Selective HDAC6 inhibitor TubA offers neuroprotection after intracerebral hemorrhage via inhibiting neuronal apoptosis

Animals and experimental groups

One hundred and ninety-six adult male Sprague-Dawley rats (250–350 g) purchased from Hunan Slack Jingda Animal Laboratory (Changsha, China) were used for this experimental research. Sprague–Dawley rats were raised three per cage in the central laboratory of Hunan Provincial People’s Hospital at a fixed temperature (25 °C) and relative humidity (60%). The rats were fed with free water and food, and dwelt in a 12-h light/dark cycle condition. Animals were stochastically allocated to each experiment group. All experimental programs were authorized by the Institutional Animal Care and Use Committee of Central South University (IACUC approval No: 2020079). The SD rats were stochastically divided into four experimental groups: the sham-treated group (no ICH, control group), vehicle group (ICH + DMSO), low dosage treatment group (ICH + TubA 25 mg/kg), and high dosage group (ICH + TubA 40 mg/kg). All rats were euthanized humanely using an intraperitoneal injection of pentobarbital sodium.

Rat model of ICH

The ICH procedure was performed by injecting collagenase type IV (Sigma-Aldrich, St. Louis, MO, USA) through stereotaxic intrastriatal implantation, as previously described with some small alterations (Rosenberg et al., 1990; Zeng et al., 2017). In simple terms, animals were deeply anesthetized with pentobarbital sodium (40 mg/kg) via intraperitoneal injection. Under deep anesthesia, the SD rats were placed carefully in the stereotactic apparatus and an incision was made along the midline to expose the bregma. We stereotactically inserted a microsyringe (10 µL) into the right striatum through the cranial drill hole with coordinates of 0.2 mm (anterior–posterior), 3 mm (right, medial–lateral), and 6 mm ventral (from pia, dorsal–ventral (DV)) from the bregma. Collagenase IV (0.2U) dissolved in 2 μl saline was injected slowly for 5 min, placed for another 5 min, and then gradually removed. The rats’ body temperatures were maintained at 37 ± 0.5 °C using a homoeothermic blanket control unit. The sham operation was performed using all the above-mentioned procedures but only an equal volume of saline was injected.

Experimental design

The allocation and use of rats is illustrated in Table S1.

Experiment I: time-course of HDAC6 (n = 30)

The variations of HDAC6 protein levels in perihematomal tissues over time were observed. Thirty rats were stochastically assigned into six experimental groups: sham, ICH 4 h, ICH 6 h, ICH 12 h, ICH 1 d, and ICH 3 d. At each time point post-ICH, three rats were used for Western blot (WB) assay.

Experiment II: effect of Tubastatin A (n = 92)

We investigated the effect of TubA on neuronal injuries induced by ICH. Rats (n = 92) were stochastically allocated into four groups: sham, ICH+ vehicle, ICH+TubA 25 mg/kg, and ICH+TubA 40 mg/kg. We determined the neurological scores and calculated the brain water content at 1 d and 3 d after operation in each group (n = 10/group). Western blot assay was conducted in each group (n = 3) to quantitatively analyze the level of HDAC6 at 3 d after ICH.

Experiment III: mechanisms of the TubA effect (n = 48)

We attempted to further probe the mechanisms of the neuroprotective effect executed by TubA. First, for the sham, ICH+ vehicle, ICH+TubA 25 mg/kg, and ICH+TubA 40 mg/kg groups, we applied Western blot assay to explore the level of acetylation α-tubulin in each group (n = 3). Second, the protein levels of Bcl-2, Bax, and cleaved caspase-3 were also detected using Western blot assay (n = 3/group) in each group. Brain sections from the sham (n = 6), ICH+ vehicle (n = 6), ICH+TubA 25 mg/kg (n = 6), and ICH+TubA 40 mg/kg (n = 6) groups were prepared for HE staining and TUNEL staining.

Drug administration

Tubastatin A (TubA, #A4101; ApexBio Technology, Houston, TX, USA) was first dissolved in 4% dimethylsulfoxide (DMSO) and then diluted with 40% Polyethylene glycol 300 (PEG300) in double-distilled water (ddH2O). TubA was injected intraperitoneally 30 min before ICH induction with two dosages: 25 and 40 mg/kg, according to previous studies (Wang et al., 2016). DMSO solution in ddH2O with PEG300 was injected in the sham group.

Behavioral testing

Behavioral testing was conducted at 1 and 3 d post-ICH in the light of the previous study (Garcia et al., 1995). The modified Garcia scoring system included six individual tests: voluntary movement, symmetry of limbs, forelimb extension, body proprioception, climbing ability, and tentacle touch. Each individual trial was scored from 0 to 3. The modified Garcia test scores ranged from 3 to 18. A minimum score of three points suggested severe behavior impairment and the highest score of 18 points represented proper neurological function. All trials, assessments, and computations were blindly monitored by two independent investigators.

Brain water content (BWC) assessment

The BWC of each group was measured at 1 and 3 d post-ICH, following the previous methods (Suzuki et al., 2010). The brains of rats were rapidly dissociated after deep anesthetization and sacrifice. The brains were then dissected into five sections: contralateral and ipsilateral basal ganglia, contralateral and ipsilateral cortex, and cerebellum. All parts were quickly weighed on a precision electronic autobalance for wet weight (WW) and dried at 100 °C for dry weight (DW). The calculation formula was: BWC = [(WW-DW)/WW] × 100%.

Hematoxylin and eosin staining

First, we perfused the rats with 4% paraformaldehyde, decapitated and collected the rat brains, fixed the rat brains in 4% paraformaldehyde for 1 day, paraffin-embedded the brains, and cut them into slices (4 μm thick). The sections were flushed with tap water and stained with hematoxylin for 5 min, differentiated using 0.1% hydrochloric acid and ethanol solution for 25 s, blued in phosphate buffered saline for 45 min, counterstained with Eosin for 5 min, dehydrated with 95% alcohol, and finally mounted with neutral balsam. Areas with intact or lysed red blood cells were identified as hematoma. The perihematomal images were obtained using an optical microscope (Nikon Eclipse E100).

Double staining of TUNEL and NeuN

To confirm neuronal apoptosis in perihematomal tissues, we performed double immunofluorescent staining of the neuron-specific nuclear protein (NeuN) with TdT-mediated dUTP-biotin nick end labeling (TUNEL). TUNEL staining was conducted following the manual of the Cell Apoptosis Detection Kit (G1501-50T; ServiceBio, Wuhan, China), followed by staining against the neuronal marker NeuN (1:200; Abcam, Cambridge, UK). Finally, the slides were blanketed with DAPI (Beyotime, Shanghai, China). The TUNEL-positive neurons were captured using a fluorescence microscope (Nikon Eclipse C1) and analyzed with Image J software.

In vivo Western blot

Perihematomal tissues from the ipsilateral hemisphere were homogenized and centrifuged, then separated using SDS-poly-acrylamide gel electrophoresis, and transmitted to nitrocellulose membranes. Next, we blocked the membrane with defatted milk (5%) for 1 h and incubated it overnight at 4 °C with the following primary antibodies: rabbit anti-HDAC6 antibody (1:2,000, #ab239362; Abcam, Cambridge, UK), mouse anti-ace-α-tubulin (1:2,000, #ab24610; Abcam, Cambridge, UK), rabbit anti-cleaved caspase-3 antibody (1:1,000, #9661; Cell Signaling Technology, Danvers, MA, USA), rabbit anti-Bax antibody (1:1,000, #ab32503; Abcam, Cambridge, UK), rabbit anti-Bcl-2 antibody (1:1,000, #ab196495; Abcam, Cambridge, UK), or mouse anti-β-actin antibody (1:5,000, #66009-1-Ig; Proteintech, Chicago, IL, USA). Corresponding secondary antibodies were chosen to incubate for 90 min. The data were analyzed using Image J software.

Cell culture and ICH model

The SH-SY5Y cells (American Type Culture Collection, Manassas, VA, USA) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Corning, NY, USA), accompanied with FBS (10%, Gibco) and penicillin-streptomycin (1%; Gibco) at 37 °C with 5% CO₂ in a moist environment. Treatment of 10 μM hemin (#C3984; ApexBio, TX, USA) for 24 h was used to induce an in vitro ICH model.

Cell viability: CCK-8 assay

Measurement of cell viability was determined using a CCK-8 detection kit (Apexbio, USA). In short, SH-SY5Y cells (1 × 10⁴ cells/mL) were sowed in 96-well plates with culture medium (100 μl/well) for 24 h. Subsequently, a CCK-8 solution (10 μL) was mixed into each well. Absorbance was gauged at 450 nm with a Bio Tek microplate reader.

Drug administration

SH-SY5Y cells were pretreated with Tubastatin A (3 μM) dissolved in DMSO for 6 h as described previously (Yu et al., 2016), and then co-treated with 10 μM hemin (Goldstein et al., 2003) for an additional 24 h. The vehicle group added an equal volume of PBS.

HDAC6 knockdown in SH-SY5Y cells

HDAC6 siRNA or siRNA-NC was transfected into SH-SY5Y cells using Lipofectamine 3000 (L3000-015; Invitrogen, Waltham, MA, USA). After transfection for 48 h, SH-SY5Y cells were treated with hemin for 24 h and collected for further analysis.

Cell apoptosis: flow cytometry

The culture medium was discarded and SH-SY5Y cells were collected using EDTA-free trypsin, then rinsed twice with PBS. We centrifuged the suspension, obtained about 3.2 × 10⁵ cells, and added 500 μL of binding buffer suspension to the collected SH-SY5Y cells. The suspension was blended with Annexin V-APC (5 μl) and we added PI (5 μl) to be reincubated for 10 min. Finally, we analyzed the mixture instantly with an FACS flow cytometer C6 (FCM; FACSCanto II; BD Biosciences) and the results were analyzed using FlowJo v10.7.1 (Tree Star, Ashland, OR, USA).

In vitro Western blot

We threw away the culture medium and washed the SH-SY5Y cells with PBS twice. Next, we added 200 μl of cold (4 °C) RIPA (P0013B; Beyotime Institute of Biotechnology) buffer for cell lysis. The lysate was centrifuged and the supernatant was harvested for protein concentration standardization. The subsequent experimental procedures were the same as the animal experiments.

Statistical analysis

All values are presented as means ± SEM. A two-tailed unpaired student t-test was used to compare the two groups; for multiple groups, we applied the one-way ANOVA followed by Bonferroni’s post-hoc tests. Probability values of p < 0.05 were regarded as statistically significant. All statistical analyses were performed using SPSS 24 software.