studied, the ways in which these proteins specifically contribute to PD pathogenesis are still largely unknown (Didonna and Opal, 2015). Despite this, HDIs are considered to have promise as disease-modifying therapies, since some in vivo studies have shown that pan-HDIs can exert neuroprotective effects in animal models of PD (Hou et al., 2021;Lai et al., 2019). However, since distinct HDACs subtypes are differentially expressed within specific cells and tissues, there is a need to investigate which class-specific HDIs have neuroprotective effects, particularly when administered peripherally.
In this study, we examined the effects of 6-OHDA treatment on neurite length, as a single cell readout of neurodegeneration, in cultured DA neurons. We also investigated the effects of 6-OHDA on HDAC5 shuttling between the cytoplasm and nucleus in these cells. We found that neurotoxic insult induced by 6-OHDA administration has a detrimental effect on neurite outgrowth in DA neurons, which is accompanied by nuclear accumulation of HDAC5. Since Class IIa HDACs nucleocytoplasmic shuttling is known to be controlled by the canonical Ca 2+ / CaMk/PKC pathway (McKinsey et al., 2001), we investigated whether HDAC5 nuclear accumulation could be altered by the activation of canonical Ca 2+ /CaMk/PKC pathway. Pre-treatment for the PKC activator, PMA, partially prevented 6-OHDA-induced nuclear accumulation increased neurite outgrowth in both cell types. A previous study reported HDAC4 nuclear accumulation following MPP + treatment of cultured DA neurons overexpressing A53T α-synuclein (Wu et al., 2017).
However, in contrast to our findings, that study also found that HDAC5 localisation in vitro in A53T cells was unchanged after MPP + administration (Wu et al., 2017). In agreement with our results, another study demonstrated that following neuronal injury, HDAC5 is exported from the nucleus in a PKC-dependent manner in dorsal root ganglia (DRG) neurons, and that this is a fundamental step in triggerring proregenerative gene expression in these cells (Cho et al., 2013). While there are some differences between these studies, collectively these data support the hypothesis that nuclear accumulation of Class IIa HDACs is associated with neurodegeneration.
Next we assessed the neuroprotective effect of the Class IIa HDI, MC1568, which we had previously showed to have beneficial effects in other cellular models of PD (Mazzocchi et al., 2019), in two in vitro PD models. We found that concurrent, but not delayed, treatment protected both SH-SY5Y cells and cultured DA neurons against 6-OHDA-induced decreases in cell viability. Furthermore, both concurrent and delayed MC1568 treatment partially protected against 6-OHDA-induced neurite degeneration in both cell types. The rational for testing delayed treatment with MC1568 is that in clinical settings, pharmacological interventions are always applied after disease onset, in a context of ongoing neurodegeneration (Löhle et al., 2014). In a previous study, we demonstrated that when administered before the toxin, MC1568 can protect cultured DA neurons from axonal degeneration (Mazzocchi et al., 2019). We also previously reported neuroprotective effects of MC1568 on cultured sympathetic neurons, including preservation of neuronal branching (Collins et al., 2015), showing that MC1568 has effects on other neuronal populations which are affected in PD.
Following positive results in in vitro studies, we examined the effects of peripheral administration of MC1568 in adult rats with intrastriatal 6-OHDA lesions. This is the first study, to our knowledge, to use MC1568 in an in vivo model of PD. Class IIa HDAC-specific inhibition has demonstrated neuroprotective effects in animal models of other brain disorders, but its therapeutic potential in PD models has not been previously examined. In a rat model of reperfusion ischemia, i.p. administration of TMP-269, a specific class-IIa HDI, at 30 min prior to the insult, exerted a neuroprotective effect by reducing infarct volume (Su et al., 2020). This indicates the potential neuroprotective benefits of Class IIa HDAC inhibition in vivo. MC1568 partially prevented behavioural impairments induced by 6-OHDA, as assessed using the stepping test. In support of our data, oral treatment with either 25 mg/kg or 50 mg/kg VPA for 15 days has been reported to significantly reduce apomorphineinduced rotational behaviour in rats with intrastriatal 6-OHDA lesions (Ximenes et al., 2015). Another study found that treatment with NaB (150 and 300 mg/kg i.p. for 14 days) significantly improved motor function in rats with 6-OHDA lesions of the medial forebrain bundle (MFB) (Sharma, Taliyan, & Singh, 2015). In that study, motor performance was assessed using several behavioural tests, including apomorphine-induced rotations, spontaneous locomotor activity, narrow beam test and rotarod test. In addition to motor effects, we found that MC1568 administration prevented 6-OHDA-induced loss of striatal DA innervation, and partially preserved DA neuronal cell bodies in the SN. These data demonstrate that MC1568 partially protects the nigrostriatal DA system from 6-OHDA-induced anatomical and functional degeneration. These findings are in agreement with several studies showing the neuroprotective potential of the pan-HDIs, VPA and NaB, in in vivo PD models (Lai et al., 2019;Sharma et al., 2015;Ximenes et al., 2015). For example, i.p. administration of 300 mg/kg NaB for 14 days protected against decreases in striatal dopamine induced by MFB 6-OHDA lesions (Yuan et al., 2005). Another study found that 6-OHDAinduced increases in pyknotic nuclei in the ST of adult rats were prevented by NaB treatment (Sharma et al., 2015). Another pan-HDI, VPA, prevented loss of striatal DA neurotransmission and protected against nigral DA neuronal cell, in rats with MFB 6-OHDA lesions (Lai et al., 2019). Collectively, these studies show that inhibition of Class IIa HDACs may, at least partially, underlie the neuroprotective effects of pan-HDI in vivo in models of PD.
Since HDAC inhibition has previously been demonstrated to have an anti-inflammatory effect in an in vitro model of PD (Harrison et al., 2018), we also investigated potential anti-inflammatory effects of MC1568 in vivo. We found that MC1568 treatment prevented 6-OHDAinduced increases in microglia, in the ST and SN, as well as preventing 6-OHDA-induced increases in nuclear levels of HDAC5 in DA neurons in the SN. A previous study showed that oral administration of VPA for 15 days reduced microglia inflammation in rats with 6-OHDA striatal lesions (Ximenes et al., 2015).
In conclusion, we have shown that both concurrent and delayed treatment with MC1568 can partially protect DA neurons from 6-OHDAinduced neurodegeneration in vitro. Further, we found that peripheral administration of the Class IIa HDI MC1568 exerts neuroprotection and behavioural improvements in an in vivo rat model of PD. Our data suggest that MC1568 may exert its effect by acting on DA neurons to limit nuclear accumulation of HDAC5, and also by preventing microglial activation. Elucidating the contributions of each of these mechanisms to the neuroprotective effects of MC1568 will be important for future research. The data rationalise the further study of Class IIa HDACs, and of pharmacological agents that target them, in developing new therapeutic approaches for neuroprotection in PD.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.