Development of Improved HDAC6 Inhibitors as Pharmacological Therapy for Axonal Charcot-Marie-Tooth Disease

Abstract

Charcot-Marie-Tooth disease (CMT) is the most common inherited peripheral neuropathy, with an estimated prevalence of 1 in 2500. The degeneration of motor and sensory nerve axons leads to motor and sensory symptoms that progress over time and have an important impact on the daily life of these patients. Currently, there is no curative treatment available. Recently, we identified histone deacetylase 6 (HDAC6), which deacetylates α-tubulin, as a potential therapeutic target in axonal CMT (CMT2). Pharmacological inhibition of the deacetylating function of HDAC6 reversed the motor and sensory deficits in a mouse model for mutant "small heat shock protein B1" (HSPB1)-induced CMT2 at the behavioral and electrophysiological level. In order to translate this potential therapeutic strategy into a clinical application, small drug-like molecules that are potent and selective HDAC6 inhibitors are essential. To screen for these, we developed a method that consisted of 3 distinct phases and that was based on the pathological findings in the mutant HSPB1-induced CMT2 mouse model. Three different inhibitors (ACY-738, ACY-775, and ACY-1215) were tested and demonstrated to be both potent and selective HDAC6 inhibitors. Moreover, these inhibitors increased the innervation of the neuromuscular junctions in the gastrocnemius muscle and improved the motor and sensory nerve conduction, confirming that HDAC6 inhibition is a potential therapeutic strategy in CMT2. Furthermore, ACY-1215 is an interesting lead molecule as it is currently tested in clinical trials for cancer. Taken together, these results may speed up the translation of pharmacological inhibition of HDAC6 into a therapy against CMT2.

Keywords: Acetylated α-tubulin; Charcot–Marie–Tooth disease; Compound screening; HSPB1; Histone deacetylase 6; Mitochondria.

Fig. 1

ACY-738 and ACY-775 are potent and selective histone deacetylase 6 inhibitors. (a) N2a cells were incubated with 1 μM ACY-738, ACY-775, or trichostatin A (TSA) as a positive control. By Western blot the acetylation of α-tubulin (acet α-tub) was assessed, as well as the acetylation of histone 3 (acetH3). Total α-tubulin levels, glyceraldehyde 3-phosphate dehydrogenase (gapdh) and histone 4 (H4) were used as a reference for equal loading. (b) The ratio of acetylated α-tubulin to total tubulin levels was calculated and normalized to TSA-treated cells. One-way analysis of variance (ANOVA); ***p < 0.0001; n = 4. (c) The ratio of acetylated histone 3 to histone 4 levels was assessed and subsequently normalized to vehicle-treated cells. One-way ANOVA; **p < 0.01; n = 3–4. (d–g) Immunocytochemistry for acetylated α-tubulin (red) and acetylated histone 3 (green) on N2a cells treated with 1 μM TSA, ACY-738, and ACY-775

Fig. 2

Pharmacokinetic properties of ACY-738 and ACY-775 in N2a cells. (a) Dose–response of trichostatin A (TSA), ACY-738, or ACY-775 on acetylation of α-tubulin was established in N2a cells and determined by Western blot. (b) Quantification of the ratio of acetylated α-tubulin to total tubulin levels on Western blot. Values were normalized to the effect of 2.5 μM (n = 3)

Fig. 3

Histone deacetylase 6 (HDAC6) inhibition using ACY-738 and ACY-775 rescued the mitochondrial defects in dorsal root ganglion (DRG) neurons cultured from symptomatic HSPB1^S135F mice. (a–c) Nontransgenic DRG neurons were incubated with 2.5 μM ACY-738 or ACY-775. Immunocytochemical labeling of acetylated α-tubulin shows increased immunofluorescence in neurons upon HDAC6 inhibition. Scale bar = 40 μm. (d) The intensity of acetylated α-tubulin was quantified in the neurites from DRG neuron cultures and corrected for the length of the signal. All values within each experiment were normalized to vehicle-treated cells; n = 5 with 43–61 cells per condition. (e) Kymographs representing the axonal movement of mitochondria in HSPB1^S135F DRG neurons treated with ACY-738 or ACY-775 (2.5 μM). Vertical lines representing stationary mitochondria and lines deflecting to the right or left indicate anterograde or retrograde movement of mitochondria, respectively. (f) The number of moving mitochondria per 100 μm in the neurites was quantified based on the kymographs from HSPB1^S135F DRG neurons treated with ACY-738 or ACY-775; n = 25–35 from 3 different transgenic mice. (g) The total number of mitochondria per 100 μm in the neurites was quantified based on the kymographs from HSPB1^S135F DRG neurons treated with ACY-738 or ACY-775; n = 25–35 from 3 different transgenic mice. (h) Additional analysis was performed to assess the number of anterogradely or retrogradely moving mitochondria per 100 μm in the neurites based on the kymographs from HSPB1^S135F DRG neurons treated with ACY-738 or ACY-775; n = 3 with 7–10 cells per experiment. One-way analysis of variance (ANOVA); *p < 0.05, ***p < 0.0001. Two-way ANOVA; *p < 0.05, ***p < 0.0001

Fig. 4

Histone deacetylase 6 (HDAC6) inhibition using ACY-738, ACY-775, or ACY-1215 reversed the axonal deficits in motor and sensory nerves and induced reinnervation of neuromuscular junction. (a) Symptomatic (12-month-old) HSPB1S135F mice received daily intraperitoneal injections of ACY-738, ACY-775 (3 mg/kg), ACY-1215 (30 mg/kg), or vehicle for 21 days. The compound muscle action potential (CMAP) amplitudes were recorded from the sciatic nerve after the treatment period. Dashed line indicates values obtained from nontransgenic mice (NTG) mice; n = 5–8 mice. (b) The SNAP amplitudes recorded over the sensory tail nerve after the treatment. Dashed line indicates values obtained from NTG mice; n = 5–8 mice. Dunnett's multiple comparison test, *p < 0.05, **p < 0.01, ***p < 0.001. (c) After treatment, the neuromuscular junctions (NMJs) were visualized from the gastrocnemius muscle by immunofluorescence. The motor endplates were labeled by α-bungarotoxin and the axon terminals were visualized by neurofilament light chain-directed antibody. Dashed line indicates values obtained from NTG mice. (d) The number of innervated NMJs was quantified by determining the percentage of overlap between α-bungarotoxin and neurofilament light chain immunofluorescence. Dashed line indicates values obtained from NTG mice; n = 5–8 mice. Dunnett's multiple comparison test, ** p <0.01, ***p < 0.001